C167SR-LM [INFINEON]
16-Bi t Single-Chip Microcont roller; 16碧吨单芯片Microcont辊型号: | C167SR-LM |
厂家: | Infineon |
描述: | 16-Bi t Single-Chip Microcont roller |
文件: | 总464页 (文件大小:3723K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
User’s Manual, V 3.1, Mar. 2000
C167CR Derivatives
16-Bit Single-Chip Microcontroller
Microcontrollers
N e v e r s t o p t h i n k i n g .
Edition 2000-03
Published by Infineon Technologies AG,
St.-Martin-Strasse 53,
D-81541 München, Germany
© Infineon Technologies AG 2000.
All Rights Reserved.
Attention please!
The information herein is given to describe certain components and shall not be considered as warranted
characteristics.
Terms of delivery and rights to technical change reserved.
We hereby disclaim any and all warranties, including but not limited to warranties of non-infringement, regarding
circuits, descriptions and charts stated herein.
Infineon Technologies is an approved CECC manufacturer.
Information
For further information on technology, delivery terms and conditions and prices please contact your nearest
Infineon Technologies Office in Germany or our Infineon Technologies Representatives worldwide (see address
list).
Warnings
Due to technical requirements components may contain dangerous substances. For information on the types in
question please contact your nearest Infineon Technologies Office.
Infineon Technologies Components may only be used in life-support devices or systems with the express written
approval of Infineon Technologies, if a failure of such components can reasonably be expected to cause the failure
of that life-support device or system, or to affect the safety or effectiveness of that device or system. Life support
devices or systems are intended to be implanted in the human body, or to support and/or maintain and sustain
and/or protect human life. If they fail, it is reasonable to assume that the health of the user or other persons may
be endangered.
User’s Manual, V 3.1, Mar. 2000
C167CR Derivatives
16-Bit Single-Chip Microcontroller
Microcontrollers
N e v e r s t o p t h i n k i n g .
C167CR
Revision History:
V 3.1, 2000-03
Previous Version:
Version 3.0, 2000-02 (intermediate version)
Version 2.1, 1999-03
Version 2.0, 03.96
Version 1.0, 08.94
Preliminary User’s Manual Revision 1.0, 07.92
Page
Subjects (major changes since last revision)
Converted to new company layout
Figures have been redrawn (compared to V3.0)
List of derivatives reworked
all
all
1-1
2-2, 4-1
6-1
Figure corrected
RTC hints removed
9-29 … 34
9-35 … 37
10-6, 10-27
10-16
Description of bus arbitration improved
Description of XBUS interface improved
Timer frequency tables improved
Figure 10-11 corrected (compared to V3.0)
Description of T5M corrected (compared to V3.0)
Figure 10-21 corrected (compared to V3.0)
Baudrate tables improved
10-30
10-33
11-12 … 14
12-5
Description of transmission timing improved
Baudrate tables improved
12-13
13-5
Time range table improved
13-6
Reset indication table improved
14-7
Baudrate table added
15-3
Port connection table corrected
15-7
Input frequency table improved
17-3
Bitfield ADSTC added
18-33
Separate section for Busoff Recovery Sequence
Status after reset regrouped
19-5 … 8
19-9 … 19
19-13 … 20
Description of initialization routine improved
Better structure for section “Startup Configuration”
Controller Area Network (CAN): License of Robert Bosch GmbH
We Listen to Your Comments
Any information within this document that you feel is wrong, unclear or missing at all?
Your feedback will help us to continuously improve the quality of this document.
Please send your proposal (including a reference to this document) to:
mcdocu.comments@infineon.com
C167CR
Derivatives
Table of Contents
Page
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
The Members of the 16-bit Microcontroller Family . . . . . . . . . . . . . . . . . 1-2
Summary of Basic Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
1.1
1.2
1.3
2
Architectural Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Basic CPU Concepts and Optimizations . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
High Instruction Bandwidth/Fast Execution . . . . . . . . . . . . . . . . . . . . . 2-3
Programmable Multiple Priority Interrupt System . . . . . . . . . . . . . . . . 2-7
The On-chip System Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
The On-chip Peripheral Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
Protected Bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18
2.1
2.1.1
2.1.2
2.2
2.3
2.4
3
Memory Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
Internal ROM Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
Internal RAM and SFR Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
The On-Chip XRAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
External Memory Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-11
Crossing Memory Boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
Protection of the On-chip Mask ROM . . . . . . . . . . . . . . . . . . . . . . . . . . 3-13
3.1
3.2
3.3
3.4
3.5
3.6
4
The Central Processing Unit (CPU) . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
Instruction Pipelining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
Particular Pipeline Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
Bit-Handling and Bit-Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10
Instruction State Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
CPU Special Function Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-12
4.1
4.2
4.3
4.4
4.5
5
Interrupt and Trap Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
Interrupt System Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
Interrupt Control Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6
Operation of the PEC Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-12
Prioritization of Interrupt and PEC Service Requests . . . . . . . . . . . . . . 5-16
Saving the Status During Interrupt Service . . . . . . . . . . . . . . . . . . . . . . 5-18
Interrupt Response Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-20
PEC Response Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-23
External Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-25
Trap Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-28
5.1
5.1.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
6
Clock Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
Frequency Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
Oscillator Watchdog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
Clock Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-9
6.1
6.2
6.3
6.4
User’s Manual
I-1
V3.1, 2000-03
C167CR
Derivatives
Table of Contents
Page
7
Parallel Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
Input Threshold Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
Output Driver Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
Alternate Port Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-7
PORT0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
PORT1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-13
Port 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18
Port 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-23
Port 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-28
Port 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-32
Port 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-36
Port 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-42
Port 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-47
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
7.10
7.11
7.12
8
Dedicated Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
9
The External Bus Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
Single Chip Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
External Bus Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-3
Programmable Bus Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-12
READY Controlled Bus Cycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-17
Controlling the External Bus Controller . . . . . . . . . . . . . . . . . . . . . . . . . 9-19
EBC Idle State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-28
External Bus Arbitration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-29
The XBUS Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-35
Accessing the On-chip XBUS Peripherals . . . . . . . . . . . . . . . . . . . . . 9-36
External Accesses to XBUS Peripherals . . . . . . . . . . . . . . . . . . . . . . 9-37
9.1
9.2
9.3
9.4
9.5
9.6
9.7
9.8
9.8.1
9.8.2
10
10.1
The General Purpose Timer Units . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Timer Block GPT1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
GPT1 Core Timer T3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-3
GPT1 Auxiliary Timers T2 and T4 . . . . . . . . . . . . . . . . . . . . . . . . . . 10-12
Interrupt Control for GPT1 Timers . . . . . . . . . . . . . . . . . . . . . . . . . . 10-21
Timer Block GPT2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-22
GPT2 Core Timer T6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24
GPT2 Auxiliary Timer T5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-30
Interrupt Control for GPT2 Timers and CAPREL . . . . . . . . . . . . . . . 10-38
10.1.1
10.1.2
10.1.3
10.2
10.2.1
10.2.2
10.2.3
11
The Asynchronous/Synchronous Serial Interface . . . . . . . . . . . . . . 11-1
Asynchronous Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-5
Synchronous Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-8
Hardware Error Detection Capabilities . . . . . . . . . . . . . . . . . . . . . . . . 11-10
ASC0 Baud Rate Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-11
ASC0 Interrupt Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-15
11.1
11.2
11.3
11.4
11.5
User’s Manual
I-2
V3.1, 2000-03
C167CR
Derivatives
Table of Contents
Page
12
The High-speed Synchronous Serial Interface . . . . . . . . . . . . . . . . . 12-1
Full-duplex Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-7
Half-duplex Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-10
Continuous Transfers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-11
Port Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-12
Baud Rate Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-13
Error Detection Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-15
SSC Interrupt Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-17
12.1
12.2
12.3
12.4
12.5
12.6
12.7
13
13.1
13.2
The Watchdog Timer (WDT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-1
Operation of the Watchdog Timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-3
Reset Source Indication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-6
14
The Bootstrap Loader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-1
15
The Capture/Compare Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-1
The CAPCOM Timers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-4
CAPCOM Unit Timer Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-9
Capture/Compare Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-10
Capture Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-13
Compare Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-14
Capture/Compare Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-22
15.1
15.2
15.3
15.4
15.5
15.6
16
The Pulse Width Modulation Module . . . . . . . . . . . . . . . . . . . . . . . . . 16-1
Operating Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-2
PWM Module Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-10
Interrupt Request Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-14
PWM Output Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-15
16.1
16.2
16.3
16.4
17
The Analog/Digital Converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-1
Mode Selection and Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-2
Conversion Timing Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-12
A/D Converter Interrupt Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-14
17.1
17.2
17.3
18
The On-chip CAN Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-1
Functional Blocks of the CAN Module . . . . . . . . . . . . . . . . . . . . . . . . . . 18-2
General Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-7
The Message Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-18
Controlling the CAN Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-30
Configuration Examples for Message Objects . . . . . . . . . . . . . . . . . . . 18-34
The CAN Application Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-36
18.1
18.2
18.3
18.4
18.5
18.6
19
19.1
19.2
System Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-1
Reset Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-2
Status After Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-5
User’s Manual
I-3
V3.1, 2000-03
C167CR
Derivatives
Table of Contents
Page
19.3
19.4
19.4.1
19.4.2
Application-specific Initialization Routine . . . . . . . . . . . . . . . . . . . . . . . . 19-9
System Startup Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-12
System Startup Configuration upon an External Reset . . . . . . . . . . 19-13
System Startup Configuration upon a Single-chip Mode Reset . . . . 19-20
20
Power Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20-1
Idle Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20-2
Power Down Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20-4
Status of Output Pins During Power Reduction Modes . . . . . . . . . . . 20-5
20.1
20.2
20.2.1
21
System Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-1
Stack Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-4
Register Banking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-9
Procedure Call Entry and Exit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-9
Table Searching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-12
Floating Point Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-12
Peripheral Control and Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-13
Trap/Interrupt Entry and Exit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-13
Unseparable Instruction Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . 21-14
Overriding the DPP Addressing Mechanism . . . . . . . . . . . . . . . . . . . . 21-14
Handling the Internal Code Memory . . . . . . . . . . . . . . . . . . . . . . . . . . 21-16
Pits, Traps and Mines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-18
21.1
21.2
21.3
21.4
21.5
21.6
21.7
21.8
21.9
21.10
21.11
22
The Register Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22-1
Register Description Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22-1
CPU General Purpose Registers (GPRs) . . . . . . . . . . . . . . . . . . . . . . . 22-2
Special Function Registers Ordered by Name . . . . . . . . . . . . . . . . . . . 22-4
Special Function Registers Ordered by Address . . . . . . . . . . . . . . . . . 22-12
Special Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22-20
22.1
22.2
22.3
22.4
22.5
23
24
25
Instruction Set Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23-1
Device Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24-1
Keyword Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25-1
User’s Manual
I-4
V3.1, 2000-03
C167CR
Derivatives
Introduction
1
Introduction
The rapidly growing area of embedded control applications is representing one of the
most time-critical operating environments for today’s microcontrollers. Complex control
algorithms have to be processed based on a large number of digital as well as analog
input signals, and the appropriate output signals must be generated within a defined
maximum response time. Embedded control applications also are often sensitive to
board space, power consumption, and overall system cost.
Embedded control applications therefore require microcontrollers, which:
• offer a high level of system integration
• eliminate the need for additional peripheral devices and the associated software overhead
• provide system security and fail-safe mechanisms
• provide effective means to control (and reduce) the device’s power consumption.
With the increasing complexity of embedded control applications, a significant increase
in CPU performance and peripheral functionality over conventional 8-bit controllers is
required from microcontrollers for high-end embedded control systems. In order to
achieve this high performance goal Infineon has decided to develop its family of 16-bit
CMOS microcontrollers without the constraints of backward compatibility.
Of course the architecture of the 16-bit microcontroller family pursues successful
hardware and software concepts, which have been established in Infineons popular 8-
bit controller families.
About this Manual
This manual describes the functionality of a number of 16-bit microcontrollers of the
Infineon C166 Family, the C167-class.
As these microcontrollers provide a great extent of identical functionality it makes sense
to describe a superset of the provided features. For this reason some sections of this
manual do not refer to all the C167 derivatives that are offered (e.g. devices without a
CAN interface). These sections contain respective notes wherever possible.
The descriptions in this manual refer to the following derivatives of the C167-class:
• C167CR-LM
• C167CR-4RM
Version with PLL, 2 KByte XRAM, CAN module
Version with PLL, 2 KByte XRAM, 32 KByte ROM, CAN module
• C167CR-16RM Version with PLL, 2 KByte XRAM, 128 KByte ROM, CAN module
• C167SR-LM Version with PLL, 2 KByte XRAM
This manual is valid for the versions with on-chip ROM or Flash memory of the
mentioned derivatives as well as for the ROMless versions. Of course it refers to all
devices of the different available temperature ranges and packages.
For simplicity all these various versions are referred to by the term C167CR throughout
this manual. The complete pro-electron conforming designations are listed in the
respective data sheets.
User’s Manual
1-1
V3.1, 2000-03
C167CR
Derivatives
Introduction
1.1
The Members of the 16-bit Microcontroller Family
The microcontrollers of the Infineon 16-bit family have been designed to meet the high
performance requirements of real-time embedded control applications. The architecture
of this family has been optimized for high instruction throughput and minimum response
time to external stimuli (interrupts). Intelligent peripheral subsystems have been
integrated to reduce the need for CPU intervention to a minimum extent. This also
minimizes the need for communication via the external bus interface. The high flexibility
of this architecture allows to serve the diverse and varying needs of different application
areas such as automotive, industrial control, or data communications.
The core of the 16-bit family has been developed with a modular family concept in mind.
All family members execute an efficient control-optimized instruction set (additional
instructions for members of the second generation). This allows an easy and quick
implementation of new family members with different internal memory sizes and
technologies, different sets of on-chip peripherals and/or different numbers of IO pins.
The XBUS concept opens a straight forward path for the integration of application
specific peripheral modules in addition to the standard on-chip peripherals in order to
build application specific derivatives.
As programs for embedded control applications become larger, high level languages are
favored by programmers, because high level language programs are easier to write, to
debug and to maintain.
The 80C166-type microcontrollers were the first generation of the 16-bit controller
family. These devices have established the C166 architecture.
The C165-type and C167-type devices are members of the second generation of this
family. This second generation is even more powerful due to additional instructions for
HLL support, an increased address space, increased internal RAM and highly efficient
management of various resources on the external bus.
Enhanced derivatives of this second generation provide additional features like
additional internal high-speed RAM, an integrated CAN-Module, an on-chip PLL, etc.
Utilizing integration to design efficient systems may require the integration of application
specific peripherals to boost system performance, while minimizing the part count.
These efforts are supported by the so-called XBUS, defined for the Infineon 16-bit
microcontrollers (second generation). This XBUS is an internal representation of the
external bus interface that opens and simplifies the integration of peripherals by
standardizing the required interface. One representative taking advantage of this
technology is the integrated CAN module.
The C165-type devices are reduced versions of the C167 which provide a smaller
package and reduced power consumption at the expense of the A/D converter, the
CAPCOM units and the PWM module.
User’s Manual
1-2
V3.1, 2000-03
C167CR
Derivatives
Introduction
The C164-type devices and some of the C161-type devices are further enhanced by a
flexible power management and form the third generation of the 16-bit controller family.
This power management mechanism provides effective means to control the power that
is consumed in a certain state of the controller and thus allows the minimization of the
overall power consumption with respect to a given application.
A variety of different versions is provided which offer various kinds of on-chip program
memory:
• Mask-programmable ROM
• Flash memory
• OTP memory
• ROMless with no non-volatile memory at all.
Also there are devices with specific functional units.
The devices may be offered in different packages, temperature ranges and speed
classes.
More standard and application-specific derivatives are planned and in development.
Note: Not all derivatives will be offered in any temperature range, speed class, package
or program memory variation.
Information about specific versions and derivatives will be made available with the
devices themselves. Contact your Infineon representative for up-to-date material.
Note: As the architecture and the basic features (i.e. CPU core and built in peripherals)
are identical for most of the currently offered versions of the C167CR, the
descriptions within this manual that refer to the “C167CR” also apply to the other
variations, unless otherwise noted.
User’s Manual
1-3
V3.1, 2000-03
C167CR
Derivatives
Introduction
1.2
Summary of Basic Features
The C167CR is an improved representative of the Infineon family of full featured 16-bit
single-chip CMOS microcontrollers. It combines high CPU performance (up to 12.5/16.5
million instructions per second) with high peripheral functionality and means for power
reduction.
Several key features contribute to the high performance of the C167CR (the indicated
timings refer to a CPU clock of 25/33 MHz).
High Performance 16-bit CPU with Four-Stage Pipeline
• 80/60 ns minimum instruction cycle time, with most instructions executed in 1 cycle
• 400/300 ns multiplication (16-bit × 16-bit), 800/600 ns division (32-bit/16-bit)
• Multiple high bandwidth internal data buses
• Register based design with multiple variable register banks
• Single cycle context switching support
• 16 MBytes linear address space for code and data (Von Neumann architecture)
• System stack cache support with automatic stack overflow/underflow detection
Control Oriented Instruction Set with High Efficiency
• Bit, byte, and word data types
• Flexible and efficient addressing modes for high code density
• Enhanced boolean bit manipulation with direct addressability of 6 Kbits
for peripheral control and user defined flags
• Hardware traps to identify exception conditions during runtime
• HLL support for semaphore operations and efficient data access
Integrated On-chip Memory
• 2 KByte internal RAM for variables, register banks, system stack and code
• 2 KByte on-chip high-speed XRAM for variables, user stack and code (not on all
derivatives)
• 128 KByte or 32 KByte on-chip ROM (not for ROMless devices)
External Bus Interface
• Multiplexed or demultiplexed bus configurations
• Segmentation capability and chip select signal generation
• 8-bit or 16-bit data bus
• Bus cycle characteristics selectable for five programmable address areas
User’s Manual
1-4
V3.1, 2000-03
C167CR
Derivatives
Introduction
16-Priority-Level Interrupt System
• 56 interrupt nodes with separate interrupt vectors
• 240/180 ns typical interrupt latency (400/300 ns maximum)
in case of internal program execution
• Fast external interrupts
8-Channel Peripheral Event Controller (PEC)
• Interrupt driven single cycle data transfer
• Transfer count option (std. CPU interrupt after programmable number of PEC transfers)
• Eliminates overhead of saving and restoring system state for interrupt requests
Intelligent On-chip Peripheral Subsystems
• 16-channel 10-bit A/D Converter with programmable conversion time
(7.76 µs minimum), auto scan modes, channel injection mode
• Two 16-channel Capture/Compare Units with 2 independent time bases each,
very flexible PWM unit/event recording unit with different operating modes,
includes four 16-bit timers/counters, maximum resolution fCPU/8
• 4-channel PWM unit
• Two Multifunctional General Purpose Timer Units
GPT1: Three 16-bit timers/counters, maximum resolution fCPU/8
GPT2: Two 16-bit timers/counters, maximum resolution fCPU/4
• Asynchronous/Synchronous Serial Channels (USART)
with baud rate generator, parity, framing, and overrun error detection
• High Speed Synchronous Serial Channel
programmable data length and shift direction
• On-chip CAN Bus Module, Rev. 2.0B active (not on all derivatives)
• Watchdog Timer with programmable time intervals
• Bootstrap Loader for flexible system initialization
111 IO Lines with Individual Bit Addressability
• Tri-stated in input mode
• Selectable input thresholds (not on all pins)
• Push/pull or open drain output mode
• Programmable port driver control (fast/reduced edge)
Different Temperature Ranges
• 0 to + 70 °C, – 40 to + 85 °C, – 40 to + 125 °C
User’s Manual
1-5
V3.1, 2000-03
C167CR
Derivatives
Introduction
Infineon CMOS Process
• Low power CMOS technology including power saving Idle and Power Down modes.
144-pin Plastic Metric Quad Flat Pack (MQFP) Package
• P-MQFP, 28 × 28 mm body, 0.65 mm (25.6 mil) lead spacing,
surface mount technology
Complete Development Support
For the development tool support of its microcontrollers, Infineon follows a clear third
party concept. Currently around 120 tool suppliers world-wide, ranging from local niche
manufacturers to multinational companies with broad product portfolios, offer powerful
development tools for the Infineon C500 and C166 microcontroller families,
guaranteeing a remarkable variety of price-performance classes as well as early
availability of high quality key tools such as compilers, assemblers, simulators,
debuggers or in-circuit emulators.
Infineon incorporates its strategic tool partners very early into the product development
process, making sure embedded system developers get reliable, well-tuned tool
solutions, which help them unleash the power of Infineon microcontrollers in the most
effective way and with the shortest possible learning curve.
The tool environment for the Infineon 16-bit microcontrollers includes the following tools:
• Compilers (C, MODULA2, FORTH)
• Macro-assemblers, linkers, locators, library managers, format-converters
• Architectural simulators
• HLL debuggers
• Real-time operating systems
• VHDL chip models
• In-circuit emulators (based on bondout or standard chips)
• Plug-in emulators
• Emulation and clip-over adapters, production sockets
• Logic analyzer disassemblers
• Starter kits
• Evaluation boards with monitor programs
• Industrial boards (also for CAN, FUZZY, PROFIBUS, FORTH applications)
• Network driver software (CAN, PROFIBUS)
User’s Manual
1-6
V3.1, 2000-03
C167CR
Derivatives
Introduction
1.3
Abbreviations
The following acronyms and terms are used within this document:
ADC
ALE
ALU
ASC
CAN
Analog Digital Converter
Address Latch Enable
Arithmetic and Logic Unit
Asynchronous/synchronous Serial Controller
Controller Area Network (License Bosch)
CAPCOM CAPture and COMpare unit
CISC
CMOS
CPU
EBC
ESFR
Flash
GPR
GPT
HLL
Complex Instruction Set Computing
Complementary Metal Oxide Silicon
Central Processing Unit
External Bus Controller
Extended Special Function Register
Non-volatile memory that may be electrically erased
General Purpose Register
General Purpose Timer unit
High Level Language
IO
Input/Output
OTP
PEC
PLA
One Time Programmable memory
Peripheral Event Controller
Programmable Logic Array
Phase Locked Loop
PLL
PWM
RAM
RISC
ROM
SFR
Pulse Width Modulation
Random Access Memory
Reduced Instruction Set Computing
Read Only Memory
Special Function Register
SSC
XBUS
XRAM
Synchronous Serial Controller
Internal representation of the External Bus
On-chip extension RAM
User’s Manual
1-7
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
2
Architectural Overview
The architecture of the C167CR combines the advantages of both RISC and CISC
processors in a very well-balanced way. The sum of the features which are combined
result in a high performance microcontroller, which is the right choice not only for today’s
applications, but also for future engineering challenges. The C167CR not only integrates
a powerful CPU core and a set of peripheral units into one chip, but also connects the
units in a very efficient way. One of the four buses used concurrently on the C167CR is
the XBUS, an internal representation of the external bus interface. This bus provides a
standardized method of integrating application-specific peripherals to produce
derivatives of the standard C167CR.
C166-Core
ProgMem
IRAM
Internal
RAM
16
16
Data
Data
32
16
ROM
128/32
KByte
Instr. / Data
CPU
2 KByte
XTAL
Osc / PLL
WDT
XRAM
2 KByte
PEC
External Instr. / Data
16-Level
Priority
Interrupt Controller
16
Interrupt Bus
Peripheral Data Bus
16
CAN
Rev 2.0B active
ADC ASC0 SSC
GPT PWM CCOM2CCOM1
10-Bit
(USART)
(SPI)
T2
T3
T4
T7
T8
T0
T1
16
Channels
EBC
8
8
XBUS Control
External Bus
Control
16
T5
T6
BRGen
BRGen
Port 0
Port 1
Port 5
Port 3
Port 7
8
Port 8
8
16
16
16
15
Figure 2-1
C167CR Functional Block Diagram
User’s Manual
2-1
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
2.1
Basic CPU Concepts and Optimizations
The main core of the CPU consists of a 4-stage instruction pipeline, a 16-bit arithmetic
and logic unit (ALU) and dedicated SFRs. Additional hardware is provided for a separate
multiply and divide unit, a bit-mask generator and a barrel shifter.
CPU
16
Internal
RAM
SP
STKOV
STKUN
MDH
MDL
R15
Exec. Unit
Instr. Ptr.
Instr. Reg.
Mul/Div-HW
Bit-Mask Gen
General
Purpose
Registers
R15
ALU
32
4-Stage
Pipeline
(16-bit)
ROM
Barrel - Shifter
Context Ptr.
R0
PSW
SYSCON
BUSCON 0
BUSCON 1
BUSCON 2
BUSCON 3
BUSCON 4
16
R0
ADDRSEL 1
ADDRSEL 2
ADDRSEL 3
ADDRSEL 4
Data Page Ptr.
Code Seg. Ptr.
MCB02147
Figure 2-2
CPU Block Diagram
To meet the demand for greater performance and flexibility, a number of areas has been
optimized in the processor core. Functional blocks in the CPU core are controlled by
signals from the instruction decode logic. These are summarized below, and described
in detail in the following sections:
1) High Instruction Bandwidth/Fast Execution
2) High Function 8-bit and 16-bit Arithmetic and Logic Unit
3) Extended Bit Processing and Peripheral Control
4) High Performance Branch-, Call-, and Loop Processing
5) Consistent and Optimized Instruction Formats
6) Programmable Multiple Priority Interrupt Structure
User’s Manual
2-2
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
2.1.1
High Instruction Bandwidth/Fast Execution
Based on the hardware provisions, most of the C167CR’s instructions can be executed
in just one machine cycle, which requires 2 CPU clock cycles (2 × 1 / fCPU = 4 TCL). For
example, shift and rotate instructions are always processed within one machine cycle,
independent of the number of bits to be shifted.
Branch-, multiply- and divide instructions normally take more than one machine cycle.
These instructions, however, have also been optimized. For example, branch
instructions only require an additional machine cycle, when a branch is taken, and most
branches taken in loops require no additional machine cycles at all, due to the so-called
‘Jump Cache’.
A 32-bit/16-bit division takes 20 CPU clock cycles, a 16-bit × 16-bit multiplication takes
10 CPU clock cycles.
The instruction cycle time has been dramatically reduced through the use of instruction
pipelining. This technique allows the core CPU to process portions of multiple sequential
instruction stages in parallel. The following four stage pipeline provides the optimum
balancing for the CPU core:
FETCH: In this stage, an instruction is fetched from the internal ROM or RAM or from the
external memory, based on the current IP value.
DECODE: In this stage, the previously fetched instruction is decoded and the required
operands are fetched.
EXECUTE: In this stage, the specified operation is performed on the previously fetched
operands.
WRITE BACK: In this stage, the result is written to the specified location.
If this technique were not used, each instruction would require four machine cycles. This
increased performance allows a greater number of tasks and interrupts to be processed.
Instruction Decoder
Instruction decoding is primarily generated from PLA outputs based on the selected
opcode. No microcode is used and each pipeline stage receives control signals staged
in control registers from the decode stage PLAs. Pipeline holds are primarily caused by
wait states for external memory accesses and cause the holding of signals in the control
registers. Multiple-cycle instructions are performed through instruction injection and
simple internal state machines which modify required control signals.
User’s Manual
2-3
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
High Function 8-bit and 16-bit Arithmetic and Logic Unit
All standard arithmetic and logical operations are performed in a 16-bit ALU. In addition,
for byte operations, signals are provided from bits six and seven of the ALU result to
correctly set the condition flags. Multiple precision arithmetic is provided through a
‘CARRY-IN’ signal to the ALU from previously calculated portions of the desired operation.
Most internal execution blocks have been optimized to perform operations on either 8-bit
or 16-bit quantities. Once the pipeline has been filled, one instruction is completed per
machine cycle, except for multiply and divide. An advanced Booth algorithm has been
incorporated to allow four bits to be multiplied and two bits to be divided per machine
cycle. Thus, these operations use two coupled 16-bit registers, MDL and MDH, and
require four and nine machine cycles, respectively, to perform a 16-bit by 16-bit (or 32-bit
by 16-bit) calculation plus one machine cycle to setup and adjust the operands and the
result. Even these longer multiply and divide instructions can be interrupted during their
execution to allow for very fast interrupt response. Instructions have also been provided
to allow byte packing in memory while providing sign extension of bytes for word wide
arithmetic operations. The internal bus structure also allows transfers of bytes or words
to or from peripherals based on the peripheral requirements.
A set of consistent flags is automatically updated in the PSW after each arithmetic,
logical, shift, or movement operation. These flags allow branching on specific conditions.
Support for both signed and unsigned arithmetic is provided through user-specifiable
branch tests. These flags are also preserved automatically by the CPU upon entry into
an interrupt or trap routine.
All targets for branch calculations are also computed in the central ALU.
A 16-bit barrel shifter provides multiple bit shifts in a single cycle. Rotates and arithmetic
shifts are also supported.
Extended Bit Processing and Peripheral Control
A large number of instructions has been dedicated to bit processing. These instructions
provide efficient control and testing of peripherals while enhancing data manipulation.
Unlike other microcontrollers, these instructions provide direct access to two operands
in the bit-addressable space without requiring to move them into temporary flags.
The same logical instructions available for words and bytes are also supported for bits.
This allows the user to compare and modify a control bit for a peripheral in one
instruction. Multiple bit shift instructions have been included to avoid long instruction
streams of single bit shift operations. These are also performed in a single machine
cycle.
In addition, bit field instructions have been provided, which allow the modification of
multiple bits from one operand in a single instruction.
User’s Manual
2-4
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
High Performance Branch-, Call-, and Loop Processing
Due to the high percentage of branching in controller applications, branch instructions
have been optimized to require one extra machine cycle only when a branch is taken.
This is implemented by precalculating the target address while decoding the instruction.
To decrease loop execution overhead, three enhancements have been provided:
• The first solution provides single cycle branch execution after the first iteration of a
loop. Thus, only one machine cycle is lost during the execution of the entire loop. In
loops which fall through upon completion, no machine cycles are lost when exiting the
loop. No special instructions are required to perform loops, and loops are
automatically detected during execution of branch instructions.
• The second loop enhancement allows the detection of the end of a table and avoids
the use of two compare instructions embedded in loops. One simply places the lowest
negative number at the end of the specific table, and specifies branching if neither this
value nor the compared value have been found. Otherwise the loop is terminated if
either condition has been met. The terminating condition can then be tested.
• The third loop enhancement provides a more flexible solution than the Decrement and
Skip on Zero instruction which is found in other microcontrollers. Through the use of
Compare and Increment or Decrement instructions, the user can make comparisons
to any value. This allows loop counters to cover any range. This is particularly
advantageous in table searching.
Saving of system state is automatically performed on the internal system stack avoiding
the use of instructions to preserve state upon entry and exit of interrupt or trap routines.
Call instructions push the value of the IP on the system stack, and require the same
execution time as branch instructions.
Instructions have also been provided to support indirect branch and call instructions.
This supports implementation of multiple CASE statement branching in assembler
macros and high level languages.
User’s Manual
2-5
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
Consistent and Optimized Instruction Formats
To obtain optimum performance in a pipelined design, an instruction set has been
designed which incorporates concepts from Reduced Instruction Set Computing (RISC).
These concepts primarily allow fast decoding of the instructions and operands while
reducing pipeline holds. These concepts, however, do not preclude the use of complex
instructions, which are required by microcontroller users. The following goals were used
to design the instruction set:
1. Provide powerful instructions to perform operations which currently require
sequences of instructions and are frequently used. Avoid transfer into and out of
temporary registers such as accumulators and carry bits. Perform tasks in parallel
such as saving state upon entry into interrupt routines or subroutines.
2. Avoid complex encoding schemes by placing operands in consistent fields for each
instruction. Also avoid complex addressing modes which are not frequently used. This
decreases the instruction decode time while also simplifying the development of
compilers and assemblers.
3. Provide most frequently used instructions with one-word instruction formats. All other
instructions are placed into two-word formats. This allows all instructions to be placed
on word boundaries, which alleviates the need for complex alignment hardware. It
also has the benefit of increasing the range for relative branching instructions.
The high performance offered by the hardware implementation of the CPU can efficiently
be utilized by a programmer via the highly functional C167CR instruction set which
includes the following instruction classes:
• Arithmetic Instructions
• Logical Instructions
• Boolean Bit Manipulation Instructions
• Compare and Loop Control Instructions
• Shift and Rotate Instructions
• Prioritize Instruction
• Data Movement Instructions
• System Stack Instructions
• Jump and Call Instructions
• Return Instructions
• System Control Instructions
• Miscellaneous Instructions
Possible operand types are bits, bytes and words. Specific instruction support the
conversion (extension) of bytes to words. A variety of direct, indirect or immediate
addressing modes are provided to specify the required operands.
User’s Manual
2-6
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
2.1.2
Programmable Multiple Priority Interrupt System
The following enhancements have been included to allow processing of a large number
of interrupt sources:
1. Peripheral Event Controller (PEC): This processor is used to off-load many interrupt
requests from the CPU. It avoids the overhead of entering and exiting interrupt or trap
routines by performing single-cycle interrupt-driven byte or word data transfers
between any two locations in segment 0 with an optional increment of either the PEC
source or the destination pointer. Just one cycle is ‘stolen’ from the current CPU
activity to perform a PEC service.
2. Multiple Priority Interrupt Controller: This controller allows all interrupts to be placed at
any specified priority. Interrupts may also be grouped, which provides the user with
the ability to prevent similar priority tasks from interrupting each other. For each of the
possible interrupt sources there is a separate control register, which contains an
interrupt request flag, an interrupt enable flag and an interrupt priority bitfield. Once
having been accepted by the CPU, an interrupt service can only be interrupted by a
higher prioritized service request. For standard interrupt processing, each of the
possible interrupt sources has a dedicated vector location.
3. Multiple Register Banks: This feature allows the user to specify up to sixteen general
purpose registers located anywhere in the internal RAM. A single one-machine-cycle
instruction allows to switch register banks from one task to another.
4. Interruptable Multiple Cycle Instructions: Reduced interrupt latency is provided by
allowing multiple-cycle instructions (multiply, divide) to be interruptable.
With an interrupt response time within a range from just 5 to 10 CPU clock cycles (in case
of internal program execution), the C167CR is capable of reacting very fast on non-
deterministic events.
Its fast external interrupt inputs are sampled every CPU clock cycle and allow to
recognize even very short external signals.
The C167CR also provides an excellent mechanism to identify and to process
exceptions or error conditions that arise during run-time, so called ‘Hardware Traps’.
Hardware traps cause an immediate non-maskable system reaction which is similar to a
standard interrupt service (branching to a dedicated vector table location). The
occurrence of a hardware trap is additionally signified by an individual bit in the trap flag
register (TFR). Except for another higher prioritized trap service being in progress, a
hardware trap will interrupt any current program execution. In turn, hardware trap
services can normally not be interrupted by standard or PEC interrupts.
Software interrupts are supported by means of the ‘TRAP’ instruction in combination with
an individual trap (interrupt) number.
User’s Manual
2-7
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
2.2
The On-chip System Resources
The C167CR controllers provide a number of powerful system resources designed
around the CPU. The combination of CPU and these resources results in the high
performance of the members of this controller family.
Peripheral Event Controller (PEC) and Interrupt Control
The Peripheral Event Controller allows to respond to an interrupt request with a single
data transfer (word or byte) which only consumes one instruction cycle and does not
require to save and restore the machine status. Each interrupt source is prioritized every
machine cycle in the interrupt control block. If PEC service is selected, a PEC transfer is
started. If CPU interrupt service is requested, the current CPU priority level stored in the
PSW register is tested to determine whether a higher priority interrupt is currently being
serviced. When an interrupt is acknowledged, the current state of the machine is saved
on the internal system stack and the CPU branches to the system specific vector for the
peripheral.
The PEC contains a set of SFRs which store the count value and control bits for eight
data transfer channels. In addition, the PEC uses a dedicated area of RAM which
contains the source and destination addresses. The PEC is controlled similar to any
other peripheral through SFRs containing the desired configuration of each channel.
An individual PEC transfer counter is implicitly decremented for each PEC service
except forming in the continuous transfer mode. When this counter reaches zero, a
standard interrupt is performed to the vector location related to the corresponding
source. PEC services are very well suited, for example, to move register contents to/from
a memory table. The C167CR has 8 PEC channels each of which offers such fast
interrupt-driven data transfer capabilities.
Memory Areas
The memory space of the C167CR is configured in a Von Neumann architecture which
means that code memory, data memory, registers and IO ports are organized within the
same linear address space which covers up to 16 MBytes. The entire memory space can
be accessed bytewise or wordwise. Particular portions of the on-chip memory have
additionally been made directly bit addressable.
A 2 KByte 16-bit wide internal RAM (IRAM) provides fast access to General Purpose
Registers (GPRs), user data (variables) and system stack. The internal RAM may also
be used for code. A unique decoding scheme provides flexible user register banks in the
internal memory while optimizing the remaining RAM for user data.
User’s Manual
2-8
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
The CPU has an actual register context consisting of up to 16 wordwide and/or bytewide
GPRs at its disposal, which are physically located within the on-chip RAM area. A
Context Pointer (CP) register determines the base address of the active register bank to
be accessed by the CPU at a time. The number of register banks is only restricted by the
available internal RAM space. For easy parameter passing, a register bank may overlap
others.
A system stack of up to 1024 words is provided as a storage for temporary data. The
system stack is also located within the on-chip RAM area, and it is accessed by the CPU
via the stack pointer (SP) register. Two separate SFRs, STKOV and STKUN, are
implicitly compared against the stack pointer value upon each stack access for the
detection of a stack overflow or underflow.
Hardware detection of the selected memory space is placed at the internal memory
decoders and allows the user to specify any address directly or indirectly and obtain the
desired data without using temporary registers or special instructions.
A 2 KByte 16-bit wide on-chip XRAM provides fast access to user data (variables),
user stacks and code. The on-chip XRAM is realized as an X-Peripheral and appears to
the software as an external RAM. Therefore it cannot store register banks and is not
bitaddressable. The XRAM allows 16-bit accesses with maximum speed.
For Special Function Registers 1024 Bytes of the address space are reserved. The
standard Special Function Register area (SFR) uses 512 Bytes, while the Extended
Special Function Register area (ESFR) uses the other 512 Bytes. (E)SFRs are wordwide
registers which are used for controlling and monitoring functions of the different on-chip
units. Unused (E)SFR addresses are reserved for future members of the C166 Family
with enhanced functionality.
An optional internal ROM provides for both code and constant data storage. This
memory area is connected to the CPU via a 32-bit-wide bus. Thus, an entire double-word
instruction can be fetched in just one machine cycle.
Program execution from on-chip program memory is the fastest of all possible
alternatives.
The size of the on-chip ROM depends on the chosen derivative.
User’s Manual
2-9
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
External Bus Interface
In order to meet the needs of designs where more memory is required than is provided
on chip, up to 16 MBytes of external RAM and/or ROM can be connected to the
microcontroller via its external bus interface. The integrated External Bus Controller
(EBC) allows to access external memory and/or peripheral resources in a very flexible
way. For up to five address areas the bus mode (multiplexed/demultiplexed), the data
bus width (8-bit/16-bit) and even the length of a bus cycle (waitstates, signal delays) can
be selected independently. This allows to access a variety of memory and peripheral
components directly and with maximum efficiency. If the device does not run in Single
Chip Mode, where no external memory is required, the EBC can control external
accesses in one of the following external access modes:
• 16-/18-/20-/24-bit Addresses, 16-bit Data, Demultiplexed
• 16-/18-/20-/24-bit Addresses, 8-bit Data, Demultiplexed
• 16-/18-/20-/24-bit Addresses, 16-bit Data, Multiplexed
• 16-/18-/20-/24-bit Addresses, 8-bit Data, Multiplexed
The demultiplexed bus modes use PORT1 for addresses and PORT0 for data input/
output. The multiplexed bus modes use PORT0 for both addresses and data input/
output. Port 4 is used for the upper address lines (A16 …) if selected.
Important timing characteristics of the external bus interface (waitstates, ALE length and
Read/Write Delay) have been made programmable to allow the user the adaption of a
wide range of different types of memories and/or peripherals. Access to very slow
memories or peripherals is supported via a particular ‘Ready’ function.
For applications which require less than 64 KBytes of address space, a non-segmented
memory model can be selected, where all locations can be addressed by 16-bits, and
thus Port 4 is not needed as an output for the upper address bits (Axx … A16), as is the
case when using the segmented memory model.
The on-chip XBUS is an internal representation of the external bus and allows to access
integrated application-specific peripherals/modules in the same way as external
components. It provides a defined interface for these customized peripherals.
The on-chip XRAM and the on-chip CAN-Module are examples for these X-Peripherals.
User’s Manual
2-10
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
2.3
The On-chip Peripheral Blocks
The C166 Family clearly separates peripherals from the core. This structure permits the
maximum number of operations to be performed in parallel and allows peripherals to be
added or deleted from family members without modifications to the core. Each functional
block processes data independently and communicates information over common
buses. Peripherals are controlled by data written to the respective Special Function
Registers (SFRs). These SFRs are located either within the standard SFR area
(00’FE00 … 00’FFFF ) or within the extended ESFR area (00’F000 … 00’F1FF ).
H
H
H
H
These built in peripherals either allow the CPU to interface with the external world, or
provide functions on-chip that otherwise were to be added externally in the respective
system.
The C167CR generic peripherals are:
• Two General Purpose Timer Blocks (GPT1 and GPT2)
• Two Serial Interfaces (ASC0 and SSC)
• A Watchdog Timer
• Two 16-channel Capture/Compare units (CAPCOM1 and CAPCOM2)
• A 4-channel Pulse Width Modulation unit
• A 10-bit Analog/Digital Converter
• Nine IO ports with a total of 111 IO lines
Each peripheral also contains a set of Special Function Registers (SFRs), which control
the functionality of the peripheral and temporarily store intermediate data results. Each
peripheral has an associated set of status flags. Individually selected clock signals are
generated for each peripheral from binary multiples of the CPU clock.
Peripheral Interfaces
The on-chip peripherals generally have two different types of interfaces, an interface to
the CPU and an interface to external hardware. Communication between CPU and
peripherals is performed through Special Function Registers (SFRs) and interrupts. The
SFRs serve as control/status and data registers for the peripherals. Interrupt requests
are generated by the peripherals based on specific events which occur during their
operation (e.g. operation complete, error, etc.).
For interfacing with external hardware, specific pins of the parallel ports are used, when
an input or output function has been selected for a peripheral. During this time, the port
pins are controlled by the peripheral (when used as outputs) or by the external hardware
which controls the peripheral (when used as inputs). This is called the ‘alternate (input
or output) function’ of a port pin, in contrast to its function as a general purpose IO pin.
User’s Manual
2-11
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
Peripheral Timing
Internal operation of CPU and peripherals is based on the CPU clock (fCPU). The on-chip
oscillator derives the CPU clock from the crystal or from the external clock signal. The
clock signal which is gated to the peripherals is independent from the clock signal which
feeds the CPU. During Idle mode the CPU’s clock is stopped while the peripherals
continue their operation. Peripheral SFRs may be accessed by the CPU once per state.
When an SFR is written to by software in the same state where it is also to be modified
by the peripheral, the software write operation has priority. Further details on peripheral
timing are included in the specific sections about each peripheral.
Programming Hints
Access to SFRs
All SFRs reside in data page 3 of the memory space. The following addressing
mechanisms allow to access the SFRs:
• Indirect or direct addressing with 16-bit (mem) addresses must guarantee that the
used data page pointer (DPP0 … DPP3) selects data page 3.
• Accesses via the Peripheral Event Controller (PEC) use the SRCPx and DSTPx
pointers instead of the data page pointers.
• Short 8-bit (reg) addresses to the standard SFR area do not use the data page
pointers but directly access the registers within this 512 Byte area.
• Short 8-bit (reg) addresses to the extended ESFR area require switching to the
512 Byte extended SFR area. This is done via the EXTension instructions EXTR,
EXTP(R), EXTS(R).
Byte write operations to word wide SFRs via indirect or direct 16-bit (mem) addressing
or byte transfers via the PEC force zeros in the non-addressed byte. Byte write
operations via short 8-bit (reg) addressing can only access the low byte of an SFR and
force zeros in the high byte. It is therefore recommended, to use the bit field instructions
(BFLDL and BFLDH) to write to any number of bits in either byte of an SFR without
disturbing the non-addressed byte and the unselected bits.
Reserved Bits
Some of the bits which are contained in the C167CR’s SFRs are marked as ‘Reserved’.
User software should never write ‘1’s to reserved bits. These bits are currently not
implemented and may be used in future products to invoke new functions. In this case,
the active state for these functions will be ‘1’, and the inactive state will be ‘0’. Therefore
writing only ‘0’s to reserved locations provides portability of the current software to future
devices. After read accesses reserved bits should be ignored or masked out.
User’s Manual
2-12
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
Serial Channels
Serial communication with other microcontrollers, processors, terminals or external
peripheral components is provided by two serial interfaces with different functionality, an
Asynchronous/Synchronous Serial Channel (ASC0) and a High-Speed Synchronous
Serial Channel (SSC).
The ASC0 is upward compatible with the serial ports of the Infineon 8-bit microcontroller
families. It supports full-duplex asynchronous communication at up to 780/1030 KBaud
and half-duplex synchronous communication at up to 3.1/4.1 MBaud @ 25/33 MHz CPU
clock.
A dedicated baud rate generator allows to set up all standard baud rates without
oscillator tuning. For transmission, reception and error handling 4 separate interrupt
vectors are provided. In asynchronous mode, 8- or 9-bit data frames are transmitted or
received, preceded by a start bit and terminated by one or two stop bits. For
multiprocessor communication, a mechanism to distinguish address from data bytes has
been included (8-bit data plus wake up bit mode).
In synchronous mode, the ASC0 transmits or receives bytes (8 bits) synchronously to a
shift clock which is generated by the ASC0. The ASC0 always shifts the LSB first. A loop
back option is available for testing purposes.
A number of optional hardware error detection capabilities has been included to increase
the reliability of data transfers. A parity bit can automatically be generated on
transmission or be checked on reception. Framing error detection allows to recognize
data frames with missing stop bits. An overrun error will be generated, if the last
character received has not been read out of the receive buffer register at the time the
reception of a new character is complete.
The SSC supports full-duplex synchronous communication at up to 6.25/8.25 Mbaud @
25/33 MHz CPU clock. It may be configured so it interfaces with serially linked peripheral
components. A dedicated baud rate generator allows to set up all standard baud rates
without oscillator tuning. For transmission, reception and error handling 3 separate
interrupt vectors are provided.
The SSC transmits or receives characters of 2 … 16-bits length synchronously to a shift
clock which can be generated by the SSC (master mode) or by an external master (slave
mode). The SSC can start shifting with the LSB or with the MSB and allows the selection
of shifting and latching clock edges as well as the clock polarity.
A number of optional hardware error detection capabilities has been included to increase
the reliability of data transfers. Transmit and receive error supervise the correct handling
of the data buffer. Phase and baudrate error detect incorrect serial data.
User’s Manual
2-13
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
The On-chip CAN Module
The integrated CAN Module handles the completely autonomous transmission and
reception of CAN frames in accordance with the CAN specification V2.0 part B (active),
i.e. the on-chip CAN Module can receive and transmit standard frames with 11-bit
identifiers as well as extended frames with 29-bit identifiers.
The module provides Full CAN functionality on up to 15 message objects. Message
object 15 may be configured for Basic CAN functionality. Both modes provide separate
masks for acceptance filtering which allows to accept a number of identifiers in Full CAN
mode and also allows to disregard a number of identifiers in Basic CAN mode. All
message objects can be updated independent from the other objects and are equipped
for the maximum message length of 8 Bytes.
The bit timing is derived from the XCLK and is programmable up to a data rate of
1 MBaud. The CAN Module uses two pins to interface to a bus transceiver.
Note: The CAN Module is not part of all C167 derivatives. This description, of course,
refers to those devices only which incorporate a CAN Module.
Parallel Ports
The C167CR provides up to 111 IO lines which are organized into eight input/output
ports and one input port. All port lines are bit-addressable, and all input/output lines are
individually (bit-wise) programmable as inputs or outputs via direction registers. The IO
ports are true bidirectional ports which are switched to high impedance state when
configured as inputs. The output drivers of five IO ports can be configured (pin by pin)
for push/pull operation or open-drain operation via control registers. During the internal
reset, all port pins are configured as inputs.
All port lines have programmable alternate input or output functions associated with
them. PORT0 and PORT1 may be used as address and data lines when accessing
external memory, while Port 4 outputs the additional segment address bits A23/19/17 …
A16 in systems where segmentation is used to access more than 64 KBytes of memory.
Port 6 provides the optional bus arbitration signals (BREQ, HLDA, HOLD) and the chip
select signals CS4 … CS0. Port 2 accepts the fast external interrupt inputs and provides
inputs/outputs for the CAPCOM1 unit. Port 3 includes alternate functions of timers, serial
interfaces, the optional bus control signal BHE and the system clock output (CLKOUT).
Port 5 is used for timer control signals and for the analog inputs to the A/D Converter.
Port 7 provides the output signals from the PWM unit and inputs/outputs for the
CAPCOM2 unit (more on P1H). Port 8 provides inputs/outputs for the CAPCOM2 unit.
Four pins of PORT1 may also be used as inputs for the CAPCOM2 unit. All port lines
that are not used for these alternate functions may be used as general purpose IO lines.
User’s Manual
2-14
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
A/D Converter
For analog signal measurement, a 10-bit A/D converter with 16 multiplexed input
channels and a sample and hold circuit has been integrated on-chip. It uses the method
of successive approximation. The sample time (for loading the capacitors) and the
conversion time is programmable and can so be adjusted to the external circuitry.
Overrun error detection/protection is provided for the conversion result register
(ADDAT): either an interrupt request will be generated when the result of a previous
conversion has not been read from the result register at the time the next conversion is
complete, or the next conversion is suspended in such a case until the previous result
has been read.
For applications which require less analog input channels, the remaining channel inputs
can be used as digital input port pins.
The A/D converter of the C167CR supports four different conversion modes. In the
standard Single Channel conversion mode, the analog level on a specified channel is
sampled once and converted to a digital result. In the Single Channel Continuous mode,
the analog level on a specified channel is repeatedly sampled and converted without
software intervention. In the Auto Scan mode, the analog levels on a prespecified
number of channels are sequentially sampled and converted. In the Auto Scan
Continuous mode, the number of prespecified channels is repeatedly sampled and
converted. In addition, the conversion of a specific channel can be inserted (injected) into
a running sequence without disturbing this sequence. This is called Channel Injection
Mode.
The Peripheral Event Controller (PEC) may be used to automatically store the
conversion results into a table in memory for later evaluation, without requiring the
overhead of entering and exiting interrupt routines for each data transfer.
User’s Manual
2-15
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
General Purpose Timer (GPT) Unit
The GPT units represent a very flexible multifunctional timer/counter structure which
may be used for many different time related tasks such as event timing and counting,
pulse width and duty cycle measurements, pulse generation, or pulse multiplication.
The five 16-bit timers are organized in two separate modules, GPT1 and GPT2. Each
timer in each module may operate independently in a number of different modes, or may
be concatenated with another timer of the same module.
Each timer can be configured individually for one of four basic modes of operation, which
are Timer, Gated Timer, Counter Mode and Incremental Interface Mode (GPT1 timers).
In Timer Mode the input clock for a timer is derived from the internal CPU clock divided
by a programmable prescaler, while Counter Mode allows a timer to be clocked in
reference to external events (via TxIN).
Pulse width or duty cycle measurement is supported in Gated Timer Mode where the
operation of a timer is controlled by the ‘gate’ level on its external input pin TxIN.
In Incremental Interface Mode the GPT1 timers can be directly connected to the
incremental position sensor signals A and B via the respective inputs TxIN and TxEUD.
Direction and count signals are internally derived from these two input signals, so the
contents of timer Tx corresponds to the sensor position. The third position sensor signal
TOP0 can be connected to an interrupt input.
The count direction (up/down) for each timer is programmable by software or may
additionally be altered dynamically by an external signal (TxEUD) to facilitate e.g.
position tracking.
The core timers T3 and T6 have output toggle latches (TxOTL) which change their state
on each timer over-flow/underflow. The state of these latches may be output on port pins
(TxOUT) or may be used internally to concatenate the core timers with the respective
auxiliary timers resulting in 32/33-bit timers/counters for measuring long time periods
with high resolution.
Various reload or capture functions can be selected to reload timers or capture a timer’s
contents triggered by an external signal or a selectable transition of toggle latch TxOTL.
The maximum resolution of the timers in module GPT1 is 8 CPU clock cycles (= 16 TCL).
With their maximum resolution of 4 CPU clock cycles (= 8 TCL) the GPT2 timers provide
precise event control and time measurement.
User’s Manual
2-16
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
Capture/Compare (CAPCOM) Units
The two CAPCOM units support generation and control of timing sequences on up to 32
channels with a maximum resolution of 8 CPU clock cycles. The CAPCOM units are
typically used to handle high speed IO tasks such as pulse and waveform generation,
pulse width modulation (PWM), Digital to Analog (D/A) conversion, software timing, or
time recording relative to external events.
Four 16-bit timers (T0/T1, T7/T8) with reload registers provide two independent time
bases for the capture/compare register array.
The input clock for the timers is programmable to several prescaled values of the internal
CPU clock, or may be derived from an overflow/underflow of timer T6 in module GPT2.
This provides a wide range of variation for the timer period and resolution and allows
precise adjustments to the application specific requirements. In addition, external count
inputs for CAPCOM timers T0 and T7 allow event scheduling for the capture/compare
registers relative to external events.
Both of the two capture/compare register arrays contain 16 dual purpose capture/
compare registers, each of which may be individually allocated to either CAPCOM timer
T0 or T1 (T7 or T8, respectively), and programmed for capture or compare function.
Each register has one port pin associated with it which serves as an input pin for
triggering the capture function, or as an output pin (except for CC24 … CC27) to indicate
the occurrence of a compare event.
When a capture/compare register has been selected for capture mode, the current
contents of the allocated timer will be latched (captured) into the capture/compare
register in response to an external event at the port pin which is associated with this
register. In addition, a specific interrupt request for this capture/compare register is
generated. Either a positive, a negative, or both a positive and a negative external signal
transition at the pin can be selected as the triggering event. The contents of all registers
which have been selected for one of the five compare modes are continuously compared
with the contents of the allocated timers. When a match occurs between the timer value
and the value in a capture/compare register, specific actions will be taken based on the
selected compare mode.
User’s Manual
2-17
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
Pulse Width Modulation Unit
The PWM Unit supports the generation of up to four independent high-speed PWM
signals. It allows to generate standard (edge aligned) PWM signals as well as
symmetrical (center aligned) PWM signals. In Burst Mode two channels may be
combined with their output signals ANDed, where one channel gates the output signal of
the other channel. Single Shot Mode allows to generate single output pulses
(retriggerable) under software control. Each PWM channel is controlled by an up/down
counter with associated reload and compare registers. The polarity of the PWM output
signals may be controlled via the respective port output latch (combination via EXOR).
Watchdog Timer
The Watchdog Timer represents one of the fail-safe mechanisms which have been
implemented to prevent the controller from malfunctioning for longer periods of time.
The Watchdog Timer is always enabled after a reset of the chip, and can only be
disabled in the time interval until the EINIT (end of initialization) instruction has been
executed. Thus, the chip’s start-up procedure is always monitored. The software has to
be designed to service the Watchdog Timer before it overflows. If, due to hardware or
software related failures, the software fails to do so, the Watchdog Timer overflows and
generates an internal hardware reset and pulls the RSTOUT pin low in order to allow
external hardware components to reset.
The Watchdog Timer is a 16-bit timer, clocked with the CPU clock divided either by 2 or by
128. The high byte of the Watchdog Timer register can be set to a prespecified reload
value (stored in WDTREL) in order to allow further variation of the monitored time interval.
Each time it is serviced by the application software, the high byte of the Watchdog Timer
is reloaded. Thus, time intervals between 21 ms and 335 ms can be monitored @
25 MHz (16 ms and 254 ms @ 33 MHz). The default Watchdog Timer interval after reset
is 5.2/4.0 ms (@ 25/33 MHz).
2.4
Protected Bits
The C167CR provides a special mechanism to protect bits which can be modified by the
on-chip hardware from being changed unintentionally by software accesses to related
bits (see also Chapter 4).
User’s Manual
2-18
V3.1, 2000-03
C167CR
Derivatives
Architectural Overview
The following bits are protected
Table 2-1
C167CR Protected Bits
Register
Bit Name
Notes
T2IC, T3IC, T4IC
T5IC, T6IC
CRIC
T2IR, T3IR, T4IR
T5IR, T6IR
CRIR
GPT1 timer interrupt request flags
GPT2 timer interrupt request flags
GPT2 CAPREL interrupt request flag
GPTx timer output toggle latches
T3CON, T6CON
T0IC, T1IC
T7IC, T8IC
S0TIC, S0TBIC
S0RIC, S0EIC
S0CON
T3OTL, T6OTL
T0IR, T1IR
T7IR, T8IR
S0TIR, S0TBIR
S0RIR, S0EIR
S0REN
CAPCOM1 timer interrupt request flags
CAPCOM2 timer interrupt request flags
ASC0 transmit(buffer) interrupt request flags
ASC0 receive/error interrupt request flags
ASC0 receiver enable flag
SSCTIC, SSCRIC SSCTIR, SSCRIR SSC transmit/receive interrupt request flags
SSCEIC
SSCEIR
SSC error interrupt request flag
SSC busy flag
SSCCON
SSCCON
SSCCON
ADCIC, ADEIC
ADCON
SSCBSY
SSCBE, SSCPE
SSCRE, SSCTE
ADCIR, ADEIR
ADST, ADCRQ
SSC error flags
SSC error flags
ADC end-of-conv./overrun intr. request flag
ADC start flag/injection request flag
CC31IC … CC16IC CC31IR … CC16IR CAPCOM2 interrupt request flags
CC15IC … CC0IC CC15IR … CC0IR CAPCOM1 interrupt request flags
PWMIC
PWMIR
PWM module interrupt request flag
All bits of PWMCON0
All bits of PWMCON1
Class A trap flags
PWMCON0
PIR3 … PTR0
PS3 … PEN0
TFR.15,14,13
TFR.7,3,2,1,0
P2.15 … P2.0
P7.7 … P7.0
P8.7 … P8.0
XP3IR … XP0IC
PWMCON1
TFR
TFR
Class B trap flags
P2
All bits of Port 2
P7
All bits of Port 7
P8
All bits of Port 8
XP3IC … XP0IC
X-Peripheral interrupt request flags
Σ = 133 protected bits.
User’s Manual
2-19
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
3
Memory Organization
The memory space of the C167CR is configured in a “Von Neumann” architecture. This
means that code and data are accessed within the same linear address space. All of the
physically separated memory areas, including internal ROM/Flash/OTP (where
integrated), internal RAM, the internal Special Function Register Areas (SFRs and
ESFRs), the address areas for integrated XBUS peripherals and external memory are
mapped into one common address space.
The C167CR provides a total addressable memory space of 16 MBytes. This address
space is arranged as 256 segments of 64 KBytes each, and each segment is again
subdivided into four data pages of 16 KBytes each (see Figure 3-1).
FF’FFFF
H
255
255...2
254...129
01’FFFF
H
128
80’0000
H
127
Begin of
Prog. Memory
above 32 KB
126...65
64
40’0000
H
63
01’8000
01’0000
H
H
62...12
0A’FFFF
H
Alternate
ROM
Area
11
10
9
8
08’0000
Data Page 3
Data Page 2
H
7
6
5
4
3
Internal
ROM
Area
02’FFFF
01’FFFF
H
H
2
1
00’0000
00’0000
0
H
H
Total Address Space
16 MByte, Segments 255...0
Segments 1 and 0
64 + 64 Kbyte
MCA04325
Figure 3-1
Address Space Overview
User’s Manual
3-1
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
Most internal memory areas are mapped into segment 0, the system segment. The
upper 4 KByte of segment 0 (00’F000 … 00’FFFF ) hold the Internal RAM and Special
H
H
Function Register Areas (SFR and ESFR). The lower 32 KByte of segment 0
(00’0000 … 00’7FFF ) may be occupied by a part of the on-chip ROM/Flash/OTP
H
H
memory and is called the Internal ROM area. This ROM area can be remapped to
segment 1 (01’0000 … 01’7FFF ), to enable external memory access in the lower half
H
H
of segment 0, or the internal ROM may be disabled at all.
Code and data may be stored in any part of the internal memory areas, except for the
SFR blocks, which may be used for control/data, but not for instructions.
Note: Accesses to the internal ROM area on ROMless devices will produce
unpredictable results.
Bytes are stored at even or odd byte addresses. Words are stored in ascending memory
locations with the low byte at an even byte address being followed by the high byte at
the next odd byte address. Double words (code only) are stored in ascending memory
locations as two subsequent words. Single bits are always stored in the specified bit
position at a word address. Bit position 0 is the least significant bit of the byte at an even
byte address, and bit position 15 is the most significant bit of the byte at the next odd
byte address. Bit addressing is supported for a part of the Special Function Registers, a
part of the internal RAM and for the General Purpose Registers.
xxxx6
xxxx5
xxxx4
xxxx3
xxxx2
xxxx1
xxxx0
xxxxF
H
H
H
H
H
H
H
H
15
7
14
6
Bits
Bits
8
0
Byte
Byte
Word (High Byte)
Word (Low Byte)
MCD01996
Figure 3-2
Storage of Words, Bytes, and Bits in a Byte Organized Memory
Note: Byte units forming a single word or a double word must always be stored within
the same physical (internal, external, ROM, RAM) and organizational (page,
segment) memory area.
User’s Manual
3-2
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
3.1
Internal ROM Area
The C167CR may reserve an address area of variable size (depending on the version)
for on-chip mask-programmable ROM/Flash/OTP memory (organized as X × 32). The
lower 32 KByte of this on-chip memory block are referred to as “Internal ROM Area”.
Internal ROM accesses are globally enabled or disabled via bit ROMEN in register
SYSCON. This bit is set during reset according to the level on pin EA, or may be altered
via software. If enabled, the internal ROM area occupies the lower 32 KByte of either
segment 0 or segment 1 (alternate ROM area). This mapping is controlled by bit ROMS1
in register SYSCON.
Note: The size of the internal ROM area is independent of the size of the actual
implemented Program Memory. Also devices with less than 32 KByte of Program
Memory or with no Program Memory at all will have this 32 KByte area occupied,
if the Program Memory is enabled. Devices with a larger Program Memory provide
the mapping option only for the internal ROM area.
Devices with a Program Memory size above 32 KByte expand the ROM area from the
middle of segment 1, i.e. starting at address 01’8000 .
H
The internal Program Memory can be used for both code (instructions) and data
(constants, tables, etc.) storage.
Code fetches are always made on even byte addresses. The highest possible code
storage location in the internal Program Memory is either xx’xxFE for single word
H
instructions, or xx’xxFC for double word instructions. The respective location must contain
H
a branch instruction (unconditional), because sequential boundary crossing from internal
Program Memory to external memory is not supported and causes erroneous results.
Any word and byte data read accesses may use the indirect or long 16-bit addressing
modes. There is no short addressing mode for internal ROM operands. Any word data
access is made to an even byte address. The highest possible word data storage
location in the internal ROM is xx’xxFE . For PEC data transfers the internal Program
H
Memory can be accessed independent of the contents of the DPP registers via the PEC
source and destination pointers.
The internal Program Memory is not provided for single bit storage, and therefore it is not
bit addressable.
Note: The ‘x’ in the locations above depend on the available Program Memory and on
the mapping.
The internal ROM may be enabled, disabled or mapped into segment 0 or segment 1
under software control. Chapter 21 shows how to do this and reminds of the precautions
that must be taken in order to prevent the system from crashing.
User’s Manual
3-3
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
3.2
Internal RAM and SFR Area
The RAM/SFR area is located within data page 3 and provides access to the internal
RAM (IRAM, organized as X × 16) and to two 512 Byte blocks of Special Function
Registers (SFRs).
The C167CR provides 2 KByte of IRAM.
00’FFFF
00’FFFF
H
H
SFR Area
IRAM/SFR
00’F000
H
X-Peripherals
XRAM
IRAM
00’F600
H
Reserved
00’F200
00’F000
H
H
ESFR Area
CAN1
00’C000
H
00’EF00
H
Ext. Memory
Reserved
00’E7FF
H
XRAM
00’8000
00’E000
H
H
Note: New XBUS peripherals will be preferably placed into the shaded areas,
which now access external memory (bus cycles executed).
MCA04326
Figure 3-3
System Memory Map
User’s Manual
3-4
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
Note: The upper 256 Bytes of SFR area, ESFR area and internal RAM are bit-
addressable (see shaded blocks in Figure 3-3).
Code accesses are always made on even byte addresses. The highest possible code
storage location in the internal RAM is either 00’FDFE for single word instructions or
H
00’FDFC for double word instructions. The respective location must contain a branch
H
instruction (unconditional), because sequential boundary crossing from internal RAM to
the SFR area is not supported and causes erroneous results.
Any word and byte data in the internal RAM can be accessed via indirect or long 16-bit
addressing modes, if the selected DPP register points to data page 3. Any word data
access is made on an even byte address. The highest possible word data storage
location in the internal RAM is 00’FDFE . For PEC data transfers, the internal RAM can
H
be accessed independent of the contents of the DPP registers via the PEC source and
destination pointers.
The upper 256 Byte of the internal RAM (00’FD00 through 00’FDFF ) and the GPRs of
H
H
the current bank are provided for single bit storage, and thus they are bitaddressable.
System Stack
The system stack may be defined within the internal RAM. The size of the system stack
is controlled by bitfield STKSZ in register SYSCON (see Table 3-1).
Table 3-1
<STKSZ>
System Stack Size Encoding
Stack Size (words) Internal RAM Addresses (words)
0 0 0
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1
256
128
64
00’FBFE … 00’FA00 (Default after Reset)
B
B
B
B
B
B
B
B
H
H
H
H
00’FBFE … 00’FB00
H
00’FBFE … 00’FB80
H
32
00’FBFE … 00’FBC0
H H
512
–
00’FBFE … 00’F800
H H
Reserved. Do not use this combination.
Reserved. Do not use this combination.
–
1024
00’FDFE … 00’F600 (Note: No circular stack)
H H
For all system stack operations the on-chip RAM is accessed via the Stack Pointer (SP)
register. The stack grows downward from higher towards lower RAM address locations.
Only word accesses are supported to the system stack. A stack overflow (STKOV) and
a stack underflow (STKUN) register are provided to control the lower and upper limits of
the selected stack area. These two stack boundary registers can be used not only for
protection against data destruction, but also allow to implement a circular stack with
hardware supported system stack flushing and filling (except for option ‘111’). The
technique of implementing this circular stack is described in Chapter 21.
User’s Manual
3-5
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
General Purpose Registers
The General Purpose Registers (GPRs) use a block of 16 consecutive words within the
internal RAM. The Context Pointer (CP) register determines the base address of the
currently active register bank. This register bank may consist of up to 16 Word-GPRs
(R0, R1, … R15) and/or of up to 16 Byte-GPRs (RL0, RH0, … RL7, RH7). The sixteen
Byte-GPRs are mapped onto the first eight Word-GPRs (see Table 3-2).
In contrast to the system stack, a register bank grows from lower towards higher address
locations and occupies a maximum space of 32 Byte. The GPRs are accessed via short
2-, 4-, or 8-bit addressing modes using the Context Pointer (CP) register as base
address (independent of the current DPP register contents). Additionally, each bit in the
currently active register bank can be accessed individually.
Table 3-2
Mapping of General Purpose Registers to RAM Addresses
Internal RAM Address Byte Registers
Word Register
<CP> + 1E
–
R15
R14
R13
R12
R11
R10
R9
H
<CP> + 1C
–
H
H
H
H
H
H
H
<CP> + 1A
–
<CP> + 18
<CP> + 16
<CP> + 14
<CP> + 12
<CP> + 10
–
–
–
–
–
R8
<CP> + 0E
RH7
RH6
RH5
RH4
RH3
RH2
RH1
RH0
RL7
RL6
RL5
RL4
RL3
RL2
RL1
RL0
R7
H
<CP> + 0C
R6
H
H
H
H
H
H
H
<CP> + 0A
R5
<CP> + 08
<CP> + 06
<CP> + 04
<CP> + 02
<CP> + 00
R4
R3
R2
R1
R0
The C167CR supports fast register bank (context) switching. Multiple register banks can
physically exist within the internal RAM at the same time. Only the register bank selected
by the Context Pointer register (CP) is active at a given time, however. Selecting a new
active register bank is simply done by updating the CP register. A particular Switch
Context (SCXT) instruction performs register bank switching and an automatic saving of
User’s Manual
3-6
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
the previous context. The number of implemented register banks (arbitrary sizes) is only
limited by the size of the available internal RAM.
Details on using, switching and overlapping register banks are described in Chapter 21.
PEC Source and Destination Pointers
The 16 word locations in the internal RAM from 00’FCE0 to 00’FCFE (just below the
H
H
bit-addressable section) are provided as source and destination address pointers for
data transfers on the eight PEC channels. Each channel uses a pair of pointers stored
in two subsequent word locations with the source pointer (SRCPx) on the lower and the
destination pointer (DSTPx) on the higher word address (x = 7 … 0).
00’FD00
00’FCFE
H
H
00’FCFE
00’FCFC
DSTP7
SRCP7
H
H
00’FCE0
00’FCDE
H
H
PEC
Source
and
Destination
Pointers
Internal
RAM
00’FCE2
00’FCE0
DSTP0
SRCP0
00’F600
H
H
H
H
00’F5FE
MCD03903
Figure 3-4
Location of the PEC Pointers
Whenever a PEC data transfer is performed, the pair of source and destination pointers,
which is selected by the specified PEC channel number, is accessed independent of the
current DPP register contents and also the locations referred to by these pointers are
accessed independent of the current DPP register contents. If a PEC channel is not
used, the corresponding pointer locations area available and can be used for word or
byte data storage.
For more details about the use of the source and destination pointers for PEC data
transfers see Section 5.
User’s Manual
3-7
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
Special Function Registers
The functions of the CPU, the bus interface, the IO ports and the on-chip peripherals of
the C167CR are controlled via a number of so-called Special Function Registers (SFRs).
These SFRs are arranged within two areas of 512 Byte size each. The first register block,
the SFR area, is located in the 512 Bytes above the internal RAM (00’FFFF …
H
00’FE00 ), the second register block, the Extended SFR (ESFR) area, is located in the
H
512 Bytes below the internal RAM (00’F1FF … 00’F000 ).
H
H
Special function registers can be addressed via indirect and long 16-bit addressing
modes. Using an 8-bit offset together with an implicit base address allows to address
word SFRs and their respective low bytes. However, this does not work for the
respective high bytes!
Note: Writing to any byte of an SFR causes the non-addressed complementary byte to
be cleared!
The upper half of each register block is bit-addressable, so the respective control/status
bits can directly be modified or checked using bit addressing.
When accessing registers in the ESFR area using 8-bit addresses or direct bit
addressing, an Extend Register (EXTR) instruction is required before, to switch the short
addressing mechanism from the standard SFR area to the Extended SFR area. This is
not required for 16-bit and indirect addresses. The GPRs R15 … R0 are duplicated, i.e.
they are accessible within both register blocks via short 2-, 4- or 8-bit addresses without
switching.
ESFR_SWITCH_EXAMPLE:
EXTR
MOV
#4
;Switch to ESFR area for next 4 instr.
;ODP2 uses 8-bit reg addressing
ODP2, #data16
BFLDL DP6, #mask, #data8 ;Bit addressing for bit fields
BSET
MOV
DP1H.7
T8REL, R1
;Bit addressing for single bits
;T8REL uses 16-bit mem address,
;R1 is duplicated into the ESFR space
;(EXTR is not required for this access)
;---- ;----------------- ;The scope of the EXTR #4 instruction…
;… ends here!
MOV
T8REL, R1
;T8REL uses 16-bit mem address,
;R1 is accessed via the SFR space
In order to minimize the use of the EXTR instructions the ESFR area mostly holds
registers which are mainly required for initialization and mode selection. Registers that
need to be accessed frequently are allocated to the standard SFR area, wherever
possible.
Note: The tools are equipped to monitor accesses to the ESFR area and will
automatically insert EXTR instructions, or issue a warning in case of missing or
excessive EXTR instructions.
User’s Manual
3-8
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
3.3
The On-Chip XRAM
The C167CR provides access to 2 KByte of on-chip extension RAM. The XRAM is
located within data page 3 (organized as 1 K × 16). As the XRAM is connected to the
internal XBUS it is accessed like external memory, however, no external bus cycles are
executed for these accesses. XRAM accesses are globally enabled or disabled via bit
XPEN in register SYSCON. This bit is cleared after reset and may be set via software
during the initialization to allow accesses to the on-chip XRAM. When the XRAM is
disabled (default after reset) all accesses to the XRAM area are mapped to external
locations. The XRAM may be used for both code (instructions) and data (variables, user
stack, tables, etc.) storage.
Code fetches are always made on even byte addresses. The highest possible code
storage location in the XRAM is either 00’E7FE for single word instructions, or
H
00’E7FC for double word instructions. The respective location must contain a branch
H
instruction (unconditional), because sequential boundary crossing from XRAM to
external memory is not supported and causes erroneous results.
Any word and byte data read accesses may use the indirect or long 16-bit addressing
modes. There is no short addressing mode for XRAM operands. Any word data access
is made to an even byte address. The highest possible word data storage location in the
XRAM is 00’E7FE . For PEC data transfers the XRAM can be accessed independent of
H
the contents of the DPP registers via the PEC source and destination pointers.
Note: As the XRAM appears like external memory it cannot be used for the C167CR’s
system stack or register banks. The XRAM is not provided for single bit storage
and therefore is not bitaddressable.
The on-chip XRAM is accessed with the following bus cycles:
• Normal ALE
• No cycle time waitstates (no READY control)
• No tristate time waitstate
• No Read/Write delay
• 16-bit demultiplexed bus cycles (4 TCL)
Even if the XRAM is used like external memory it does not occupy BUSCONx/
ADDRSELx registers but rather is selected via additional dedicated XBCON/XADRS
registers. These registers are mask-programmed and are not user accessible. With
these registers the address area 00’E000 to 00’E7FF is reserved for XRAM accesses.
H
H
User’s Manual
3-9
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
XRAM Access via External Masters
In X-Peripheral Share mode (bit XPER-SHARE in register SYSCON is set) the on-chip
XRAM of the C167CR can be accessed by an external master during hold mode via the
C167CR’s bus interface. These external accesses must use the same configuration as
internally programmed (see above). No waitstates are required. In X-Peripheral Share
mode the C167CR bus interface reverses its direction, i.e. address lines (PORT1,
Port 4), control signals (RD, WR), and BHE must be driven by the external master.
Note: The configuration in register SYSCON cannot be changed after the execution of
the EINIT instruction.
User’s Manual
3-10
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
3.4
External Memory Space
The C167CR is capable of using an address space of up to 16 MByte. Only parts of this
address space are occupied by internal memory areas. All addresses which are not used
for on-chip memory (ROM/Flash/OTP or RAM) or for registers may reference external
memory locations. This external memory is accessed via the C167CR’s external bus
interface.
Four memory bank sizes are supported:
• Non-segmented mode: 64 KByte with A15 … A0 on PORT0 or PORT1
• 2-bit segmented mode: 256 KByte with A17 … A16 on Port 4
and A15 … A0 on PORT0 or PORT1
• 4-bit segmented mode: 1 MByte with A19 … A16 on Port 4
and A15 … A0 on PORT0 or PORT1
• 8-bit segmented mode: 16 MByte with A22 … A16 on Port 4
and A15 … A0 on PORT0 or PORT1
Each bank can be directly addressed via the address bus, while the programmable chip
select signals can be used to select various memory banks.
The C167CR also supports four different bus types:
• Multiplexed 16-bit Bus
• Multiplexed 8-bit Bus
with address and data on PORT0 (Default after Reset)
with address and data on PORT0/P0L
• Demultiplexed 16-bit Bus with address on PORT1 and data on PORT0
• Demultiplexed 8-bit Bus with address on PORT1 and data on P0L
Memory model and bus mode are selected during reset by pin EA and PORT0 pins. For
further details about the external bus configuration and control please refer to Chapter 9.
External word and byte data can only be accessed via indirect or long 16-bit addressing
modes using one of the four DPP registers. There is no short addressing mode for
external operands. Any word data access is made to an even byte address.
For PEC data transfers the external memory in segment 0 can be accessed independent
of the contents of the DPP registers via the PEC source and destination pointers.
The external memory is not provided for single bit storage and therefore it is not
bitaddressable.
User’s Manual
3-11
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
3.5
Crossing Memory Boundaries
The address space of the C167CR is implicitly divided into equally sized blocks of
different granularity and into logical memory areas. Crossing the boundaries between
these blocks (code or data) or areas requires special attention to ensure that the
controller executes the desired operations.
Memory Areas are partitions of the address space that represent different kinds of
memory (if provided at all). These memory areas are the internal RAM/SFR area, the
internal ROM/Flash/OTP (if available), the on-chip X-Peripherals (if integrated) and the
external memory.
Accessing subsequent data locations that belong to different memory areas is no
problem. However, when executing code, the different memory areas must be switched
explicitly via branch instructions. Sequential boundary crossing is not supported and
leads to erroneous results.
Note: Changing from the external memory area to the internal RAM/SFR area takes
place within segment 0.
Segments are contiguous blocks of 64 KByte each. They are referenced via the code
segment pointer CSP for code fetches and via an explicit segment number for data
accesses overriding the standard DPP scheme.
During code fetching segments are not changed automatically, but rather must be
switched explicitly. The instructions JMPS, CALLS and RETS will do this.
In larger sequential programs make sure that the highest used code location of a
segment contains an unconditional branch instruction to the respective following
segment, to prevent the prefetcher from trying to leave the current segment.
Data Pages are contiguous blocks of 16 KByte each. They are referenced via the data
page pointers DPP3 … 0 and via an explicit data page number for data accesses
overriding the standard DPP scheme. Each DPP register can select one of the possible
1024 data pages. The DPP register that is used for the current access is selected via the
two upper bits of the 16-bit data address. Subsequent 16-bit data addresses that cross
the 16 KByte data page boundaries therefore will use different data page pointers, while
the physical locations need not be subsequent within memory.
User’s Manual
3-12
V3.1, 2000-03
C167CR
Derivatives
Memory Organization
3.6
Protection of the On-chip Mask ROM
The on-chip mask ROM of the C167CR can be protected against read accesses of both
code and data. ROM protection is established during the production process of the
device (a ROM mask can be ordered with a ROM protection or without it). No software
control is possible, i.e. the ROM protection cannot be disabled or enabled by software.
When a device has been produced with ROM protection active, the ROM contents are
protected against unauthorized access by the following measures:
• No data read accesses to the internal ROM by any instruction which is executed from
any location outside the on-chip mask ROM (including IRAM, XRAM, and external
memory).
A program cannot read any data out of the protected ROM from outside.
The read data will be replaced by the default value 009B for any read access to any
H
location.
• No codes fetches from the internal ROM by any instruction which is executed from
any location outside the on-chip mask ROM (including IRAM, XRAM, and external
memory).
A program cannot branch to a location within the protected ROM from outside. This
applies to JUMPs as well as to RETurns, i.e. a called routine within RAM or external
memory can never return to the protected ROM.
The fetched code will be replaced by the default value 009BH for any access to any
location. This default value will be decoded as the instruction “TRAP #00” which will
restart program execution at location 00’0000 .
H
Note: ROM protection may be used for applications where the complete software fits into
the on-chip ROM, or where the on-chip ROM holds an initialization software which
is then replaced by an external (e.g.) application software. In the latter case no
data (constants, tables, etc.) can be stored within the ROM. The ROM itself should
be mapped to segment 1 before branching outside, so an interrupt vector table
can established in external memory.
User’s Manual
3-13
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
4
The Central Processing Unit (CPU)
Basic tasks of the CPU are to fetch and decode instructions, to supply operands for the
arithmetic and logic unit (ALU), to perform operations on these operands in the ALU, and
to store the previously calculated results. As the CPU is the main engine of the C167CR
controller, it is also affected by certain actions of the peripheral subsystem.
Since a four stage pipeline is implemented in the C167CR, up to four instructions can be
processed in parallel. Most instructions of the C167CR are executed in one machine
cycle (2 CPU clock periods) due to this parallelism.
This chapter describes how the pipeline works for sequential and branch instructions in
general, and which hardware provisions have been made to speed the execution of jump
instructions in particular. The general instruction timing is described including standard
and exceptional timing.
While internal memory accesses are normally performed by the CPU itself, external
peripheral or memory accesses are performed by a particular on-chip External Bus
Controller (EBC), which is automatically invoked by the CPU whenever a code or data
address refers to the external address space.
CPU
16
Internal
RAM
SP
STKOV
STKUN
MDH
MDL
R15
Exec. Unit
Instr. Ptr.
Instr. Reg.
Mul/Div-HW
Bit-Mask Gen
General
Purpose
Registers
R15
ALU
32
4-Stage
Pipeline
(16-bit)
ROM
Barrel - Shifter
Context Ptr.
R0
PSW
SYSCON
BUSCON 0
BUSCON 1
BUSCON 2
BUSCON 3
BUSCON 4
16
R0
ADDRSEL 1
ADDRSEL 2
ADDRSEL 3
ADDRSEL 4
Data Page Ptr.
Code Seg. Ptr.
MCB02147
Figure 4-1
CPU Block Diagram
User’s Manual
4-1
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
If possible, the CPU continues operating while an external memory access is in
progress. If external data are required but are not yet available, or if a new external
memory access is requested by the CPU, before a previous access has been completed,
the CPU will be held by the EBC until the request can be satisfied. The EBC is described
in a dedicated chapter.
The on-chip peripheral units of the C167CR work nearly independent of the CPU with a
separate clock generator. Data and control information is interchanged between the
CPU and these peripherals via Special Function Registers (SFRs).
Whenever peripherals need a non-deterministic CPU action, an on-chip Interrupt
Controller compares all pending peripheral service requests against each other and
prioritizes one of them. If the priority of the current CPU operation is lower than the
priority of the selected peripheral request, an interrupt will occur.
Basically, there are two types of interrupt processing:
• Standard interrupt processing forces the CPU to save the current program status
and the return address on the stack before branching to the interrupt vector jump
table.
• PEC interrupt processing steals just one machine cycle from the current CPU
activity to perform a single data transfer via the on-chip Peripheral Event Controller
(PEC).
System errors detected during program execution (so-called hardware traps) or an
external non-maskable interrupt are also processed as standard interrupts with a very
high priority.
In contrast to other on-chip peripherals, there is a closer conjunction between the
watchdog timer and the CPU. If enabled, the watchdog timer expects to be serviced by
the CPU within a programmable period of time, otherwise it will reset the chip. Thus, the
watchdog timer is able to prevent the CPU from going totally astray when executing
erroneous code. After reset, the watchdog timer starts counting automatically, but it can
be disabled via software, if desired.
Beside its normal operation there are the following particular CPU states:
• Reset state: Any reset (hardware, software, watchdog) forces the CPU into a
predefined active state.
• IDLE state: The clock signal to the CPU itself is switched off, while the clocks for the
on-chip peripherals keep running.
• POWER DOWN state: All of the on-chip clocks are switched off (RTC clock selectable),
all inputs are disregarded.
A transition into an active CPU state is forced by an interrupt (if being in IDLE) or by a
reset (if being in POWER DOWN mode).
The IDLE, POWER DOWN, and RESET states can be entered by particular C167CR
system control instructions.
User’s Manual
4-2
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
A set of Special Function Registers is dedicated to the functions of the CPU core:
• General System Configuration:
SYSCON (RP0H)
• CPU Status Indication and Control: PSW
• Code Access Control:
IP, CSP
• Data Paging Control:
DPP0, DPP1, DPP2, DPP3
CP
SP, STKUN, STKOV
MDL, MDH, MDC
ZEROS, ONES
• GPRs Access Control:
• System Stack Access Control:
• Multiply and Divide Support:
• ALU Constants Support:
4.1
Instruction Pipelining
The instruction pipeline of the C167CR partitiones instruction processing into four stages
of which each one has its individual task:
1st → FETCH: In this stage the instruction selected by the Instruction Pointer (IP) and
the Code Segment Pointer (CSP) is fetched from either the internal ROM, internal RAM,
or external memory.
2nd → DECODE: In this stage the instructions are decoded and, if required, the operand
addresses are calculated and the respective operands are fetched. For all instructions,
which implicitly access the system stack, the SP register is either decremented or
incremented, as specified. For branch instructions the Instruction Pointer and the Code
Segment Pointer are updated with the desired branch target address (provided that the
branch is taken).
3rd → EXECUTE: In this stage an operation is performed on the previously fetched
operands in the ALU. Additionally, the condition flags in the PSW register are updated
as specified by the instruction. All explicit writes to the SFR memory space and all auto-
increment or auto-decrement writes to GPRs used as indirect address pointers are
performed during the execute stage of an instruction, too.
4th → WRITE BACK: In this stage all external operands and the remaining operands
within the internal RAM space are written back.
A particularity of the C167CR are the so-called injected instructions. These injected
instructions are generated internally by the machine to provide the time needed to
process instructions, which cannot be processed within one machine cycle. They are
automatically injected into the decode stage of the pipeline, and then they pass through
the remaining stages like every standard instruction. Program interrupts are performed
by means of injected instructions, too. Although these internally injected instructions will
not be noticed in reality, they are introduced here to ease the explanation of the pipeline
in the following.
User’s Manual
4-3
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
Sequential Instruction Processing
Each single instruction has to pass through each of the four pipeline stages regardless
of whether all possible stage operations are really performed or not. Since passing
through one pipeline stage takes at least one machine cycle, any isolated instruction
takes at least four machine cycles to be completed. Pipelining, however, allows parallel
(i.e. simultaneous) processing of up to four instructions. Thus, most of the instructions
seem to be processed during one machine cycle as soon as the pipeline has been filled
once after reset (see Figure 4-2).
Instruction pipelining increases the average instruction throughput considered over a
certain period of time. In the following, any execution time specification of an instruction
always refers to the average execution time due to pipelined parallel instruction
processing.
1 Machine Cycle
I1
I2
I1
I3
I2
I1
I4
I3
I2
I1
I5
I4
I3
I2
I6
I5
I4
I3
FETCH
DECODE
EXECUTE
WRITEBACK
Time
MCT04327
Figure 4-2
Sequential Instruction Pipelining
Standard Branch Instruction Processing
Instruction pipelining helps to speed sequential program processing. In the case that a
branch is taken, the instruction which has already been fetched providently is mostly not
the instruction which must be decoded next. Thus, at least one additional machine cycle
is normally required to fetch the branch target instruction. This extra machine cycle is
provided by means of an injected instruction (see Figure 4-3).
1 Machine Cycle
Injection
BRANCH
In+2
ITARGET
ITARGET+1 ITARGET+2 ITARGET+3
ITARGET ITARGET+1 ITARGET+2
ITARGET ITARGET+1
(IINJECT ITARGET
FETCH
BRANCH
In
...
...
(IINJECT
BRANCH
In
)
DECODE
EXECUTE
WRITEBACK
In
(IINJECT)
...
BRANCH
)
Time
MCT04328
Figure 4-3
Standard Branch Instruction Pipelining
User’s Manual
4-4
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
If a conditional branch is not taken, there is no deviation from the sequential program
flow, and thus no extra time is required. In this case the instruction after the branch
instruction will enter the decode stage of the pipeline at the beginning of the next
machine cycle after decode of the conditional branch instruction.
Cache Jump Instruction Processing
The C167CR incorporates a jump cache to optimize conditional jumps, which are
processed repeatedly within a loop. Whenever a jump on cache is taken, the extra time
to fetch the branch target instruction can be saved and thus the corresponding cache
jump instruction in most cases takes only one machine cycle.
This performance is achieved by the following mechanism:
Whenever a cache jump instruction passes through the decode stage of the pipeline for
the first time (and provided that the jump condition is met), the jump target instruction is
fetched as usual, causing a time delay of one machine cycle. In contrast to standard
branch instructions, however, the target instruction of a cache jump instruction (JMPA,
JMPR, JB, JBC, JNB, JNBS) is additionally stored in the cache after having been fetched.
After each repeatedly following execution of the same cache jump instruction, the jump
target instruction is not fetched from program memory but taken from the cache and
immediately injected into the decode stage of the pipeline (see Figure 4-4).
A time saving jump on cache is always taken after the second and any further occurrence
of the same cache jump instruction, unless an instruction which, has the fundamental
capability of changing the CSP register contents (JMPS, CALLS, RETS, TRAP, RETI),
or any standard interrupt has been processed during the period of time between two
following occurrences of the same cache jump instruction.
Injection of Cached
Injection
1 Machine Cycle
Target Instruction
In+2
ITARGET
ITARGET+1
ITARGET
(IINJECT
Cache Jmp
In+2
ITARGET+1
ITARGET+2
ITARGET+1
ITARGET
FETCH
Cache Jmp
Cache Jmp
(IINJECT
Cache Jmp
In
)
ITARGET
Cache Jmp
In
DECODE
EXECUTE
WRITEBACK
In
)
In
...
...
Cache Jmp
1st Loop Iteration
Repeated Loop Iteration
MCT04329
Figure 4-4
Cache Jump Instruction Pipelining
User’s Manual
4-5
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
4.2
Particular Pipeline Effects
Since up to four different instructions are processed simultaneously, additional hardware
has been spent in the C167CR to consider all causal dependencies which may exist on
instructions in different pipeline stages without a loss of performance. This extra
hardware (i.e. for ‘forwarding’ operand read and write values) resolves most of the
possible conflicts (e.g. multiple usage of buses) in a time optimized way and thus avoids
that the pipeline becomes noticeable for the user in most cases. However, there are
some very rare cases, where the circumstance that the C167CR is a pipelined machine
requires attention by the programmer. In these cases the delays caused by pipeline
conflicts can be used for other instructions in order to optimize performance.
Context Pointer Updating
An instruction, which calculates a physical GPR operand address via the CP register, is
mostly not capable of using a new CP value, which is to be updated by an immediately
preceding instruction. Thus, to make sure that the new CP value is used, at least one
instruction must be inserted between a CP-changing and a subsequent GPR-using
instruction, as shown in the following example:
I
I
I
:SCXT CP,#0FC00h ;select a new context
n
:…
;must not be an instruction using a GPR
;write to GPR 0 in the new context
n + 1
n + 2
:MOV R0,#dataX
Data Page Pointer Updating
An instruction, which calculates a physical operand address via a particular DPPn
(n = 0 to 3) register, is mostly not capable of using a new DPPn register value, which is
to be updated by an immediately preceding instruction. Thus, to make sure that the new
DPPn register value is used, at least one instruction must be inserted between a DPPn-
changing instruction and a subsequent instruction which implicitly uses DPPn via a long
or indirect addressing mode, as shown in the following example:
I
I
I
:MOV DPP0,#4
:…
;select data page 4 via DPP0
;must not be an instruction using DPP0
n
n + 1
n + 2
:MOV DPP0:0000H,R1;move contents of R1 to address location
01’0000
H
;(in data page 4) supposed segment. is
enabled
User’s Manual
4-6
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
Explicit Stack Pointer Updating
None of the RET, RETI, RETS, RETP or POP instructions is capable of correctly using
a new SP register value, which is to be updated by an immediately preceding instruction.
Thus, in order to use the new SP register value without erroneously performed stack
accesses, at least one instruction must be inserted between an explicitly SP-writing and
any subsequent of the just mentioned implicitly SP-using instructions, as shown in the
following example:
I
I
:MOV SP,#0FA40H ;select a new top of stack
n
:…
;must not be an instruction popping operands
;from the system stack
n + 1
I
:POP R0
;pop word value from new top of stack into
n + 2
R0
Note: Conflicts with instructions writing to the stack (PUSH, CALL, SCXT) are solved
internally by the CPU logic.
Controlling Interrupts
Software modifications (implicit or explicit) of the PSW are done in the execute phase of
the respective instructions. In order to maintain fast interrupt responses, however, the
current interrupt prioritization round does not consider these changes, i.e. an interrupt
request may be acknowledged after the instruction that disables interrupts via IEN or
ILVL or after the following instructions. Timecritical instruction sequences therefore
should not begin directly after the instruction disabling interrupts, as shown in the
following examples:
INTERRUPTS_OFF:
BCLR IEN
;globally disable interrupts
<Instr non-crit>
<Instr 1st-crit>
…
;non-critical instruction
;begin of uninterruptable critical sequence
<Instr last-crit>
INTERRUPTS_ON:
BSET IEN
;end of uninterruptable critical sequence
;globally re-enable interrupts
CRITICAL_SEQUENCE:
ATOMIC #3
BCLR IEN
;immediately block interrupts
;globally disable interrupts
…
;here is the uninterruptable sequence
;globally re-enable interrupts
BSET IEN
Note: The described delay of 1 instruction also applies for enabling the interrupts system
i.e. no interrupt requests are acknowledged until the instruction following the
enabling instruction.
User’s Manual
4-7
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
External Memory Access Sequences
The effect described here will only become noticeable, when watching the external
memory access sequences on the external bus (e.g. by means of a Logic Analyzer).
Different pipeline stages can simultaneously put a request on the External Bus Controller
(EBC). The sequence of instructions processed by the CPU may diverge from the
sequence of the corresponding external memory accesses performed by the EBC, due
to the predefined priority of external memory accesses:
1st
2nd
3rd
Write Data
Fetch Code
Read Data.
Initialization of Port Pins
Modifications of the direction of port pins (input or output) become effective only after the
instruction following the modifying instruction. As bit instructions (BSET, BCLR) use
internal read-modify-write sequences accessing the whole port, instructions modifying
the port direction should be followed by an instruction that does not access the same port
(see example below).
PORT_INIT_WRONG:
BSET DP3.13
BSET P3.9
;change direction of P3.13 to output
;P3.13 is still input,
;rd-mod-wr reads pin P3.13
PORT_INIT_RIGHT:
BSET DP3.13
NOP
;change direction of P3.13 to output
;any instruction not accessing port 3
;P3.13 is now output,
BSET P3.9
;rd-mod-wr reads P3.13’s output latch
Note: Special attention must be paid to interrupt service routines that modify the same
port as the software they have interrupted.
Changing the System Configuration
The instruction following an instruction that changes the system configuration via register
SYSCON (e.g. the mapping of the internal ROM, segmentation, stack size) cannot use
the new resources (e.g. ROM or stack). In these cases an instruction that does not
access these resources should be inserted. Code accesses to the new ROM area are
only possible after an absolute branch to this area.
Note: As a rule, instructions that change ROM mapping should be executed from
internal RAM or external memory.
User’s Manual
4-8
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
BUSCON/ADDRSEL
The instruction following an instruction that changes the properties of an external
address area cannot access operands within the new area. In these cases an instruction
that does not access this address area should be inserted. Code accesses to the new
address area should be made after an absolute branch to this area.
Note: As a rule, instructions that change external bus properties should not be executed
from the respective external memory area.
Timing
Instruction pipelining reduces the average instruction processing time in a wide scale
(from four to one machine cycles, mostly). However, there are some rare cases, where
a particular pipeline situation causes the processing time for a single instruction to be
extended either by a half or by one machine cycle. Although this additional time
represents only a tiny part of the total program execution time, it might be of interest to
avoid these pipeline-caused time delays in time critical program modules.
Besides a general execution time description, Section 4.3 provides some hints on how
to optimize time-critical program parts with regard to such pipeline-caused timing
particularities.
User’s Manual
4-9
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
4.3
Bit-Handling and Bit-Protection
The C167CR provides several mechanisms to manipulate bits. These mechanisms
either manipulate software flags within the internal RAM, control on-chip peripherals via
control bits in their respective SFRs or control IO functions via port pins.
The instructions BSET, BCLR, BAND, BOR, BXOR, BMOV, BMOVN explicitly set or
clear specific bits. The instructions BFLDL and BFLDH allow to manipulate up to 8 bits
of a specific byte at one time. The instructions JBC and JNBS implicitly clear or set the
specified bit when the jump is taken. The instructions JB and JNB (also conditional jump
instructions that refer to flags) evaluate the specified bit to determine if the jump is to be
taken.
Note: Bit operations on undefined bit locations will always read a bit value of ‘0’, while
the write access will not effect the respective bit location.
All instructions that manipulate single bits or bit groups internally use a read-modify-write
sequence that accesses the whole word, which contains the specified bit(s).
This method has several consequences:
• Bits can only be modified within the internal address areas, i.e. internal RAM and
SFRs. External locations cannot be used with bit instructions.
The upper 256 Bytes of the SFR area, the ESFR area and the internal RAM are bit-
addressable (see Chapter 3), i.e. those register bits located within the respective
sections can be directly manipulated using bit instructions. The other SFRs must be
accessed byte/word wise.
Note: All GPRs are bit-addressable independent of the allocation of the register bank via
the context pointer CP. Even GPRs which are allocated to not bit-addressable
RAM locations provide this feature.
• The read-modify-write approach may be critical with hardware-effected bits. In these
cases the hardware may change specific bits while the read-modify-write operation is
in progress, where the writeback would overwrite the new bit value generated by the
hardware. The solution is either the implemented hardware protection (see below) or
realized through special programming (see Chapter 4.2).
Protected bits are not changed during the read-modify-write sequence, i.e. when
hardware sets e.g. an interrupt request flag between the read and the write of the read-
modify-write sequence. The hardware protection logic guarantees that only the intended
bit(s) is/are effected by the write-back operation.
Note: If a conflict occurs between a bit manipulation generated by hardware and an
intended software access the software access has priority and determines the
final value of the respective bit.
A summary of the protected bits implemented in the C167CR can be found at the end of
Chapter 2.
User’s Manual
4-10
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
4.4
Instruction State Times
Basically, the time to execute an instruction depends on where the instruction is fetched
from, and where possible operands are read from or written to. The fastest processing
mode of the C167CR is to execute a program fetched from the internal code memory. In
that case most of the instructions can be processed within just one machine cycle, which
is also the general minimum execution time.
All external memory accesses are performed by the C167CR’s on-chip External Bus
Controller (EBC), which works in parallel with the CPU.
This section summarizes the execution times in a very condensed way. A detailed
description of the execution times for the various instructions and the specific exceptions
can be found in the “C166 Family Instruction Set Manual”.
Table 4-1 shows the minimum execution times required to process a C167CR
instruction fetched from the internal code memory, the internal RAM or from external
memory. These execution times apply to most of the C167CR instructions - except some
of the branches, the multiplication, the division and a special move instruction. In case
of internal ROM program execution there is no execution time dependency on the
instruction length except for some special branch situations. The numbers in the table
are in units of CPU clock cycles and assume no waitstates.
Table 4-1
Minimum Execution Times
Instruction Fetch
Word Operand Access
Memory Area
Word
Instruction
Doubleword
Instruction
Read from
Write to
Internal code memory
Internal RAM
2
6
2
3
4
6
2
8
2
0/1
2
–
0
2
3
4
6
16-bit Demux Bus
16-bit Mux Bus
8-bit Demux Bus
8-bit Mux Bus
4
6
3
8
4
12
6
Execution from the internal RAM provides flexibility in terms of loadable and modifyable
code on the account of execution time.
Execution from external memory strongly depends on the selected bus mode and the
programming of the bus cycles (waitstates).
User’s Manual
4-11
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The operand and instruction accesses listed below can extend the execution time of an
instruction:
• Internal code memory operand reads (same for byte and word operand reads)
• Internal RAM operand reads via indirect addressing modes
• Internal SFR operand reads immediately after writing
• External operand reads
• External operand writes
• Jumps to non-aligned double word instructions in the internal ROM space
• Testing Branch Conditions immediately after PSW writes
4.5
CPU Special Function Registers
The core CPU requires a set of Special Function Registers (SFRs) to maintain the
system state information, to supply the ALU with register-addressable constants and to
control system and bus configuration, multiply and divide ALU operations, code memory
segmentation, data memory paging, and accesses to the General Purpose Registers
and the System Stack.
The access mechanism for these SFRs in the CPU core is identical to the access
mechanism for any other SFR. Since all SFRs can simply be controlled by means of any
instruction, which is capable of addressing the SFR memory space, a lot of flexibility has
been gained, without the need to create a set of system-specific instructions.
Note, however, that there are user access restrictions for some of the CPU core SFRs
to ensure proper processor operations. The instruction pointer IP and code segment
pointer CSP cannot be accessed directly at all. They can only be changed indirectly via
branch instructions.
The PSW, SP, and MDC registers can be modified not only explicitly by the programmer,
but also implicitly by the CPU during normal instruction processing. Note that any explicit
write request (via software) to an SFR supersedes a simultaneous modification by
hardware of the same register.
Note: Any write operation to a single byte of an SFR clears the non-addressed
complementary byte within the specified SFR.
Non-implemented (reserved) SFR bits cannot be modified, and will always supply
a read value of ‘0’.
User’s Manual
4-12
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The System Configuration Register SYSCON
This bit-addressable register provides general system configuration and control
functions. The reset value for register SYSCON depends on the state of the PORT0 pins
during reset (see hardware effectable bits).
SYSCON
System Control Register
SFR (FF12 /89 )
Reset value: 0XX0
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
OW BD
RST
DIS EN
ROM SGT ROM BYT CLK WR CS
S1 DIS EN DIS EN CFG CFG
XPE VISI
STKSZ
-
D
N
BLE
rw
rw
rw rwh rwh rw rwh rw
-
rw rw
rw
rw
rw
Bit
XPER-SHARE XBUS Peripheral Share Mode Control
0: External accesses to XBUS peripherals are disabled
Function
1: XBUS peripherals are accessible via the external bus during hold
mode
VISIBLE
XPEN
Visible Mode Control
0: Accesses to XBUS peripherals are done internally
1: XBUS peripheral accesses are made visible on the external pins
XBUS Peripheral Enable Bit
0: Accesses to the on-chip X-Peripherals and their functions are disabled
1: The on-chip X-Peripherals are enabled and can be accessed
BDRSTEN
Bidirectional Reset Enable Bit
0: Pin RSTIN is an input only.
1: Pin RSTIN is pulled low during the internal reset sequence
after any reset.
OWDDIS
CSCFG
Oscillator Watchdog Disable Bit (Cleared after reset)
0: The on-chip oscillator watchdog is enabled and active.
1: The on-chip oscillator watchdog is disabled and the CPU clock is
always fed from the oscillator input.
Chip Select Configuration Control (Cleared after reset)
0: Latched CS mode. The CS signals are latched internally
and driven to the (enabled) port pins synchronously.
1: Unlatched CS mode. The CS signals are directly derived from
the address and driven to the (enabled) port pins.
WRCFG
Write Configuration Control (Set according to pin P0H.0 during reset)
0: Pins WR and BHE retain their normal function
1: Pin WR acts as WRL, pin BHE acts as WRH
User’s Manual
4-13
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
Bit
Function
CLKEN
System Clock Output Enable (CLKOUT, cleared after reset)
0: CLKOUT disabled: pin may be used for general purpose IO
1: CLKOUT enabled: pin outputs the system clock signal
BYTDIS
ROMEN
Disable/Enable Control for Pin BHE (Set according to data bus width)
0: Pin BHE enabled
1: Pin BHE disabled, pin may be used for general purpose IO
Internal ROM Enable (Set according to pin EA during reset)
0: Internal program memory disabled, accesses to the ROM area use
the external bus
1: Internal program memory enabled
SGTDIS
Segmentation Disable/Enable Control (Cleared after reset)
0: Segmentation enabled
(CSP is saved/restored during interrupt entry/exit)
1: Segmentation disabled (Only IP is saved/restored)
ROMS1
STKSZ
Internal ROM Mapping
0: Internal ROM area mapped to segment 0 (00’0000 … 00’7FFF )
1: Internal ROM area mapped to segment 1 (01’0000 … 01’7FFF )
H
H
H
H
System Stack Size
Selects the size of the system stack (in the internal RAM)
from 32 to 512 words
Note: Register SYSCON cannot be changed after execution of the EINIT instruction.
The function of bits XPER-SHARE, VISIBLE, WRCFG, BYTDIS, ROMEN and
ROMS1 is described in more detail in Chapter 9.
User’s Manual
4-14
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
System Clock Output Enable (CLKEN)
The system clock output function is enabled by setting bit CLKEN in register SYSCON
to ‘1’. If enabled, port pin P3.15 takes on its alternate function as CLKOUT output pin.
The clock output is a 50% duty cycle clock (except for direct drive operation where
CLKOUT reflects the clock input signal, and for slowdown operation where CLKOUT
mirrors the CPU clock signal) whose frequency equals the CPU operating frequency
(fOUT = fCPU).
Note: The output driver of port pin P3.15 is switched on automatically, when the
CLKOUT function is enabled. The port direction bit is disregarded.
After reset, the clock output function is disabled (CLKEN = ‘0’).
In emulation mode the CLKOUT function is enabled automatically.
Segmentation Disable/Enable Control (SGTDIS)
Bit SGTDIS allows to select either the segmented or non-segmented memory mode.
In non-segmented memory mode (SGTDIS = ‘1’) it is assumed that the code address
space is restricted to 64 KBytes (segment 0) and thus 16 bits are sufficient to represent
all code addresses. For implicit stack operations (CALL or RET) the CSP register is
totally ignored and only the IP is saved to and restored from the stack.
In segmented memory mode (SGTDIS = ‘0’) it is assumed that the whole address
space is available for instructions. For implicit stack operations (CALL or RET) the CSP
register and the IP are saved to and restored from the stack. After reset the segmented
memory mode is selected.
Note: Bit SGTDIS controls if the CSP register is pushed onto the system stack in addition
to the IP register before an interrupt service routine is entered, and it is repopped
when the interrupt service routine is left again.
System Stack Size (STKSZ)
This bitfield defines the size of the physical system stack, which is located in the internal
RAM of the C167CR. An area of 32 … 512 words or all of the internal RAM may be
dedicated to the system stack. A so-called “circular stack” mechanism allows to use a
bigger virtual stack than this dedicated RAM area.
These techniques as well as the encoding of bitfield STKSZ are described in more detail
in Chapter 21.
User’s Manual
4-15
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Processor Status Word PSW
This bit-addressable register reflects the current state of the microcontroller. Two groups
of bits represent the current ALU status, and the current CPU interrupt status. A separate
bit (USR0) within register PSW is provided as a general purpose user flag.
PSW
Program Status Word
SFR (FF10 /88 )
Reset value: 0000
H
H
H
15
14
ILVL
rwh
13
12
11
10
9
8
7
6
5
4
3
2
1
0
HLD
EN
MUL
IP
IEN
-
-
-
USR0
E
Z
V
C
N
rw
rw
-
-
-
rw rwh rwh rwh rwh rwh rwh
Bit
Function
Negative Result
N
Set, when the result of an ALU operation is negative.
C
Carry Flag
Set, when the result of an ALU operation produces a carry bit.
V
Overflow Result
Set, when the result of an ALU operation produces an overflow.
Z
Zero Flag
Set, when the result of an ALU operation is zero.
E
End of Table Flag
Set, when the source operand of an instruction is 8000 or 80 .
H
H
MULIP
Multiplication/Division In Progress
0: There is no multiplication/division in progress.
1: A multiplication/division has been interrupted.
USR0
User General Purpose Flag
May be used by the application software.
HLDEN, ILVL, Interrupt and EBC Control Fields
IEN
Define the response to interrupt requests and enable external bus
arbitration. (Described in Section 5)
ALU Status (N, C, V, Z, E, MULIP)
The condition flags (N, C, V, Z, E) within the PSW indicate the ALU status due to the last
recently performed ALU operation. They are set by most of the instructions due to
specific rules, which depend on the ALU or data movement operation performed by an
instruction.
User’s Manual
4-16
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
After execution of an instruction which explicitly updates the PSW register, the condition
flags cannot be interpreted as described in the following, because any explicit write to
the PSW register supersedes the condition flag values, which are implicitly generated by
the CPU. Explicitly reading the PSW register supplies a read value which represents the
state of the PSW register after execution of the immediately preceding instruction.
Note: After reset, all of the ALU status bits are cleared.
N-Flag: For most of the ALU operations, the N-flag is set to ‘1’, if the most significant bit
of the result contains a ‘1’, otherwise it is cleared. In the case of integer operations the
N-flag can be interpreted as the sign bit of the result (negative: N = ‘1’, positive: N = ‘0’).
Negative numbers are always represented as the 2’s complement of the corresponding
positive number. The range of signed numbers extends from ‘– 8000 ’ to ‘+ 7FFF ’ for
H
H
the word data type, or from ‘– 80 ’ to ‘+ 7F ’ for the byte data type.For Boolean bit
H
H
operations with only one operand the N-flag represents the previous state of the
specified bit. For Boolean bit operations with two operands the N-flag represents the
logical XORing of the two specified bits.
C-Flag: After an addition the C-flag indicates that a carry from the most significant bit of
the specified word or byte data type has been generated. After a subtraction or a
comparison the C-flag indicates a borrow, which represents the logical negation of a
carry for the addition.
This means that the C-flag is set to ‘1’, if no carry from the most significant bit of the
specified word or byte data type has been generated during a subtraction, which is
performed internally by the ALU as a 2’s complement addition, and the C-flag is cleared
when this complement addition caused a carry.
The C-flag is always cleared for logical, multiply and divide ALU operations, because
these operations cannot cause a carry anyhow.
For shift and rotate operations the C-flag represents the value of the bit shifted out last.
If a shift count of zero is specified, the C-flag will be cleared. The C-flag is also cleared
for a prioritize ALU operation, because a ‘1’ is never shifted out of the MSB during the
normalization of an operand.
For Boolean bit operations with only one operand the C-flag is always cleared. For
Boolean bit operations with two operands the C-flag represents the logical ANDing of the
two specified bits.
V-Flag: For addition, subtraction and 2’s complementation the V-flag is always set to ‘1’,
if the result overflows the maximum range of signed numbers, which are representable
by either 16 bits for word operations (‘– 8000 ’ to ‘+ 7FFF ’), or by 8 bits for byte
H
H
operations (‘– 80 ’ to ‘+ 7F ’), otherwise the V-flag is cleared. Note that the result of an
H
H
integer addition, integer subtraction, or 2’s complement is not valid, if the V-flag indicates
an arithmetic overflow.
For multiplication and division the V-flag is set to ‘1’, if the result cannot be represented
in a word data type, otherwise it is cleared. Note that a division by zero will always cause
an overflow. In contrast to the result of a division, the result of a multiplication is valid
User’s Manual
4-17
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
regardless of whether the V-flag is set to ‘1’ or not.
Since logical ALU operations cannot produce an invalid result, the V-flag is cleared by
these operations.
The V-flag is also used as ‘Sticky Bit’ for rotate right and shift right operations. With only
using the C-flag, a rounding error caused by a shift right operation can be estimated up
to a quantity of one half of the LSB of the result. In conjunction with the V-flag, the C-flag
allows evaluating the rounding error with a finer resolution (see Table 4-2).
For Boolean bit operations with only one operand the V-flag is always cleared. For
Boolean bit operations with two operands the V-flag represents the logical ORing of the
two specified bits.
Table 4-2
C-flag
Shift Right Rounding Error Evaluation
V-flag
Rounding Error Quantity
0
0
1
1
0
1
0
1
–
0 <
No rounding error
–
1
Rounding error
Rounding error
Rounding error
< / LSB
= / LSB
> / LSB
2
1
2
1
2
Z-Flag: The Z-flag is normally set to ‘1’, if the result of an ALU operation equals zero,
otherwise it is cleared.
For the addition and subtraction with carry the Z-flag is only set to ‘1’, if the Z-flag already
contains a ‘1’ and the result of the current ALU operation additionally equals zero. This
mechanism is provided for the support of multiple precision calculations.
For Boolean bit operations with only one operand the Z-flag represents the logical
negation of the previous state of the specified bit. For Boolean bit operations with two
operands the Z-flag represents the logical NORing of the two specified bits. For the
prioritize ALU operation the Z-flag indicates, if the second operand was zero or not.
E-Flag: The E-flag can be altered by instructions, which perform ALU or data movement
operations. The E-flag is cleared by those instructions which cannot be reasonably used
for table search operations. In all other cases the E-flag is set depending on the value of
the source operand to signify whether the end of a search table is reached or not. If the
value of the source operand of an instruction equals the lowest negative number, which
is representable by the data format of the corresponding instruction (‘8000 ’ for the word
H
data type, or ‘80 ’ for the byte data type), the E-flag is set to ‘1’, otherwise it is cleared.
H
MULIP-Flag: The MULIP-flag will be set to ‘1’ by hardware upon the entrance into an
interrupt service routine, when a multiply or divide ALU operation was interrupted before
completion. Depending on the state of the MULIP bit, the hardware decides whether a
multiplication or division must be continued or not after the end of an interrupt service.
The MULIP bit is overwritten with the contents of the stacked MULIP-flag when the
return-from-interrupt-instruction (RETI) is executed. This normally means that the
MULIP-flag is cleared again after that.
User’s Manual
4-18
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
Note: The MULIP flag is a part of the task environment! When the interrupting service
routine does not return to the interrupted multiply/divide instruction (i.e. in case of
a task scheduler that switches between independent tasks), the MULIP flag must
be saved as part of the task environment and must be updated accordingly for the
new task before this task is entered.
CPU Interrupt Status (IEN, ILVL)
The Interrupt Enable bit allows to globally enable (IEN = ‘1’) or disable (IEN = ‘0’)
interrupts. The four-bit Interrupt Level field (ILVL) specifies the priority of the current CPU
activity. The interrupt level is updated by hardware upon entry into an interrupt service
routine, but it can also be modified via software to prevent other interrupts from being
acknowledged. In case an interrupt level ‘15’ has been assigned to the CPU, it has the
highest possible priority, and thus the current CPU operation cannot be interrupted
except by hardware traps or external non-maskable interrupts. For details please refer
to Chapter 5.
After reset all interrupts are globally disabled, and the lowest priority (ILVL = 0) is
assigned to the initial CPU activity.
The Instruction Pointer IP
This register determines the 16-bit intra-segment address of the currently fetched
instruction within the code segment selected by the CSP register. The IP register is not
mapped into the C167CR’s address space, and thus it is not directly accessable by the
programmer. The IP can, however, be modified indirectly via the stack by means of a
return instruction.
The IP register is implicitly updated by the CPU for branch instructions and after
instruction fetch operations.
IP
Instruction Pointer
- - - (- - - -/- -)
Reset value: 0000
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ip
(r)(w)h
Bit
ip
Function
Specifies the intra segment offset, from where the current instruction is
to be fetched. IP refers to the current segment <SEGNR>.
User’s Manual
4-19
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Code Segment Pointer CSP
This non-bit addressable register selects the code segment being used at run-time to
access instructions. The lower 8 bits of register CSP select one of up to 256 segments
of 64 KBytes each, while the upper 8 bits are reserved for future use.
CSP
Code Segment Pointer
SFR (FE08 /04 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
-
-
-
-
-
-
SEGNR
-
-
-
-
-
-
-
-
r(w)h
Bit
SEGNR
Function
Segment Number
Specifies the code segment, from where the current instruction is to be
fetched. SEGNR is ignored, when segmentation is disabled.
Code memory addresses are generated by directly extending the 16-bit contents of the
IP register by the contents of the CSP register as shown in Figure 4-5.
In case of the segmented memory mode the selected number of segment address bits
(via bitfield SALSEL) of register CSP is output on the respective segment address pins
of Port 4 for all external code accesses. For non-segmented memory mode or Single
Chip Mode the content of this register is not significant, because all code acccesses are
automatically restricted to segment 0.
Note: The CSP register can only be read but not written by data operations. It is,
however, modified either directly by means of the JMPS and CALLS instructions,
or indirectly via the stack by means of the RETS and RETI instructions.
Upon the acceptance of an interrupt or the execution of a software TRAP
instruction, the CSP register is automatically set to zero.
User’s Manual
4-20
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
CSP Register
IP Register
Code Segment
15
0
15
0
FF’FFFF
H
255
254
FE’0000
H
1
0
01’0000
00’0000
H
H
24/20/18-Bit Physical Code Address
MCA02265
Figure 4-5
Addressing via the Code Segment Pointer
Note: When segmentation is disabled, the IP value is used directly as the 16-bit address.
User’s Manual
4-21
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Data Page Pointers DPP0, DPP1, DPP2, DPP3
These four non-bit addressable registers select up to four different data pages being
active simultaneously at run-time. The lower 10 bits of each DPP register select one of
the 1024 possible 16-KByte data pages while the upper 6 bits are reserved for future use.
The DPP registers allow to access the entire memory space in pages of 16 KBytes each.
DPP0
Data Page Pointer 0
SFR (FE00 /00 )
Reset value: 0000
H
H
H
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
-
-
-
-
DPP0PN
-
-
-
-
-
-
rw
DPP1
Data Page Pointer 1
SFR (FE02 /01 )
Reset value: 0001
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
-
-
-
-
DPP1PN
-
-
-
-
-
-
rw
DPP2
Data Page Pointer 2
SFR (FE04 /02 )
Reset value: 0002
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
-
-
-
-
DPP2PN
-
-
-
-
-
-
rw
DPP3
Data Page Pointer 3
SFR (FE06 /03 )
Reset value: 0003
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
-
-
-
-
DPP3PN
-
-
-
-
-
-
rw
User’s Manual
4-22
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
Bit
Function
DPPxPN
Data Page Number of DPPx
Specifies the data page selected via DPPx. Only the least significant two
bits of DPPx are significant, when segmentation is disabled.
The DPP registers are implicitly used, whenever data accesses to any memory location
are made via indirect or direct long 16-bit addressing modes (except for override
accesses via EXTended instructions and PEC data transfers). After reset, the Data Page
Pointers are initialized in a way that all indirect or direct long 16-bit addresses result in
identical 18-bit addresses. This allows to access data pages 3 … 0 within segment 0 as
shown in Figure 4-6. If the user does not want to use any data paging, no further action
is required.
Data paging is performed by concatenating the lower 14 bits of an indirect or direct long
16-bit address with the contents of the DPP register selected by the upper two bits of the
16-bit address. The contents of the selected DPP register specify one of the
1024 possible data pages. This data page base address together with the 14-bit page
offset forms the physical 24-bit address (selectable part is driven to the address pins).
In case of non-segmented memory mode, only the two least significant bits of the
implicitly selected DPP register are used to generate the physical address. Thus,
extreme care should be taken when changing the content of a DPP register, if a non-
segmented memory model is selected, because otherwise unexpected results could
occur.
In case of the segmented memory mode the selected number of segment address bits
(via bitfield SALSEL) of the respective DPP register is output on the respective segment
address pins of Port 4 for all external data accesses.
A DPP register can be updated via any instruction, which is capable of modifying an SFR.
Note: Due to the internal instruction pipeline, a new DPP value is not yet usable for the
operand address calculation of the instruction immediately following the
instruction updating the DPP register.
User’s Manual
4-23
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
16-Bit Data Address
15 14
0
1023
1022
1021
DPP Registers
3
2
1
0
DPP3-11
DPP2-10
DPP1-01
DPP0-00
14-Bit
Intra-Page Address
(concatenated with
content of DPPx).
Affer reset or with segmentation disabled the DPP registers select data pages 3...0.
All of the internal memory is accessible in these cases.
MCA02264
Figure 4-6
Addressing via the Data Page Pointers
User’s Manual
4-24
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Context Pointer CP
This non-bit addressable register is used to select the current register context. This
means that the CP register value determines the address of the first General Purpose
Register (GPR) within the current register bank of up to 16 wordwide and/or bytewide
GPRs.
CP
Context Pointer
SFR (FE10 /08 )
Reset value: FC00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1
1
1
1
cp
0
r
r
r
r
rw
r
Bit
cp
Function
Modifiable portion of register CP
Specifies the (word) base address of the current register bank.
When writing a value to register CP with bits CP.11 … CP.9 = ‘000’, bits
CP.11 … CP.10 are set to ‘11’ by hardware, in all other cases all bits of
bit field “cp” receive the written value.
Note: It is the user’s responsibility that the physical GPR address specified via CP
register plus short GPR address must always be an internal RAM location. If this
condition is not met, unexpected results may occur.
• Do not set CP below the IRAM start address, i.e. 00’FA00 /00’F600 /00’F200
H
H
H
(referring to an IRAM size of 1/2/3 KByte)
• Do not set CP above 00’FDFE
H
• Be careful using the upper GPRs with CP above 00’FDE0
H
The CP register can be updated via any instruction which is capable of modifying an SFR.
Note: Due to the internal instruction pipeline, a new CP value is not yet usable for GPR
address calculations of the instruction immediately following the instruction
updating the CP register.
The Switch Context instruction (SCXT) allows to save the content of register CP on the
stack and updating it with a new value in just one machine cycle.
User’s Manual
4-25
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
Internal RAM
R15
R14
R13
R12
R11
R10
R9
(CP) + 30
(CP) + 28
Context
Pointer
.
.
.
R8
R7
R6
R5
R4
R3
R2
R1
(CP) + 2
(CP)
R0
MCD02003
Figure 4-7
Register Bank Selection via Register CP
Several addressing modes use register CP implicitly for address calculations. The
addressing modes mentioned below are described in Chapter 23.
Short 4-Bit GPR Addresses (mnemonic: Rw or Rb) specify an address relative to the
memory location specified by the contents of the CP register, i.e. the base of the current
register bank.
Depending on whether a relative word (Rw) or byte (Rb) GPR address is specified, the
short 4-bit GPR address is either multiplied by two or not before it is added to the
content of register CP (see Figure 4-8). Thus, both byte and word GPR accesses are
possible in this way.
GPRs used as indirect address pointers are always accessed wordwise. For some
instructions only the first four GPRs can be used as indirect address pointers. These
GPRs are specified via short 2-bit GPR addresses. The respective physical address
calculation is identical to that for the short 4-bit GPR addresses.
Short 8-Bit Register Addresses (mnemonic: reg or bitoff) within a range from F0 to
H
FF interpret the four least significant bits as short 4-bit GPR address, while the four
H
most significant bits are ignored. The respective physical GPR address calculation is
identical to that for the short 4-bit GPR addresses. For single bit accesses on a GPR, the
GPR’s word address is calculated as just described, but the position of the bit within the
word is specified by a separate additional 4-bit value.
User’s Manual
4-26
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
Specified by reg or bitoff
Context
Pointer
4-Bit GPR
1111
Address
Internal
RAM
2
*
Control
+
Must be
within the
internal
GPRs
RAM area
For byte GPR
accesses
For word GPR
accesses
MCD02005
Figure 4-8
Implicit CP Use by Short GPR Addressing Modes
The Stack Pointer SP
This non-bit addressable register is used to point to the top of the internal system stack
(TOS). The SP register is pre-decremented whenever data is to be pushed onto the
stack, and it is post-incremented whenever data is to be popped from the stack. Thus,
the system stack grows from higher toward lower memory locations.
Since the least significant bit of register SP is tied to ‘0’ and bits 15 through 12 are tied
to ‘1’ by hardware, the SP register can only contain values from F000 to FFFE . This
H
H
allows to access a physical stack within the internal RAM of the C167CR. A virtual stack
(usually bigger) can be realized via software. This mechanism is supported by registers
STKOV and STKUN (see respective descriptions below).
The SP register can be updated via any instruction, which is capable of modifying an SFR.
Note: Due to the internal instruction pipeline, a POP or RETURN instruction must not
immediately follow an instruction updating the SP register.
SP
Stack Pointer Register
SFR (FE12 /09 )
Reset value: FC00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1
1
1
1
sp
0
r
r
r
r
rwh
r
Bit
sp
Function
Modifiable portion of register SP
Specifies the top of the internal system stack.
User’s Manual
4-27
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Stack Overflow Pointer STKOV
This non-bit addressable register is compared against the SP register after each
operation, which pushes data onto the system stack (e.g. PUSH and CALL instructions
or interrupts) and after each subtraction from the SP register. If the content of the SP
register is less than the content of the STKOV register, a stack overflow hardware trap
will occur.
Since the least significant bit of register STKOV is tied to ‘0’ and bits 15 through 12 are
tied to ‘1’ by hardware, the STKOV register can only contain values from F000 to
H
FFFE .
H
STKOV
Stack Overflow Reg.
SFR (FE14 /0A )
Reset value:FA00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1
1
1
1
stkov
0
r
r
r
r
rw
Bit
stkov
Function
Modifiable portion of register STKOV
Specifies the lower limit of the internal system stack.
The Stack Overflow Trap (entered when (SP) < (STKOV)) may be used in two different
ways:
• Fatal error indication treats the stack overflow as a system error through the
associated trap service routine. Under these circumstances data in the bottom of the
stack may have been overwritten by the status information stacked upon servicing the
stack overflow trap.
• Automatic system stack flushing allows to use the system stack as a ‘Stack Cache’
for a bigger external user stack. In this case register STKOV should be initialized to a
value, which represents the desired lowest Top of Stack address plus 12 according to
the selected maximum stack size. This considers the worst case that will occur, when
a stack overflow condition is detected just during entry into an interrupt service routine.
Then, six additional stack word locations are required to push IP, PSW, and CSP for
both the interrupt service routine and the hardware trap service routine.
More details about the stack overflow trap service routine and virtual stack management
are given in Chapter 21.
User’s Manual
4-28
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Stack Underflow Pointer STKUN
This non-bit addressable register is compared against the SP register after each
operation, which pops data from the system stack (e.g. POP and RET instructions) and
after each addition to the SP register. If the content of the SP register is greater than the
content of the STKUN register, a stack underflow hardware trap will occur.
Since the least significant bit of register STKUN is tied to ‘0’ and bits 15 through 12 are
tied to ‘1’ by hardware, the STKUN register can only contain values from F000 to
H
FFFE .
H
STKUN
Stack Underflow Reg.
SFR (FE16 /0B )
Reset value: FC00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1
1
1
1
stkun
0
r
r
r
r
rw
r
Bit
stkun
Function
Modifiable portion of register STKUN
Specifies the upper limit of the internal system stack.
The Stack Underflow Trap (entered when (SP) > (STKUN)) may be used in two different
ways:
• Fatal error indication treats the stack underflow as a system error through the
associated trap service routine.
• Automatic system stack refilling allows to use the system stack as a ‘Stack Cache’
for a bigger external user stack. In this case register STKUN should be initialized to a
value, which represents the desired highest Bottom of Stack address.
More details about the stack underflow trap service routine and virtual stack
management are given in Chapter 21.
Scope of Stack Limit Control
The stack limit control realized by the register pair STKOV and STKUN detects cases
where the stack pointer SP is moved outside the defined stack area either by ADD or
SUB instructions or by PUSH or POP operations (explicit or implicit, i.e. CALL or RET
instructions).
This control mechanism is not triggered, i.e. no stack trap is generated, when
• the stack pointer SP is directly updated via MOV instructions
• the limits of the stack area (STKOV, STKUN) are changed, so that SP is outside of the
new limits.
User’s Manual
4-29
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Multiply/Divide High Register MDH
This register is a part of the 32-bit multiply/divide register, which is implicitly used by the
CPU, when it performs a multiplication or a division. After a multiplication, this non-bit
addressable register represents the high order 16 bits of the 32-bit result. For long
divisions, the MDH register must be loaded with the high order 16 bits of the 32-bit
dividend before the division is started. After any division, register MDH represents the
16-bit remainder.
MDH
Multiply/Divide High Reg.
SFR (FE0C /06 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
mdh
rwh
Bit
mdh
Function
Specifies the high order 16 bits of the 32-bit multiply and divide reg. MD.
Whenever this register is updated via software, the Multiply/Divide Register In Use
(MDRIU) flag in the Multiply/Divide Control register (MDC) is set to ‘1’.
When a multiplication or division is interrupted before its completion and when a new
multiply or divide operation is to be performed within the interrupt service routine, register
MDH must be saved along with registers MDL and MDC to avoid erroneous results.
A detailed description of how to use the MDH register for programming multiply and
divide algorithms can be found in Chapter 21.
User’s Manual
4-30
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Multiply/Divide Low Register MDL
This register is a part of the 32-bit multiply/divide register, which is implicitly used by the
CPU, when it performs a multiplication or a division. After a multiplication, this non-bit
addressable register represents the low order 16 bits of the 32-bit result. For long
divisions, the MDL register must be loaded with the low order 16 bits of the 32-bit
dividend before the division is started. After any division, register MDL represents the 16-
bit quotient.
MDL
Multiply/Divide Low Reg.
SFR (FE0E /07 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
mdl
rwh
Bit
mdl
Function
Specifies the low order 16 bits of the 32-bit multiply and divide reg. MD.
Whenever this register is updated via software, the Multiply/Divide Register In Use
(MDRIU) flag in the Multiply/Divide Control register (MDC) is set to ‘1’. The MDRIU flag
is cleared, whenever the MDL register is read via software.
When a multiplication or division is interrupted before its completion and when a new
multiply or divide operation is to be performed within the interrupt service routine, register
MDL must be saved along with registers MDH and MDC to avoid erroneous results.
A detailed description of how to use the MDL register for programming multiply and
divide algorithms can be found in Chapter 21.
User’s Manual
4-31
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Multiply/Divide Control Register MDC
This bit addressable 16-bit register is implicitly used by the CPU, when it performs a
multiplication or a division. It is used to store the required control information for the
corresponding multiply or divide operation. Register MDC is updated by hardware during
each single cycle of a multiply or divide instruction.
MDC
Multiply/Divide Control Reg.
SFR (FF0E /87 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
MDR
IU
-
-
-
-
-
-
-
-
!!
!!
!!
!!
!!
!!
!!
-
-
-
-
-
-
-
-
r(w)h r(w)h r(w)h r(w)h r(w)h r(w)h r(w)h r(w)h
Bit
Function
Multiply/Divide Register In Use
MDRIU
0:
1:
Cleared, when register MDL is read via software.
Set when register MDL or MDH is written via software, or when
a multiply or divide instruction is executed.
!!
Internal Machine Status
The multiply/divide unit uses these bits to control internal operations.
Never modify these bits without saving and restoring register MDC.
When a division or multiplication was interrupted before its completion and the multiply/
divide unit is required, the MDC register must first be saved along with registers MDH
and MDL (to be able to restart the interrupted operation later), and then it must be
cleared prepare it for the new calculation. After completion of the new division or
multiplication, the state of the interrupted multiply or divide operation must be restored.
The MDRIU flag is the only portion of the MDC register which might be of interest for the
user. The remaining portions of the MDC register are reserved for dedicated use by the
hardware, and should never be modified by the user in another way than described
above. Otherwise, a correct continuation of an interrupted multiply or divide operation
cannot be guaranteed.
A detailed description of how to use the MDC register for programming multiply and
divide algorithms can be found in Chapter 21.
User’s Manual
4-32
V3.1, 2000-03
C167CR
Derivatives
The Central Processing Unit (CPU)
The Constant Zeros Register ZEROS
All bits of this bit-addressable register are fixed to ‘0’ by hardware. This register can be
read only. Register ZEROS can be used as a register-addressable constant of all zeros,
i.e. for bit manipulation or mask generation. It can be accessed via any instruction, which
is capable of addressing an SFR.
ZEROS
Zeros Register
SFR (FF1C /8E )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
The Constant Ones Register ONES
All bits of this bit-addressable register are fixed to ‘1’ by hardware. This register can be
read only. Register ONES can be used as a register-addressable constant of all ones,
i.e. for bit manipulation or mask generation. It can be accessed via any instruction, which
is capable of addressing an SFR.
ONES
Ones Register
SFR (FF1E /8F )
Reset Value: FFFF
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
User’s Manual
4-33
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5
Interrupt and Trap Functions
The architecture of the C167CR supports several mechanisms for fast and flexible
response to service requests that can be generated from various sources internal or
external to the microcontroller.
These mechanisms include:
Normal Interrupt Processing
The CPU temporarily suspends the current program execution and branches to an
interrupt service routine in order to service an interrupt requesting device. The current
program status (IP, PSW, in segmentation mode also CSP) is saved on the internal
system stack. A prioritization scheme with 16 priority levels allows the user to specify the
order in which multiple interrupt requests are to be handled.
Interrupt Processing via the Peripheral Event Controller (PEC)
A faster alternative to normal software controlled interrupt processing is servicing an
interrupt requesting device with the C167CR’s integrated Peripheral Event Controller
(PEC). Triggered by an interrupt request, the PEC performs a single word or byte data
transfer between any two locations in segment 0 (data pages 0 through 3) through one
of eight programmable PEC Service Channels. During a PEC transfer the normal
program execution of the CPU is halted for just 1 instruction cycle. No internal program
status information needs to be saved. The same prioritization scheme is used for PEC
service as for normal interrupt processing. PEC transfers share the 2 highest priority
levels.
Trap Functions
Trap functions are activated in response to special conditions that occur during the
execution of instructions. A trap can also be caused externally by the Non-Maskable
Interrupt pin NMI. Several hardware trap functions are provided for handling erroneous
conditions and exceptions that arise during the execution of an instruction. Hardware
traps always have highest priority and cause immediate system reaction. The software
trap function is invoked by the TRAP instruction, which generates a software interrupt for
a specified interrupt vector. For all types of traps the current program status is saved on
the system stack.
External Interrupt Processing
Although the C167CR does not provide dedicated interrupt pins, it allows to connect
external interrupt sources and provides several mechanisms to react on external events,
including standard inputs, non-maskable interrupts and fast external interrupts. These
interrupt functions are alternate port functions, except for the non-maskable interrupt and
the reset input.
User’s Manual
5-1
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5.1
Interrupt System Structure
The C167CR provides 56 separate interrupt nodes that may be assigned to 16 priority
levels. In order to support modular and consistent software design techniques, most
sources of an interrupt or PEC request are supplied with a separate interrupt control
register and interrupt vector. The control register contains the interrupt request flag, the
interrupt enable bit, and the interrupt priority of the associated source. Each source
request is then activated by one specific event, depending on the selected operating
mode of the respective device. For efficient usage of the resources also multi-source
interrupt nodes are incorporated. These nodes can be activated by several source
requests, e.g. as different kinds of errors in the serial interfaces. However, specific status
flags which identify the type of error are implemented in the serial channels’ control
registers.
The C167CR provides a vectored interrupt system. In this system specific vector
locations in the memory space are reserved for the reset, trap, and interrupt service
functions. Whenever a request occurs, the CPU branches to the location that is
associated with the respective interrupt source. This allows direct identification of the
source that caused the request. The only exceptions are the class B hardware traps,
which all share the same interrupt vector. The status flags in the Trap Flag Register
(TFR) can then be used to determine which exception caused the trap. For the special
software TRAP instruction, the vector address is specified by the operand field of the
instruction, which is a seven bit trap number.
The reserved vector locations build a jump table in the low end of the C167CR’s address
space (segment 0). The jump table is made up of the appropriate jump instructions that
transfer control to the interrupt or trap service routines, which may be located anywhere
within the address space. The entries of the jump table are located at the lowest
addresses in code segment 0 of the address space. Each entry occupies 2 words,
except for the reset vector and the hardware trap vectors, which occupy 4 or 8 words.
Table 5-1 lists all sources that are capable of requesting interrupt or PEC service in the
C167CR, the associated interrupt vectors, their locations and the associated trap
numbers. It also lists the mnemonics of the affected Interrupt Request flags and their
corresponding Interrupt Enable flags. The mnemonics are composed of a part that
specifies the respective source, followed by a part that specifies their function
(IR = Interrupt Request flag, IE = Interrupt Enable flag).
Note: Each entry of the interrupt vector table provides room for two word instructions or
one doubleword instruction. The respective vector location results from multiplying
the trap number by 4 (4 Bytes per entry).
All interrupt nodes that are currently not used by their associated modules or are
not connected to a module in the actual derivative may be used to generate
software controlled interrupt requests by setting the respective IR flag.
User’s Manual
5-2
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Table 5-1
C167CR Interrupt Notes and Vectors
Source of Interrupt or Request
Enable
Flag
Interrupt Vector
Trap
PEC Service Request
CAPCOM Register 0
CAPCOM Register 1
CAPCOM Register 2
CAPCOM Register 3
CAPCOM Register 4
CAPCOM Register 5
CAPCOM Register 6
CAPCOM Register 7
CAPCOM Register 8
CAPCOM Register 9
CAPCOM Register 10
CAPCOM Register 11
CAPCOM Register 12
CAPCOM Register 13
CAPCOM Register 14
CAPCOM Register 15
CAPCOM Register 16
CAPCOM Register 17
CAPCOM Register 18
CAPCOM Register 19
CAPCOM Register 20
CAPCOM Register 21
CAPCOM Register 22
CAPCOM Register 23
CAPCOM Register 24
CAPCOM Register 25
CAPCOM Register 26
CAPCOM Register 27
CAPCOM Register 28
CAPCOM Register 29
Flag
Vector
Location Number
CC0IR
CC0IE
CC0INT
CC1INT
CC2INT
CC3INT
CC4INT
CC5INT
CC6INT
CC7INT
CC8INT
CC9INT
00’0040
00’0044
00’0048
10 /16
H
H
H
H
D
D
D
D
D
D
D
D
D
D
CC1IR
CC1IE
11 /17
H
CC2IR
CC2IE
12 /18
H
CC3IR
CC3IE
00’004C
13 /19
H
H
CC4IR
CC4IE
00’0050
00’0054
00’0058
14 /20
H
H
CC5IR
CC5IE
15 /21
H
H
H
CC6IR
CC6IE
16 /22
H
CC7IR
CC7IE
00’005C
17 /23
H
H
CC8IR
CC8IE
00’0060
00’0064
18 /24
H
H
CC9IR
CC9IE
19 /25
H
H
H
CC10IR
CC11IR
CC12IR
CC13IR
CC14IR
CC15IR
CC16IR
CC17IR
CC18IR
CC19IR
CC20IR
CC21IR
CC22IR
CC23IR
CC24IR
CC25IR
CC26IR
CC27IR
CC28IR
CC29IR
CC10IE
CC11IE
CC12IE
CC13IE
CC14IE
CC15IE
CC16IE
CC17IE
CC18IE
CC19IE
CC20IE
CC21IE
CC22IE
CC23IE
CC24IE
CC25IE
CC26IE
CC27IE
CC28IE
CC29IE
CC10INT 00’0068
1A /26
H
D
D
CC11INT 00’006C
1B /27
H
H
CC12INT 00’0070
CC13INT 00’0074
CC14INT 00’0078
1C /28
H
H
D
D
D
D
D
D
D
D
D
D
D
D
D
D
1D /29
H
H
H
1E /30
H
CC15INT 00’007C
CC16INT 00’00C0
CC17INT 00’00C4
CC18INT 00’00C8
1F /31
H
H
H
H
H
30 /48
H
31 /49
H
32 /50
H
CC19INT 00’00CC
33 /51
H
H
CC20INT 00’00D0
CC21INT 00’00D4
CC22INT 00’00D8
34 /52
H
H
35 /53
H
H
H
36 /54
H
CC23INT 00’00DC
37 /55
H
H
CC24INT 00’00E0
CC25INT 00’00E4
CC26INT 00’00E8
38 /56
H
H
39 /57
H
H
H
3A /58
H
D
D
CC27INT 00’00EC
3B /59
H
H
CC28INT 00’00F0
3C /60
H D
H
CC29INT 00’0110
44 /68
H D
H
User’s Manual
5-3
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Table 5-1
C167CR Interrupt Notes and Vectors (cont’d)
Source of Interrupt or Request
Enable
Flag
Interrupt Vector
Vector Location Number
Trap
PEC Service Request
CAPCOM Register 30
CAPCOM Register 31
CAPCOM Timer 0
CAPCOM Timer 1
CAPCOM Timer 7
CAPCOM Timer 8
GPT1 Timer 2
Flag
CC30IR
CC31IR
T0IR
CC30IE
CC31IE
T0IE
CC30INT 00’0114
45 /69
H
H
H
H
H
D
D
D
D
CC31INT 00’0118
46 /70
H
T0INT
T1INT
T7INT
T8INT
T2INT
T3INT
T4INT
T5INT
T6INT
CRINT
ADCINT
ADEINT
S0TINT
00’0080
00’0084
20 /32
H
T1IR
T1IE
21 /33
H
T7IR
T7IE
00’00F4
00’00F8
3D /61
H
H
H
H
D
D
D
D
D
D
D
D
D
D
T8IR
T8IE
3E /62
H
T2IR
T2IE
00’0088
22 /34
H
GPT1 Timer 3
T3IR
T3IE
00’008C
23 /35
H
H
GPT1 Timer 4
T4IR
T4IE
00’0090
00’0094
00’0098
24 /36
H
H
GPT2 Timer 5
T5IR
T5IE
25 /37
H
H
H
GPT2 Timer 6
T6IR
T6IE
26 /38
H
GPT2 CAPREL Register CRIR
A/D Conversion Complete ADCIR
CRIE
00’009C
27 /39
H
H
H
H
H
ADCIE
ADEIE
S0TIE
S0TBIE
S0RIE
S0EIE
SCTIE
SCRIE
SCEIE
PWMIE
XP0IE
XP1IE
XP2IE
XP3IE
00’00A0
00’00A4
00’00A8
28 /40
H
A/D Overrun Error
ASC0 Transmit
ASC0 Transmit Buffer
ASC0 Receive
ASC0 Error
ADEIR
S0TIR
S0TBIR
S0RIR
S0EIR
SCTIR
SCRIR
SCEIR
PWMIR
XP0IR
XP1IR
XP2IR
XP3IR
29 /41
H
2A /42
H
D
D
S0TBINT 00’011C
47 /71
H
H
S0RINT
S0EINT
SCTINT
SCRINT
SCEINT
00’00AC
2B /43
H D
H
00’00B0
00’00B4
00’00B8
2C /44
H
H
D
D
D
D
D
D
D
D
D
SSC Transmit
SSC Receive
2D /45
H
H
H
2E /46
H
SSC Error
00’00BC
2F /47
H
H
H
PWM Channel 0 … 3
CAN1
PWMINT 00’00FC
3F /63
H
XP0INT
XP1INT
XP2INT
XP3INT
00’0100
00’0104
00’0108
40 /64
H
H
Unassigned node
Unassigned node
PLL/OWD
41 /65
H
H
H
42 /66
H
00’010C
43 /67
H
H
User’s Manual
5-4
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Table 5-2 lists the vector locations for hardware traps and the corresponding status flags
in register TFR. It also lists the priorities of trap service for cases, where more than one
trap condition might be detected within the same instruction. After any reset (hardware
reset, software reset instruction SRST, or reset by watchdog timer overflow) program
execution starts at the reset vector at location 00’0000 . Reset conditions have priority
H
over every other system activity and therefore have the highest priority (trap priority III).
Software traps may be initiated to any vector location between 00’0000 and 00’01FC .
H
H
A service routine entered via a software TRAP instruction is always executed on the
current CPU priority level which is indicated in bit field ILVL in register PSW. This means
that routines entered via the software TRAP instruction can be interrupted by all
hardware traps or higher level interrupt requests.
Table 5-2
Hardware Trap Summary
Exception Condition
Trap
Flag
Trap
Vector
Vector
Trap
Trap
Location Number Prio
Reset Functions
Hardware Reset
Software Reset
RESET
RESET
RESET
00’0000
00’0000
00’0000
00
00
00
III
III
III
H
H
H
H
H
H
Watchdog Timer Overflow
Class A Hardware Traps
Non-Maskable Interrupt
Stack Overflow
NMI
STKOF
STKUF
NMITRAP 00’0008
STOTRAP 00’0010
STUTRAP 00’0018
02
04
06
II
II
II
H
H
H
H
H
H
Stack Underflow
Class B Hardware Traps
Undefined Opcode
Protected Instruction Fault
Illegal Word Operand Access ILLOPA
Illegal Instruction Access
Illegal External Bus Access
UNDOPC BTRAP
PRTFLT BTRAP
00’0028
00’0028
00’0028
00’0028
00’0028
0A
0A
0A
0A
0A
I
I
I
I
I
H
H
H
H
H
H
H
H
H
H
BTRAP
BTRAP
BTRAP
ILLINA
ILLBUS
Reserved
[2C -3C ] [0B -0F ]
H H H H
Software Traps
Any
Any
Current
TRAP Instruction
[00’0000 - [00 -7F ] CPU
H H H
00’01FC ]
Priority
H
in steps
of 4
H
User’s Manual
5-5
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Normal Interrupt Processing and PEC Service
During each instruction cycle one out of all sources which require PEC or interrupt
processing is selected according to its interrupt priority. This priority of interrupts and
PEC requests is programmable in two levels. Each requesting source can be assigned
to a specific priority. A second level (called “group priority”) allows to specify an internal
order for simultaneous requests from a group of different sources on the same priority
level. At the end of each instruction cycle the one source request with the highest current
priority will be determined by the interrupt system. This request will then be serviced, if
its priority is higher than the current CPU priority in register PSW.
Interrupt System Register Description
Interrupt processing is controlled globally by register PSW through a general interrupt
enable bit (IEN) and the CPU priority field (ILVL). Additionally the different interrupt
sources are controlled individually by their specific interrupt control registers (… IC).
Thus, the acceptance of requests by the CPU is determined by both the individual
interrupt control registers and the PSW. PEC services are controlled by the respective
PECCx register and the source and destination pointers, which specify the task of the
respective PEC service channel.
5.1.1
Interrupt Control Registers
All interrupt control registers are organized identically. The lower 8 bits of an interrupt
control register contain the complete interrupt status information of the associated
source, which is required during one round of prioritization, the upper 8 bits of the
respective register are reserved. All interrupt control registers are bit-addressable and all
bits can be read or written via software. This allows each interrupt source to be
programmed or modified with just one instruction. When accessing interrupt control
registers through instructions which operate on word data types, their upper 8 bits
(15 … 8) will return zeros, when read, and will discard written data.
The layout of the Interrupt Control registers shown below applies to each xxIC register,
where xx stands for the mnemonic for the respective source.
User’s Manual
5-6
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
xxIC
SFR (yyyy /zz )
Reset value: - - 00
H
H
H
15
14
-
13
-
12
-
11
-
10
-
9
8
7
6
5
4
3
2
1
0
xxIR xxIE
ILVL
GLVL
rw
-
-
-
rwh rw
Bit
Function
GLVL
Group Level
Defines the internal order for simultaneous requests of the same priority.
3: Highest group priority
0: Lowest group priority
ILVL
Interrupt Priority Level
Defines the priority level for the arbitration of requests.
F : Highest priority level
H
0 : Lowest priority level
H
xxIE
xxIR
Interrupt Enable Control Bit (individually enables/disables a specific source)
‘0’: Interrupt request is disabled
‘1’: Interrupt Request is enabled
Interrupt Request Flag
‘0’: No request pending
‘1’: This source has raised an interrupt request
The Interrupt Request Flag is set by hardware whenever a service request from the
respective source occurs. It is cleared automatically upon entry into the interrupt service
routine or upon a PEC service. In the case of PEC service the Interrupt Request flag
remains set, if the COUNT field in register PECCx of the selected PEC channel
decrements to zero. This allows a normal CPU interrupt to respond to a completed PEC
block transfer.
Note: Modifying the Interrupt Request flag via software causes the same effects as if it
had been set or cleared by hardware.
User’s Manual
5-7
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Interrupt Priority Level and Group Level
The four bits of bit field ILVL specify the priority level of a service request for the
arbitration of simultaneous requests. The priority increases with the numerical value of
ILVL, so 0000 is the lowest and 1111 is the highest priority level.
B
B
When more than one interrupt request on a specific level gets active at the same time,
the values in the respective bit fields GLVL are used for second level arbitration to select
one request for being serviced. Again the group priority increases with the numerical
value of GLVL, so 00 is the lowest and 11 is the highest group priority.
B
B
Note: All interrupt request sources that are enabled and programmed to the same
priority level must always be programmed to different group priorities. Otherwise
an incorrect interrupt vector will be generated.
Upon entry into the interrupt service routine, the priority level of the source that won the
arbitration and who’s priority level is higher than the current CPU level, is copied into bit
field ILVL of register PSW after pushing the old PSW contents on the stack.
The interrupt system of the C167CR allows nesting of up to 15 interrupt service routines
of different priority levels (level 0 cannot be arbitrated).
Interrupt requests that are programmed to priority levels 15 or 14 (i.e. ILVL = 111X ) will
B
be serviced by the PEC, unless the COUNT field of the associated PECC register
contains zero. In this case the request will instead be serviced by normal interrupt
processing. Interrupt requests that are programmed to priority levels 13 through 1 will
always be serviced by normal interrupt processing.
Note: Priority level 0000 is the default level of the CPU. Therefore a request on level 0
B
will never be serviced, because it can never interrupt the CPU. However, an
enabled interrupt request on level 0000 will terminate the C167CR’s Idle mode
B
and reactivate the CPU.
For interrupt requests which are to be serviced by the PEC, the associated PEC channel
number is derived from the respective ILVL (LSB) and GLVL (see Figure 5-1). So
programming a source to priority level 15 (ILVL = 1111 ) selects the PEC channel group
B
7 … 4, programming a source to priority level 14 (ILVL = 1110 ) selects the PEC channel
B
group 3 … 0. The actual PEC channel number is then determined by the group priority
field GLVL.
Simultaneous requests for PEC channels are prioritized according to the PEC channel
number, where channel 0 has lowest and channel 8 has highest priority.
Note: All sources that request PEC service must be programmed to different PEC
channels. Otherwise an incorrect PEC channel may be activated.
User’s Manual
5-8
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Interrupt
Control Register
ILVL
GLVL
PEC Control
PEC Channel #
MCA04330
Figure 5-1
Priority Levels and PEC Channels
Table 5-3 shows in a few examples, which action is executed with a given programming
of an interrupt control register.
Table 5-3
Priority Level
ILVL GLVL COUNT = 00H
Interrupt Priority Examples
Type of Service
COUNT ≠ 00
H
1 1 1 1 1 1
1 1 1 1 1 0
1 1 1 0 1 0
1 1 0 1 1 0
0 0 0 1 1 1
0 0 0 1 0 0
0 0 0 0 X X
CPU interrupt,
level 15, group priority 3
PEC service,
channel 7
CPU interrupt,
level 15, group priority 2
PEC service,
channel 6
CPU interrupt,
level 14, group priority 2
PEC service,
channel 2
CPU interrupt,
level 13, group priority 2
CPU interrupt,
level 13, group priority 2
CPU interrupt,
level 1, group priority 3
CPU interrupt,
level 1, group priority 3
CPU interrupt,
level 1, group priority 0
CPU interrupt,
level 1, group priority 0
No service!
No service!
Note: All requests on levels 13 … 1 cannot initiate PEC transfers. They are always
serviced by an interrupt service routine. No PECC register is associated and no
COUNT field is checked.
User’s Manual
5-9
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Interrupt Control Functions in the PSW
The Processor Status Word (PSW) is functionally divided into 2 parts: the lower byte of
the PSW basically represents the arithmetic status of the CPU, the upper byte of the
PSW controls the interrupt system of the C167CR and the arbitration mechanism for the
external bus interface.
Note: Pipeline effects have to be considered when enabling/disabling interrupt requests
via modifications of register PSW (see Chapter 4).
PSW
Processor Status Word
SFR (FF10 /88 )
Reset value: 0000
H
H
H
15
14
ILVL
rw
13
12
11
10
9
8
7
6
5
4
3
2
1
0
HLD
EN
USR MUL
0
IEN
-
-
-
-
-
-
E
Z
V
C
N
IP
rwh rwh rwh rwh rwh rwh
rw
rw
rw
Bit
Function
N, C, V, Z, E, CPU status flags (Described in Section 4)
MULIP, USR0 Define the current status of the CPU (ALU, multiplication unit).
HLDEN
HOLD Enable (Enables External Bus Arbitration)
0:
Bus arbitration disabled,
P6.7 … P6.5 may be used for general purpose IO
Bus arbitration enabled,
1:
P6.7 … P6.5 serve as BREQ, HLDA, HOLD, resp.
IEN
Interrupt Enable Control Bit (globally enables/disables interrupt
requests)
0:
1:
Interrupt requests are disabled
Interrupt requests are enabled
ILVL
CPU Priority Level
Defines the current priority level for the CPU
F : Highest priority level
H
0 : Lowest priority level
H
User’s Manual
5-10
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
CPU priority ILVL defines the current level for the operation of the CPU. This bit field
reflects the priority level of the routine that is currently executed. Upon the entry into an
interrupt service routine this bit field is updated with the priority level of the request that
is being serviced. The PSW is saved on the system stack before. The CPU level
determines the minimum interrupt priority level that will be serviced. Any request on the
same or a lower level will not be acknowledged.
The current CPU priority level may be adjusted via software to control which interrupt
request sources will be acknowledged.
PEC transfers do not really interrupt the CPU, but rather “steal” a single cycle, so PEC
services do not influence the ILVL field in the PSW.
Hardware traps switch the CPU level to maximum priority (i.e. 15) so no interrupt or PEC
requests will be acknowledged while an exception trap service routine is executed.
Note: The TRAP instruction does not change the CPU level, so software invoked trap
service routines may be interrupted by higher requests.
Interrupt Enable bit IEN globally enables or disables PEC operation and the
acceptance of interrupts by the CPU. When IEN is cleared, no new interrupt requests are
accepted by the CPU. Requests that already have entered the pipeline at that time will
process, however. When IEN is set to ‘1’, all interrupt sources, which have been
individually enabled by the interrupt enable bits in their associated control registers, are
globally enabled.
Note: Traps are non-maskable and are therefore not affected by the IEN bit.
User’s Manual
5-11
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5.2
Operation of the PEC Channels
The C167CR’s Peripheral Event Controller (PEC) provides 8 PEC service channels,
which move a single byte or word between two locations in segment 0 (data pages
3 … 0). This is the fastest possible interrupt response and in many cases is sufficient to
service the respective peripheral request (e.g. serial channels, etc.). Each channel is
controlled by a dedicated PEC Channel Counter/Control register (PECCx) and a pair of
pointers for source (SRCPx) and destination (DSTPx) of the data transfer.
The PECC registers control the action that is performed by the respective PEC channel.
PECCx
PEC Control Reg.
SFR (FECy /6z , see Table 5-4)
Reset value: 0000
H
H
H
15
14
13
12
11
-
10
9
8
7
6
5
4
3
2
1
0
INC
BWT
COUNT
-
-
-
-
rw
rw
rw
Bit
Function
PEC Transfer Count
COUNT
Counts PEC transfers and influences the channel’s action (see Table 5-5)
BWT
Byte/Word Transfer Selection
0:
1:
Transfer a Word
Transfer a Byte
INC
Increment Control (Modification of SRCPx or DSTPx)
0 0: Pointers are not modified
0 1: Increment DSTPx by 1 or 2 (BWT)
1 0: Increment SRCPx by 1 or 2 (BWT)
1 1: Reserved. Do not use this combination.
(changed to ‘10’ by hardware)
Table 5-4
Register
PECC0
PECC1
PECC2
PECC3
PEC Control Register Addresses
Address Reg. Space Register
FEC0 /60
Address
FEC8 /64
H
Reg. Space
SFR
SFR
SFR
SFR
SFR
PECC4
PECC5
PECC6
PECC7
H
H
H
H
H
H
FEC2 /61
FECA /65
SFR
H
H
H
H
H
FEC4 /62
FECC /66
SFR
H
H
FEC6 /63
FECE /67
SFR
H
H
User’s Manual
5-12
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Byte/Word Transfer bit BWT controls, if a byte or a word is moved during a PEC service
cycle. This selection controls the transferred data size and the increment step for the
modified pointer.
Increment Control field INC controls, if one of the PEC pointers is incremented after
the PEC transfer. It is not possible to increment both pointers, however. If the pointers
are not modified (INC = ‘00’), the respective channel will always move data from the
same source to the same destination.
Note: The reserved combination ‘11’ is changed to ‘10’ by hardware. However, it is not
recommended to use this combination.
The PEC Transfer Count Field COUNT controls the action of a respective PEC channel,
where the content of bit field COUNT at the time the request is activated selects the
action. COUNT may allow a specified number of PEC transfers, unlimited transfers or no
PEC service at all.
Table 5-5 summarizes, how the COUNT field itself, the interrupt requests flag IR and the
PEC channel action depends on the previous content of COUNT.
Table 5-5
Influence of Bitfield COUNT
Previous
COUNT
Modified
COUNT
IRafterPEC Action of PEC Channel
Service
and Comments
FF
FF
‘0’
Move a Byte/Word
H
H
Continuous transfer mode, i.e. COUNT is not
modified
FE … 02 FD … 01 ‘0’
Move a Byte/Word and decrement COUNT
H
H
H
H
01
00
‘1’
Move a Byte/Word
H
H
Leave request flag set, which triggers another
request
00
00
(‘1’)
No action!
H
H
Activate interrupt service routine rather than
PEC channel.
The PEC transfer counter allows to service a specified number of requests by the
respective PEC channel, and then (when COUNT reaches 00 ) activate the interrupt
H
service routine, which is associated with the priority level. After each PEC transfer the
COUNT field is decremented and the request flag is cleared to indicate that the request
has been serviced.
User’s Manual
5-13
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Continuous transfers are selected by the value FF in bit field COUNT. In this case
H
COUNT is not modified and the respective PEC channel services any request until it is
disabled again.
When COUNT is decremented from 01 to 00 after a transfer, the request flag is not
H
H
cleared, which generates another request from the same source. When COUNT already
contains the value 00 , the respective PEC channel remains idle and the associated
H
interrupt service routine is activated instead. This allows to choose, if a level 15 or 14
request is to be serviced by the PEC or by the interrupt service routine.
Note: PEC transfers are only executed, if their priority level is higher than the CPU level,
i.e. only PEC channels 7 … 4 are processed, while the CPU executes on level 14.
All interrupt request sources that are enabled and programmed for PEC service
should use different channels. Otherwise only one transfer will be performed for
all simultaneous requests. When COUNT is decremented to 00 , and the CPU is
H
to be interrupted, an incorrect interrupt vector will be generated.
The source and destination pointers specifiy the locations between which the data is
to be moved. A pair of pointers (SRCPx and DSTPx) is associated with each of the 8
PEC channels. These pointers do not reside in specific SFRs, but are mapped into the
internal RAM of the C167CR just below the bit-addressable area (see Figure 5-2).
DSTP7
SRCP7
DSTP6
SRCP6
DSTP5
SRCP5
DSTP4
SRCP4
DSTP3
SRCP3
DSTP2
SRCP2
DSTP1
SRCP1
DSTP0
SRCP0
00’FCFE
00’FCFC
00’FCFA
00’FCF8
00’FCF6
00’FCF4
00’FCF2
00’FCF0
00’FCEE
00’FCEC
00’FCEA
00’FCE8
00’FCE6
00’FCE4
00’FCE2
00’FCE0
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
MCA04331
Figure 5-2
Mapping of PEC Pointers into the Internal RAM
User’s Manual
5-14
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
PEC data transfers do not use the data page pointers DPP3 … DPP0. The PEC source
and destination pointers are used as 16-bit intra-segment addresses within segment 0,
so data can be transferred between any two locations within the first four data pages
3 … 0.
The pointer locations for inactive PEC channels may be used for general data storage.
Only the required pointers occupy RAM locations.
Note: If word data transfer is selected for a specific PEC channel (i.e. BWT = ‘0’), the
respective source and destination pointers must both contain a valid word address
which points to an even byte boundary. Otherwise the Illegal Word Access trap will
be invoked, when this channel is used.
User’s Manual
5-15
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5.3
Prioritization of Interrupt and PEC Service Requests
Interrupt and PEC service requests from all sources can be enabled, so they are
arbitrated and serviced (if they win), or they may be disabled, so their requests are
disregarded and not serviced.
Enabling and disabling interrupt requests may be done via three mechanisms:
Control bits allow to switch each individual source “ON” or “OFF”, so it may generate a
request or not. The control bits (xxIE) are located in the respective interrupt control
registers. All interrupt requests may be enabled or disabled generally via bit IEN in
register PSW. This control bit is the “main switch” that selects, if requests from any
source are accepted or not.
For a specific request to be arbitrated the respective source’s enable bit and the global
enable bit must both be set.
The priority level automatically selects a certain group of interrupt requests that will be
acknowledged, disclosing all other requests. The priority level of the source that won the
arbitration is compared against the CPU’s current level and the source is only serviced,
if its level is higher than the current CPU level. Changing the CPU level to a specific value
via software blocks all requests on the same or a lower level. An interrupt source that is
assigned to level 0 will be disabled and never be serviced.
The ATOMIC and EXTend instructions automatically disable all interrupt requests for
the duration of the following 1 … 4 instructions. This is useful e.g. for semaphore
handling and does not require to re-enable the interrupt system after the unseparable
instruction sequence (see Chapter 21).
Interrupt Class Management
An interrupt class covers a set of interrupt sources with the same importance, i.e. the
same priority from the system’s viewpoint. Interrupts of the same class must not interrupt
each other. The C167CR supports this function with two features:
Classes with up to 4 members can be established by using the same interrupt priority
(ILVL) and assigning a dedicated group level (GLVL) to each member. This functionality
is built-in and handled automatically by the interrupt controller.
Classes with more than 4 members can be established by using a number of adjacent
interrupt priorities (ILVL) and the respective group levels (4 per ILVL). Each interrupt
service routine within this class sets the CPU level to the highest interrupt priority within
the class. All requests from the same or any lower level are blocked now, i.e. no request
of this class will be accepted.
User’s Manual
5-16
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
The example below establishes 3 interrupt classes which cover 2 or 3 interrupt priorities,
depending on the number of members in a class. A level 6 interrupt disables all other
sources in class 2 by changing the current CPU level to 8, which is the highest priority
(ILVL) in class 2. Class 1 requests or PEC requests are still serviced in this case.
The 24 interrupt sources (excluding PEC requests) are so assigned to 3 classes of
priority rather than to 7 different levels, as the hardware support would do.
Table 5-6
Software Controlled Interrupt Classes (Example)
ILVL
(Priority)
GLVL
1
Interpretation
3
2
0
15
14
13
12
11
10
9
PEC service on up to 8 channels
X
X
X
X
X
Interrupt Class 1
5 sources on 2 levels
8
X
X
X
X
X
X
X
X
X
X
X
Interrupt Class 2
9 sources on 3 levels
7
6
5
X
X
X
Interrupt Class 3
5 sources on 2 levels
4
3
2
1
0
No service!
User’s Manual
5-17
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5.4
Saving the Status During Interrupt Service
Before an interrupt request that has been arbitrated is actually serviced, the status of the
current task is automatically saved on the system stack. The CPU status (PSW) is saved
along with the location, where the execution of the interrupted task is to be resumed after
returning from the service routine. This return location is specified through the Instruction
Pointer (IP) and, in case of a segmented memory model, the Code Segment Pointer
(CSP). Bit SGTDIS in register SYSCON controls, how the return location is stored.
The system stack receives the PSW first, followed by the IP (unsegmented) or followed
by CSP and then IP (segmented mode). This optimizes the usage of the system stack,
if segmentation is disabled.
The CPU priority field (ILVL in PSW) is updated with the priority of the interrupt request
that is to be serviced, so the CPU now executes on the new level. If a multiplication or
division was in progress at the time the interrupt request was acknowledged, bit MULIP
in register PSW is set to ‘1’. In this case the return location that is saved on the stack is
not the next instruction in the instruction flow, but rather the multiply or divide instruction
itself, as this instruction has been interrupted and will be completed after returning from
the service routine.
High
Addresses
Status of
Interrupted
Task
SP
--
--
--
PSW
IP
PSW
CSP
IP
SP
--
SP
Low
Addresses
a) System Stack before
Interrupt Entry
b) System Stack after
Interrupt Entry
b) System Stack after
Interrupt Entry
(Unsegmented)
(Segmented)
MCD02226
Figure 5-3
Task Status saved on the System Stack
The interrupt request flag of the source that is being serviced is cleared. The IP is loaded
with the vector associated with the requesting source (the CSP is cleared in case of
segmentation) and the first instruction of the service routine is fetched from the
respective vector location, which is expected to branch to the service routine itself. The
data page pointers and the context pointer are not affected.
When the interrupt service routine is left (RETI is executed), the status information is
popped from the system stack in the reverse order, taking into account the value of bit
SGTDIS.
User’s Manual
5-18
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Context Switching
An interrupt service routine usually saves all the registers it uses on the stack, and
restores them before returning. The more registers a routine uses, the more time is
wasted with saving and restoring. The C167CR allows to switch the complete bank of
CPU registers (GPRs) with a single instruction, so the service routine executes within its
own, separate context.
The instruction “SCXT CP, #New_Bank” pushes the content of the context pointer (CP)
on the system stack and loads CP with the immediate value “New_Bank”, which selects
a new register bank. The service routine may now use its “own registers”. This register
bank is preserved, when the service routine terminates, i.e. its contents are available on
the next call.
Before returning (RETI) the previous CP is simply POPped from the system stack, which
returns the registers to the original bank.
Note: The first instruction following the SCXT instruction must not use a GPR.
Resources that are used by the interrupting program must eventually be saved and
restored, e.g. the DPPs and the registers of the MUL/DIV unit.
User’s Manual
5-19
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5.5
Interrupt Response Times
The interrupt response time defines the time from an interrupt request flag of an enabled
interrupt source being set until the first instruction (I1) being fetched from the interrupt
vector location. The basic interrupt response time for the C167CR is 3 instruction cycles.
Pipeline Stage
FETCH
Cycle 1
N
Cycle 2
N + 1
N
Cycle 3
N + 2
TRAP (1)
N
Cycle 4
I1
DECODE
N - 1
N - 2
N - 3
TRAP (2)
TRAP
N
EXECUTE
WRITEBACK
N - 1
N - 2
N - 1
Interrupt Response Time
1
IR-Flag
0
MCT04332
Figure 5-4
Pipeline Diagram for Interrupt Response Time
All instructions in the pipeline including instruction N (during which the interrupt request
flag is set) are completed before entering the service routine. The actual execution time
for these instructions (e.g. waitstates) therefore influences the interrupt response time.
In Figure 5-4 the respective interrupt request flag is set in cycle 1 (fetching of instruction
N). The indicated source wins the prioritization round (during cycle 2). In cycle 3 a TRAP
instruction is injected into the decode stage of the pipeline, replacing instruction N + 1
and clearing the source’s interrupt request flag to ‘0’. Cycle 4 completes the injected
TRAP instruction (save PSW, IP and CSP, if segmented mode) and fetches the first
instruction (I1) from the respective vector location.
All instructions that entered the pipeline after setting of the interrupt request flag (N + 1,
N + 2) will be executed after returning from the interrupt service routine.
The minimum interrupt response time is 5 states (10 TCL). This requires program
execution from the internal code memory, no external operand read requests and setting
the interrupt request flag during the last state of an instruction cycle. When the interrupt
request flag is set during the first state of an instruction cycle, the minimum interrupt
response time under these conditions is 6 state times (12 TCL).
The interrupt response time is increased by all delays of the instructions in the pipeline
that are executed before entering the service routine (including N).
User’s Manual
5-20
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
• When internal hold conditions between instruction pairs N – 2/N – 1 or N – 1/N occur,
or instruction N explicitly writes to the PSW or the SP, the minimum interrupt response
time may be extended by 1 state time for each of these conditions.
• When instruction N reads an operand from the internal code memory, or when N is a
call, return, trap, or MOV Rn, [Rm+ #data16] instruction, the minimum interrupt
response time may additionally be extended by 2 state times during internal code
memory program execution.
• In case instruction N reads the PSW and instruction N – 1 has an effect on the
condition flags, the interrupt response time may additionally be extended by 2 state
times.
The worst case interrupt response time during internal code memory program execution
adds to 12 state times (24 TCL).
Any reference to external locations increases the interrupt response time due to pipeline
related access priorities. The following conditions have to be considered:
• Instruction fetch from an external location
• Operand read from an external location
• Result write-back to an external location
Depending on where the instructions, source and destination operands are located,
there are a number of combinations. Note, however, that only access conflicts contribute
to the delay.
A few examples illustrate these delays:
• The worst case interrupt response time including external accesses will occur, when
instructions N, N + 1 and N + 2 are executed out of external memory, instructions N – 1
and N require external operand read accesses, instructions N – 3 through N write
back external operands, and the interrupt vector also points to an external location. In
this case the interrupt response time is the time to perform 9 word bus accesses,
because instruction I1 cannot be fetched via the external bus until all write, fetch and
read requests of preceding instructions in the pipeline are terminated.
• When the above example has the interrupt vector pointing into the internal code
memory, the interrupt response time is 7 word bus accesses plus 2 states, because
fetching of instruction I1 from internal code memory can start earlier.
• When instructions N, N + 1 and N + 2 are executed out of external memory and the
interrupt vector also points to an external location, but all operands for instructions N – 3
through N are in internal memory, then the interrupt response time is the time to
perform 3 word bus accesses.
• When the above example has the interrupt vector pointing into the internal code
memory, the interrupt response time is 1 word bus access plus 4 states.
User’s Manual
5-21
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
After an interrupt service routine has been terminated by executing the RETI instruction,
and if further interrupts are pending, the next interrupt service routine will not be entered
until at least two instruction cycles have been executed of the program that was
interrupted. In most cases two instructions will be executed during this time. Only one
instruction will typically be executed, if the first instruction following the RETI instruction
is a branch instruction (without cache hit), or if it reads an operand from internal code
memory, or if it is executed out of the internal RAM.
Note: A bus access in this context includes all delays which can occur during an external
bus cycle.
User’s Manual
5-22
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5.6
PEC Response Times
The PEC response time defines the time from an interrupt request flag of an enabled
interrupt source being set until the PEC data transfer being started. The basic PEC
response time for the C167CR is 2 instruction cycles.
Pipeline Stage
FETCH
Cycle 1
N
Cycle 2
N + 1
N
Cycle 3
N + 2
PEC
N
Cycle 4
N + 2
N + 1
PEC
N
DECODE
N - 1
N - 2
N - 3
EXECUTE
WRITEBACK
N - 1
N - 2
N - 1
PEC Response Time
1
IR-Flag
0
MCT04333
Figure 5-5
Pipeline Diagram for PEC Response Time
In Figure 5-5 above the respective interrupt request flag is set in cycle 1 (fetching of
instruction N). The indicated source wins the prioritization round (during cycle 2). In
cycle 3 a PEC transfer “instruction” is injected into the decode stage of the pipeline,
suspending instruction N + 1 and clearing the source’s interrupt request flag to ‘0’.
Cycle 4 completes the injected PEC transfer and resumes the execution of instruction
N + 1.
All instructions that entered the pipeline after setting of the interrupt request flag (N + 1,
N + 2) will be executed after the PEC data transfer.
Note: When instruction N reads any of the PEC control registers PECC7 … PECC0,
while a PEC request wins the current round of prioritization, this round is repeated
and the PEC data transfer is started one cycle later.
The minimum PEC response time is 3 states (6 TCL). This requires program execution
from the internal code memory, no external operand read requests and setting the
interrupt request flag during the last state of an instruction cycle. When the interrupt
request flag is set during the first state of an instruction cycle, the minimum PEC
response time under these conditions is 4 state times (8 TCL).
The PEC response time is increased by all delays of the instructions in the pipeline that
are executed before starting the data transfer (including N).
User’s Manual
5-23
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
• When internal hold conditions between instruction pairs N – 2/N– 1 or N – 1/N occur,
the minimum PEC response time may be extended by 1 state time for each of these
conditions.
• When instruction N reads an operand from the internal code memory, or when N is a
call, return, trap, or MOV Rn, [Rm+ #data16] instruction, the minimum PEC response
time may additionally be extended by 2 state times during internal code memory
program execution.
• In case instruction N reads the PSW and instruction N – 1 has an effect on the
condition flags, the PEC response time may additionally be extended by 2 state times.
The worst case PEC response time during internal code memory program execution
adds to 9 state times (18 TCL).
Any reference to external locations increases the PEC response time due to pipeline
related access priorities. The following conditions have to be considered:
• Instruction fetch from an external location
• Operand read from an external location
• Result write-back to an external location
Depending on where the instructions, source and destination operands are located,
there are a number of combinations. Note, however, that only access conflicts contribute
to the delay.
A few examples illustrate these delays:
• The worst case interrupt response time including external accesses will occur, when
instructions N and N + 1 are executed out of external memory, instructions N – 1 and
N require external operand read accesses and instructions N – 3, N – 2 and N – 1
write back external operands. In this case the PEC response time is the time to
perform 7 word bus accesses.
• When instructions N and N + 1 are executed out of external memory, but all operands
for instructions N – 3 through N – 1 are in internal memory, then the PEC response
time is the time to perform 1 word bus access plus 2 state times.
Once a request for PEC service has been acknowledged by the CPU, the execution of
the next instruction is delayed by 2 state times plus the additional time it might take to
fetch the source operand from internal code memory or external memory and to write the
destination operand over the external bus in an external program environment.
Note: A bus access in this context includes all delays which can occur during an external
bus cycle.
User’s Manual
5-24
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5.7
External Interrupts
Although the C167CR has no dedicated INTR input pins, it provides many possibilities
to react on external asynchronous events by using a number of IO lines for interrupt
input. The interrupt function may either be combined with the pin’s main function or may
be used instead of it, i.e. if the main pin function is not required.
Interrupt signals may be connected to:
• CC31IO … CC16IO, the capture input/compare output lines of the CAPCOM2 unit,
• CC15IO … CC0IO, the capture input/compare output lines of the CAPCOM1 unit,
• T4IN, T2IN, the timer input pins,
• CAPIN, the capture input of GPT2
For each of these pins either a positive, a negative, or both a positive and a negative
external transition can be selected to cause an interrupt or PEC service request. The
edge selection is performed in the control register of the peripheral device associated
with the respective port pin. The peripheral must be programmed to a specific operating
mode to allow generation of an interrupt by the external signal. The priority of the
interrupt request is determined by the interrupt control register of the respective
peripheral interrupt source, and the interrupt vector of this source will be used to service
the external interrupt request.
Note: In order to use any of the listed pins as external interrupt input, it must be switched
to input mode via its direction control bit DPx.y in the respective port direction
control register DPx.
Table 5-7
Port Pin
Pins to be Used as External Interrupt Inputs
Original Function
Control Register
CC31-CC28
CC27-CC24
CC23-CC16
CC15-CC0
T2CON
P7.7-4/CC31-28IO
CAPCOM register 31-28 capture input
P1H.7-4/CC27-24IO CAPCOM register 27-24 capture input
P8.7-0/CC23-16IO
P2.15-0/CC15-0IO
P3.7/T2IN
CAPCOM register 23-16 capture input
CAPCOM register 15-0 capture input
Auxiliary timer T2 input pin
P3.5/T4IN
Auxiliary timer T4 input pin
T4CON
P3.2/CAPIN
GPT2 capture input pin
T5CON
When port pins CCxIO are to be used as external interrupt input pins, bit field CCMODx
in the control register of the corresponding capture/compare register CCx must select
capture mode. When CCMODx is programmed to 001 , the interrupt request flag CCxIR
B
in register CCxIC will be set on a positive external transition at pin CCxIO. When
CCMODx is programmed to 010 , a negative external transition will set the interrupt
B
request flag. When CCMODx = 011 , both a positive and a negative transition will set
B
User’s Manual
5-25
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
the request flag. In all three cases, the contents of the allocated CAPCOM timer will be
latched into capture register CCx, independent whether the timer is running or not. When
the interrupt enable bit CCxIE is set, a PEC request or an interrupt request for vector
CCxINT will be generated.
Pins T2IN or T4IN can be used as external interrupt input pins when the associated
auxiliary timer T2 or T4 in block GPT1 is configured for capture mode. This mode is
selected by programming the mode control fields T2M or T4M in control registers
T2CON or T4CON to 101 . The active edge of the external input signal is determined by
B
bit fields T2I or T4I. When these fields are programmed to X01 , interrupt request flags
B
T2IR or T4IR in registers T2IC or T4IC will be set on a positive external transition at pins
T2IN or T4IN, respectively. When T2I or T4I are programmed to X10 , then a negative
B
external transition will set the corresponding request flag. When T2I or T4I are
programmed to X11 , both a positive and a negative transition will set the request flag.
B
In all three cases, the contents of the core timer T3 will be captured into the auxiliary
timer registers T2 or T4 based on the transition at pins T2IN or T4IN. When the interrupt
enable bits T2IE or T4IE are set, a PEC request or an interrupt request for vector T2INT
or T4INT will be generated.
Pin CAPIN differs slightly from the timer input pins as it can be used as external interrupt
input pin without affecting peripheral functions. When the capture mode enable bit T5SC
in register T5CON is cleared to ‘0’, signal transitions on pin CAPIN will only set the
interrupt request flag CRIR in register CRIC, and the capture function of register
CAPREL is not activated.
So register CAPREL can still be used as reload register for GPT2 timer T5, while pin
CAPIN serves as external interrupt input. Bit field CI in register T5CON selects the
effective transition of the external interrupt input signal. When CI is programmed to 01 ,
B
a positive external transition will set the interrupt request flag. CI = 10 selects a negative
B
transition to set the interrupt request flag, and with CI = 11 , both a positive and a
B
negative transition will set the request flag. When the interrupt enable bit CRIE is set, an
interrupt request for vector CRINT or a PEC request will be generated.
Note: The non-maskable interrupt input pin NMI and the reset input RSTIN provide
another possibility for the CPU to react on an external input signal. NMIand RSTIN
are dedicated input pins, which cause hardware traps.
User’s Manual
5-26
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Fast External Interrupts
The input pins that may be used for external interrupts are sampled every 16 TCL, i.e.
external events are scanned and detected in timeframes of 16 TCL. The C167CR
provides 8 interrupt inputs that are sampled every 2 TCL, so external events are
captured faster than with standard interrupt inputs.
The upper 8 pins of Port 2 (P2.15-P2.8) can individually be programmed to this fast
interrupt mode, where also the trigger transition (rising, falling or both) can be selected.
The External Interrupt Control register EXICON controls this feature for all 8 pins.
EXICON
Ext. Interrupt Control Reg.
ESFR (F1C0 /E0 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
EXI7ES
EXI6ES
EXI5ES
EXI4ES
EXI3ES
EXI2ES
EXI1ES
EXI0ES
rw
rw
rw
rw
rw
rw
rw
rw
Bit
EXIxES
Function
External Interrupt x Edge Selection Field (x = 7 … 0)
0 0: Fast external interrupts disabled: standard mode
0 1: Interrupt on positive edge (rising)
1 0: Interrupt on negative edge (falling)
1 1: Interrupt on any edge (rising or falling)
Note: The fast external interrupt inputs are sampled every 2 TCL. The interrupt request
arbitration and processing, however, is executed every 8 TCL.
These fast external interrupts use the interrupt nodes and vectors of the CAPCOM
channels CC8-CC15, so the capture/compare function cannot be used on the respective
Port 2 pins (with EXIxES ≠ 00 ). However, general purpose IO is possible in all cases.
B
User’s Manual
5-27
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
5.8
Trap Functions
Traps interrupt the current execution similar to standard interrupts. However, trap
functions offer the possibility to bypass the interrupt system’s prioritization process in
cases where immediate system reaction is required. Trap functions are not maskable
and always have priority over interrupt requests on any priority level.
The C167CR provides two different kinds of trapping mechanisms. Hardware traps are
triggered by events that occur during program execution (e.g. illegal access or undefined
opcode), software traps are initiated via an instruction within the current execution flow.
Software Traps
The TRAP instruction is used to cause a software call to an interrupt service routine. The
trap number that is specified in the operand field of the trap instruction determines which
vector location in the address range from 00’0000 through 00’01FC will be branched to.
H
H
Executing a TRAP instruction causes a similar effect as if an interrupt at the same vector
had occurred. PSW, CSP (in segmentation mode), and IP are pushed on the internal
system stack and a jump is taken to the specified vector location. When segmentation is
enabled and a trap is executed, the CSP for the trap service routine is set to code
segment 0. No Interrupt Request flags are affected by the TRAP instruction. The
interrupt service routine called by a TRAP instruction must be terminated with a RETI
(return from interrupt) instruction to ensure correct operation.
Note: The CPU level in register PSW is not modified by the TRAP instruction, so the
service routine is executed on the same priority level from which it was invoked.
Therefore, the service routine entered by the TRAP instruction can be interrupted
by other traps or higher priority interrupts, other than when triggered by a
hardware trap.
Hardware Traps
Hardware traps are issued by faults or specific system states that occur during runtime
of a program (not identified at assembly time). A hardware trap may also be triggered
intentionally, e.g. to emulate additional instructions by generating an Illegal Opcode trap.
The C167CR distinguishes eight different hardware trap functions. When a hardware
trap condition has been detected, the CPU branches to the trap vector location for the
respective trap condition. Depending on the trap condition, the instruction which caused
the trap is either completed or cancelled (i.e. it has no effect on the system state) before
the trap handling routine is entered.
Hardware traps are non-maskable and always have priority over every other CPU activity.
If several hardware trap conditions are detected within the same instruction cycle, the
highest priority trap is serviced (see Table 5-2).
User’s Manual
5-28
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
PSW, CSP (in segmentation mode), and IP are pushed on the internal system stack and
the CPU level in register PSW is set to the highest possible priority level (i.e. level 15),
disabling all interrupts. The CSP is set to code segment zero, if segmentation is enabled.
A trap service routine must be terminated with the RETI instruction.
The eight hardware trap functions of the C167CR are divided into two classes:
Class A traps are
• external Non-Maskable Interrupt (NMI)
• Stack Overflow
• Stack Underflow Trap
These traps share the same trap priority, but have an individual vector address.
Class B traps are
• Undefined Opcode
• Protection Fault
• Illegal Word Operand Access
• Illegal Instruction Access
• Illegal External Bus Access Trap
These traps share the same trap priority, and the same vector address.
The bit-addressable Trap Flag Register (TFR) allows a trap service routine to identify the
kind of trap which caused the exception. Each trap function is indicated by a separate
request flag. When a hardware trap occurs, the corresponding request flag in register
TFR is set to ‘1’.
The reset functions (hardware, software, watchdog) may be regarded as a type of trap.
Reset functions have the highest system priority (trap priority III).
Class A traps have the second highest priority (trap priority II), on the 3rd rank are class
B traps, so a class A trap can interrupt a class B trap. If more than one class A trap occur
at a time, they are prioritized internally, with the NMI trap on the highest and the stack
underflow trap on the lowest priority.
All class B traps have the same trap priority (trap priority I). When several class B traps
get active at a time, the corresponding flags in the TFR register are set and the trap
service routine is entered. Since all class B traps have the same vector, the priority of
service of simultaneously occurring class B traps is determined by software in the trap
service routine.
A class A trap occurring during the execution of a class B trap service routine will be
serviced immediately. During the execution of a class A trap service routine, however,
any class B trap occurring will not be serviced until the class A trap service routine is
exited with a RETI instruction. In this case, the occurrence of the class B trap condition
is stored in the TFR register, but the IP value of the instruction which caused this trap is
lost.
User’s Manual
5-29
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
TFR
Trap Flag Register
SFR (FFAC /D6 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
STK STK
OF UF
UND
OPC
PRT ILL ILL ILL
FLT OPA INA BUS
NMI
-
-
-
-
-
-
-
-
rwh rwh rwh
-
-
-
-
rwh
-
-
-
rwh rwh rwh rwh
Bit
Function
Illegal External Bus Access Flag
An external access has been attempted with no external bus defined.
ILLBUS
ILLINA
ILLOPA
PRTFLT
UNDOPC
STKUF
STKOF
NMI
Illegal Instruction Access Flag
A branch to an odd address has been attempted.
Illegal Word Operand Access Flag
A word operand access (read or write) to an odd address has been attempted.
Protection Fault Flag
A protected instruction with an illegal format has been detected.
Undefined Opcode Flag
The currently decoded instruction has no valid C167CR opcode.
Stack Underflow Flag
The current stack pointer value exceeds the content of register STKUN.
Stack Overflow Flag
The current stack pointer value falls below the content of reg. STKOV.
Non Maskable Interrupt Flag
A negative transition (falling edge) has been detected on pin NMI.
Note: The trap service routine must clear the respective trap flag, otherwise a new trap
will be requested after exiting the service routine. Setting a trap request flag by
software causes the same effects as if it had been set by hardware.
In the case where e.g. an Undefined Opcode trap (class B) occurs simultaneously with
an NMI trap (class A), both the NMI and the UNDOPC flag is set, the IP of the instruction
with the undefined opcode is pushed onto the system stack, but the NMI trap is executed.
After return from the NMI service routine, the IP is popped from the stack and
immediately pushed again because of the pending UNDOPC trap.
User’s Manual
5-30
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
External NMI Trap
Whenever a high to low transition on the dedicated external NMI pin (Non-Maskable
Interrupt) is detected, the NMI flag in register TFR is set and the CPU will enter the NMI
trap routine. The IP value pushed on the system stack is the address of the instruction
following the one after which normal processing was interrupted by the NMI trap.
Note: The NMI pin is sampled with every CPU clock cycle to detect transitions.
Stack Overflow Trap
Whenever the stack pointer is decremented to a value which is less than the value in the
stack overflow register STKOV, the STKOF flag in register TFR is set and the CPU will
enter the stack overflow trap routine. Which IP value will be pushed onto the system
stack depends on which operation caused the decrement of the SP. When an implicit
decrement of the SP is made through a PUSH or CALL instruction, or upon interrupt or
trap entry, the IP value pushed is the address of the following instruction. When the SP
is decremented by a subtract instruction, the IP value pushed represents the address of
the instruction after the instruction following the subtract instruction.
For recovery from stack overflow it must be ensured that there is enough excess space
on the stack for saving the current system state (PSW, IP, in segmented mode also CSP)
twice. Otherwise, a system reset should be generated.
Stack Underflow Trap
Whenever the stack pointer is incremented to a value which is greater than the value in
the stack underflow register STKUN, the STKUF flag is set in register TFR and the CPU
will enter the stack underflow trap routine. Again, which IP value will be pushed onto the
system stack depends on which operation caused the increment of the SP. When an
implicit increment of the SP is made through a POP or return instruction, the IP value
pushed is the address of the following instruction. When the SP is incremented by an
add instruction, the pushed IP value represents the address of the instruction after the
instruction following the add instruction.
Undefined Opcode Trap
When the instruction currently decoded by the CPU does not contain a valid C167CR
opcode, the UNDOPC flag is set in register TFR and the CPU enters the undefined
opcode trap routine. The IP value pushed onto the system stack is the address of the
instruction that caused the trap.
This can be used to emulate unimplemented instructions. The trap service routine can
examine the faulting instruction to decode operands for unimplemented opcodes based
on the stacked IP. In order to resume processing, the stacked IP value must be
incremented by the size of the undefined instruction, which is determined by the user,
before a RETI instruction is executed.
User’s Manual
5-31
V3.1, 2000-03
C167CR
Derivatives
Interrupt and Trap Functions
Protection Fault Trap
Whenever one of the special protected instructions is executed where the opcode of that
instruction is not repeated twice in the second word of the instruction and the byte
following the opcode is not the complement of the opcode, the PRTFLT flag in register
TFR is set and the CPU enters the protection fault trap routine. The protected
instructions include DISWDT, EINIT, IDLE, PWRDN, SRST, and SRVWDT. The IP value
pushed onto the system stack for the protection fault trap is the address of the instruction
that caused the trap.
Illegal Word Operand Access Trap
Whenever a word operand read or write access is attempted to an odd byte address, the
ILLOPA flag in register TFR is set and the CPU enters the illegal word operand access
trap routine. The IP value pushed onto the system stack is the address of the instruction
following the one which caused the trap.
Illegal Instruction Access Trap
Whenever a branch is made to an odd byte address, the ILLINA flag in register TFR is
set and the CPU enters the illegal instruction access trap routine. The IP value pushed
onto the system stack is the illegal odd target address of the branch instruction.
Illegal External Bus Access Trap
Whenever the CPU requests an external instruction fetch, data read or data write, and
no external bus configuration has been specified, the ILLBUS flag in register TFR is set
and the CPU enters the illegal bus access trap routine. The IP value pushed onto the
system stack is the address of the instruction following the one which caused the trap.
User’s Manual
5-32
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
6
Clock Generation
All activities of the C167CR’s controller hardware and its on-chip peripherals are
controlled via the system clock signal fCPU
.
This reference clock is generated in three stages (see also Figure 6-1):
Oscillator
The on-chip Pierce oscillator can either run with an external crystal and appropriate
oscillator circuitry or it can be driven by an external oscillator.
Frequency Control
The input clock signal feeds the controller hardware:
• directly, providing phase coupled operation on not too high input frequency
• divided by 2 in order to get 50% duty cycle clock signal
• via an on-chip phase locked loop (PLL) providing max. performance on low input
frequency
The resulting internal clock signal is referred to as “CPU clock” fCPU
.
Clock Drivers
The CPU clock is distributed via separate clock drivers which feed the CPU itself and two
groups of peripheral modules.
Oscillator
Frequency Control
Clock Drivers
Idle Mode
CPU
CCD
Prescaler
PLL
OSC
Power Down
PCD
Peripherals,
Ports, Intr. Ctrl.
Interfaces
MCS04334
Figure 6-1
CPU Clock Generation Stages
User’s Manual
6-1
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
6.1
Oscillator
The main oscillator of the C167CR is a power optimized Pierce oscillator providing an
inverter and a feedback element. Pins XTAL1 and XTAL2 connect the inverter to the
external crystal. The standard external oscillator circuitry (see Figure 6-2) comprises the
crystal, two low end capacitors and series resistor (Rx2) to limit the current through the
crystal. The additional LC combination is only required for 3rd overtone crystals to
suppress oscillation in the fundamental mode. A test resistor (RQ) may be temporarily
inserted to measure the oscillation allowance of the oscillator circuitry.
XTAL1
XTAL2
RQ
Rx2
MCS04335
Figure 6-2
External Oscillator Circuitry
The on-chip oscillator is optimized for an input frequency range of 4 to 40 MHz.
An external clock signal (e.g. from an external oscillator or from a master device) may
be fed to the input XTAL1. The Pierce oscillator then is not required to support the
oscillation itself but is rather driven by the input signal. In this case the input frequency
range may be 0 to 50 MHz (please note that the maximum applicable input frequency is
limited by the device’s maximum CPU frequency).
User’s Manual
6-2
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
For input frequencies above 25 … 30 MHz the oscillator’s output should be terminated
as shown in Figure 6-3, at lower frequencies it may be left open. This termination
improves the operation of the oscillator by filtering out frequencies above the intended
oscillator frequency.
XTAL1
XTAL2
15 pF
3 kΩ
Input Clock
MCS04336
Figure 6-3
Oscillator Output Termination
Note: It is strongly recommended to measure the oscillation allowance (or margin) in the
final target system (layout) to determine the optimum parameters for the oscillator
operation.
User’s Manual
6-3
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
6.2
Frequency Control
The CPU clock is generated from the oscillator clock:
The basic clock is the standard operating clock for the C167CR and is required to
deliver the intended maximum performance. The clock configuration in register RP0H
(bitfield CLKCFG = RP0H.7-5) determines one of three possible basic clock generation
modes:
• Direct Drive: the oscillator clock is directly fed to the controller hardware.
• Prescaler: the oscillator clock is divided by 2 to achieve a 50% duty cycle.
• PLL: the oscillator clock is multiplied by a configurable factor of F = 1.5 … 5.
Configuration
PLL
Oscillator Clock
CPU Clock
fOSC
fCPU
2:1
MCS04337
Figure 6-4
Frequency Control Paths
Note: The configuration register RP0H is loaded with the logic levels present on the
upper half of PORT0 (P0H) after a long hardware reset, i.e. bitfield CLKCFG
represents the logic levels on pins P0.15-13 (P0H.7-5).
The internal operation of the C167CR is controlled by the internal CPU clock fCPU. Both
edges of the CPU clock can trigger internal (e.g. pipeline) or external (e.g. bus cycles)
operations (see Figure 6-5).
User’s Manual
6-4
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
Phase Locked Loop Operation
fOSC
TCL
fCPU
TCL
Direct Clock Drive
fOSC
TCL
fCPU
TCL
Prescaler Operation
fOSC
TCL
fCPU
MCT04338
TCL
Figure 6-5
Direct Drive
Generation Mechanisms for the CPU Clock
When direct drive is configured (CLKCFG = 011 ) the C167CR’s clock system is directly
B
fed from the external clock input, i.e. fCPU = fOSC. This allows operation of the C167CR
with a reasonably small fundamental mode crystal. The specified minimum values for the
CPU clock phases (TCLs) must be respected. Therefore the maximum input clock
frequency depends on the clock signal’s duty cycle.
Prescaler Operation
When prescaler operation is configured (CLKCFG = 001 ) the C167CR’s input clock is
B
divided by 2 to generate then CPU clock signal, i.e. fCPU = fOSC / 2. This requires the
oscillator (or input clock) to run on 2 times the intended operating frequency but
guarantees a 50% duty cycle for the internal clock system independent of the input clock
signal’s waveform.
User’s Manual
6-5
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
PLL Operation
When PLL operation is configured (via CLKCFG) the C167CR’s input clock is fed to the
on-chip phase locked loop circuit which multiplies its frequency by a factor ofF = 1.5 … 5
(selectable via CLKCFG, see Table 6-1) and generates a CPU clock signal with 50%
duty cycle, i.e. fCPU = fOSC × F.
The on-chip PLL circuit allows operation of the C167CR on a low frequency external
clock while still providing maximum performance. The PLL also provides fail safe
mechanisms which allow the detection of frequency deviations and the execution of
emergency actions in case of an external clock failure.
When the PLL detects a missing input clock signal it generates an interrupt request. This
warning interrupt indicates that the PLL frequency is no more locked, i.e. no more stable.
This occurs when the input clock is unstable and especially when the input clock fails
completely, e.g. due to a broken crystal. In this case the synchronization mechanism will
reduce the PLL output frequency down to the PLL’s base frequency (2 … 5 MHz). The
base frequency is still generated and allows the CPU to execute emergency actions in
case of a loss of the external clock.
On power-up the PLL provides a stable clock signal within ca. 1 ms after VDD has
reached the specified valid range, even if there is no external clock signal (in this case
the PLL will run on its base frequency of 2 … 5 MHz). The PLL starts synchronizing with
the external clock signal as soon as it is available. Within ca. 1 ms after stable
oscillations of the external clock within the specified frequency range the PLL will be
synchronous with this clock at a frequency of F × fOSC, i.e. the PLL locks to the external
clock.
When PLL operation is selected the CPU clock is a selectable multiple of the oscillator
frequency, i.e. the input frequency.
Table 6-1 lists the possible selections.
User’s Manual
6-6
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
Table 6-1
C167CR Clock Generation Modes
RP0H.7-5 CPU Frequency External Clock
Notes
1)
(P0H.7-5)
1 1 1
1 1 0
1 0 1
1 0 0
0 1 1
0 1 0
0 0 1
fCPU = fOSC × F Input Range
fOSC × 4
fOSC × 3
fOSC × 2
fOSC × 5
fOSC × 1
fOSC × 1.5
fOSC / 2
2.5 to 8.25 MHz
3.33 to 11 MHz
5 to 16.5 MHz
2 to 6.6 MHz
1 to 33 MHz
Default configuration
–
–
–
2)
Direct drive
6.66 to 22 MHz
2 to 66 MHz
–
CPU clock via prescaler
–
0 0 0
fOSC × 2.5
4 to 13.2 MHz
1)
The external clock input range refers to a CPU clock range of 10 … 33 MHz.
2)
The maximum frequency depends on the duty cycle of the external clock signal. In emulation mode pin P0.15
(P0H.7) is inverted, i.e. the configuration ‘111’ would select direct drive in emulation mode.
The PLL constantly synchronizes to the external clock signal. Due to the fact that the
external frequency is 1/F’th of the PLL output frequency the output frequency may be
slightly higher or lower than the desired frequency. This jitter is irrelevant for longer time
periods. For short periods (1 … 4 CPU clock cycles) it remains below 4%.
PLL Circuit
fPLL
fIN
fCPU
fPLL = F × fIN
Reset
Reset
Sleep
F
PWRDN
Lock
OWD
XP3INT
CLKCFG
(RP0H.7-5)
MCB04339
Figure 6-6
PLL Block Diagram
User’s Manual
6-7
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
6.3
Oscillator Watchdog
The C167CR provides an Oscillator Watchdog (OWD) which monitors the clock signal
fed to input XTAL1 of the on-chip oscillator (either with a crystal or via external clock
drive) in prescaler or direct drive mode (not if the PLL provides the basic clock). For this
operation the PLL provides a clock signal (base frequency) which is used to supervise
transitions on the oscillator clock. This PLL clock is independent from the XTAL1 clock.
When the expected oscillator clock transitions are missing the OWD activates the PLL
Unlock/OWD interrupt node and supplies the CPU with the PLL clock signal instead of
the selected oscillator clock. Under these circumstances the PLL will oscillate with its
base frequency.
In direct drive mode the PLL base frequency is used directly (fCPU = 2 … 5 MHz).
In prescaler mode the PLL base frequency is divided by 2 (fCPU = 1 … 2.5 MHz).
If the oscillator clock fails while the PLL provides the basic clock the system will be
supplied with the PLL base frequency anyway.
With this PLL clock signal the CPU can either execute a controlled shutdown sequence
bringing the system into a defined and safe idle state, or it can provide an emergency
operation of the system with reduced performance based on this (normally slower)
emergency clock.
Note: The CPU clock source is only switched back to the oscillator clock after a
hardware reset.
The oscillator watchdog can be disabled by setting bit OWDDIS in register SYSCON.
In this case the PLL remains idle and provides no clock signal, while the CPU clock
signal is derived directly from the oscillator clock or via prescaler. Also no interrupt
request will be generated in case of a missing oscillator clock.
Note: The oscillator watchdog may also be disabled via hardware by (externally) pulling
the OWE line low. However, this is mainly provided for testing purposes and not
recommended for application systems.
User’s Manual
6-8
V3.1, 2000-03
C167CR
Derivatives
Clock Generation
6.4
Clock Drivers
The operating clock signal fCPU is distributed to the controller hardware via several clock
drivers which are disabled under certain circumstances. Table 6-2 summarizes the
different clock drivers and their function, especially in power reduction modes:
Table 6-2
Clock Drivers Description
Clock Driver Clock Active
Signal mode
Idle
mode
P. Down Connected Circuitry
mode
CCD
fCPU
ON
Off
Off
CPU,
CPU
Clock Driver
internal memory modules
(IRAM, ROM/OTP/Flash)
PCD
Peripheral
Clock Driver
fCPU
ON
ON
Off
(X)Peripherals (timers, etc.),
interrupt controller, ports
User’s Manual
6-9
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7
Parallel Ports
In order to accept or generate single external control signals or parallel data, the
C167CR provides up to 111 parallel IO lines organized into one 16-bit IO port (Port 2),
eight 8-bit IO ports (PORT0 made of P0H and P0L, PORT1 made of P1H and P1L,
Port 4, Port 6, Port 7, Port 8), one 15-bit IO port (Port 3), and one 16-bit input port
(Port 5).
These port lines may be used for general purpose Input/Output controlled via software
or may be used implicitly by the C167CR’s integrated peripherals or the External Bus
Controller.
All port lines are bit addressable, and all input/output lines are individually (bit-wise)
programmable as inputs or outputs via direction registers (except Port 5, of course). The
IO ports are true bidirectional ports which are switched to high impedance state when
configured as inputs. The output drivers of five IO ports (2, 3, 6, 7, 8) can be configured
(pin by pin) for push/pull operation or open-drain operation via control registers.
The logic level of a pin is clocked into the input latch once per state time, regardless
whether the port is configured for input or output.
Data Input / Output
Registers
Direction Control
Registers
Port Driver Control
Register
Diverse Control
Registers
P0L
P0H
P1L
P1H
P2
E
E
E
E
E
E
DP0L
DP0H
DP1L
DP1H
DP2
PDCR
PICON
E
E
ODP2
ODP3
P3
DP3
P4
DP4
P5
E
E
P6
DP6
DP7
DP8
ODP6
ODP7
ODP8
P7
E
P8
MCA04340
Figure 7-1
SFRs and Pins associated with the Parallel Ports
User’s Manual
7-1
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
A write operation to a port pin configured as an input causes the value to be written into
the port output latch, while a read operation returns the latched state of the pin itself. A
read-modify-write operation reads the value of the pin, modifies it, and writes it back to
the output latch.
Writing to a pin configured as an output (DPx.y = ‘1’) causes the output latch and the pin
to have the written value, since the output buffer is enabled. Reading this pin returns the
value of the output latch. A read-modify-write operation reads the value of the output
latch, modifies it, and writes it back to the output latch, thus also modifying the level at
the pin.
7.1
Input Threshold Control
The standard inputs of the C167CR determine the status of input signals according to
TTL levels. In order to accept and recognize noisy signals, CMOS-like input thresholds
can be selected instead of the standard TTL thresholds for all pins of specific ports.
These special thresholds are defined above the TTL thresholds and feature a defined
hysteresis to prevent the inputs from toggling while the respective input signal level is
near the thresholds.
The Port Input Control register PICON allows to select these thresholds for each byte of
the indicated ports, i.e. 8-bit ports are controlled by one bit each while 16-bit ports are
controlled by two bits each.
PICON
Port Input Control Reg.
SFR (F1C4 /E2 )
Reset value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
P8L P7L P6L
IN
P3H P3L P2H P2L
IN
-
IN
IN
IN
IN
IN
-
-
-
-
-
-
-
rw
rw
rw
-
rw
rw
rw
rw
Bit
Function
Port x Low Byte Input Level Selection
PxLIN
0:
1:
Pins Px.7 … Px.0 switch on standard TTL input levels
Pins Px.7 … Px.0 switch on special threshold input levels
PxHIN
Port x High Byte Input Level Selection
0:
1:
Pins Px.15 … Px.8 switch on standard TTL input levels
Pins Px.15 … Px.8 switch on special threshold input levels
User’s Manual
7-2
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
All options for individual direction and output mode control are available for each pin
independent from the selected input threshold.
The input hysteresis provides stable inputs from noisy or slowly changing external
signals.
Hysteresis
Input Level
Bit State
MCT04341
Figure 7-2
Hysteresis for Special Input Thresholds
User’s Manual
7-3
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.2
Output Driver Control
The output driver of a port pin is activated by switching the respective pin to output, i.e.
DPx.y = ’1’. The value that is driven to the pin is determined by the port output latch or
by the associated alternate function (e.g. address, peripheral IO, etc.). The user software
can control the characteristics of the output driver via the following mechanisms:
• Open Drain Mode: The upper (push) transistor is always disabled. Only ‘0’ is driven
actively, an external pullup is required.
• Edge Characteristic: The rise/fall time of an output signal can be selected.
Open Drain Mode
In the C167CR certain ports provide Open Drain Control, which allows to switch the
output driver of a port pin from a push/pull configuration to an open drain configuration.
In push/pull mode a port output driver has an upper and a lower transistor, thus it can
actively drive the line either to a high or a low level. In open drain mode the upper
transistor is always switched off, and the output driver can only actively drive the line to
a low level. When writing a ‘1’ to the port latch, the lower transistor is switched off and
the output enters a high-impedance state. The high level must then be provided by an
external pullup device. With this feature, it is possible to connect several port pins
together to a Wired-AND configuration, saving external glue logic and/or additional
software overhead for enabling/disabling output signals.
This feature is controlled through the respective Open Drain Control Registers ODPx
which are provided for each port that has this feature implemented. These registers allow
the individual bit-wise selection of the open drain mode for each port line.
If the respective control bit ODPx.y is ‘0’ (default after reset), the output driver is in the
push/pull mode. If ODPx.y is ‘1’, the open drain configuration is selected. Note that all
ODPx registers are located in the ESFR space.
External
Pullup
Pin
Pin
Q
Q
Push/Pull Output Driver
Open Drain Output Driver
MCS01975
Figure 7-3
Output Drivers in Push/Pull Mode and in Open Drain Mode
User’s Manual
7-4
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Edge Characteristic
This defines the rise/fall time for the respective output, i.e. the output transition time.
Slow edges reduce the peak currents that are drawn when changing the voltage level of
an external capacitive load. For a bus interface, however, fast edges may still be
required. Edge characteristic effects the pre-driver which controls the final output driver
stage.
Control Signals Driver Control Logic
Driver Stage
Open Drain Control
Edge Control
Push
Pin
Data Signal
Pull
MCS04342
Figure 7-4
Structure of Output Driver with Edge Control
User’s Manual
7-5
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
The Port Driver Control Register PDCR provides the corresponding control bits. A
separate control bit is provided for bus pins as well as for non-bus pins.
PDCR
Port Driver Control Reg.
ESFR (F0AA /55 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
NBP
EC
BIP
EC
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
rw
-
-
-
rw
Bit
Function
BIPEC Bus Interface Pins Edge Characteristic (Defines the outp. rise/fall time tRF)
0:
1:
Fast edge mode, rise/fall times depend on the driver’s dimensioning.
Reduced edge mode.
BIPEC controls: PORT0, PORT1, Port 4, Port 6, RD, WR, ALE, CLKOUT,
BHE/WRH, READY (emulation mode only).
NBPEC Non-Bus Pins Edge Characteristic (Defines the output rise/fall time tRF)
0:
Fast edge mode, rise/fall times depend on the driver’s dimensioning.
1:
Reduced edge mode.
NBPEC controls: Port 2, Port 3, Port 7, Port 8, RSTOUT, RSTIN
(bidirectional reset mode only).
Figure 7-5 summarizes the effects of the driver characteristics:
Edge characteristic generally influences the output signal’s shape.
Fast Edge
Slow Edge
MCD04343
Figure 7-5
General Output Signal Waveforms
User’s Manual
7-6
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.3
Alternate Port Functions
In order to provide a maximum of flexibility for different applications and their specific IO
requirements port lines have programmable alternate input or output functions
associated with them.
Table 7-1
Summary of Alternate Port Functions
Alternate Function(s)
Port
Alternate Signal(s)
PORT0 Address and data lines when accessing
external resources (e.g. memory)
AD15 … AD0
PORT1 Address lines when accessing ext. resources A15 … A0,
Capture inputs of the CAPCOM units
CC27IO … CC24IO
Port 2 Capture inputs or compare outputs of the
CAPCOM units, CAPCOM timer input
Fast external interrupt inputs
CC15IO … CC0IO,
T7IN,
EX7IN … EX0IN
Port 3 System clock output
Optional bus control signal
Input/output functions of serial interfaces,
timers
CLKOUT, BHE/WRH,
RxD0, TxD0, MTSR, MRST,
SCLK, T2IN, T3IN, T4IN,
T3EUD, T3OUT, CAPIN,
T6OUT, T0IN
Port 4 Selected segment address lines in systems
with more than 64 KBytes of external resources
CAN interface (where implemented)
A23 … A16,
CAN1_TxD, CAN1_RxD
Port 5 Analog input channels to the A/D converter
Timer control signal inputs
AN15 … AN0,
T2EUD, T4EUD, T5IN, T6IN
Port 6 Bus arbitration signals,
Chip select output signals
BREQ, HLDA, HOLD,
CS4 … CS0
Port_7 Capture inputs or compare outputs of the
CAPCOM units
CC31IO … CC28IO,
PWM output signals
POUT3 … POUT0
CC23IO … CC16IO
Port 8 Capture inputs or compare outputs of the
CAPCOM units
If an alternate output function of a pin is to be used, the direction of this pin must be
programmed for output (DPx.y = ‘1’), except for some signals that are used directly after
reset and are configured automatically. Otherwise the pin remains in the high-impedance
state and is not effected by the alternate output function. The respective port latch should
hold a ‘1’, because its output is combined with the alternate output data.
X-Peripherals (peripherals connected to the on-chip XBUS) control their associated IO
pins directly via separate control lines.
User’s Manual
7-7
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
If an alternate input function of a pin is used, the direction of the pin must be
programmed for input (DPx.y = ‘0’) if an external device is driving the pin. The input
direction is the default after reset. If no external device is connected to the pin, however,
one can also set the direction for this pin to output. In this case, the pin reflects the state
of the port output latch. Thus, the alternate input function reads the value stored in the
port output latch. This can be used for testing purposes to allow a software trigger of an
alternate input function by writing to the port output latch.
On most of the port lines, the user software is responsible for setting the proper direction
when using an alternate input or output function of a pin. This is done by setting or
clearing the direction control bit DPx.y of the pin before enabling the alternate function.
There are port lines, however, where the direction of the port line is switched
automatically. For instance, in the multiplexed external bus modes of PORT0, the
direction must be switched several times for an instruction fetch in order to output the
addresses and to input the data. Obviously, this cannot be done through instructions. In
these cases, the direction of the port line is switched automatically by hardware if the
alternate function of such a pin is enabled.
To determine the appropriate level of the port output latches check how the alternate
data output is combined with the respective port latch output.
There is one basic structure for all port lines with only an alternate input function. Port
lines with only an alternate output function, however, have different structures due to the
way the direction of the pin is switched and depending on whether the pin is accessible
by the user software or not in the alternate function mode.
All port lines that are not used for these alternate functions may be used as general
purpose IO lines. When using port pins for general purpose output, the initial output
value should be written to the port latch prior to enabling the output drivers, in order to
avoid undesired transitions on the output pins. This applies to single pins as well as to
pin groups (see examples below).
OUTPUT_ENABLE_SINGLE_PIN:
BSET
BSET
P4.0
DP4.0
;Initial output level is ‘high’
;Switch on the output driver
OUTPUT_ENABLE_PIN_GROUP:
BFLDL
BFLDL
P4, #05H, #05H
DP4, #05H, #05H
;Initial output level is ‘high’
;Switch on the output drivers
Note: When using several BSET pairs to control more pins of one port, these pairs must
be separated by instructions, which do not reference the respective port (see
Chapter 4.2).
Each of these ports and the alternate input and output functions are described in detail
in the following subsections.
User’s Manual
7-8
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.4
PORT0
The two 8-bit ports P0H and P0L represent the higher and lower part of PORT0,
respectively. Both halfs of PORT0 can be written (e.g. via a PEC transfer) without
effecting the other half.
If this port is used for general purpose IO, the direction of each line can be configured
via the corresponding direction registers DP0H and DP0L.
P0L
PORT0 Low Register
SFR (FF00 /80 )
Reset value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
P0L P0L P0L P0L P0L P0L P0L P0L
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
P0H
PORT0 High Register
SFR (FF02 /81 )
Reset value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
P0H P0H P0H P0H P0H P0H P0H P0H
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
P0X.y
Function
Port data register P0H or P0L bit y
User’s Manual
7-9
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
DP0L
P0L Direction Ctrl. Register
ESFR (F100 /80 )
Reset value: - - 00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
DP0L DP0L DP0L DP0L DP0L DP0L DP0L DP0L
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
DP0H
P0H Direction Ctrl. Register
ESFR (F102 /81 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
DP0H DP0H DP0H DP0H DP0H DP0H DP0H DP0H
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
DP0X.y
Function
Port direction register DP0H or DP0L bit y
DP0X.y = 0: Port line P0X.y is an input (high-impedance)
DP0X.y = 1: Port line P0X.y is an output
Alternate Functions of PORT0
When an external bus is enabled, PORT0 is used as data bus or address/data bus.
Note that an external 8-bit demultiplexed bus only uses P0L, while P0H is free for IO
(provided that no other bus mode is enabled).
PORT0 is also used to select the system startup configuration. During reset, PORT0 is
configured to input, and each line is held high through an internal pullup device. Each
line can now be individually pulled to a low level (see DC-level specifications in the
respective Data Sheets) through an external pulldown device. A default configuration is
selected when the respective PORT0 lines are at a high level. Through pulling individual
lines to a low level, this default can be changed according to the needs of the
applications.
The internal pullup devices are designed such that an external pulldown resistors (see
Data Sheet specification) can be used to apply a correct low level. These external
pulldown resistors can remain connected to the PORT0 pins also during normal
operation, however, care has to be taken such that they do not disturb the normal
function of PORT0 (this might be the case, for example, if the external resistor is too
strong).
With the end of reset, the selected bus configuration will be written to the BUSCON0
register. The configuration of the high byte of PORT0 will be copied into the special
register RP0H. This read-only register holds the selection for the number of chip selects
User’s Manual
7-10
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
and segment addresses. Software can read this register in order to react according to
the selected configuration, if required.
When the reset is terminated, the internal pullup devices are switched off, and PORT0
will be switched to the appropriate operating mode.
During external accesses in multiplexed bus modes PORT0 first outputs the 16-bit
intra-segment address as an alternate output function. PORT0 is then switched to
high-impedance input mode to read the incoming instruction or data. In 8-bit data bus
mode, two memory cycles are required for word accesses, the first for the low byte and
the second for the high byte of the word. During write cycles PORT0 outputs the data
byte or word after outputting the address.
During external accesses in demultiplexed bus modes PORT0 reads the incoming
instruction or data word or outputs the data byte or word.
Alternate Function
P0H.7
a)
b)
c)
d)
D15
D14
D13
D12
D11
D10
D9
D8
D7
D6
D5
A15
A14
A13
A12
A11
A10
A9
AD15
AD14
AD13
AD12
AD11
AD10
AD9
AD8
AD7
AD6
AD5
P0H.6
P0H.5
P0H.4
P0H.3
P0H.2
P0H.1
P0H.0
P0L.7
P0L.6
P0L.5
P0L.4
P0L.3
P0L.2
P0L.1
P0L.0
P0H
A8
Port 0
D7
D6
D5
D4
D3
D2
D1
D0
AD7
AD6
AD5
AD4
AD3
AD2
AD1
AD0
D4
D3
D2
D1
AD4
AD3
AD2
AD1
P0L
D0
AD0
General Purpose
Input/Output
8-Bit
Demux Bus
16-Bit
Demux Bus
8-Bit
MUX Bus
16-Bit
MUX Bus
MCA04344
Figure 7-6
PORT0 IO and Alternate Functions
When an external bus mode is enabled, the direction of the port pin and the loading of
data into the port output latch are controlled by the bus controller hardware. The input of
the port output latch is disconnected from the internal bus and is switched to the line
labeled “Alternate Data Output” via a multiplexer. The alternate data can be the 16-bit
intrasegment address or the 8/16-bit data information. The incoming data on PORT0 is
read on the line “Alternate Data Input”. While an external bus mode is enabled, the user
software should not write to the port output latch, otherwise unpredictable results may
occur. When the external bus modes are disabled, the contents of the direction register
last written by the user becomes active.
Figure 7-7 shows the structure of a PORT0 pin.
User’s Manual
7-11
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
1 0
0
1
AltDir
AltEN
Pin
0
1
AltDataOut
Driver
Clock
AltDataIN
Input
Latch
MCB04345
P0H.7-0, P0L.7-0
Figure 7-7
Block Diagram of a PORT0 Pin
User’s Manual
7-12
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.5
PORT1
The two 8-bit ports P1H and P1L represent the higher and lower part of PORT1,
respectively. Both halfs of PORT1 can be written (e.g. via a PEC transfer) without
effecting the other half.
If this port is used for general purpose IO, the direction of each line can be configured
via the corresponding direction registers DP1H and DP1L.
P1L
PORT1 Low Register
SFR (FF04 /82 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
P1L P1L P1L P1L P1L P1L P1L P1L
.7
6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
P1H
PORT1 High Register
SFR (FF06 /83 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
P1H P1H P1H P1H P1H P1H P1H P1H
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
P1X.y
Function
Port data register P1H or P1L bit y
User’s Manual
7-13
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
DP1L
P1L Direction Ctrl. Register
ESFR (F104 /82 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
DP1 DP1 DP1 DP1 DP1 DP1 DP1 DP1
L.7 L.6 L.5 L.4 L.3 L.2 L.1 L.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
DP1H
P1H Direction Ctrl. Register
ESFR (F106 /83 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
DP1 DP1 DP1 DP1 DP1 DP1 DP1 DP1
H.7 H.6 H.5 H.4 H.3 H.2 H.1 H.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
DP1X.y
Function
Port direction register DP1H or DP1L bit y
DP1X.y = 0: Port line P1X.y is an input (high-impedance)
DP1X.y = 1: Port line P1X.y is an output
Alternate Functions of PORT1
When a demultiplexed external bus is enabled, PORT1 is used as address bus.
Note that demultiplexed bus modes use PORT1 as a 16-bit port. Otherwise all 16 port
lines can be used for general purpose IO.
The upper four pins of PORT1 (P1H.7 … P1H.4) also serve as capture inputs
(CC27IO … CC24IO) for the CAPCOM2 unit.
The usage of the port lines by the CAPCOM unit, its accessibility via software, and the
precautions are the same as described for the Port 2 lines.
As all other capture inputs, the capture input function of pins P1H.7 … P1H.4 can also
be used as external interrupt inputs (sample rate 16 TCL).
As a side effect, the capture input capability of these lines can also be used in the
address bus mode. Hereby changes of the upper address lines could be detected and
trigger an interrupt request in order to perform some special service routines. External
capture signals can only be applied if no address output is selected for PORT1.
User’s Manual
7-14
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Alternate Function
P1H.7
a)
b)
A15
A14
A13
A12
A11
A10
A9
A8
A7
A6
A5
CC27IO
P1H.6
P1H.5
P1H.4
P1H.3
P1H.2
P1H.1
P1H.0
P1L.7
P1L.6
P1L.5
P1L.4
P1L.3
P1L.2
P1L.1
P1L.0
CC26IO
CC25IO
CC24IO
P1H
Port 1
A4
A3
A2
A1
P1L
A0
General Purpose
Input/Output
8/16-Bit
Demux Bus
CAPCOM2
Capture Inputs
MCA04346
Figure 7-8
PORT1 IO and Alternate Functions
During external accesses in demultiplexed bus modes PORT1 outputs the 16-bit
intra-segment address as an alternate output function.
During external accesses in multiplexed bus modes, when no BUSCON register selects
a demultiplexed bus mode, PORT1 is not used and is available for general purpose IO.
When an external bus mode is enabled, the direction of the port pin and the loading of
data into the port output latch are controlled by the bus controller hardware. The input of
the port output latch is disconnected from the internal bus and is switched to the line
labeled “Alternate Data Output” via a multiplexer. The alternate data is the 16-bit
intrasegment address. While an external bus mode is enabled, the user software should
not write to the port output latch, otherwise unpredictable results may occur. When the
external bus modes are disabled, the contents of the direction register last written by the
user becomes active.
Figure 7-9 show the structure of PORT1 pins. The upper 4 pins of PORT1 combine
internal bus data and alternate data output before the port latch input.
User’s Manual
7-15
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
1 0
0
1
AltDir = '1'
AltEN
Pin
0
1
AltDataOut
Driver
Clock
AltDataIN (Pin)
Input
Latch
MCB04347
P1H.7-4
Figure 7-9
Block Diagram of a PORT1 Pin with Address and CAPCOM Function
User’s Manual
7-16
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
1 0
0
1
AltDir = '1'
AltEN
Pin
0
1
AltDataOut
Driver
Clock
Input
Latch
MCB04348
P1H.3-0, P1L.7-0
Figure 7-10 Block Diagram of a PORT1 Pin with Address Function
User’s Manual
7-17
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.6
Port 2
If this 16-bit port is used for general purpose IO, the direction of each line can be
configured via the corresponding direction register DP2. Each port line can be switched
into push/pull or open drain mode via the open drain control register ODP2.
P2
Port 2 Data Register
SFR (FFC0 /E0 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
P2
P2
P2 P2
P2
P2
P2.9 P2.8 P2.7 P2.6 P2.5 P2.4 P2.3 P2.2 P2.1 P2.0
rw rw rw rw rw rw rw rw rw rw
.15 .14 .13 .12 .11 .10
rw
rw
rw
rw
rw
rw
Bit
Function
Port data register P2 bit y
P2.y
DP2
P2 Direction Ctrl. Register
SFR (FFC2 /E1 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2 DP2
.15 .14 .13 .12 .11 .10
.9
.8
.7
.6
.5
.4
.3
.2
.1
.0
rw rw rw rw rw rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
Bit
DP2.y
Function
Port direction register DP2 bit y
DP2.y = 0: Port line P2.y is an input (high-impedance)
DP2.y = 1: Port line P2.y is an output
User’s Manual
7-18
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
ODP2
P2 Open Drain Ctrl. Reg.
ESFR (F1C2 /E1 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2
.15 .14 .13 .12 .11 .10
.9
.8
.7
.6
.5
.4
.3
.2
.1
.0
rw rw rw rw rw rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
Bit
ODP2.y
Function
Port 2 Open Drain control register bit y
ODP2.y = 0: Port line P2.y output driver in push/pull mode
ODP2.y = 1: Port line P2.y output driver in open drain mode
User’s Manual
7-19
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Alternate Functions of Port 2
All Port 2 lines (P2.15 … P2.0) serve as capture inputs or compare outputs
(CC15IO … CC0IO) for the CAPCOM1 unit. The upper eight Port 2 lines (P2.15 … P2.8)
also serve as external interrupt inputs EX7IN … EX0IN (16 TCL sample rate).
When a Port 2 line is used as a capture input, the state of the input latch, which
represents the state of the port pin, is directed to the CAPCOM unit via the line “Alternate
Pin Data Input”. If an external capture trigger signal is used, the direction of the
respective pin must be set to input. If the direction is set to output, the state of the port
output latch will be read since the pin represents the state of the output latch. This can
be used to trigger a capture event through software by setting or clearing the port latch.
Note that in the output configuration, no external device may drive the pin, otherwise
conflicts would occur.
When a Port 2 line is used as a compare output (compare modes 1 and 3), the compare
event (or the timer overflow in compare mode 3) directly effects the port output latch. In
compare mode 1, when a valid compare match occurs, the state of the port output latch
is read by the CAPCOM control hardware via the line “Alternate Latch Data Input”,
inverted, and written back to the latch via the line “Alternate Data Output”. The port
output latch is clocked by the signal “Compare Trigger” which is generated by the
CAPCOM unit. In compare mode 3, when a match occurs, the value ‘1’ is written to the
port output latch via the line “Alternate Data Output”. When an overflow of the
corresponding timer occurs, a ‘0’ is written to the port output latch. In both cases, the
output latch is clocked by the signal “Compare Trigger”. The direction of the pin should
be set to output by the user, otherwise the pin will be in the high-impedance state and
will not reflect the state of the output latch.
As can be seen from the port structure below, the user software always has free access
to the port pin even when it is used as a compare output. This is useful for setting up the
initial level of the pin when using compare mode 1 or the double-register mode. In these
modes, unlike in compare mode 3, the pin is not set to a specific value when a compare
match occurs, but is toggled instead.
When the user wants to write to the port pin at the same time a compare trigger tries to
clock the output latch, the write operation of the user software has priority. Each time a
CPU write access to the port output latch occurs, the input multiplexer of the port output
latch is switched to the line connected to the internal bus. The port output latch will
receive the value from the internal bus and the hardware triggered change will be lost.
As all other capture inputs, the capture input function of pins P2.15 … P2.0 can also be
used as external interrupt inputs (sample rate 16 TCL) or as Fast External Interrupt
inputs (sample rate 2 TCL).
P2.15 in addition serves as input for CAPCOM2 timer T7 (T7IN).
Table 7-2 summarizes the alternate functions of Port 2.
User’s Manual
7-20
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Table 7-2
Alternate Functions of Port 2
Port 2 Pin Alternate
Function a)
Alternate Function b)
Alternate Function c)
P2.0
P2.1
P2.2
P2.3
P2.4
P2.5
P2.6
P2.7
CC0IO
CC1IO
CC2IO
CC3IO
CC4IO
CC5IO
CC6IO
CC7IO
CC8IO
CC9IO
CC10IO
CC11IO
CC12IO
CC13IO
CC14IO
CC15IO
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
P2.8
P2.9
EX0IN Fast External Interrupt 0 Inp. –
EX1IN Fast External Interrupt 1 Inp. –
EX2IN Fast External Interrupt 2 Inp. –
EX3IN Fast External Interrupt 3 Inp. –
EX4IN Fast External Interrupt 4 Inp. –
EX5IN Fast External Interrupt 5 Inp. –
EX6IN Fast External Interrupt 6 Inp. –
EX7IN Fast External Interrupt 7 Inp. T7IN Timer T7
P2.10
P2.11
P2.12
P2.13
P2.14
P2.15
Ext. Count Input
Alternate Function
P2.15
a)
b)
c)
CC15IO
CC14IO
CC13IO
CC12IO
CC11IO
CC10IO
CC9IO
CC8IO
CC7IO
CC6IO
CC5IO
CC4IO
CC3IO
CC2IO
CC1IO
CC0IO
EX7IN
EX6IN
EX5IN
EX4IN
EX3IN
EX2IN
EX1IN
EX0IN
T7IN
P2.14
P2.13
P2.12
P2.11
P2.10
P2.9
P2.8
P2.7
P2.6
P2.5
P2.4
P2.3
P2.2
P2.1
P2.0
Port 2
General Purpose
Input/Output
CAPCOM1
Capt. Inp./
Fast External
Interrupt Input
CAPCOM2
Timer T7 Input
Comp. Output
MCA04349
Figure 7-11 Port 2 IO and Alternate Functions
The pins of Port 2 combine internal bus data and alternate data output before the port
latch input.
User’s Manual
7-21
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
CCx
0 1
Port Output
Latch
Direction
Latch
Open Drain
Latch
1 0
Pin
Driver
Clock
AltDataIn (Latch)
AltDataIn (Pin)
EXzIN
Input
Latch
MCB04350
P2.15-0, x = 15-0, z = 7-0
Figure 7-12 Block Diagram of a Port 2 Pin
Note: Fast external interrupt inputs only on the upper eight pins of Port2.
User’s Manual
7-22
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.7
Port 3
If this 15-bit port is used for general purpose IO, the direction of each line can be
configured via the corresponding direction register DP3. Most port lines can be switched
into push/pull or open drain mode via the open drain control register ODP3 (pins P3.15
and P3.12 do not support open drain mode!).
P3
Port 3 Data Register
SFR (FFC4 /E2 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
P3
P3
P3
P3
P3
P3
.9
P3
.8
-
P3.7 P3.6 P3.5 P3.4 P3.3 P3.2 P3.1 P3.0
rw rw rw rw rw rw rw rw
.15
.13 .12 .11 .10
rw
-
rw
rw
rw
rw
rw
rw
Bit
Function
Port data register P3 bit y
P3.y
DP3
P3 Direction Ctrl. Register
SFR (FFC6 /E3 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
DP3
.15
DP3 DP3 DP3 DP3 DP3 DP3 DP3 DP3 DP3 DP3 DP3 DP3 DP3 DP3
-
.13 .12 .11 .10
.9
.8
.7
.6
.5
.4
.3
.2
.1
.0
rw
-
rw rw rw rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
Bit
DP3.y
Function
Port direction register DP3 bit y
DP3.y = 0: Port line P3.y is an input (high-impedance)
DP3.y = 1: Port line P3.y is an output
User’s Manual
7-23
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
ODP3
P3 Open Drain Ctrl. Reg.
ESFR (F1C6 /E3 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ODP
3.13
ODP ODP ODP ODP ODP ODP ODP ODP ODP ODP ODP ODP
3.11 3.10 3.9 3.8 3.7 3.6 3.5 3.4 3.3 3.2 3.1 3.0
-
-
-
-
-
rw
-
rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
Bit
ODP3.y
Function
Port 3 Open Drain control register bit y
ODP3.y = 0: Port line P3.y output driver in push/pull mode
ODP3.y = 1: Port line P3.y output driver in open drain mode
Note: Due to pin limitations register bit P3.14 is not connected to an IO pin.
Pins P3.15 and P3.12 do not support open drain mode.
Alternate Functions of Port 3
The pins of Port 3 serve for various functions which include external timer control lines,
the two serial interfaces, and the control lines BHE/WRH and clock output.
Table 7-3 summarizes the alternate functions of Port 3.
Table 7-3
Port 3 Pin
Alternate Functions of Port 3
Alternate Function
P3.0
P3.1
P3.2
P3.3
P3.4
P3.5
P3.6
P3.7
P3.8
P3.9
P3.10
P3.11
P3.12
P3.13
–
T0IN
CAPCOM1 Timer T0 Count Input
GPT2 Timer T6 Toggle Latch Output
GPT2 Capture Input
GPT1 Timer T3 Toggle Latch Output
GPT1 Timer T3 External Up/Down Input
GPT1 Timer T4 Count Input
GPT1 Timer T3 Count Input
GPT1 Timer T2 Count Input
SSC Master Receive/Slave Transmit
SSC Master Transmit/Slave Receive
ASC0 Transmit Data Output
T6OUT
CAPIN
T3OUT
T3EUD
T4IN
T3IN
T2IN
MRST
MTSR
TxD0
RxD0
ASC0 Receive Data Input
Byte High Enable/Write High Output
SSC Shift Clock Input/Output
BHE/WRH
SCLK
–
P3.15
CLKOUT
System Clock Output
7-24
User’s Manual
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Alternate Function
P3.15
a)
b)
CLKOUT
No Pin
P3.13
P3.12
P3.11
P3.10
P3.9
P3.8
P3.7
P3.6
P3.5
P3.4
P3.3
P3.2
P3.1
P3.0
SCLK
BHE
RxD0
TxD0
MTSR
MRST
T2IN
T3IN
T4IN
T3EUD
T3OUT
CAPIN
T6OUT
T0IN
WRH
Port 3
General Purpose
Input/Output
MCA04351
Figure 7-13 Port 3 IO and Alternate Functions
The port structure of the Port 3 pins depends on their alternate function (see
Figure 7-14).
When the on-chip peripheral associated with a Port 3 pin is configured to use the
alternate input function, it reads the input latch, which represents the state of the pin, via
the line labeled “Alternate Data Input”. Port 3 pins with alternate input functions are:
T0IN, T2IN, T3IN, T4IN, T3EUD, and CAPIN.
When the on-chip peripheral associated with a Port 3 pin is configured to use the
alternate output function, its “Alternate Data Output” line is ANDed with the port output
latch line. When using these alternate functions, the user must set the direction of the
port line to output (DP3.y = 1) and must set the port output latch (P3.y = 1). Otherwise
the pin is in its high-impedance state (when configured as input) or the pin is stuck at ‘0’
(when the port output latch is cleared). When the alternate output functions are not used,
the “Alternate Data Output” line is in its inactive state, which is a high level (‘1’). Port 3
pins with alternate output functions are:
T6OUT, T3OUT, TxD0, and CLKOUT.
When the on-chip peripheral associated with a Port 3 pin is configured to use both the
alternate input and output function, the descriptions above apply to the respective
current operating mode. The direction must be set accordingly. Port 3 pins with alternate
input/output functions are:
MTSR, MRST, RxD0, and SCLK.
Note: Enabling the CLKOUT function automatically enables the P3.15 output driver.
Setting bit DP3.15 = ‘1’ is not required.
The CLKOUT function is automatically enabled in emulation mode.
User’s Manual
7-25
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
Open Drain
Latch
1 0
Pin
&
AltDataOut
AltDataIn
Driver
Clock
Input
Latch
MCB04352
P3.13, P3.11-0
Figure 7-14 Block Diagram of a Port 3 Pin with Alternate Input or Alternate
Output Function
Pin P3.12 (BHE/WRH) is one more pin with an alternate output function. However, its
structure is slightly different (see Figure 7-15), because after reset the BHE or WRH
function must be used depending on the system startup configuration. In these cases
there is no possibility to program any port latches before. Thus the appropriate alternate
function is selected automatically. If BHE/WRH is not used in the system, this pin can be
used for general purpose IO by disabling the alternate function (BYTDIS = ‘1’/
WRCFG = ‘0’).
User’s Manual
7-26
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
1 0
0
1
AltDir = '1'
AltEN
Pin
0
1
AltDataOut
Driver
Clock
Input
Latch
MCB04353
P3.15, P3.12
Figure 7-15 Block Diagram of Pins P3.15 (CLKOUT) and P3.12 (BHE/WRH)
Note: Enabling the BHE or WRH function automatically enables the P3.12 output driver.
Setting bit DP3.12 = ‘1’ is not required.
During bus hold pin P3.12 is switched back to its standard function and is then
controlled by DP3.12 and P3.12. Keep DP3.12 = ‘0’ in this case to ensure floating
in hold mode.
Enabling the CLKOUT function automatically enables the P3.15 output driver.
Setting bit DP3.15 = ‘1’ is not required.
User’s Manual
7-27
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.8
Port 4
If this 8-bit port is used for general purpose IO, the direction of each line can be
configured via the corresponding direction register DP4.
P4
Port 4 Data Register
SFR (FFC8 /E4 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
P4.7 P4.6 P4.5 P4.4 P4.3 P4.2 P4.1 P4.0
rw rw rw rw rw rw rw rw
-
-
-
-
-
-
Bit
Function
Port data register P4 bit y
P4.y
DP4
P4 Direction Ctrl. Register
SFR (FFCA /E5 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
DP4 DP4 DP4 DP4 DP4 DP4 DP4 DP4
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
DP4.y
Function
Port direction register DP4 bit y
DP4.y = 0: Port line P4.y is an input (high-impedance)
DP4.y = 1: Port line P4.y is an output
User’s Manual
7-28
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Alternate Functions of Port 4
During external bus cycles that use segmentation (i.e. an address space above
64 KByte) a number of Port 4 pins may output the segment address lines. The number
of pins that is used for segment address output determines the external address space
which is directly accessible. The other pins of Port 4 (if any) may be used for general
purpose IO or for the CAN interface.
If segment address lines are selected, the alternate function of Port 4 may be necessary
to access e.g. external memory directly after reset. For this reason Port 4 will be
switched to this alternate function automatically.
The number of segment address lines is selected via PORT0 during reset. The selected
value can be read from bitfield SALSEL in register RP0H (read only) e.g. in order to
check the configuration during run time.
Devices with CAN interface use 2 pins of Port 4 to interface the CAN module to an
external CAN transceiver. In this case the number of possible segment address lines is
reduced.
Table 7-4 summarizes the alternate functions of Port 4 depending on the number of
selected segment address lines (coded via bitfield SALSEL).
Table 7-4
Alternate Functions of Port 4
Altern. Function Altern. Function Altern. Function
Port 4 Std. Function
Pin
SALSEL = 01
64 KB
SALSEL = 11
256KB
SALSEL = 00
1 MB
SALSEL = 10
16 MB
P4.0
P4.1
P4.2
P4.3
P4.4
P4.5
P4.6
P4.7
Gen. purpose IO Seg. Address A16 Seg. Address A16 Seg. Address A16
Gen. purpose IO Seg. Address A17 Seg. Address A17 Seg. Address A17
Gen. purpose IO Gen. purpose IO Seg. Address A18 Seg. Address A18
Gen. purpose IO Gen. purpose IO Seg. Address A19 Seg. Address A19
Gen. purpose IO Gen. purpose IO Gen. purpose IO Seg. Address A20
Gen. p. IO or CAN Gen. p. IO or CAN Gen. p. IO or CAN Seg. Address A21
Gen. p. IO or CAN Gen. p. IO or CAN Gen. p. IO or CAN Seg. Address A22
Gen. purpose IO Gen. purpose IO Gen. purpose IO Seg. Address A23
Note: Port 4 pins that are neither used for segment address output nor for the CAN
interface may be used for general purpose IO.
If more than one function is selected for a Port 4 pin, the segment address takes
preference over the CAN interface lines.
User’s Manual
7-29
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Alternate Function
a)
b)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Port 4
P4.7
A23
A22
A21
A20
A19
A18
A17
A16
P4.7
CAN_TxD
CAN_RxD
P4.4
A19
A18
A17
A16
P4.6
P4.5
P4.4
P4.3
P4.2
P4.1
P4.0
General Purpose
Input/Output
Full Segment
Address (16 MB)
CAN Interface
MCA04354
Figure 7-16 Port 4 IO and Alternate Functions
User’s Manual
7-30
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
1 0
0
1
AltDir = '1'
AltEN
Pin
0
1
AltDataOut
Driver
Clock
AltDataIN
Input
Latch
MCB04355
P4.7-0
Figure 7-17 Block Diagram of a Port 4 Pin
User’s Manual
7-31
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.9
Port 5
This 16-bit input port can only read data. There is no output latch and no direction
register. Data written to P5 will be lost.
P5
Port 5 Data Register
SFR (FFA2 /D1 )
Reset Value: XXXX
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
P5
P5
P5 P5
P5
P5
P5
.9
P5
.8
P5.7 P5.6 P5.5 P5.4 P5.3 P5.2 P5.1 P5.0
.15 .14 .13 .12 .11 .10
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
Bit
Function
Port data register P5 bit y (Read only)
P5.y
Alternate Functions of Port 5
Each line of Port 5 is also connected to the input multiplexer of the Analog/Digital
Converter. All port lines can accept analog signals (ANx) that can be converted by the
ADC. For pins that shall be used as analog inputs it is recommended to disable the digital
input stage via register P5DIDIS (see description below). This avoids undesired cross
currents and switching noise while the (analog) input signal level is between VIL and VIH.
Some pins of Port 5 also serve as external GPT timer control lines.
Table 7-5 summarizes the alternate functions of Port 5.
User’s Manual
7-32
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Table 7-5
Port 5 Pin
Alternate Functions of Port 5
Alternate Function a) Alternate Function b)
P5.0
P5.1
P5.2
P5.3
P5.4
P5.5
P5.6
P5.7
Analog Input AN0
Analog Input AN1
Analog Input AN2
Analog Input AN3
Analog Input AN4
Analog Input AN5
Analog Input AN6
Analog Input AN7
Analog Input AN8
Analog Input AN9
Analog Input AN10
Analog Input AN11
Analog Input AN12
Analog Input AN13
Analog Input AN14
Analog Input AN15
–
–
–
–
–
–
–
–
P5.8
P5.7
–
–
P5.10
P5.11
P5.12
P5.13
P5.14
P5.15
T6EUD
T5EUD
T6IN
T5IN
T4EUD
T2EUD
Timer 6 ext. Up/Down Input
Timer 5 ext. Up/Down Input
Timer 6 Count Input
Timer 5 Count Input
Timer 4 ext. Up/Down Input
Timer 2 ext. Up/Down Input
Alternate Function
a)
b)
P5.15
P5.14
P5.13
P5.12
P5.11
P5.10
P5.9
AN15
T2EUD
T4EUD
T5IN
T6IN
T5EUD
T6EUD
AN14
AN13
AN12
AN11
AN10
AN9
AN8
AN7
AN6
AN5
AN4
AN3
AN2
AN1
AN0
P5.8
P5.7
Port 5
P5.6
P5.5
P5.4
P5.3
P5.2
P5.1
P5.0
General Purpose
Input
A/D Converter
Input
Timer Control
Input
MCA04356
Figure 7-18 Port 5 IO and Alternate Functions
User’s Manual
7-33
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Port 5 Digital Input Control
Port 5 pins may be used for both digital an analog input. By setting the respective bit in
register P5DIDIS the digital input stage of the respective Port 5 pin can be disconnected
from the pin. This is recommended when the pin is to be used as analog input, as it
reduces the current through the digital input stage and prevents it from toggling while the
(analog) input level is between the digital low and high thresholds. So the consumed
power and the generated noise can be reduced.
After reset all digital inputs are enabled.
P5DIDIS
Port 5 Digital Inp.Disable Reg. SFR (FFA4 /D2 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
P5D P5D P5D P5D P5D P5D P5D P5D P5D P5D P5D P5D P5D P5D P5D P5D
.15 .14 .13 .12 .11 .10
.9
.8
.7
.6
.5
.4
.3
.2
.1
.0
rw rw rw rw rw rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
Bit
P5D.y
Function
Port 5 Bit y Digital Input Control
P5D.y = 0:
P5D.y = 1:
Digital input stage connected to port line P5.y
Digital input stage disconnected from port line P5.y
When being read or used as alternate input this line
appears as ‘1’.
Port 5 pins have a special port structure (see Figure 7-19), first because it is an input
only port, and second because the analog input channels are directly connected to the
pins rather than to the input latches.
User’s Manual
7-34
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
DigInputEN
AltDataIn
Clock
Pin
Input
Latch
ChannelSelect
AnalogInput
MCB04357
P5.15-0
Figure 7-19 Block Diagram of a Port 5 Pin
Note: The “AltDataIn” line does not exist on all Port 5 inputs.
User’s Manual
7-35
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.10
Port 6
If this 8-bit port is used for general purpose IO, the direction of each line can be
configured via the corresponding direction register DP6. Each port line can be switched
into push/pull or open drain mode via the open drain control register ODP6.
P6
Port 6 Data Register
SFR (FFCC /E6 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
P6.7 P6.6 P6.5 P6.4 P6.3 P6.2 P6.1 P6.0
rw rw rw rw rw rw rw rw
-
-
-
-
-
-
Bit
Function
Port data register P6 bit y
P6.y
DP6
P6 Direction Ctrl. Register
SFR (FFCE /E7 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
DP6 DP6 DP6 DP6 DP6 DP6 DP6 DP6
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
DP6.y
Function
Port direction register DP6 bit y
DP6.y = 0: Port line P6.y is an input (high-impedance)
DP6.y = 1: Port line P6.y is an output
User’s Manual
7-36
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
ODP6
P6 Open Drain Ctrl. Reg.
ESFR (F1CE /E7 )
Reset value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
ODP6 ODP6 ODP6 ODP6 ODP6 ODP6 ODP6 ODP6
.7
.6
.5
.4
.3
.2
2.1
.0
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
ODP6.y
Function
Port 6 Open Drain control register bit y
ODP6.y = 0: Port line P6.y output driver in push/pull mode
ODP6.y = 1: Port line P6.y output driver in open drain mode
Alternate Functions of Port 6
A programmable number of chip select signals (CS4 … CS0) derived from the bus
control registers (BUSCON4 … BUSCON0) can be output on 5 pins of Port 6. The other
3 pins may be used for bus arbitration to accomodate additional masters in a C167CR
system.
The number of chip select signals is selected via PORT0 during reset. The selected
value can be read from bitfield CSSEL in register RP0H (read only) e.g. in order to check
the configuration during run time.
Table 7-6 summarizes the alternate functions of Port 6 depending on the number of
selected chip select lines (coded via bitfield CSSEL).
Table 7-6
Port 6 Pin Altern. Function Altern. Function Altern. Function Altern. Function
CSSEL = 10 CSSEL = 01 CSSEL = 00 CSSEL = 11
Alternate Functions of Port 6
P6.0
P6.1
P6.2
P6.3
P6.4
Gen. purpose IO Chip select CS0 Chip select CS0 Chip select CS0
Gen. purpose IO Chip select CS1 Chip select CS1 Chip select CS1
Gen. purpose IO Gen. purpose IO Chip select CS2 Chip select CS2
Gen. purpose IO Gen. purpose IO Gen. purpose IO Chip select CS3
Gen. purpose IO Gen. purpose IO Gen. purpose IO Chip select CS4
P6.5
P6.6
P6.7
HOLD
HLDA
BREQ
External hold request input
Hold acknowledge output (master mode) or input (slave mode)
Bus request output
User’s Manual
7-37
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Alternate Function
a)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Port 6
P6.7
BREQ
HLDA
HOLD
CS4
CS3
CS2
CS1
CS0
P6.6
P6.5
P6.4
P6.3
P6.2
P6.1
P6.0
General Purpose
Input/Output
MCA04358
Figure 7-20 Port 6 IO and Alternate Functions
The chip select lines of Port 6 additionally have an internal weak pullup device. This
device is switched on under the following conditions:
• always during reset for all potential CS output pins
• if the Port 6 line is used as a chip select output,
and the C167CR is in Hold mode (invoked through HOLD),
and the respective pin driver is in push/pull mode (ODP6.x = ‘0’).
This feature is implemented to drive the chip select lines high during reset in order to
avoid multiple chip selection, and to allow another master to access the external memory
via the same chip select lines (Wired-AND), while the C167CR is in Hold mode.
With ODP6.x = ‘1’ (open drain output selected), the internal pullup device will not be
active during Hold mode; external pullup devices must be used in this case.
When entering Hold mode the CS lines are actively driven high for one clock phase, then
the output level is controlled by the pullup devices (if activated).
After reset the CS function must be used, if selected so. In this case there is no possibility
to program any port latches before. Thus the alternate function (CS) is selected
automatically in this case.
Note: The open drain output option can only be selected via software earliest during the
initialization routine; the configured chip select lines (via CSSEL) will be in push/
pull output driver mode directly after reset.
User’s Manual
7-38
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
Open Drain
Latch
1 0
0
1
AltDir = '1'
AltEN
Pin
0
1
AltDataOut
Driver
Clock
Input
Latch
MCB04359
P6.7, P6.4-0
Figure 7-21 Block Diagram of Port 6 Pins with an Alternate Output Function
User’s Manual
7-39
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
The bus arbitration signals HOLD, HLDA and BREQ are selected with bit HLDEN in
register PSW. When the bus arbitration signals are enabled via HLDEN, also these pins
are switched automatically to the appropriate direction. Note that the pin drivers for
HLDA and BREQ are automatically controlled, while the pin driver for HOLD is
automatically disabled.
Internal Bus
Port Output
Latch
Direction
Latch
Open Drain
Latch
1 0
0
1
AltDir
AltEN
Pin
0
1
AltDataOut
Driver
Clock
AltDataIn
Input
Latch
MCB04360
P6.6
Figure 7-22 Block Diagram of Pin P6.6 (HLDA)
User’s Manual
7-40
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
Open Drain
Latch
1 0
0
1
AltDir = '0'
AltEN
Pin
Driver
Clock
AltDataIn
Input
Latch
MCB04361
P6.5
Figure 7-23 Block Diagram of Pin P6.5 (HOLD)
User’s Manual
7-41
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.11
Port 7
If this 8-bit port is used for general purpose IO, the direction of each line can be
configured via the corresponding direction register DP7. Each port line can be switched
into push/pull or open drain mode via the open drain control register ODP7.
P7
Port 7 Data Register
SFR (FFD0 /E8 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
P7.7 P7.6 P7.5 P7.4 P7.3 P7.2 P7.1 P7.0
rw rw rw rw rw rw rw rw
-
-
-
-
-
-
Bit
Function
Port data register P7 bit y
P7.y
DP7
P7 Direction Ctrl. Register
SFR (FFD2 /E9 )
Reset Value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
DP7 DP7 DP7 DP7 DP7 DP7 DP7 DP7
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
DP7.y
Function
Port direction register DP7 bit y
DP7.y = 0: Port line P7.y is an input (high-impedance)
DP7.y = 1: Port line P7.y is an output
User’s Manual
7-42
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
ODP7
P7 Open Drain Ctrl. Reg.
ESFR (F1D2 /E9 )
Reset value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
ODP7 ODP7 ODP7 ODP7 ODP7 ODP7 ODP7 ODP7
.7
.6
.5
.4
.3
.2
2.1
.0
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
ODP7.y
Function
Port 7 Open Drain control register bit y
ODP7.y = 0: Port line P7.y output driver in push/pull mode
ODP7.y = 1: Port line P7.y output driver in open drain mode
Alternate Functions of Port 7
The upper four lines of Port 7 (P7.7-4) serve as capture inputs or compare outputs
(CC31IO … CC28IO) for the CAPCOM2 unit.
The usage of the port lines by the CAPCOM unit, its accessibility via software, and the
precautions are the same as described for the Port 2 lines.
As all other capture inputs, the capture input function of pins P7.7-4 can also be used as
external interrupt inputs (sample rate 16 TCL).
The lower 4 lines of Port 7 (P7.3-0) serve as outputs from the PWM module (POUT3 …
POUT0). At these pins the value of the respective port output latch is XORed with the
value of the PWM output rather than ANDed, as the other pins do. This allows to use the
alternate output value either as it is (port latch holds a ‘0’) or invert its level at the pin (port
latch holds a ‘1’).
Note that the PWM outputs must be enabled via the respective PENx bits in PWMCON1.
Table 7-7 summarizes the alternate functions of Port 7.
Table 7-7
Port 7 Pin
Alternate Functions of Port 7
Alternate Function
P7.0
P7.1
P7.2
P7.3
P7.4
P7.5
P7.6
P7.7
POUT0
POUT1
POUT2
POUT3
CC28IO
CC29IO
CC30IO
CC31IO
PWM model channel 0 output
PWM model channel 1 output
PWM model channel 2 output
PWM model channel 3 output
Capture input/compare output channel 28
Capture input/compare output channel 29
Capture input/compare output channel 30
Capture input/compare output channel 31
User’s Manual
7-43
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Port 7
P7.7
P7.6
P7.5
P7.4
P7.3
P7.2
P7.1
P7.0
CC31IO
CC30IO
CC29IO
CC28IO
POUT3
POUT2
POUT1
POUT0
General Purpose
Input/Output
Alternate Function
MCA04362
Figure 7-24 Port 7 IO and Alternate Functions
The port structures of Port 7 differ in the way the output latches are connected to the
internal bus and to the pin driver (see Figure 7-25 and Figure 7-26).
Pins P7.3-0 (POUT3 … POUT0) of Port 7 XOR the alternate data output with the port
latch output, which allows to use the alternate data directly or inverted at the pin driver.
User’s Manual
7-44
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
Port Output
Latch
Direction
Latch
Open Drain
Latch
1 0
Pin
XOR
AltDataOut
Driver
Clock
Input
Latch
MCB04363
P7.3-0
Figure 7-25 Block Diagram of Port 7 Pins P7.3-0
User’s Manual
7-45
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Pins P7.7-4 of Port 7 combine internal bus data and alternate data output before the port
latch input, as do the Port 2 pins.
Internal Bus
CCx
0 1
Port Output
Latch
Direction
Latch
Open Drain
Latch
1 0
Pin
Driver
Clock
AltDataIn (Latch)
AltDataIn (Pin)
Input
Latch
MCB04364
P7.7-4
Figure 7-26 Block Diagram of Port 7 Pins P7.7-4
User’s Manual
7-46
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
7.12
Port 8
If this 8-bit port is used for general purpose IO, the direction of each line can be
configured via the corresponding direction register DP8. Each port line can be switched
into push/pull or open drain mode via the open drain control register ODP8.
P8
Port 8 Data Register
SFR (FFD4 /EA )
Reset Value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
P8.7 P8.6 P8.5 P8.4 P8.3 P8.2 P8.1 P8.0
-
-
-
-
-
-
rwh rwh rwh rwh rwh rwh rwh rwh
Bit
Function
Port data register P8 bit y
P8.y
DP8
P8 Direction Ctrl. Register
SFR (FFD6 /EB )
Reset Value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
DP8 DP8 DP8 DP8 DP8 DP8 DP8 DP8
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
DP8.y
Function
Port direction register DP8 bit y
DP8.y = 0: Port line P8.y is an input (high-impedance)
DP8.y = 1: Port line P8.y is an output
User’s Manual
7-47
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
ODP8
P8 Open Drain Ctrl. Reg.
ESFR (F1D6 /EB )
Reset Value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
ODP8 ODP8 ODP8 ODP8 ODP8 ODP8 ODP8 ODP8
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
rw
rw
rw
Bit
ODP8.y
Function
Port 8 Open Drain control register bit y
ODP8.y = 0: Port line P8.y output driver in push/pull mode
ODP8.y = 1: Port line P8.y output driver in open drain mode
Alternate Functions of Port 8
All Port 8 lines (P8.7-0) serve as capture inputs or compare outputs (CC23IO …
CC16IO) for the CAPCOM2 unit (see Table 7-8).
The usage of the port lines by the CAPCOM unit, its accessibility via software and the
precautions are the same as described for the Port 2 lines.
As all other capture inputs, the capture input function of pins P8.7-0 can also be used as
external interrupt inputs (sample rate 16 TCL).
Table 7-8
Port 8 Pin
Alternate Functions of Port 8
Alternate Function
P8.0
P8.1
P8.2
P8.3
P8.4
P8.5
P8.6
P8.7
CC16IO
CC17IO
CC18IO
CC19IO
CC20IO
CC21IO
CC22IO
CC23IO
Capture input/compare output channel 16
Capture input/compare output channel 17
Capture input/compare output channel 18
Capture input/compare output channel 19
Capture input/compare output channel 20
Capture input/compare output channel 21
Capture input/compare output channel 22
Capture input/compare output channel 23
User’s Manual
7-48
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Alternate Function
a)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Port 8
P8.7
CC23IO
CC22IO
CC21IO
CC20IO
CC19IO
CC18IO
CC17IO
CC16IO
P8.6
P8.5
P8.4
P8.3
P8.2
P8.1
P8.0
General Purpose
Input/Output
MCA04365
Figure 7-27 Port 8 IO and Alternate Functions
The pins of Port 8 combine internal bus data and alternate data output before the port
latch input, as do the Port 2 pins.
User’s Manual
7-49
V3.1, 2000-03
C167CR
Derivatives
Parallel Ports
Internal Bus
CCx
0 1
Port Output
Latch
Direction
Latch
Open Drain
Latch
1 0
Pin
Driver
Clock
AltDataIn (Latch)
AltDataIn (Pin)
Input
Latch
MCB04366
P8.7-0
Figure 7-28 Block Diagram of Port 8 Pins
User’s Manual
7-50
V3.1, 2000-03
C167CR
Derivatives
Dedicated Pins
8
Dedicated Pins
Most of the input/output or control signals of the functional the C167CR are realized as
alternate functions of pins of the parallel ports. There is, however, a number of signals
that use separate pins, including the oscillator, special control signals and, of course, the
power supply.
Table 8-1 summarizes the 33 dedicated pins of the C167CR.
Table 8-1
Pin(s)
C167CR Dedicated Pins
Function
ALE
Address Latch Enable
External Read Strobe
External Write/Write Low Strobe
Ready Input
RD
WR/WRL
READY
EA
External Access Enable
Non-Maskable Interrupt Input
Oscillator Input/Output
Reset Input
NMI
XTAL1, XTAL2
RSTIN
RSTOUT
Reset Output
VAREF, VAGND Power Supply for Analog/Digital Converter
VDD
VSS
OWE
Digital Power Supply (10 pins)
Digital Reference Ground (10 pins)
Oscillator Watchdog Enable
The Address Latch Enable signal ALE controls external address latches that provide
a stable address in multiplexed bus modes.
ALE is activated for every external bus cycle independent of the selected bus mode,
i.e. it is also activated for bus cycles with a demultiplexed address bus. When an external
bus is enabled (one or more of the BUSACT bits set) also X-Peripheral accesses will
generate an active ALE signal.
ALE is not activated for internal accesses, i.e. accesses to ROM/OTP/Flash (if provided),
the internal RAM and the special function registers. In single chip mode, i.e. when no
external bus is enabled (no BUSACT bit set), ALE will also remain inactive for
X-Peripheral accesses.
During reset an internal pulldown ensures an inactive (low) level on the ALE output.
User’s Manual
8-1
V3.1, 2000-03
C167CR
Derivatives
Dedicated Pins
The External Read Strobe RD controls the output drivers of external memory or
peripherals when the C167CR reads data from these external devices. During accesses
to on-chip X-Peripherals RD remains inactive (high).
During reset an internal pullup ensures an inactive (high) level on the RD output.
The External Write Strobe WR/WRL controls the data transfer from the C167CR to an
external memory or peripheral device. This pin may either provide an general WR signal
activated for both byte and word write accesses, or specifically control the low byte of an
external 16-bit device (WRL) together with the signal WRH (alternate function of P3.12/
BHE). During accesses to on-chip X-Peripherals WR/WRL remains inactive (high).
During reset an internal pullup ensures an inactive (high) level on the WR/WRL output.
The Ready Input READY receives a control signal from an external memory or
peripheral device that is used to terminate an external bus cycle, provided that this
function is enabled for the current bus cycle. READY may be used as synchronous
READY or may be evaluated asynchronously. When waitstates are defined for a READY
controlled address window the READY input is not evaluated during these waitstates.
An internal pullup ensures an inactive (high) level on the READY input.
The External Access Enable Pin EA determines if the C167CR after reset starts
fetching code from the internal ROM area (EA = ‘1’) or via the external bus interface
(EA = ‘0’). Be sure to hold this input low for ROMless devices. At the end of the internal
reset sequence the EA signal is latched together with the PORT0 configuration.
The Non-Maskable Interrupt Input NMI allows to trigger a high priority trap via an
external signal (e.g. a power-fail signal). It also serves to validate the PWRDN instruction
that switches the C167CR into Power-Down mode. The NMI pin is sampled with every
CPU clock cycle to detect transitions.
The Oscillator Input XTAL1 and Output XTAL2 connect the internal Main Oscillator
to the external crystal. The oscillator provides an inverter and a feedback element. The
standard external oscillator circuitry (see Chapter 6) comprises the crystal, two low end
capacitors and series resistor to limit the current through the crystal. The main oscillator
is intended for the generation of the basic operating clock signal of the C167CR.
An external clock signal may be fed to the input XTAL1, leaving XTAL2 open or
terminating it for higher input frequencies.
User’s Manual
8-2
V3.1, 2000-03
C167CR
Derivatives
Dedicated Pins
The Reset Input RSTIN allows to put the C167CR into the well defined reset condition
either at power-up or external events like a hardware failure or manual reset. The input
voltage threshold of the RSTIN pin is raised compared to the standard pins in order to
minimize the noise sensitivity of the reset input.
In bidirectional reset mode the C167CR’s line RSTIN may be driven active by the chip
logic e.g. in order to support external equipment which is required for startup (e.g. flash
memory).
Bidirectional reset reflects internal reset sources (software, watchdog) also to the RSTIN
pin and converts short hardware reset pulses to a minimum duration of the internal reset
sequence. Bidirectional reset is enabled by setting bit BDRSTEN in register SYSCON
and changes RSTIN from a pure input to an open drain IO line. When an internal reset
is triggered by the SRST instruction or by a watchdog timer overflow or a low level is
applied to the RSTIN line, an internal driver pulls it low for the duration of the internal
reset sequence. After that it is released and is then controlled by the external circuitry
alone.
The bidirectional reset function is useful in applications where external devices require
a defined reset signal but cannot be connected to the C167CR’s RSTOUT signal, e.g.
an external flash memory which must come out of reset and deliver code well before
RSTOUT can be deactivated via EINIT.
The following behavior differences must be observed when using the bidirectional reset
feature in an application:
• Bit BDRSTEN in register SYSCON cannot be changed after EINIT and is cleared
automatically after a reset.
• The reset indication flags always indicate a long hardware reset.
• The PORT0 configuration is treated like on a hardware reset. Especially the bootstrap
loader may be activated when P0L.4 is low.
• PinRSTINmay only be connected to external reset devices with an open drain output driver.
• A short hardware reset is extended to the duration of the internal reset sequence.
The Reset Output RSTOUT provides a special reset signal for external circuitry.
RSTOUT is activated at the beginning of the reset sequence, triggered via RSTIN, a
watchdog timer overflow or by the SRST instruction. RSTOUT remains active (low) until
the EINIT instruction is executed. This allows to initialize the controller before the
external circuitry is activated.
Note: During emulation mode pin RSTOUT is used as an input and therefore must be
driven by the external circuitry.
User’s Manual
8-3
V3.1, 2000-03
C167CR
Derivatives
Dedicated Pins
The Power Supply pins for the Analog/Digital Converter VAREF and VAGND
provide a separate power supply (reference voltage) for the on-chip ADC. This reduces
the noise that is coupled to the analog input signals from the digital logic sections and so
improves the stability of the conversion results, when VAREF and VAGND are properly
discoupled from VDD and VSS.
The Power Supply pins VDD and VSS provide the power supply for the digital logic of
the C167CR. The respective VDD/VSS pairs should be decoupled as close to the pins
as possible. For best results it is recommended to implement two-level decoupling, e.g.
(the widely used) 100 nF in parallel with 30 … 40 pF capacitors which deliver the peak
currents.
Note: All VDD pins and all VSS pins must be connected to the power supply and ground,
respectively.
The Oscillator Watchdog Enable input OWE enables the oscillator watchdog, when
driven high or disables it, when driven low e.g. for testing purposes. An internal pullup
device holds this input high if nothing is driving it.
For normal operation pin OWE should be high or not connected. The oscillator watchdog
can be disabled via software by setting bit OWDDIS in register SYSCON.
User’s Manual
8-4
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9
The External Bus Interface
Although the C167CR provides a powerful set of on-chip peripherals and on-chip RAM
and ROM/OTP/Flash (except for ROMless versions) areas, these internal units only
cover a small fraction of its address space of up to 16 MByte. The external bus interface
allows to access external peripherals and additional volatile and non-volatile memory.
The external bus interface provides a number of configurations, so it can be tailored to
fit perfectly into a given application system.
Ports & Direction Control
Alternate Functions
Address Registers
Mode Registers
Control Registers
P0L / P0H
P1L / P1H
DP3
BUSCON0
BUSCON1
BUSCON2
BUSCON3
BUSCON4
SYSCON
RP0H
ADDRSEL1
ADDRSEL2
ADDRSEL3
ADDRSEL4
P3
P4
E
ODP6
DP6
P6
PORT0 EA
PORT1 RSTIN
ALE READY
RD
WR/WRL
BHE/WRH
P0L/P0H PORT0 Data Register
P1L/P1H PORT1 Data Register
ADDRSELx Address Range Select Register 1...4
BUSCONx Bus Mode Control Register 0...4
DP3
P3
Port 3 Direction Control Register
Port 3 Data Register
SYSCON
RP0H
System Control Register
Port P0H Reset Configuration Register
P4
Port 4 Data Register
ODP6
DP6
P6
Port 6 Open Drain Control Register
Port 6 Direction Control Register
Port 6 Data Register
MCA04367
Figure 9-1
SFRs and Port Pins Associated with the External Bus Interface
Accesses to external memory or peripherals are executed by the integrated External Bus
Controller (EBC). The function of the EBC is controlled via the SYSCON register and the
BUSCONx and ADDRSELx registers. The BUSCONx registers specify the external bus
cycles in terms of address (mux/demux), data width (16-bit/8-bit), chip selects and length
(waitstates/READY control/ALE/RW delay). These parameters are used for accesses
within a specific address area which is defined via the corresponding register
ADDRSELx.
The four pairs BUSCON1/ADDRSEL1 … BUSCON4/ADDRSEL4 allow to define four
independent “address windows”, while all external accesses outside these windows are
controlled via register BUSCON0.
User’s Manual
9-1
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.1
Single Chip Mode
Single chip mode is entered, when pin EA is high during reset. In this case register
BUSCON0 is initialized with 0000 , which also resets bit BUSACT0, so no external bus
H
is enabled.
In single chip mode the C167CR operates only with and out of internal resources. No
external bus is configured and no external peripherals and/or memory can be accessed.
Also no port lines are occupied for the bus interface. When running in single chip mode,
however, external access may be enabled by configuring an external bus under software
control. Single chip mode allows the C167CR to start execution out of the internal
program memory (Mask-ROM, OTP or Flash memory).
Note: Any attempt to access a location in the external memory space in single chip mode
results in the hardware trap ILLBUS if no external bus has been explicitly enabled
by software.
User’s Manual
9-2
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.2
External Bus Modes
When the external bus interface is enabled (bit BUSACTx = ‘1’) and configured (bitfield
BTYP), the C167CR uses a subset of its port lines together with some control lines to
build the external bus.
Table 9-1
Summary of External Bus Modes
External Data Bus Width
BTYP
External Address Bus Mode
Encoding
0 0
0 1
1 0
1 1
8-bit Data
8-bit Data
16-bit Data
16-bit Data
Demultiplexed Addresses
Multiplexed Addresses
Demultiplexed Addresses
Multiplexed Addresses
The bus configuration (BTYP) for the address windows (BUSCON4 … BUSCON1) is
selected via software typically during the initialization of the system.
The bus configuration (BTYP) for the default address range (BUSCON0) is selected via
PORT0 during reset, provided that pin EA is low during reset. Otherwise BUSCON0 may
be programmed via software just like the other BUSCON registers.
The 16 MByte address space of the C167CR is divided into 256 segments of 64 KByte
each. The 16-bit intra-segment address is output on PORT0 for multiplexed bus modes
or on PORT1 for demultiplexed bus modes. When segmentation is disabled, only one
64 KByte segment can be used and accessed. Otherwise additional address lines may
be output on Port 4 (addressing up to 16 MByte) and/or several chip select lines may be
used to select different memory banks or peripherals. These functions are selected
during reset via bitfields SALSEL and CSSEL of register RP0H, respectively.
Note: Bit SGTDIS of register SYSCON defines, if the CSP register is saved during
interrupt entry (segmentation active) or not (segmentation disabled).
User’s Manual
9-3
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Multiplexed Bus Modes
In the multiplexed bus modes the 16-bit intra-segment address as well as the data use
PORT0. The address is time-multiplexed with the data and has to be latched externally.
The width of the required latch depends on the selected data bus width, i.e. an 8-bit data
bus requires a byte latch (the address bits A15 … A8 on P0H do not change, while P0L
multiplexes address and data), a 16-bit data bus requires a word latch (the least
significant address line A0 is not relevant for word accesses).
The upper address lines (An … A16) are permanently output on Port 4 (if segmentation
is enabled) and do not require latches.
The EBC initiates an external access by generating the Address Latch Enable signal
(ALE) and then placing an address on the bus. The falling edge of ALE triggers an
external latch to capture the address. After a period of time during which the address
must have been latched externally, the address is removed from the bus. The EBC now
activates the respective command signal (RD, WR, WRL, WRH). Data is driven onto the
bus either by the EBC (for write cycles) or by the external memory/peripheral (for read
cycles). After a period of time, which is determined by the access time of the memory/
peripheral, data become valid.
Read cycles: Input data is latched and the command signal is now deactivated. This
causes the accessed device to remove its data from the bus which is then tri-stated
again.
Write cycles: The command signal is now deactivated. The data remain valid on the bus
until the next external bus cycle is started.
Bus Cycle
Segment (P4)
Address
ALE
BUS (P0)
Address
Address
Data/Instr.
RD
BUS (P0)
WR
Data
MCT02060
Figure 9-2
Multiplexed Bus Cycle
User’s Manual
9-4
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Demultiplexed Bus Modes
In the demultiplexed bus modes the 16-bit intra-segment address is permanently output
on PORT1, while the data uses PORT0 (16-bit data) or P0L (8-bit data).
The upper address lines are permanently output on Port 4 (if selected via SALSEL
during reset). No address latches are required.
The EBC initiates an external access by placing an address on the address bus. After a
programmable period of time the EBC activates the respective command signal (RD,
WR, WRL, WRH). Data is driven onto the data bus either by the EBC (for write cycles)
or by the external memory/peripheral (for read cycles). After a period of time, which is
determined by the access time of the memory/peripheral, data become valid.
Read cycles: Input data is latched and the command signal is now deactivated. This
causes the accessed device to remove its data from the data bus which is then tri-stated
again.
Write cycles: The command signal is now deactivated. If a subsequent external bus
cycle is required, the EBC places the respective address on the address bus. The data
remain valid on the bus until the next external bus cycle is started.
Figure 9-3
Demultiplexed Bus Cycle
User’s Manual
9-5
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Switching Between the Bus Modes
The EBC allows to switch between different bus modes dynamically, i.e. subsequent
external bus cycles may be executed in different ways. Certain address areas may use
multiplexed or demultiplexed buses or use READY control or predefined waitstates.
A change of the external bus characteristics can be initiated in two different ways:
Reprogramming the BUSCON and/or ADDRSEL registers allows to either change
the bus mode for a given address window, or change the size of an address window that
uses a certain bus mode. Reprogramming allows to use a great number of different
address windows (more than BUSCONs are available) on the expense of the overhead
for changing the registers and keeping appropriate tables.
Switching between predefined address windows automatically selects the bus mode
that is associated with the respective window. Predefined address windows allow to use
different bus modes without any overhead, but restrict their number to the number of
BUSCONs. However, as BUSCON0 controls all address areas, which are not covered
by the other BUSCONs, this allows to have gaps between these windows, which use the
bus mode of BUSCON0.
PORT1 will output the intra-segment address, when any of the BUSCON registers
selects a demultiplexed bus mode, even if the current bus cycle uses a multiplexed bus
mode. This allows to have an external address decoder connected to PORT1 only, while
using it for all kinds of bus cycles.
Note: Never change the configuration for an address area that currently supplies the
instruction stream. Due to the internal pipelining it is very difficult to determine the
first instruction fetch that will use the new configuration. Only change the
configuration for address areas that are not currently accessed. This applies to
BUSCON registers as well as to ADDRSEL registers.
The usage of the BUSCON/ADDRSEL registers is controlled via the issued addresses.
When an access (code fetch or data) is initiated, the respective generated physical
address defines, if the access is made internally, uses one of the address windows
defined by ADDRSEL4 … 1, or uses the default configuration in BUSCON0. After
initializing the active registers, they are selected and evaluated automatically by
interpreting the physical address. No additional switching or selecting is necessary
during run time, except when more than the four address windows plus the default is to
be used.
Switching from demultiplexed to multiplexed bus mode represents a special case.
The bus cycle is started by activating ALE and driving the address to Port 4 and PORT1
as usual, if another BUSCON register selects a demultiplexed bus. However, in the
multiplexed bus modes the address is also required on PORT0. In this special case the
address on PORT0 is delayed by one CPU clock cycle, which delays the complete
(multiplexed) bus cycle and extends the corresponding ALE signal (see Figure 9-4).
User’s Manual
9-6
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
This extra time is required to allow the previously selected device (via demultiplexed bus)
to release the data bus, which would be available in a demultiplexed bus cycle.
Figure 9-4
Switching from Demultiplexed to Multiplexed Bus Mode
User’s Manual
9-7
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
External Data Bus Width
The EBC can operate on 8-bit or 16-bit wide external memory/peripherals. A 16-bit data
bus uses PORT0, while an 8-bit data bus only uses P0L, the lower byte of PORT0. This
saves on address latches, bus transceivers, bus routing and memory cost on the
expense of transfer time. The EBC can control word accesses on an 8-bit data bus as
well as byte accesses on a 16-bit data bus.
Word accesses on an 8-bit data bus are automatically split into two subsequent byte
accesses, where the low byte is accessed first, then the high byte. The assembly of bytes
to words and the disassembly of words into bytes is handled by the EBC and is
transparent to the CPU and the programmer.
Byte accesses on a 16-bit data bus require that the upper and lower half of the memory
can be accessed individually. In this case the upper byte is selected with the BHEsignal,
while the lower byte is selected with the A0 signal. So the two bytes of the memory can
be enabled independent from each other, or together when accessing words.
When writing bytes to an external 16-bit device, which has a single CS input, but two WR
enable inputs (for the two bytes), the EBC can directly generate these two write control
signals. This saves the external combination of the WR signal with A0 or BHE. In this
case pin WR serves as WRL (write low byte) and pin BHE serves as WRH (write high
byte). Bit WRCFG in register SYSCON selects the operating mode for pins WR and
BHE. The respective byte will be written on both data bus halfs.
When reading bytes from an external 16-bit device, whole words may be read and the
C167CR automatically selects the byte to be input and discards the other. However, care
must be taken when reading devices that change state when being read, like FIFOs,
interrupt status registers, etc. In this case individual bytes should be selected using BHE
and A0.
Table 9-2
Bus Mode versus Performance
Bus Mode
Transfer Rate
(Speed factor for
System Requirements
Free IO
Lines
byte/word/dword access)
8-bit Multiplexed Very low
8-bit Demultipl. Low
16-bit Multiplexed High
(1.5/3/6)
(1/2/4)
Low (8-bit latch, byte bus)
P1H, P1L
Very low (no latch, byte bus) P0H
(1.5/1.5/3)
High (16-bit latch, word bus) P1H, P1L
16-bit Demultipl. Very high (1/1/2)
Low (no latch, word bus)
–
Note: PORT1 becomes available for general purpose IO, when none of the BUSCON
registers selects a demultiplexed bus mode.
Disable/Enable Control for Pin BHE (BYTDIS)
Bit BYTDIS is provided for controlling the active low Byte High Enable (BHE) pin. The
User’s Manual
9-8
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
function of the BHE pin is enabled, if the BYTDIS bit contains a ‘0’. Otherwise, it is
disabled and the pin can be used as standard IO pin. The BHE pin is implicitly used by
the External Bus Controller to select one of two byte-organized memory chips, which are
connected to the C167CR via a word-wide external data bus. After reset the BHE
function is automatically enabled (BYTDIS = ‘0’), if a 16-bit data bus is selected during
reset, otherwise it is disabled (BYTDIS = ‘1’). It may be disabled, if byte access to 16-bit
memory is not required, and the BHE signal is not used.
Segment Address Generation
During external accesses the EBC generates a (programmable) number of address lines
on Port 4, which extend the 16-bit address output on PORT0 or PORT1 and so increase
the accessible address space. The number of segment address lines is selected during
reset and coded in bit field SALSEL in register RP0 (see Table 9-3).
H
Table 9-3
SALSEL
1 1
Decoding of Segment Address Lines
Segment Address Lines
Two: A17 … A16
Eight: A23 … A16
None
Directly accessible Address Space
256
16
64
1
KByte (Default without pull-downs)
MByte (Maximum)
KByte (Minimum)
1 0
0 1
0 0
Four: A19 … A16
MByte
Note: The total accessible address space may be increased by accessing several banks
which are distinguished by individual chip select lines.
User’s Manual
9-9
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
CS Signal Generation
During external accesses the EBC can generate a (programmable) number of CS lines
on Port 6, which allow to directly select external peripherals or memory banks without
requiring an external decoder. The number of CS lines is selected during reset and
coded in bit field CSSEL in register RP0H (see Table 9-4).
Table 9-4
CSSEL
1 1
Decoding of Chip Select Lines
Chip Select Lines
Five: CS4 … CS0
None
Note
Default without pull-downs
Port 6 pins free for IO
1 0
0 1
Two: CS1 … CS0
Three: CS2 … CS0
0 0
The CSx outputs are associated with the BUSCONx registers and are driven active (low)
for any access within the address area defined for the respective BUSCON register. For
any access outside this defined address area the respective CSx signal will go inactive
(high). At the beginning of each external bus cycle the corresponding valid CS signal is
determined and activated. All other CS lines are deactivated (driven high) at the same
time.
Note: The CSx signals will not be updated for an access to any internal address area
(i.e. when no external bus cycle is started), even if this area is covered by the
respective ADDRSELx register. An access to an on-chip X-Peripheral deactivates
all external CS signals.
Upon accesses to address windows without a selected CS line all selected CS
lines are deactivated.
The chip select signals allow to be operated in four different modes (see Table 9-5)
which are selected via bits CSWENx and CSRENx in the respective BUSCONx register.
Table 9-5
Chip Select Generation Modes
CSWENx
CSRENx
Chip Select Mode
0
0
1
1
0
1
0
1
Address Chip Select (Default after Reset)
Read Chip Select
Write Chip Select
Read/Write Chip Select
User’s Manual
9-10
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Read or Write Chip Select signals remain active only as long as the associated control
signal (RD or WR) is active. This also includes the programmable read/write delay. Read
chip select is only activated for read cycles, write chip select is only activated for write
cycles, read/write chip select is activated for both read and write cycles (write cycles are
assumed, if any of the signals WRH or WRL gets active). These modes save external
glue logic, when accessing external devices like latches or drivers that only provide a
single enable input.
Address Chip Select signals remain active during the complete bus cycle. For address
chip select signals two generation modes can be selected via bit CSCFG in register
SYSCON:
• A latched address chip select signal (CSCFG = ‘0’) becomes active with the falling
edge of ALE and becomes inactive at the beginning of an external bus cycle that
accesses a different address window. No spikes will be generated on the chip select
lines and no changes occur as long as locations within the same address window or
within internal memory (excluding X-Peripherals and XRAM) are accessed.
• An early address chip select signal (CSCFG = ‘1’) becomes active together with the
address and BHE (if enabled) and remains active until the end of the current bus cycle.
Early address chip select signals are not latched internally and may toggle
intermediately while the address is changing.
Note: CS0 provides a latched address chip select directly after reset (except for single
chip mode) when the first instruction is fetched.
Internal pullup devices hold all CS lines high during reset. After the end of a reset
sequence the pullup devices are switched off and the pin drivers control the pin levels
on the selected CS lines. Not selected CS lines will enter the high-impedance state and
are available for general purpose IO.
The pullup devices are also active during bus hold on the selected CS lines, while HLDA
is active and the respective pin is switched to push/pull mode. Open drain outputs will
float during bus hold. In this case external pullup devices are required or the new bus
master is responsible for driving appropriate levels on the CS lines.
Segment Address versus Chip Select
The external bus interface of the C167CR supports many configurations for the external
memory. By increasing the number of segment address lines the C167CR can address
a linear address space of 256 KByte, 1 MByte or 16 MByte. This allows to implement a
large sequential memory area, and also allows to access a great number of external
devices, using an external decoder. By increasing the number of CS lines the C167CR
can access memory banks or peripherals without external glue logic. These two features
may be combined to optimize the overall system performance.
Note: Bit SGTDIS of register SYSCON defines, if the CSP register is saved during
interrupt entry (segmentation active) or not (segmentation disabled).
User’s Manual
9-11
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.3
Programmable Bus Characteristics
Important timing characteristics of the external bus interface have been made user
programmable to allow to adapt it to a wide range of different external bus and memory
configurations with different types of memories and/or peripherals.
The following parameters of an external bus cycle are programmable:
• ALE Control defines the ALE signal length and the address hold time after its falling
edge
• Memory Cycle Time (extendable with 1 … 15 waitstates) defines the allowable
access time
• Memory Tri-State Time (extendable with 1 waitstate) defines the time for a data
driver to float
• Read/Write Delay Time defines when a command is activated after the falling edge
of ALE
• READY Control defines, if a bus cycle is terminated internally or externally
Note: Internal accesses are executed with maximum speed and therefore are not
programmable.
External accesses use the slowest possible bus cycle after reset. The bus cycle
timing may then be optimized by the initialization software.
ALE
ADDR
RD/WR
DATA
ALE
ADDR
RD/WR
DATA
ALECTL
MCTC
MTTC
MCD02225
Figure 9-5
Programmable External Bus Cycle
User’s Manual
9-12
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
ALE Length Control
The length of the ALE signal and the address hold time after its falling edge are
controlled by the ALECTLx bits in the BUSCON registers. When bit ALECTL is set to ‘1’,
external bus cycles accessing the respective address window will have their ALE signal
prolonged by half a CPU clock (1 TCL). Also the address hold time after the falling edge
of ALE (on a multiplexed bus) will be prolonged by half a CPU clock, so the data transfer
within a bus cycle refers to the same CLKOUT edges as usual (i.e. the data transfer is
delayed by one CPU clock). This allows more time for the address to be latched.
Note: ALECTL0 is ‘1’ after reset to select the slowest possible bus cycle, the other
ALECTLx are ‘0’ after reset.
Figure 9-6
ALE Length Control
User’s Manual
9-13
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Programmable Memory Cycle Time
The C167CR allows the user to adjust the controller’s external bus cycles to the access
time of the respective memory or peripheral. This access time is the total time required
to move the data to the destination. It represents the period of time during which the
controller’s signals do not change.
Bus Cycle
Segment
Address
ALE
BUS (P0)
Data/Instr.
Address
Address
RD
BUS (P0)
WR
Data
MCTC Wait States (1...15)
MCT02063
Figure 9-7
Memory Cycle Time
The external bus cycles of the C167CR can be extended for a memory or peripheral,
which cannot keep pace with the controller’s maximum speed, by introducing wait states
during the access (see Figure 9-7). During these memory cycle time wait states, the
CPU is idle, if this access is required for the execution of the current instruction.
The memory cycle time wait states can be programmed in increments of one CPU clock
(2 TCL) within a range from 0 to 15 (default after reset) via the MCTC fields of the
BUSCON registers. 15 – <MCTC> waitstates will be inserted.
User’s Manual
9-14
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Programmable Memory Tri-state Time
The C167CR allows the user to adjust the time between two subsequent external
accesses to account for the tri-state time of the external device. The tri-state time defines,
when the external device has released the bus after deactivation of the read command
(RD).
Bus Cycle
Segment
Address
ALE
BUS (P0)
Data/Instr.
Address
RD
MTTC Wait State
MCT02065
Figure 9-8
Memory Tri-state Time
The output of the next address on the external bus can be delayed for a memory or
peripheral, which needs more time to switch off its bus drivers, by introducing a wait state
after the previous bus cycle (see Figure 9-8). During this memory tri-state time wait
state, the CPU is not idle, so CPU operations will only be slowed down if a subsequent
external instruction or data fetch operation is required during the next instruction cycle.
The memory tri-state time waitstate requires one CPU clock (2 TCL) and is controlled via
the MTTCx bits of the BUSCON registers. A waitstate will be inserted, if bit MTTCx is ‘0’
(default after reset).
Note: External bus cycles in multiplexed bus modes implicitly add one tri-state time
waitstate in addition to the programmable MTTC waitstate.
User’s Manual
9-15
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Read/Write Signal Delay
The C167CR allows the user to adjust the timing of the read and write commands to
account for timing requirements of external peripherals. The read/write delay controls the
time between the falling edge of ALE and the falling edge of the command. Without read/
write delay the falling edges of ALE and command(s) are coincident (except for
propagation delays). With the delay enabled, the command(s) become active half a CPU
clock (1 TCL) after the falling edge of ALE.
The read/write delay does not extend the memory cycle time, and does not slow down
the controller in general. In multiplexed bus modes, however, the data drivers of an
external device may conflict with the C167CR’s address, when the early RD signal is
used. Therefore multiplexed bus cycles should always be programmed with read/write
delay.
The read/write delay is controlled via the RWDCx bits in the BUSCON registers. The
command(s) will be delayed, if bit RWDCx is ‘0’ (default after reset).
Bus Cycle
Segment
Address
ALE
1)
BUS (P0)
Data/Instr.
RD
BUS (P0)
WR
Address
Data
Read/Write
Delay
1)
The Data drivers from the previous bus cycle should be disabled when the RD signal becomes active.
MCT02066
Figure 9-9
Read/Write Signal Duration Control
User’s Manual
9-16
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.4
READY Controlled Bus Cycles
For situations, where the programmable waitstates are not enough, or where the
response (access) time of a peripheral is not constant, the C167CR provides external
bus cycles that are terminated via a READY input signal (synchronous or
asynchronous). In this case the C167CR first inserts a programmable number of
waitstates (0 … 7) and then monitors the READY line to determine the actual end of the
current bus cycle. The external device drives READY low in order to indicate that data
have been latched (write cycle) or are available (read cycle).
Figure 9-10 READY Controlled Bus Cycles
The READY function is enabled via the RDYENx bits in the BUSCON registers. When
this function is selected (RDYENx = ‘1’), only the lower 3 bits of the respective MCTC bit
field define the number of inserted waitstates (0 … 7), while the MSB of bit field MCTC
selects the READY operation:
MCTC.3 = ‘0’: Synchronous READY, i.e. the READY signal must meet setup and hold
times.
MCTC.3 = ‘1’: Asynchronous READY, i.e. the READY signal is synchronized internally.
User’s Manual
9-17
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
The Synchronous READY provides the fastest bus cycles, but requires setup and hold
times to be met. The CLKOUT signal should be enabled and may be used by the
peripheral logic to control the READY timing in this case.
The Asynchronous READY is less restrictive, but requires additional waitstates caused
by the internal synchronization. As the asynchronous READY is sampled earlier (see
Figure 9-10) programmed waitstates may be necessary to provide proper bus cycles
(see also notes on “normally-ready” peripherals below).
A READY signal (especially asynchronous READY) that has been activated by an
external device may be deactivated in response to the trailing (rising) edge of the
respective command (RD or WR).
Note: When the READY function is enabled for a specific address window, each bus
cycle within this window must be terminated with an active READY signal.
Otherwise the controller hangs until the next reset. A timeout function is only
provided by the watchdog timer.
Combining the READY function with predefined waitstates is advantageous in two
cases:
Memory components with a fixed access time and peripherals operating with READY
may be grouped into the same address window. The (external) waitstate control logic in
this case would activate READY either upon the memory’s chip select or with the
peripheral’s READY output. After the predefined number of waitstates the C167CR will
check its READY line to determine the end of the bus cycle. For a memory access it will
be low already (see example a) in Figure 9-10), for a peripheral access it may be
delayed (see example b) in Figure 9-10). As memories tend to be faster than
peripherals, there should be no impact on system performance.
When using the READY function with so-called “normally-ready” peripherals, it may lead
to erroneous bus cycles, if the READY line is sampled too early. These peripherals pull
their READY output low, while they are idle. When they are accessed, they deactivate
READY until the bus cycle is complete, then drive it low again. If, however, the peripheral
deactivates READY after the first sample point of the C167CR, the controller samples
an active READY and terminates the current bus cycle, which, of course, is too early. By
inserting predefined waitstates the first READY sample point can be shifted to a time,
where the peripheral has safely controlled the READY line (e.g. after 2 waitstates in
Figure 9-10).
User’s Manual
9-18
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.5
Controlling the External Bus Controller
A set of registers controls the functions of the EBC. General features like the usage of
interface pins (WR, BHE), segmentation and internal ROM mapping are controlled via
register SYSCON. The properties of a bus cycle like chip select mode, usage of READY,
length of ALE, external bus mode, read/write delay and waitstates are controlled via
registers BUSCON4 … BUSCON0. Four of these registers (BUSCON4 … BUSCON1)
have an address select register (ADDRSEL4 … ADDRSEL1) associated with them,
which allows to specify up to four address areas and the individual bus characteristics
within these areas. All accesses that are not covered by these four areas are then
controlled via BUSCON0. This allows to use memory components or peripherals with
different interfaces within the same system, while optimizing accesses to each of them.
SYSCON
System Control Register
SFR (FF12 /89 )
Reset value: 0XX0
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
BD
RST
EN
ROM SGT ROM BYT CLK WR CS
S1 DIS EN DIS EN CFG CFG
VISI-
BLE
STKSZ
-
rw
rw rw rwh rwh rw rwh rw
-
rw
rw
rw
Bit
XPER-SHARE XBUS Peripheral Share Mode Control
0: External accesses to XBUS peripherals are disabled
Function
1: XBUS peripherals are accessible via the ext. bus during hold mode
VISIBLE
XPEN
Visible Mode Control
0: Accesses to XBUS peripherals are done internally
1: XBUS peripheral accesses are made visible on the external pins
XBUS Peripheral Enable Bit
0: Accesses to the on-chip X-Peripherals and their functions are disabled
1: The on-chip X-Peripherals are enabled and can be accessed
BDRSTEN
Bidirectional Reset Enable Bit
0: Pin RSTIN is an input only.
1: Pin RSTIN is pulled low during the internal reset sequence
after any reset.
OWDDIS
Oscillator Watchdog Disable Bit (Cleared after reset)
0: The on-chip oscillator watchdog is enabled and active.
1: The on-chip oscillator watchdog is disabled and the CPU clock is
always fed from the oscillator input.
User’s Manual
9-19
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Bit
Function
CSCFG
Chip Select Configuration Control (Cleared after reset)
0: Latched CS mode. The CS signals are latched internally
and driven to the (enabled) port pins synchronously.
1: Unlatched CS mode. The CS signals are directly derived from
the address and driven to the (enabled) port pins.
WRCFG
CLKEN
BYTDIS
ROMEN
Write Configuration Control (Set according to pin P0H.0 during reset)
0: Pins WR and BHE retain their normal function
1: Pin WR acts as WRL, pin BHE acts as WRH
System Clock Output Enable (CLKOUT, cleared after reset)
0: CLKOUT disabled: pin may be used for general purpose IO
1: CLKOUT enabled: pin outputs the system clock signal
Disable/Enable Control for Pin BHE (Set according to data bus width)
0: Pin BHE enabled
1: Pin BHE disabled, pin may be used for general purpose IO
Internal ROM Enable (Set according to pin EA during reset)
0: Internal program memory disabled,
accesses to the ROM area use the external bus
1: Internal program memory enabled
SGTDIS
Segmentation Disable/Enable Control (Cleared after reset)
0: Segmentation enabled
(CSP is saved/restored during interrupt entry/exit)
1: Segmentation disabled (Only IP is saved/restored)
ROMS1
STKSZ
Internal ROM Mapping
0: Internal ROM area mapped to segment 0 (00’0000 … 00’7FFF )
1: Internal ROM area mapped to segment 1 (01’0000 … 01’7FFF )
H
H
H
H
System Stack Size
Selects the size of the system stack (in the internal RAM)
from 32 to 512 words
Note: Register SYSCON cannot be changed after execution of the EINIT instruction.
Bit SGTDIS controls the correct stack operation (push/pop of CSP or not) during
traps and interrupts.
The layout of the BUSCON registers and ADDRSEL registers is identical (respectively).
Registers BUSCON4 … BUSCON1, which control the selected address windows, are
completely under software control, while register BUSCON0, which e.g. is also used for
the very first code access after reset, is partly controlled by hardware, i.e. it is initialized
via PORT0 during the reset sequence. This hardware control allows to define an
appropriate external bus for systems, where no internal program memory is provided.
User’s Manual
9-20
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
BUSCON0
Bus Control Register 0
SFR (FF0C /86 )
Reset value: 0XX0
H
H
H
H
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
BUS ALE
ACT CTL
CSW CSR
EN0 EN0
RDY
EN0
MTT RWD
C0 C0
-
-
-
BTYP
MCTC
rw
0
0
rw
rw
-
rw
-
rwh rwh
-
rwh
rw
rw
BUSCON1
Bus Control Register 1
SFR (FF14 /8A )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
BUS ALE
ACT CTL
CSW CSR
EN1 EN1
RDY
EN1
MTT RWD
C1 C1
-
-
-
BTYP
MCTC
rw
1
1
rw rw
-
rw
-
rw
rw
-
rw
rw rw
BUSCON2
Bus Control Register 2
SFR (FF16 /8B )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
BUS ALE
ACT CTL
CSW CSR
EN2 EN2
RDY
EN2
MTT RWD
C2 C2
-
-
-
BTYP
MCTC
rw
2
2
rw
rw
-
rw
-
rw
rw
-
rw
rw
rw
BUSCON3
Bus Control Register 3
SFR (FF18 /8C )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
BUS ALE
CSW CSR
EN3 EN3
RDY
EN3
MTT RWD
C3 C3
-
-
ACT CTL
-
BTYP
MCTC
rw
3
3
rw
rw
-
rw
-
rw
rw
-
rw
rw
rw
BUSCON4
Bus Control Register 4
SFR (FF1A /8D )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
BUS ALE
ACT CTL
CSW CSR
EN4 EN4
RDY
EN4
MTT RWD
C4 C4
-
-
-
BTYP
rw
MCTC
rw
4
4
rw
rw
-
rw
-
rw
rw
-
rw
rw
User’s Manual
9-21
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Bit
Function
MCTC
Memory Cycle Time Control (Number of memory cycle time wait states)
0000:15 waitstates
…
(Number = 15 – <MCTC>)
1111: No waitstates
Note: The definition of bitfield MCTCx changes if RDYENx = ‘1’
(see Chapter 9.4)
RWDCx
MTTCx
BTYP
Read/Write Delay Control for BUSCONx
0:
1:
With rd/wr delay: activate command 1 TCL after falling edge of ALE
No rd/wr delay: activate command with falling edge of ALE
Memory Tristate Time Control
0:
1:
1 waitstate
No waitstate
External Bus Configuration
00: 8-bit Demultiplexed Bus
01: 8-bit Multiplexed Bus
10: 16-bit Demultiplexed Bus
11: 16-bit Multiplexed Bus
Note: For BUSCON0 BTYP is defined via PORT0 during reset.
ALECTLx ALE Lengthening Control
0:
1:
Normal ALE signal
Lengthened ALE signal
BUSACTx Bus Active Control
0:
1:
External bus disabled
External bus enabled within respective address window (ADDRSEL)
RDYENx
CSRENx
READY Input Enable
0:
1:
External bus cycle is controlled by bit field MCTC only
External bus cycle is controlled by the READY input signal
Read Chip Select Enable
0:
1:
The CS signal is independent of the read command (RD)
The CS signal is generated for the duration of the read command
CSWENx Write Chip Select Enable
0:
1:
The CS signal is independent of the write cmd. (WR,WRL,WRH)
The CS signal is generated for the duration of the write command
Note: BUSCON0 is initialized with 00C0 , if pin EA is high during reset. If pin EA is low
H
during reset, bits BUSACT0 and ALECTL0 are set (‘1’) and bit field BTYP is loaded
with the bus configuration selected via PORT0.
User’s Manual
9-22
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
ADDRSEL1
Address Select Register 1
SFR (FF18 /0C )
Reset value: 0000
H
H
H
H
H
H
15
14
13
12
11
10
9
8
7
6
5
5
5
5
4
4
4
4
3
2
1
0
RGSAD
RGSZ
rw
rw
ADDRSEL2
Address Select Register 2
SFR (FE1A /0D )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
3
2
1
0
RGSAD
RGSZ
rw
rw
ADDRSEL3
Address Select Register 3
SFR (FE1C /0E )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
3
2
1
0
RGSAD
RGSZ
rw
rw
ADDRSEL4
Address Select Register 4
SFR (FE1E /0F )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
3
2
1
0
RGSAD
RGSZ
rw
rw
Bit
RGSZ
Function
Range Size Selection
Defines the size of the address area controlled by the respective
BUSCONx/ADDRSELx register pair. See Table 9-6.
RGSAD
Range Start Address
Defines the upper bits of the start address of the respective address
area. See Table 9-6.
User’s Manual
9-23
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Note: There is no register ADDRSEL0, as register BUSCON0 controls all external
accesses outside the four address windows of BUSCON4 … BUSCON1 within the
complete address space.
Definition of Address Areas
The four register pairs BUSCON4/ADDRSEL4 … BUSCON1/ADDRSEL1 allow to define
4 separate address areas within the address space of the C167CR. Within each of these
address areas external accesses can be controlled by one of the four different bus
modes, independent of each other and of the bus mode specified in register BUSCON0.
Each ADDRSELx register in a way cuts out an address window, within which the
parameters in register BUSCONx are used to control external accesses. The range start
address of such a window defines the upper address bits, which are not used within the
address window of the specified size (see Table 9-6). For a given window size only
those upper address bits of the start address are used (marked “R”), which are not
implicitly used for addresses inside the window. The lower bits of the start address
(marked “x”) are disregarded.
Table 9-6
Address Window Definition
Bit field RGSZ Resulting Window Size Relevant Bits (R) of Start Addr. (A12 …)
0 0 0 0
0 0 0 1
0 0 1 0
0 0 1 1
0 1 0 0
0 1 0 1
0 1 1 0
0 1 1 1
1 0 0 0
1 0 0 1
1 0 1 0
1 0 1 1
1 1 x x
4 KByte
8 KByte
R R R R R R R R R R R R
R R R R R R R R R R R x
R R R R R R R R R R x x
R R R R R R R R R x x x
R R R R R R R R x x x x
R R R R R R R x x x x x
R R R R R R x x x x x x
R R R R R x x x x x x x
R R R R x x x x x x x x
R R R x x x x x x x x x
R R x x x x x x x x x x
R x x x x x x x x x x x
16 KByte
32 KByte
64 KByte
128 KByte
256 KByte
512 KByte
1 MByte
2 MByte
4 MByte
8 MByte
Reserved.
User’s Manual
9-24
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Address Window Arbitration
The address windows that can be defined within the C167CR’s address space may
partly overlap each other. Thus e.g. small areas may be cut out of bigger windows in
order to effectively utilize external resources, especially within segment 0.
For each access the EBC compares the current address with all address select registers
(programmable ADDRSELx and hardwired XADRSx). This comparison is done in four
levels.
Priority 1: The hardwired XADRSx registers are evaluated first. A match with one of
these registers directs the access to the respective X-Peripheral using the
corresponding XBCONx register and ignoring all other ADDRSELx regis-
ters.
Priority 2: Registers ADDRSEL2 and ADDRSEL4 are evaluated before ADDRSEL1
and ADDRSEL3, respectively. A match with one of these registers directs
the access to the respective external area using the corresponding BUS-
CONx register and ignoring registers ADDRSEL1/3 (see Figure 9-11).
Priority 3: A match with registers ADDRSEL1 or ADDRSEL3 directs the access to the
respective external area using the corresponding BUSCONx register.
Priority 4: If there is no match with any XADRSx or ADDRSELx register the access to
the external bus uses register BUSCON0.
XBCON0
BUSCON2
BUSCON1
BUSCON0
BUSCON4
BUSCON3
Active Window
Inactive Window
MCA04368
Figure 9-11 Address Window Arbitration
Note: Only the indicated overlaps are defined. All other overlaps lead to erroneous bus
cycles. E.g. ADDRSEL4 may not overlap ADDRSEL2 or ADDRSEL1. The
hardwired XADRSx registers are defined non-overlapping.
User’s Manual
9-25
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
RP0H
Reset Value of P0H
SFR (F108 /84 )
Reset value: - - XX
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
CLKCFG
SALSEL
CSSEL WRC
rh rh
rh
rh
Bit
WRC
Function
Write Configuration
0:
1:
Pins WR and BHE operate as WRL and WRH signals
Pins WR and BHE operate as WR and BHE signals
CSSEL
Chip Select Line Selection (Number of active CS outputs)
00: 3 CS lines: CS2 …CS0
01: 2 CS lines: CS1 … CS0
10: No CS lines at all
11: 5 CS lines: CS4 …CS0 (Default without pulldowns)
SALSEL
CLKCFG
Segment Address Line Selection (nr. of active segment addr. outputs)
00: 4-bit segment address: A19 … A16
01: No segment address lines at all
10: 8-bit segment address: A23 … A16
11: 2-bit segment address: A17 … A16 (Default without pulldowns)
Clock Generation Mode Configuration
These pins define the clock generation mode, i.e. the mechanism how the
internal CPU clock is generated from the externally applied (XTAL) input
clock.
Note: RP0H cannot be changed via software, but rather allows to check the current
configuration.
User’s Manual
9-26
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Precautions and Hints
• The ext. bus interface is enabled as long as at least one of the BUSCON registers has
its BUSACT bit set.
• PORT1 will output the intra-segment addr. as long as at least one of the BUSCON
registers selects a demultiplexed external bus, even for multiplexed bus cycles.
• Not all addr. windows defined via registers ADDRSELx may overlap each other. The
operation of the EBC will be unpredictable in such a case. See “Address Window
Arbitration” on Page 9-25.
• The addr. windows defined via registers ADDRSELx may overlap internal addr. areas.
Internal accesses will be executed in this case.
• For any access to an internal addr. area the EBC will remain inactive (see EBC Idle
State).
User’s Manual
9-27
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.6
EBC Idle State
When the external bus interface is enabled, but no external access is currently executed,
the EBC is idle. As long as only internal resources (from an architecture point of view)
like IRAM, GPRs or SFRs, etc. are used the external bus interface does not change (see
Table 9-7).
Accesses to on-chip X-Peripherals are also controlled by the EBC. However, even
though an X-Peripheral appears like an external peripheral to the controller, the
respective accesses do not generate valid external bus cycles.
Due to timing constraints address and write data of an XBUS cycle are reflected on the
external bus interface (see Table 9-7). The “address” mentioned above includes
PORT1, Port 4, BHE and ALE which also pulses for an XBUS cycle. The external CS
signals on Port 6 are driven inactive (high) because the EBC switches to an internal XCS
signal.
The external control signals (RD and WR or WRL/WRH if enabled) remain inactive (high).
Table 9-7
Pins
Status of the External Bus Interface During EBC Idle State
Internal accesses only
XBUS accesses
PORT0
Tristated (floating)
Tristated (floating) for read
accesses
XBUS write data for write accesses
PORT1
Port 4
Port 6
BHE
Last used external address
(if used for the bus interface)
Last used XBUS address
(if used for the bus interface)
1)
Last used external segment address Last used XBUS segment address
(on selected pins)
(on selected pins)
Active external CS signal
corresponding to last used address signals
Inactive (high) for selected CS
Level corresponding to last external Level corresponding to last XBUS
access
access
ALE
RD
Inactive (low)
Inactive (high)
Pulses as defined for X-Peripheral
1)
Inactive (high)
1)
WR/WRL Inactive (high)
Inactive (high)
1)
WRH
Inactive (high)
Inactive (high)
1)
Used and driven in visible mode.
User’s Manual
9-28
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.7
External Bus Arbitration
In embedded systems it may be efficient to share external resources like memory banks
or peripheral devices among more than one microcontroller or processor. The C167CR
supports this approach with the possibility to arbitrate the access to its external bus, i.e.
to the external resources. Several bus masters can therefore use the same set of
resources, resulting in compact, though powerful systems.
Note: Sharing external resources is useful if these resources are used to a limited
amount only. The performance of a bus master which relies on these external
resources to a great extent (e.g. external code) will be reduced by the bandwidth
used by the other masters.
Bus arbitration uses three control signals (HOLD, HLDA/BGR, BREQ) and can be
enabled and disabled via software (PSW.HLDEN), e.g. in order to protect time-critical
code sections from being suspended by other bus masters. A bus arbiter logic can be
designed to determine which of the bus masters controls the external system at a given
time.
Using the specific master and slave modes for bus arbitration saves external glue logic
(bus arbiter) when connecting two devices of the C166 Family.
Note: Bus arbitration does not work if there is no clock signal for the EBC, i.e. during Idle
mode and Powerdown mode.
Signals and Control
The upper three pins of Port 6 are used for the bus arbitration interface. Table 9-8
summarizes the functions of these interface lines.
The external bus arbitration is enabled by setting bit HLDEN in register PSW to ‘1’. In
this case the three bus arbitration pins HOLD, HLDA and BREQ are automatically
controlled by the EBC independent of their IO configuration.
Bit HLDEN may be cleared during the execution of program sequences, where the
external resources are required but cannot be shared with other bus masters, or during
sequences which need to access on-chip XBUS resources but which shall not be
interrupted by hold states. In this case the C167CR will not answer to HOLD requests
from other external masters. If HLDEN is cleared while the C167CR is in Hold State
(code execution from internal RAM/ROM) this Hold State is left only after HOLD has
been deactivated again. I.e. in this case the current Hold State continues and only the
next HOLD request is not answered.
Note: The pins HOLD, HLDA and BREQ keep their alternate function (bus arbitration)
even after the arbitration mechanism has been switched off by clearing HLDEN.
All three pins are used for bus arbitration after bit HLDEN was set once.
User’s Manual
9-29
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Table 9-8
Interface Pins for Bus Arbitration
Pin Function Direction Operational Description
P6.5 HOLD
INput
The hold request signal requests the external bus system
from the C167CR.
P6.6 HLDA
(Master
OUTput
The hold acknowledge signal acknowledges a hold
request and indicates to the external partners that the
C167CR has withdrawn from the bus and another
external bus master may now use it.
mode)
1)
BGR
INput
The bus grant signal indicates to the C167CR (slave) that
the master has withdrawn from the external bus in
response to the slave’s BREQ. The slave may now use
the external bus system until the master requests it back.
(Slave
mode)
P6.7 BREQ
OUTput
The bus request signal indicates to the bus arbiter logic
that the C167CR requires control over the external bus,
which has been released and is currently controlled by
another bus master.
1)
In slave mode pin HLDA inverts its direction to input. The changed functionality is indicated through the
different name.
Arbitration Sequences
An external master may request the C167CR’s bus via the HOLD input. After completing
the currently running bus cycle the C167CR acknowledges this request via the HLDA
output and will then float its bus lines (internal pullups at CSx, RD, and WR, internal
pulldown at ALE). The new master may now access the peripheral devices or memory
banks via the same interface lines as the C167CR. During this time the C167CR can
keep on executing, as long as it does not need access to the external bus. All actions
that just require internal resources like instruction or data memory and on-chip generic
peripherals, may be executed in parallel.
Note: The XBUS is an internal representation of the external bus interface. Accesses to
XBUS peripherals use the EBC and therefore also cannot be executed during the
bus hold state.
When the C167CR needs access to its external bus while it is occupied by another bus
master, it demands it via the BREQ output. The arbiter can then remove the current
master from the bus. This is indicated to the C167CR by deactivating its HOLD input.
The C167CR responds by deactivating its HLDA signal, and then taking control of the
external bus itself. Of course also the request signal BREQ is deactivated after getting
control of the bus back.
User’s Manual
9-30
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Entering the Hold State
Access to the C167CR’s external bus is requested by driving its HOLD input low. After
synchronizing this signal the C167CR will complete a current external bus cycle or XBUS
cycle (if any is active), release the external bus and grant access to it by driving the
HLDA output low. During hold state the C167CR treats the external bus interface as
follows:
• Address and data bus(es) float to tri-state
• ALE is pulled low by an internal pulldown device
• Command lines are pulled high by internal pullup devices (RD, WR)
• CSx outputs are driven high for 1 TCL and then pulled high (in push/pull mode),
or float to tri-state (in open drain mode)
Should the C167CR require access to its external bus or XBUS during hold mode, it
activates its bus request output BREQ to notify the arbitration circuitry. BREQ is
activated only during hold mode. It will be inactive during normal operation.
Figure 9-12 External Bus Arbitration, Releasing the Bus
Note: The C167CR will complete the currently running bus cycle before granting bus
access as indicated by the broken lines. This may delay hold acknowledge
compared to this figure.
Figure 9-12 shows the first possibility for BREQ to get active.
During bus hold pin BHE/WRH is floating. An external pullup should be used if this
is required.
User’s Manual
9-31
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Exiting the Hold State
The external bus master returns the access rights to the C167CR by driving the HOLD
input high. After synchronizing this signal the C167CR will drive the HLDA output high,
actively drive the control signals and resume executing external bus cycles if required.
Depending on the arbitration logic, the external bus can be returned to the C167CR
under two circumstances:
• The external master does no more require access to the shared resources and gives
up its own access rights, or
• The C167CR needs access to the shared resources and demands this by activating
its BREQ output. The arbitration logic may then activate the other master’s HOLD and
so free the external bus for the C167CR, depending on the priority of the different
masters.
Note: The Hold State is not terminated by clearing bit HLDEN.
HOLD
HLDA
BREQ
CSx
Other
Signals
MCD02236
Figure 9-13 External Bus Arbitration, Regaining the Bus
Note: The falling BREQ edge shows the last chance for BREQ to trigger the indicated
regain-sequence. Even if BREQ is activated earlier the regain-sequence is
initiated by HOLD going high. BREQ and HOLD are connected via an external
arbitration circuitry. Please note that HOLD may also be deactivated without the
C167CR requesting the bus.
User’s Manual
9-32
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
Connecting Bus Masters
When multiple bus masters (C167CRs or other masters) shall share external resources
a bus arbiter is required that determines the currently active bus master and also enables
a C167CR which has surrendered its bus interface to regain control of it in case it must
access the shared external resources.
The structure of this bus arbiter defines the degree of control the C167CR has over the
external bus system. Whenever the C167CR has released the bus, there is no way of
actively regaining control of it, besides the activation of the BREQ signal. In any case the
C167CR must wait for the deactivation of its HOLD input.
Note: The full arbitration logic is required if the “other” bus master does not automatically
remove its hold request after having used the shared resources.
External Bus System
C167CR or
other Master
C167CR or
other Master
C 67CR or
other Master
HOLD
HLDA
BREQ
Bus Arbiter Logic
MCA04369
Figure 9-14 Principle Arbitration Logic
Compact Two-Master Systems (Master/Slave Mode)
When two C167CRs (or other members of the C166 Family) are to be connected in this
way the external bus arbitration logic (normally required to combine the respective output
signals HLDA and BREQ) can be left out.
In this case one of the controllers operates in Master Mode (the standard default
operating mode, DP6.7 = ‘0’), while the other one must operate in Slave Mode (selected
with DP6.7 = ‘1’). In this configuration the master-device normally controls the external
bus, while the slave-device gets control of it on demand only. In most cases this requires
that (at least) the slave-device operates out of internal resources most of the time, in
order to get an acceptable overall system performance.
User’s Manual
9-33
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
In Slave Mode the C167CR inverts the direction of its HLDA pin and uses it as the bus
grant input BGR, while the master’s HLDA pin remains an output. This permits the direct
connection of these two signals without any additional glue logic for bus arbitration. The
BREQ outputs are mutually connected to the other partner’s HOLD input (see
Figure 9-15).
C167CR
C167CR
in Master Mode
in Slave Mode
Ext. Bus
HOLD
HOLD
HLDA
BRG
BREQ
BREQ
MCS04370
The pullups provide correct signal levels during the initialization phase.
Figure 9-15 Sharing External Resources Using Slave Mode
Slave Mode is selected by intentionally switching pin BREQ to output (DP6.7 = ‘1’).
Normally the port direction register bits for the arbitration interface pins retain their reset
value which is ‘0’. Clearing bit DP6.7 (or preserving the reset value) selects Master
Mode, where the device operates compatible with earlier versions without the slave
mode feature.
Note: If the C167CR operates in slave mode and executes a loop out of external memory
which fits completely into the jump cache (e.g. JB bitaddr, $) its BREQ output may
toggle (period = 2 CPU clock cycles). BREQ is activated by the prefetcher that
wants to read the next sequential intstruction. BREQ is the deactivated, because
the target of the taken jump is found in the jump cache. A loop of a minimum length
of 3 words avoids this.
User’s Manual
9-34
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.8
The XBUS Interface
The C167CR provides an on-chip interface (the XBUS interface), via which integrated
customer/application specific peripherals can be connected to the standard controller
core. The XBUS is an internal representation of the external bus interface, i.e. it is
operated in the same way.
For each peripheral on the XBUS (X-Peripheral) there is a separate address window
controlled by a register pair XBCONx/XADRSx (similar to registers BUSCONx and
ADDRSELx). As an interface to a peripheral in many cases is represented by just a few
registers, the XADRSx registers partly select smaller address windows than the standard
ADDRSEL registers. As the XBCONx/XADRSx register pairs control integrated
peripherals rather than externally connected ones, they are fixed by mask programming
rather than being user programmable.
X-Peripheral accesses provide the same choices as external accesses, so these
peripherals may be bytewide or wordwide. Because the on-chip connection can be
realized very efficient and for performance reasons X-Peripherals are only implemented
with a separate address bus, i.e. in demultiplexed bus mode. Interrupt nodes are
provided for X-Peripherals to be integrated.
Note: If you plan to develop a peripheral of your own to be integrated into a C167CR
device to create a customer specific version, please ask for the specification of the
XBUS interface and for further support.
User’s Manual
9-35
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.8.1
Accessing the On-chip XBUS Peripherals
Enabling of XBUS Peripherals
After reset all on-chip XBUS peripherals are disabled. In order to be usable an XBUS
peripheral must be enabled via the global enable bit XPEN in register SYSCON.
Table 9-9 summarizes the XBUS peripherals and also the number of waitstates which
are used when accessing the respective peripheral.
Table 9-9
XBUS Peripherals in the C167CR
Associated XBUS Peripheral
CAN
Waitstates
2
0
XRAM 2 KByte
Visible Mode
The C167CR can mirror on-chip access cycles to its XBUS peripherals so these
accesses can be observed or recorded by the external system. This function is enabled
via bit VISIBLE in register SYSCON.
Accesses to XBUS peripherals also use the EBC. Due to timing constraints the address
bus will change for all accesses using the EBC.
Note: As XBUS peripherals use demultiplexed bus cycles, the respective address is
driven on PORT1 in visible mode, even if the external system uses MUX buses
only.
If visible mode is activated, accesses to on-chip XBUS peripherals (including control
signals RD, WR, and BHE) are mirrored to the bus interface. Accesses to internal
resources (program memory, IRAM, GPRs) do not use the EBC and cannot be mirrored
to outside.
If visible mode is deactivated, however, no control signals (RD, WR) will be activated,
i.e. there will be no valid external bus cycles.
Note: Visible mode can only work if the external bus is enabled at all.
User’s Manual
9-36
V3.1, 2000-03
C167CR
Derivatives
The External Bus Interface
9.8.2
External Accesses to XBUS Peripherals
The on-chip XBUS peripherals of the C167CR can be accessed from outside via the
external bus interface under certain circumstances. In emulation mode the XBUS
peripherals are controlled by the bondout-chip. During normal operation this external
access is accomplished selecting the XPER-Share mode.
XPER-Share Mode
The C167CR can share its on-chip XBUS peripherals with other (external) bus masters,
i.e. it can allow them to access its X-Peripherals while it is in hold mode. This external
access is enabled via bit XPERSHARE in register SYSCON and is only possible while
the host controller is in hold mode.
During XPER-Share mode the C167CR’s bus interface inverts its direction so the
external master can drive address, control, and data signals to the respective peripheral.
This can be used e.g. to install a mailbox memory in a multi-processor system.
Note: When XPER-Share mode is disabled no accesses to on-chip XBUS peripherals
can be executed from outside.
User’s Manual
9-37
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
10
The General Purpose Timer Units
The General Purpose Timer Units GPT1 and GPT2 represent very flexible
multifunctional timer structures which may be used for timing, event counting, pulse
width measurement, pulse generation, frequency multiplication, and other purposes.
They incorporate five 16-bit timers that are grouped into the two timer blocks GPT1 and
GPT2.
Block GPT1 contains 3 timers/counters with a maximum resolution of 16 TCL, while
block GPT2 contains 2 timers/counters with a maximum resolution of 8 TCL and a 16-bit
Capture/Reload register (CAPREL). Each timer in each block may operate
independently in a number of different modes such as gated timer or counter mode, or
may be concatenated with another timer of the same block. The auxiliary timers of GPT1
may optionally be configured as reload or capture registers for the core timer. In the
GPT2 block, the additional CAPREL register supports capture and reload operation with
extended functionality, and its core timer T6 may be concatenated with timers of the
CAPCOM units (T0, T1, T7, and T8). Each block has alternate input/output functions and
specific interrupts associated with it.
10.1
Timer Block GPT1
From a programmer’s point of view, the GPT1 block is composed of a set of SFRs as
summarized below. Those portions of port and direction registers which are used for
alternate functions by the GPT1 block are shaded.
Ports & Direction Control
Alternate Functions
Data Registers
Control Registers
Interrupt Control
E
ODP3
DP3
P3
T2
T2CON
T3CON
T4CON
P5DIDIS
T2IC
T3IC
T4IC
T3
T4
P5
T2IN/P3.7 T2EUD/P5.15
T3IN/P3.6 T3EUD/P3.4
T4IN/P3.5 T4EUD/P5.14
T3OUT/P3.3
P5
Port 5 Data Register
P5DIDIS Port 5 Digital Input Disable Register
ODP3
DP3
P3
Port 3 Open Drain Control Register
Port 3 Direction Control Register
Port 3 Data Register
T2
T3
T4
GPT1 Timer 2 Register
GPT1 Timer 3 Register
GPT1 Timer 4 Register
T2CON GPT1 Timer 2 Control Register
T3CON GPT1 Timer 3 Control Register
T4CON GPT1 Timer 4 Control Register
T2IC
T3IC
T4IC
GPT1 Timer 2 Interrupt Control Register
GPT1 Timer 3 Interrupt Control Register
GPT1 Timer 4 Interrupt Control Register
MCA04371
Figure 10-1 SFRs and Port Pins Associated with Timer Block GPT1
User’s Manual
10-1
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
All three timers of block GPT1 (T2, T3, T4) can run in 4 basic modes, which are timer,
gated timer, counter and incremental interface mode, and all timers can either count up
or down. Each timer has an alternate input function pin (TxIN) associated with it which
serves as the gate control in gated timer mode, or as the count input in counter mode.
The count direction (Up/Down) may be programmed via software or may be dynamically
altered by a signal at an external control input pin. Each overflow/underflow of core timer
T3 is latched in the toggle FlipFlop T3OTL and may be indicated on an alternate output
function pin. The auxiliary timers T2 and T4 may additionally be concatenated with the
core timer, or used as capture or reload registers for the core timer.
The current contents of each timer can be read or modified by the CPU by accessing the
corresponding timer registers T2, T3, or T4, which are located in the non-bitaddressable
SFR space. When any of the timer registers is written to by the CPU in the state
immediately before a timer increment, decrement, reload, or capture is to be performed,
the CPU write operation has priority in order to guarantee correct results.
T2EUD
fCPU
T2IN
U/D
Interrupt
Request
2n : 1
GPT1 Timer T2
T2
Mode
Control
Reload
Capture
Interrupt
Request
fCPU
2n : 1
Toggle FF
T3OTL
T3
Mode
Control
T3IN
GPT1 Timer T3
T3OUT
U/D
T3EUD
Other
Timers
Capture
Reload
T4IN
T4
Mode
Control
Interrupt
Request
2n : 1
GPT1 Timer T4
U/D
fCPU
T4EUD
MCT02141
Figure 10-2 GPT1 Block Diagram
User’s Manual
10-2
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
10.1.1
GPT1 Core Timer T3
The core timer T3 is configured and controlled via its bitaddressable control register
T3CON.
T3CON
Timer 3 Control Register
SFR (FF42 /A1 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
T3
T3
T3
T3
-
-
-
-
-
T3R
T3M
T3I
OTL OE UDE UD
-
-
-
-
-
rwh rw
rw
rw
rw
rw
rw
Bit
Function
Timer 3 Input Selection
T3I
Depends on the operating mode, see respective sections.
T3M
Timer 3 Mode Control (Basic Operating Mode)
000: Timer Mode
001: Counter Mode
010: Gated Timer with Gate active low
011: Gated Timer with Gate active high
100: Reserved. Do not use this combination.
101: Reserved. Do not use this combination.
110: Incremental Interface Mode
111: Reserved. Do not use this combination.
T3R
Timer 3 Run Bit
0:
1:
Timer/Counter 3 stops
Timer/Counter 3 runs
1)
T3UD
Timer 3 Up/Down Control
1)
T3UDE
T3OE
Timer 3 External Up/Down Enable
Alternate Output Function Enable
0:
1:
Alternate Output Function Disabled
Alternate Output Function Enabled
T3OTL
Timer 3 Output Toggle Latch
Toggles on each overflow/underflow of T3. Can be set or reset by
software.
1)
For the effects of bits T3UD and T3UDE refer to the direction Table 10-1.
User’s Manual
10-3
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 3 Run Bit
The timer can be started or stopped by software through bit T3R (Timer T3 Run Bit). If
T3R = ‘0’, the timer stops. Setting T3R to ‘1’ will start the timer.In gated timer mode, the
timer will only run if T3R = ‘1’ and the gate is active (high or low, as programmed).
Count Direction Control
The count direction of the core timer can be controlled either by software or by the
external input pin T3EUD (Timer T3 External Up/Down Control Input), which is the
alternate input function of port pin P3.4. These options are selected by bits T3UD and
T3UDE in control register T3CON. When the up/down control is done by software (bit
T3UDE = ‘0’), the count direction can be altered by setting or clearing bit T3UD. When
T3UDE = ‘1’, pin T3EUD is selected to be the controlling source of the count direction.
However, bit T3UD can still be used to reverse the actual count direction, as shown in
Table 10-1. If T3UD = ‘0’ and pin T3EUD shows a low level, the timer is counting up.
With a high level at T3EUD the timer is counting down. If T3UD = ‘1’, a high level at pin
T3EUD specifies counting up, and a low level specifies counting down. The count
direction can be changed regardless of whether the timer is running or not.
When pin T3EUD/P3.4 is used as external count direction control input, it must be
configured as input, i.e. its corresponding direction control bit DP3.4 must be set to ‘0’.
Table 10-1 GPT1 Core Timer T3 Count Direction Control
Pin TxEUD
Bit TxUDE
Bit TxUD
Count Direction
Count Up
X
X
0
1
0
1
0
0
1
1
1
1
0
1
0
0
1
1
Count Down
Count Up
Count Down
Count Down
Count Up
Note: The direction control works the same for core timer T3 and for auxiliary timers T2
and T4. Therefore the pins and bits are named Tx …
User’s Manual
10-4
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 3 Output Toggle Latch
An overflow or underflow of timer T3 will clock the toggle bit T3OTL in control register
T3CON. T3OTL can also be set or reset by software. Bit T3OE (Alternate Output
Function Enable) in register T3CON enables the state of T3OTL to be an alternate
function of the external output pin T3OUT. For that purpose, a ‘1’ must be written into the
respective port data latch and pin T3OUT must be configured as output by setting the
corresponding direction control bit to ‘1’. If T3OE = ‘1’, pin T3OUT then outputs the state
of T3OTL. If T3OE = ‘0’, pin T3OUT can be used as general purpose IO pin.
In addition, T3OTL can be used in conjunction with the timer over/underflows as an input
for the counter function or as a trigger source for the reload function of the auxiliary
timers T2 and T4. For this purpose, the state of T3OTL does not have to be available at
pin T3OUT, because an internal connection is provided for this option.
Timer 3 in Timer Mode
Timer mode for the core timer T3 is selected by setting bit field T3M in register T3CON
to ‘000 ’. In this mode, T3 is clocked with the internal system clock (CPU clock) divided
B
by a programmable prescaler, which is selected by bit field T3I. The input frequency fT3
for timer T3 and its resolution rT3 are scaled linearly with lower clock frequencies fCPU
,
as can be seen from the following formula:
<T3I>
fCPU
8 × 2
fT3 =
rT3 [µs] =
,
<T3I>
fCPU [MHz]
8 × 2
Txl
2n : 1
Interrupt
Request
fCPU
Core Timer Tx
Up/
Down
TxR
TxOTL
TxOUT
TxU
D
0
TxOE
MUX
EXOR
1
TxEUD
TxUDE
MCB02028
T3EUD = P3.4
T3OUT = P3.3
x = 3
Figure 10-3 Block Diagram of Core Timer T3 in Timer Mode
User’s Manual
10-5
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
The timer input frequencies, resolution and periods which result from the selected
prescaler option are listed in Table 10-2. This table also applies to the Gated Timer
Mode of T3 and to the auxiliary timers T2 and T4 in timer and gated timer mode. Note
that some numbers may be rounded to 3 significant digits.
Table 10-2 GPT1 Timer Input Frequencies, Resolution and Periods @ 20 MHz
fCPU = 20 MHz
Timer Input Selection T2I/T3I/T4I
000
001
010
32
011
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler factor 8
16
64
128
256
512
1024
Input
Frequency
2.5
MHz
1.25
MHz
625
kHz
312.5 156.25 78.125 39.06 19.53
kHz kHz kHz kHz kHz
Resolution
Period
400 ns 800 ns 1.6 µs 3.2 µs 6.4 µs 12.8 µs 25.6 µs 51.2 µs
26.2 ms 52.5ms 105 ms 210 ms 420 ms 840 ms 1.68 s 3.36 s
Table 10-3 GPT1 Timer Input Frequencies, Resolution and Periods @ 25 MHz
fCPU = 25 MHz Timer Input Selection T2I/T3I/T4I
000
001
010
011
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler factor 8
16
32
64
128
256
512
1024
Input
Frequency
3.125 1.56
MHz MHz
781.25 390.62 195.3 97.65 48.83 24.42
kHz kHz kHz kHz kHz kHz
Resolution
Period
320 ns 640 ns 1.28 µs 2.56 µs 5.12 µs 10.2 µs 20.5 µs 41.0 µs
21.0 ms 41.9ms 83.9 ms 168 ms 336 ms 671 ms 1.34 s 2.68 s
Table 10-4 GPT1 Timer Input Frequencies, Resolution and Periods @ 33 MHz
fCPU = 33 MHz Timer Input Selection T2I/T3I/T4I
000
001
010
011
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler factor 8
16
32
64
128
256
512
1024
Input
Frequency
4.125 2.0625 1.031 515.62 257.81 128.91 64.45 32.23
MHz MHz MHz kHz kHz kHz kHz kHz
Resolution
Period
242 ns 485 ns 970 ns 1.94 µs 3.88 µs 7.76 µs 15.5 µs 31.0 µs
15.9 ms 31.8ms 63.6 ms 127 ms 254 ms 508 ms 1.02 s 2.03 s
User’s Manual
10-6
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 3 in Gated Timer Mode
Gated timer mode for the core timer T3 is selected by setting bit field T3M in register
T3CON to ‘010 ’ or ‘011 ’. Bit T3M.0 (T3CON.3) selects the active level of the gate
B
B
input. In gated timer mode the same options for the input frequency as for the timer mode
are available. However, the input clock to the timer in this mode is gated by the external
input pin T3IN (Timer T3 External Input).
To enable this operation pin T3IN must be configured as input, i.e. the corresponding
direction control bit must contain ‘0’.
TxI
fCPU
2n : 1
MUX
TxM
Core Timer Tx
TxOTL
TxOUT
TxIN
Up/
Down
TxR
TxOE
TxU
D
0
Interrupt
Request
MUX
XOR
1
TxEUD
MCB02029
TxUDE
T3IN = P3.6
T3EUD = P3.4
T3OUT = P3.3
x = 3
Figure 10-4 Block Diagram of Core Timer T3 in Gated Timer Mode
If T3M.0 = ‘0’, the timer is enabled when T3IN shows a low level. A high level at this pin
stops the timer. If T3M.0 = ‘1’, pin T3IN must have a high level in order to enable the
timer. In addition, the timer can be turned on or off by software using bit T3R. The timer
will only run, if T3R = ‘1’ and the gate is active. It will stop, if either T3R = ‘0’ or the gate
is inactive.
Note: A transition of the gate signal at pin T3IN does not cause an interrupt request.
User’s Manual
10-7
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 3 in Counter Mode
Counter mode for the core timer T3 is selected by setting bit field T3M in register T3CON
to ‘001 ’. In counter mode timer T3 is clocked by a transition at the external input pin
B
T3IN. The event causing an increment or decrement of the timer can be a positive, a
negative, or both a positive and a negative transition at this pin. Bit field T3I in control
register T3CON selects the triggering transition (see Table 10-5).
Edge
Select
TxIN
Core Timer Tx
TxOTL
TxOUT
Up/
Down
TxR
TxOE
Txl
TxU
D
0
Interrupt
Request
MUX
XOR
1
TxEUD
TxUDE
MCB02030
T3IN = P3.6
T3EUD = P3.4
T3OUT = P3.3
x = 3
Figure 10-5 Block Diagram of Core Timer T3 in Counter Mode
Table 10-5 GPT1 Core Timer T3 (Counter Mode) Input Edge Selection
T3I
Triggering Edge for Counter Increment/Decrement
None. Counter T3 is disabled
0 0 0
0 0 1
0 1 0
0 1 1
1 X X
Positive transition (rising edge) on T3IN
Negative transition (falling edge) on T3IN
Any transition (rising or falling edge) on T3IN
Reserved. Do not use this combination
For counter operation, pin T3IN must be configured as input, i.e. the respective direction
control bit DPx.y must be ‘0’. The maximum input frequency which is allowed in counter
mode is fCPU/16. To ensure that a transition of the count input signal which is applied to
T3IN is correctly recognized, its level should be held high or low for at least 8 fCPU cycles
before it changes.
User’s Manual
10-8
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 3 in Incremental Interface Mode
Incremental Interface mode for the core timer T3 is selected by setting bit field T3M in
register T3CON to ‘110 ’. In incremental interface mode the two inputs associated with
B
timer T3 (T3IN, T3EUD) are used to interface to an incremental encoder. T3 is clocked
by each transition on one or both of the external input pins which gives 2-fold or 4-fold
resolution of the encoder input.
T3IN
Edge
Select
Timer T3
T3OTL
T3OUT
T3OE
T3l
T3R
Up/
Down
Interrupt
Request
T3U
D
0
MUX
Phase
Detect
XOR
1
T3EUD
MCB04000B
T3UDE
Figure 10-6 Block Diagram of Core Timer T3 in Incremental Interface Mode
Bitfield T3I in control register T3CON selects the triggering transitions (see Table 10-6).
In this mode the sequence of the transitions of the two input signals is evaluated and
generates count pulses as well as the direction signal. So T3 is modified automatically
according to the speed and the direction of the incremental encoder and its contents
therefore always represent the encoder’s current position.
Table 10-6
T3I
GPT1 Core Timer T3 (Incremental Interface Mode) Input Edge Selection
Triggering Edge for Counter Increment/Decrement
None. Counter T3 stops.
0 0 0
0 0 1
Any transition (rising or falling edge) on T3IN.
0 1 0
Any transition (rising or falling edge) on T3EUD.
Any transition (rising or falling edge) on any T3 input (T3IN or T3EUD).
Reserved. Do not use this combination
0 1 1
1 X X
User’s Manual
10-9
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
The incremental encoder can be connected directly to the C167CR without external
interface logic. In a standard system, however, comparators will be employed to convert
the encoder’s differential outputs (e.g. A, A) to digital signals (e.g. A). This greatly
increases noise immunity.
Note: The third encoder output Top0, which indicates the mechanical zero position, may
be connected to an external interrupt input and trigger a reset of timer T3 (e.g. via
PEC transfer from ZEROS).
Signal
Conditioning
Encoder
C167CR
T3Input
A
B
A
A
B
B
T3Input
Interrupt
T0
T0
T0
MCS04372
Figure 10-7 Connection of the Encoder to the C167CR
For incremental interface operation the following conditions must be met:
• Bitfield T3M must be ‘110 ’.
B
• Both pins T3IN and T3EUD must be configured as input, i.e. the respective direction
control bits must be ‘0’.
• Bit T3UDE must be ‘1’ to enable automatic direction control.
The maximum counting frequency which is allowed in incremental interface mode is
fCPU/16. To ensure that a transition of any input signal is correctly recognized, its level
should be held high or low for at least 8 fCPU cycles before it changes. As in Incremental
Interface Mode two input signals with a 90° phase shift are evaluated, their maximum
input frequency can be fCPU/32.
In Incremental Interface Mode the count direction is automatically derived from the
sequence in which the input signals change, which corresponds to the rotation direction
of the connected sensor. Table 10-7 summarizes the possible combinations.
Table 10-7 GPT1 Core Timer T3 (Incremental Interface Mode) Count Direction
Level on respective
other input
T3IN Input
Falling
T3EUD Input
Falling
Down
Rising
Down
Up
Rising
Up
High
Low
Up
Down
Down
Up
User’s Manual
10-10
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Figure 10-8 and Figure 10-9 give examples of T3’s operation, visualizing count signal
generation and direction control. It also shows how input jitter is compensated which
might occur if the sensor rests near to one of its switching points.
Forward
Jitter
Backward
Jitter
Forward
T3IN
T3EUD
Contents
of T3
Up
Down
Up
Note: This example shows the timer behaviour assuming that T3 counts upon any
transition on input, i.e. T3I = '011 B'.
MCT04373
Figure 10-8 Evaluation of the Incremental Encoder Signals
Forward
Jitter
Backward
Jitter
Forward
T3IN
T3EUD
Contents
of T3
Up
Down
Up
Note: This example shows the timer behaviour assuming that T3 counts upon any
transition on input T3IN, i.e. T3I = '001 B'.
MCT04374
Figure 10-9 Evaluation of the Incremental Encoder Signals
Note: Timer T3 operating in incremental interface mode automatically provides
information on the sensor’s current position. Dynamic information (speed,
acceleration, deceleration) may be obtained by measuring the incoming signal
periods. This is facilitated by an additional special capture mode for timer T5.
User’s Manual
10-11
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
10.1.2
GPT1 Auxiliary Timers T2 and T4
Both auxiliary timers T2 and T4 have exactly the same functionality. They can be
configured for timer, gated timer, counter, or incremental interface mode with the same
options for the timer frequencies and the count signal as the core timer T3. In addition to
these 4 counting modes, the auxiliary timers can be concatenated with the core timer, or
they may be used as reload or capture registers in conjunction with the core timer.
The individual configuration for timers T2 and T4 is determined by their bitaddressable
control registers T2CON and T4CON, which are both organized identically. Note that
functions which are present in all 3 timers of block GPT1 are controlled in the same bit
positions and in the same manner in each of the specific control registers.
Note: The auxiliary timers have no output toggle latch and no alternate output function.
T2CON
Timer 2 Control Register
SFR (FF40 /A0 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
T2
T2
-
-
-
-
-
-
-
T2R
T2M
T2I
UDE UD
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
T4CON
Timer 4 Control Register
SFR (FF44 /A2 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
T4
T4
-
-
-
-
-
-
-
T4R
T4M
T4I
UDE UD
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
User’s Manual
10-12
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Bit
TxI
Function
Timer x Input Selection
Depends on the Operating Mode, see respective sections.
TxM
Timer x Mode Control (Basic Operating Mode)
000: Timer Mode
001: Counter Mode
010: Gated Timer with Gate active low
011: Gated Timer with Gate active high
100: Reload Mode
101: Capture Mode
110: Incremental Interface Mode
111: Reserved. Do not use this combination.
TxR
Timer x Run Bit
0:
1:
Timer/Counter x stops
Timer/Counter x runs
1)
TxUD
Timer x Up/Down Control
Timer x External Up/Down Enable
1)
TxUDE
1)
For the effects of bits TxUD and TxUDE refer to Table 10-1 (see T3 section).
Count Direction Control for Auxiliary Timers
The count direction of the auxiliary timers can be controlled in the same way as for the
core timer T3. The description and the table apply accordingly.
Timers T2 and T4 in Timer Mode or Gated Timer Mode
When the auxiliary timers T2 and T4 are programmed to timer mode or gated timer
mode, their operation is the same as described for the core timer T3. The descriptions,
figures and tables apply accordingly with one exception:
• There is no output toggle latch for T2 and T4.
Timers T2 and T4 in Incremental Interface Mode
When the auxiliary timers T2 and T4 are programmed to incremental interface mode,
their operation is the same as described for the core timer T3. The descriptions, figures
and tables apply accordingly.
User’s Manual
10-13
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timers T2 and T4 in Counter Mode
Counter mode for the auxiliary timers T2 and T4 is selected by setting bit field TxM in the
respective register TxCON to ‘001 ’. In counter mode timers T2 and T4 can be clocked
B
either by a transition at the respective external input pin TxIN, or by a transition of timer
T3’s output toggle latch T3OTL.
Edge
Select
Interrupt
Request
TxIN
Auxiliary Timer Tx
Up/
Down
TxR
Txl
TxU
D
0
MUX
XOR
1
TxEUD
TxUDE
MCB02221
x = 2.4
Figure 10-10 Block Diagram of an Auxiliary Timer in Counter Mode
The event causing an increment or decrement of a timer can be a positive, a negative,
or both a positive and a negative transition at either the respective input pin, or at the
toggle latch T3OTL.
Bit field TxI in the respective control register TxCON selects the triggering transition (see
Table 10-8).
User’s Manual
10-14
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Table 10-8 GPT1 Auxiliary Timer (Counter Mode) Input Edge Selection
T2I/T4I
X 0 0
0 0 1
0 1 0
0 1 1
1 0 1
1 1 0
1 1 1
Triggering Edge for Counter Increment/Decrement
None. Counter Tx is disabled
Positive transition (rising edge) on TxIN
Negative transition (falling edge) on TxIN
Any transition (rising or falling edge) on TxIN
Positive transition (rising edge) of output toggle latch T3OTL
Negative transition (falling edge) of output toggle latch T3OTL
Any transition (rising or falling edge) of output toggle latch T3OTL
Note: Only state transitions of T3OTL which are caused by the overflows/underflows of
T3 will trigger the counter function of T2/T4. Modifications of T3OTL via software
will NOT trigger the counter function of T2/T4.
For counter operation, pin TxIN must be configured as input, i.e. the respective direction
control bit must be ‘0’. The maximum input frequency which is allowed in counter mode
is fCPU/16. To ensure that a transition of the count input signal which is applied to TxIN
is correctly recognized, its level should be held for at least 8 fCPU cycles before it
changes.
User’s Manual
10-15
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer Concatenation
Using the toggle bit T3OTL as a clock source for an auxiliary timer in counter mode
concatenates the core timer T3 with the respective auxiliary timer. Depending on which
transition of T3OTL is selected to clock the auxiliary timer, this concatenation forms a 32-
bit or a 33-bit timer/counter.
• 32-bit Timer/Counter: If both a positive and a negative transition of T3OTL is used to
clock the auxiliary timer, this timer is clocked on every overflow/underflow of the core
timer T3. Thus, the two timers form a 32-bit timer.
• 33-bit Timer/Counter: If either a positive or a negative transition of T3OTL is selected
to clock the auxiliary timer, this timer is clocked on every second overflow/underflow
of the core timer T3. This configuration forms a 33-bit timer (16-bit core timer + T3OTL
+ 16-bit auxiliary timer).
The count directions of the two concatenated timers are not required to be the same.
This offers a wide variety of different configurations.
T3 can operate in timer, gated timer or counter mode in this case.
Tyl
fCPU
2n : 1
Core Timer Ty
Up/Down
TyOTL
TyOUT
TyR
TyOE
Interrupt
Request
*)
Edge
Select
Interrupt
Request
Auxiliary Timer Tx
TxIR
TxR
Txl
*) Note: Line only affected by over/underflows of T3, but NOT by software modifications of T3OTL.
MCB02034
T3OUT = P3.3
x = 2.4, y = 3
Figure 10-11 Concatenation of Core Timer T3 and an Auxiliary Timer
User’s Manual
10-16
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Auxiliary Timer in Reload Mode
Reload mode for the auxiliary timers T2 and T4 is selected by setting bit field TxM in the
respective register TxCON to ‘100 ’. In reload mode the core timer T3 is reloaded with
B
the contents of an auxiliary timer register, triggered by one of two different signals. The
trigger signal is selected the same way as the clock source for counter mode (see
Table 10-8), i.e. a transition of the auxiliary timer’s input or the output toggle latch T3OTL
may trigger the reload.
Note: When programmed for reload mode, the respective auxiliary timer (T2 or T4) stops
independent of its run flag T2R or T4R.
Source/Edge
Select
Reload Register Tx
Interrupt
Request
TxIN
TxI
Input
Clock
Interrupt
Request
Core Timer T3
Up/Down
*)
T3OTL
T3OUT
T3OE
x = 2.4
*) Note: Line only affected by over/underflows of T3, but NOT by software modifications of T3OTL.
MCB02035
Figure 10-12 GPT1 Auxiliary Timer in Reload Mode
Upon a trigger signal T3 is loaded with the contents of the respective timer register (T2
or T4) and the interrupt request flag (T2IR or T4IR) is set.
Note: When a T3OTL transition is selected for the trigger signal, also the interrupt
request flag T3IR will be set upon a trigger, indicating T3’s overflow or underflow.
Modifications of T3OTL via software will NOT trigger the counter function of T2/T4.
User’s Manual
10-17
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
The reload mode triggered by T3OTL can be used in a number of different
configurations. Depending on the selected active transition the following functions can
be performed:
• If both a positive and a negative transition of T3OTL is selected to trigger a reload, the
core timer will be reloaded with the contents of the auxiliary timer each time it
overflows or underflows. This is the standard reload mode (reload on overflow/
underflow).
• If either a positive or a negative transition of T3OTL is selected to trigger a reload, the
core timer will be reloaded with the contents of the auxiliary timer on every second
overflow or underflow.
• Using this “single-transition” mode for both auxiliary timers allows to perform very
flexible pulse width modulation (PWM). One of the auxiliary timers is programmed to
reload the core timer on a positive transition of T3OTL, the other is programmed for a
reload on a negative transition of T3OTL. With this combination the core timer is
alternately reloaded from the two auxiliary timers.
Figure 10-13 shows an example for the generation of a PWM signal using the alternate
reload mechanism. T2 defines the high time of the PWM signal (reloaded on positive
transitions) and T4 defines the low time of the PWM signal (reloaded on negative
transitions). The PWM signal can be output on T3OUT with T3OE = ‘1’, port latch = ‘1’
and direction bit = ‘1’. With this method the high and low time of the PWM signal can be
varied in a wide range.
Note: The output toggle latch T3OTL is accessible via software and may be changed, if
required, to modify the PWM signal.
However, this will NOT trigger the reloading of T3.
User’s Manual
10-18
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Reload Register T2
Interrupt
Request
*)
T2I
Input
Clock
Core Timer T3
T3OTL
T3OUT
Up/
T3OE
Down
Interrupt
Request
*)
Interrupt
Request
Reload Register T4
T4I
*) Note: Line only affected by over/underflows of T3, but NOT by software modifications of T3OTL.
MCB02037
Figure 10-13 GPT1 Timer Reload Configuration for PWM Generation
Note: Although it is possible, it should be avoided to select the same reload trigger event
for both auxiliary timers. In this case both reload registers would try to load the
core timer at the same time. If this combination is selected, T2 is disregarded and
the contents of T4 is reloaded.
User’s Manual
10-19
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Auxiliary Timer in Capture Mode
Capture mode for the auxiliary timers T2 and T4 is selected by setting bit field TxM in the
respective register TxCON to ‘101 ’. In capture mode the contents of the core timer are
B
latched into an auxiliary timer register in response to a signal transition at the respective
auxiliary timer’s external input pin TxIN. The capture trigger signal can be a positive, a
negative, or both a positive and a negative transition.
The two least significant bits of bit field TxI are used to select the active transition (see
Table 10-8), while the most significant bit TxI.2 is irrelevant for capture mode. It is
recommended to keep this bit cleared (TxI.2 = ‘0’).
Note: When programmed for capture mode, the respective auxiliary timer (T2 or T4)
stops independent of its run flag T2R or T4R.
Edge
Select
Capture Register Tx
Interrupt
Request
TxIN
TxI
Input
Clock
Interrupt
Request
Core Timer T3
Up/Down
T3OTL
T3OUT
T3OE
x = 2.4
MCB02038
Figure 10-14 GPT1 Auxiliary Timer in Capture Mode
Upon a trigger (selected transition) at the corresponding input pin TxIN the contents of
the core timer are loaded into the auxiliary timer register and the associated interrupt
request flag TxIR will be set.
Note: The direction control bits for T2IN and T4IN must be set to ‘0’, and the level of the
capture trigger signal should be held high or low for at least 8 fCPU cycles before
it changes to ensure correct edge detection.
User’s Manual
10-20
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
10.1.3
Interrupt Control for GPT1 Timers
When a timer overflows from FFFF to 0000 (when counting up), or when it underflows
H
H
from 0000 to FFFF (when counting down), its interrupt request flag (T2IR, T3IR or
H
H
T4IR) in register TxIC will be set. This will cause an interrupt to the respective timer
interrupt vector (T2INT, T3INT or T4INT) or trigger a PEC service, if the respective
interrupt enable bit (T2IE, T3IE or T4IE in register TxIC) is set. There is an interrupt
control register for each of the three timers.
T2IC
Timer 2 Intr. Ctrl. Reg.
SFR (FF60 /B0 )
Reset value: - - 00
H
H
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
5
5
4
4
4
3
2
1
0
-
T2IR T2IE
ILVL
GLVL
rw
-
-
-
-
-
-
-
rwh rw
rw
T3IC
Timer 3 Intr. Ctrl. Reg.
SFR (FF62 /B1 )
Reset value: - - 00
H
H
15
14
13
12
11
10
-
9
8
7
6
3
2
1
0
-
T3IR T3IE
ILVL
GLVL
rw
-
-
-
-
-
-
-
rwh rw
rw
T4IC
Timer 4 Intr. Ctrl. Reg.
SFR (FF64 /B2 )
Reset value: - - 00
H
H
15
14
13
12
11
10
-
9
8
7
6
3
2
1
0
-
T4IR T4IE
ILVL
GLVL
rw
-
-
-
-
-
-
-
rwh rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
User’s Manual
10-21
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
10.2
Timer Block GPT2
From a programmer’s point of view, the GPT2 block is represented by a set of SFRs as
summarized below. Those portions of port and direction registers which are used for
alternate functions by the GPT2 block are shaded.
Ports & Direction Control
Alternate Functions
Data Registers
Control Registers
Interrupt Control
E
ODP3
DP3
P3
T5
T5CON
T6CON
T5IC
T6IC
CRIC
T6
CAPREL
P5
P5DIDIS
T5IN/P5.13 T5EUD/P5.11
T6IN/P5.12 T6EUD/P5.10
CAPIN/P3.2 T6OUT/P3.1
ODP3
DP3
P3
Port 3 Open Drain Control Register T5 GPT2
Port 3 Direction Control Register T6 GPT2
CAPRELGPT2 Capture/Reload Register
T5ICGPT2 Timer 5 Interrupt Control Register
Timer 5 Register
Timer 6 Register
Port 3 Data Register
Port 5 Data Register
P5
P5DIDIS
Port 5 Digital Input Disable Register T6ICGPT2
Timer 6 Interrupt Control Register
CAPREL Interrupt Control Register
T5CONGPT2 Timer 5 Control Register
T6CONGPT2 Timer 6 Control Register
CRICGPT2
MCA04375
Figure 10-15 SFRs and Port Pins Associated with Timer Block GPT2
Timer block GPT2 supports high precision event control with a maximum resolution of
8 TCL. It includes the two timers T5 and T6, and the 16-bit capture/reload register
CAPREL. Timer T6 is referred to as the core timer, and T5 is referred to as the auxiliary
timer of GPT2.
Each timer has an alternate input function pin associated with it which serves as the gate
control in gated timer mode, or as the count input in counter mode. The count direction
(Up/Down) may be programmed via software. An overflow/underflow of T6 is indicated
by the output toggle bit T6OTL whose state may be output on an alternate function port
pin (T6OUT). In addition, T6 may be reloaded with the contents of CAPREL.
The toggle bit also supports the concatenation of T6 with auxiliary timer T5, while
concatenation of T6 with the timers of the CAPCOM units is provided through a direct
connection. Triggered by an external signal, the contents of T5 can be captured into
register CAPREL, and T5 may optionally be cleared. Both timer T6 and T5 can count up
or down, and the current timer value can be read or modified by the CPU in the non-
bitaddressable SFRs T5 and T6.
User’s Manual
10-22
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
T5EUD
fCPU
2n : 1
T5
Mode
Control
T5IN
U/D
GPT2 Timer T5
Interrupt
Request
Clear
Capture
Interrupt
Request
T3
MUX
CT3
CAPIN
GPT2 CAPREL
Interrupt
Request
GPT2 Timer T6
U/D
T6OTL
T6OUT
T6IN
T6
Mode
Control
Other
Timers
fCPU
2n : 1
T6EUD
MCB03999
Figure 10-16 GPT2 Block Diagram
User’s Manual
10-23
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
10.2.1
GPT2 Core Timer T6
The operation of the core timer T6 is controlled by its bitaddressable control register
T6CON.
T6CON
Timer 6 Control Register
SFR (FF48 /A4 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
T6
SR
T6
T6
T6
T6
-
-
-
-
T6R
T6M
T6I
OTL OE UDE UD
rw
-
-
-
-
rwh rw
rw
rw
rw
rw
rw
Bit
Function
Timer 6 Input Selection
T6I
Depends on the Operating Mode, see respective sections.
T6M
Timer 6 Mode Control (Basic Operating Mode)
000: Timer Mode
001: Counter Mode
010: Gated Timer with Gate active low
011: Gated Timer with Gate active high
1XX: Reserved. Do not use this combination.
T6R
Timer 6 Run Bit
0:
1:
Timer/Counter 6 stops
Timer/Counter 6 runs
1)
T6UD
Timer 6 Up/Down Control
1)
T6UDE
T6OE
Timer 6 External Up/Down Enable
Alternate Output Function Enable
0:
1:
Alternate Output Function Disabled
Alternate Output Function Enabled
T6OTL
Timer 6 Output Toggle Latch
Toggles on each overflow/underflow of T6. Can be set or reset by
software.
T6SR
Timer 6 Reload Mode Enable
0:
1:
Reload from register CAPREL Disabled
Reload from register CAPREL Enabled
1)
For the effects of bits T6UD and T6UDE refer to the Table 10-9.
User’s Manual
10-24
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 6 Run Bit
The timer can be started or stopped by software through bit T6R (Timer T6 Run Bit). If
T6R = ‘0’, the timer stops. Setting T6R to ‘1’ will start the timer.
In gated timer mode, the timer will only run if T6R = ‘1’ and the gate is active (high or low,
as programmed).
Count Direction Control
The count direction of the core timer can be controlled either by software or by the
external input pin T6EUD (Timer T6 External Up/Down Control Input), which is the
alternate input function of port pin P5.10. These options are selected by bits T6UD and
T6UDE in control register T6CON. When the up/down control is done by software (bit
T6UDE = ‘0’), the count direction can be altered by setting or clearing bit T6UD. When
T6UDE = ‘1’, pin T6EUD is selected to be the controlling source of the count direction.
However, bit T6UD can still be used to reverse the actual count direction, as shown in
Table 10-9. If T6UD = ‘0’ and pin T6EUD shows a low level, the timer is counting up.
With a high level at T6EUD the timer is counting down. If T6UD = ‘1’, a high level at pin
T6EUD specifies counting up, and a low level specifies counting down. The count
direction can be changed regardless of whether the timer is running or not.
Table 10-9 GPT2 Core Timer T6 Count Direction Control
Pin TxEUD
Bit TxUDE
Bit TxUD
Count Direction
Count Up
X
X
0
1
0
1
0
0
1
1
1
1
0
1
0
0
1
1
Count Down
Count Up
Count Down
Count Down
Count Up
Note: The direction control works the same for core timer T6 and for auxiliary timer T5.
Therefore the pins and bits are named Tx …
User’s Manual
10-25
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 6 Output Toggle Latch
An overflow or underflow of timer T6 will clock the toggle bit T6OTL in control register
T6CON. T6OTL can also be set or reset by software. Bit T6OE (Alternate Output
Function Enable) in register T6CON enables the state of T6OTL to be an alternate
function of the external output pin T6OUT. For that purpose, a ‘1’ must be written into the
respective port data latch and pin T6OUT must be configured as output by setting the
respective direction control bit to ‘1’. If T6OE = ‘1’, pin T6OUT then outputs the state of
T6OTL. If T6OE = ‘0’ pin T6OUT can be used as general purpose IO pin.
In addition, T6OTL can be used in conjunction with the timer over/underflows as an input
for the counter function of the auxiliary timer T5. For this purpose, the state of T6OTL
does not have to be available at pin T6OUT, because an internal connection is provided
for this option.
An overflow or underflow of timer T6 can also be used to clock the timers in the CAPCOM
units. For this purpose, there is a direct internal connection between timer T6 and the
CAPCOM timers.
Timer 6 in Timer Mode
Timer mode for the core timer T6 is selected by setting bitfield T6M in register T6CON
to ‘000 ’. In this mode, T6 is clocked with the internal system clock divided by a
B
programmable prescaler, which is selected by bit field T6I. The input frequency fT6 for
timer T6 and its resolution rT6 are scaled linearly with lower clock frequencies fCPU, as
can be seen from the following formula:
<T6I>
fCPU
4 × 2
fT6 =
rT6 [µs] =
,
<T6I>
fCPU [MHz]
4 × 2
Txl
Interrupt
Request
fCPU
2n : 1
Core Timer Tx
Up/
Down
TxR
TxOTL
TxOUT
TxU
D
0
TxOE
MUX
EXOR
1
TxEUD
TxUDE
MCB02028
T6EUD = P5.10
T6OUT = P3.1
x = 6
Figure 10-17 Block Diagram of Core Timer T6 in Timer Mode
User’s Manual
10-26
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
The timer input frequencies, resolution and periods which result from the selected
prescaler option are listed in Table 10-10. This table also applies to the gated timer
mode of T6 and to the auxiliary timer T5 in timer and gated timer mode. Note that some
numbers may be rounded to 3 significant digits.
Table 10-10 GPT2 Timer Input Frequencies, Resolution and Periods @ 20 MHz
fCPU = 20 MHz
Timer Input Selection T5I/T6I
000
001
8
010
011
32
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler factor 4
16
64
128
256
512
Input
Frequency
5
MHz
2.5
MHz
1.25
MHz
625
kHz
312.5 156.25 78.125 39.06
kHz kHz kHz kHz
Resolution
Period
200ns 400 ns 800 ns 1.6 µs 3.2 µs 6.4 µs 12.8 ms 25.6 µs
13 ms 26 ms 52.5ms 105 ms 210 ms 420 ms 840 ms 1.68 s
Table 10-11 GPT2 Timer Input Frequencies, Resolution and Periods @ 25 MHz
fCPU = 25 MHz Timer Input Selection T5I/T6I
000
001
010
011
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler factor 4
8
16
32
64
128
256
512
Input
Frequency
6.25
MHz
3.125 1.56
MHz MHz
781.25 390.62 195.31 97.66 48.83
kHz kHz kHz kHz kHz
Resolution
Period
160ns 320 ns 640 ns 1.28 µs 2.56 µs 5.12 µs 10.2 ms 20.5 µs
10.5 ms 21.0 ms 42.0ms 83.9 ms 168 ms 336 ms 671 ms 1.34 s
Table 10-12 GPT2 Timer Input Frequencies, Resolution and Periods @ 33 MHz
fCPU = 33 MHz Timer Input Selection T5I/T6I
000
001
010
011
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler factor 4
8
16
32
64
128
256
512
Input
Frequency
2.06
MHz
4.125 2.0625 1.031 515.62 257.81 128.91 64.45
MHz MHz MHz kHz kHz kHz kHz
Resolution
Period
121 ns 242 ns 485 ns 970 ns 1.94 µs 3.88 µs 7.76 µs 15.5 µs
7.9 ms 15.9 ms 31.8ms 63.6 ms 127 ms 254 ms 508 ms 1.02 s
User’s Manual
10-27
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 6 in Gated Timer Mode
Gated timer mode for the core timer T6 is selected by setting bit field T6M in register
T6CON to ‘010 ’ or ‘011 ’. Bit T6M.0 (T6CON.3) selects the active level of the gate
B
B
input. In gated timer mode the same options for the input frequency as for the timer mode
are available. However, the input clock to the timer in this mode is gated by the external
input pin T6IN (Timer T6 External Input).
TxI
fCPU
2n : 1
MUX
TxM
Core Timer Tx
TxOTL
TxOUT
TxIN
Up/
Down
TxR
TxOE
TxU
D
0
Interrupt
Request
MUX
XOR
1
TxEUD
MCB02029
TxUDE
T6IN = P5.12
T6EUD = P5.10
T6OUT = P3.1
x = 6
Figure 10-18 Block Diagram of Core Timer T6 in Gated Timer Mode
If T6M.0 = ‘0’, the timer is enabled when T6IN shows a low level. A high level at this pin
stops the timer. If T6M.0 = ‘1’, pin T6IN must have a high level in order to enable the
timer. In addition, the timer can be turned on or off by software using bit T6R. The timer
will only run, if T6R = ‘1’ and the gate is active. It will stop, if either T6R = ‘0’ or the gate
is inactive.
Note: A transition of the gate signal at pin T6IN does not cause an interrupt request.
User’s Manual
10-28
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer 6 in Counter Mode
Counter mode for the core timer T6 is selected by setting bit field T6M in register T6CON
to ‘001 ’. In counter mode timer T6 is clocked by a transition at the external input pin
B
T6IN, which is an alternate function of P5.12. The event causing an increment or
decrement of the timer can be a positive, a negative, or both a positive and a negative
transition at this pin. Bit field T6I in control register T6CON selects the triggering
transition (see Table 10-13).
Edge
Select
TxIN
Core Timer Tx
TxOTL
TxOUT
Up/
Down
TxR
TxOE
Txl
TxU
D
0
Interrupt
Request
MUX
XOR
1
TxEUD
TxUDE
MCB02030
T6IN = P5.12
T6EUD = P5.10
T6OUT = P3.1
x = 6
Figure 10-19 Block Diagram of Core Timer T6 in Counter Mode
Table 10-13 GPT2 Core Timer T6 (Counter Mode) Input Edge Selection
T6I
Triggering Edge for Counter Increment/Decrement
None. Counter T6 is disabled
0 0 0
0 0 1
0 1 0
0 1 1
1 X X
Positive transition (rising edge) on T6IN
Negative transition (falling edge) on T6IN
Any transition (rising or falling edge) on T6IN
Reserved. Do not use this combination
The maximum input frequency which is allowed in counter mode is fCPU / 8. To ensure
that a transition of the count input signal which is applied to T6IN is correctly recognized,
its level should be held high or low for at least 4 fCPU cycles before it changes.
User’s Manual
10-29
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
10.2.2
GPT2 Auxiliary Timer T5
The auxiliary timer T5 can be configured for timer, gated timer or counter mode with the
same options for the timer frequencies and the count signal as the core timer T6. In
addition the auxiliary timer can be concatenated with the core timer (operation in counter
mode). Its contents may be captured to register CAPREL upon a selectable trigger.
The individual configuration for timer T5 is determined by its bitaddressable control
register T5CON. Note that functions which are present in both timers of block GPT2 are
controlled in the same bit positions and in the same manner in each of the specific control
registers.
Note: The auxiliary timer has no output toggle latch and no alternate output function.
T5CON
Timer 5 Control Register
SFR (FF46 /A3 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
T5
T5
T5
T5
CI
-
CT3
-
T5R
T5M
T5I
SC CLR
UDE UD
rw
rw
rw
-
rw
-
rw
rw
rw
rw
rw
Bit
Function
Timer 5 Input Selection
T5I
Depends on the Operating Mode, see respective sections.
T5M
Timer 5 Mode Control (Basic Operating Mode)
000: Timer Mode
001: Counter Mode
010: Gated Timer with Gate active low
011: Gated Timer with Gate active high
1XX: Reserved. Do not use this combination.
T5R
Timer 5 Run Bit
0:
1:
Timer / Counter 5 stops
Timer / Counter 5 runs
*)
T5UD
Timer 5 Up / Down Control
*)
T5UDE
CT3
Timer 5 External Up/Down Enable
Timer 3 Capture Trigger Enable
0:
1:
Capture trigger from pin CAPIN
Capture trigger from T3 input pins
User’s Manual
10-30
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Bit
CI
Function
Register CAPREL Capture Trigger Selection (depending on bit CT3)
00: Capture disabled
01: Positive transition (rising edge) on CAPIN or
any transition on T3IN
10: Negative transition (falling edge) on CAPIN or
any transition on T3EUD
11: Any transition (rising or falling edge) on CAPIN or
any transition on T3IN or T3EUD
T5CLR
T5SC
Timer 5 Clear Bit
0:
1:
Timer 5 not cleared on a capture
Timer 5 is cleared on a capture
Timer 5 Capture Mode Enable
0:
1:
Capture into register CAPREL disabled
Capture into register CAPREL enabled
*)
For the effects of bits T5UD and T5UDE refer to the direction table (see T6 section).
Count Direction Control for Auxiliary Timer
The count direction of the auxiliary timer can be controlled in the same way as for the
core timer T6. The description and the table apply accordingly.
Timer T5 in Timer Mode or Gated Timer Mode
When the auxiliary timer T5 is programmed to timer mode or gated timer mode, its
operation is the same as described for the core timer T6. The descriptions, figures and
tables apply accordingly with one exception:
• There is no output toggle latch for T5.
Timer T5 in Counter Mode
Counter mode for the auxiliary timer T5 is selected by setting bit field T5M in register
T5CON to ‘001 ’. In counter mode timer T5 can be clocked either by a transition at the
B
external input pin T5IN, or by a transition of timer T6’s output toggle latch T6OTL (i.e.
timer concatenation).
User’s Manual
10-31
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Edge
Select
Interrupt
Auxiliary Timer Tx
TxIN
Request
Up/
Down
TxR
Txl
TxU
D
0
MUX
XOR
1
TxEUD
TxUDE
MCB02221
x = 5
Figure 10-20 Block Diagram of Auxiliary Timer T5 in Counter Mode
The event causing an increment or decrement of the timer can be a positive, a negative,
or both a positive and a negative transition at either the input pin, or at the toggle latch
T6OTL.
Bitfield T5I in control register T5CON selects the triggering transition (see Table 10-14).
Table 10-14 GPT2 Auxiliary Timer (Counter Mode) Input Edge Selection
T5I
Triggering Edge for Counter Increment/Decrement
None. Counter T5 is disabled
X 0 0
0 0 1
0 1 0
0 1 1
1 0 1
1 1 0
1 1 1
Positive transition (rising edge) on T5IN
Negative transition (falling edge) on T5IN
Any transition (rising or falling edge) on T5IN
Positive transition (rising edge) of output toggle latch T6OTL
Negative transition (falling edge) of output toggle latch T6OTL
Any transition (rising or falling edge) of output toggle latch T6OTL
Note: Only state transitions of T6OTL which are caused by the overflows/underflows of
T6 will trigger the counter function of T5. Modifications of T6OTL via software will
NOT trigger the counter function of T5.
The maximum input frequency which is allowed in counter mode is fCPU / 8. To ensure
that a transition of the count input signal which is applied to T5IN is correctly recognized,
its level should be held high or low for at least 4 fCPU cycles before it changes.
User’s Manual
10-32
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
Timer Concatenation
Using the toggle bit T6OTL as a clock source for the auxiliary timer in counter mode
concatenates the core timer T6 with the auxiliary timer. Depending on which transition of
T6OTL is selected to clock the auxiliary timer, this concatenation forms a 32-bit or a 33-
bit timer/counter.
• 32-bit Timer/Counter: If both a positive and a negative transition of T6OTL is used to
clock the auxiliary timer, this timer is clocked on every overflow/underflow of the core
timer T6. Thus, the two timers form a 32-bit timer.
• 33-bit Timer/Counter: If either a positive or a negative transition of T6OTL is selected
to clock the auxiliary timer, this timer is clocked on every second overflow/underflow
of the core timer T6. This configuration forms a 33-bit timer (16-bit core timer + T6OTL
+ 16-bit auxiliary timer).
The count directions of the two concatenated timers are not required to be the same.
This offers a wide variety of different configurations.
T6 can operate in timer, gated timer or counter mode in this case.
Tyl
fCPU
2n : 1
Core Timer Ty
Up/Down
TyOTL
TyOUT
TyR
TyOE
Interrupt
Request
*)
Edge
Select
Interrupt
Request
Auxiliary Timer Tx
TxIR
TxR
Txl
*) Note: Line only affected by over/underflows of T3, but NOT by software modifications of T3OTL.
MCB02034
T6OUT = P3.1
x = 5, y = 6
Figure 10-21 Concatenation of Core Timer T6 and Auxiliary Timer T5
User’s Manual
10-33
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
GPT2 Capture/Reload Register CAPREL in Capture Mode
This 16-bit register can be used as a capture register for the auxiliary timer T5. This
mode is selected by setting bit T5SC = ‘1’ in control register T5CON. Bit CT3 selects the
external input pin CAPIN or the input pins of timer T3 as the source for a capture trigger.
Either a positive, a negative, or both a positive and a negative transition at pin CAPIN
can be selected to trigger the capture function, or transitions on input T3IN or input
T3EUD or both inputs T3IN and T3EUD. The active edge is controlled by bit field CI in
register T5CON. The same coding is used as in the two least significant bits of bit field
T5I (see Table 10-14).
The maximum input frequency for the capture trigger signal at CAPIN is fCPU / 8. To
ensure that a transition of the capture trigger signal is correctly recognized, its level
should be held for at least 4 fCPU cycles before it changes.
Up/Down
Input
Clock
Interrupt
Request
Auxiliary Timer T5
Edge
Select
T5CLR
T5SC
CAPIN
Interrupt
Request
CI
CAPREL Register
MCB02044B
Figure 10-22 GPT2 Register CAPREL in Capture Mode
When the timer T3 capture trigger is enabled (CT3 = ‘1’) register CAPREL captures the
contents of T5 upon transitions of the selected input(s). These values can be used to
measure T3’s input signals. This is useful e.g. when T3 operates in incremental interface
mode, in order to derive dynamic information (speed acceleration) from the input signals.
User’s Manual
10-34
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
When a selected transition at the selected input pin(s) (CAPIN, T3IN, T3EUD) is
detected, the contents of the auxiliary timer T5 are latched into register CAPREL, and
interrupt request flag CRIR is set. With the same event, timer T5 can be cleared to
0000 . This option is controlled by bit T5CLR in register T5CON. If T5CLR = ‘0’, the
H
contents of timer T5 are not affected by a capture. If T5CLR = ‘1’, timer T5 is cleared after
the current timer value has been latched into register CAPREL.
Note: Bit T5SC only controls whether a capture is performed or not. If T5SC = ‘0’, the
selected trigger event can still be used to clear timer T5 or to generate an interrupt
request. This interrupt is controlled by the CAPREL interrupt control register CRIC.
GPT2 Capture/Reload Register CAPREL in Reload Mode
This 16-bit register can be used as a reload register for the core timer T6. This mode is
selected by setting bit T6SR = ‘1’ in register T6CON. The event causing a reload in this
mode is an overflow or underflow of the core timer T6.
When timer T6 overflows from FFFF to 0000 (when counting up) or when it underflows
H
H
from 0000 to FFFF (when counting down), the value stored in register CAPREL is
H
H
loaded into timer T6. This will not set the interrupt request flag CRIR associated with the
CAPREL register. However, interrupt request flag T6IR will be set indicating the
overflow/underflow of T6.
CAPREL Register
T6OTL
T6OUT
T6SR
T6OE
Input
Clock
Interrupt
Request
Core Timer T6
Up/Down
MCB02045
Figure 10-23 GPT2 Register CAPREL in Reload Mode
User’s Manual
10-35
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
GPT2 Capture/Reload Register CAPREL in Capture-And-Reload Mode
Since the reload function and the capture function of register CAPREL can be enabled
individually by bits T5SC and T6SR, the two functions can be enabled simultaneously by
setting both bits. This feature can be used to generate an output frequency that is a
multiple of the input frequency.
Up/Down
Input
Clock
Interrupt
Request
Auxiliary Timer T5
Edge
Select
T5CLR
T5SC
CAPIN
Interrupt
Request
CI
CAPREL Register
T6OTL
T6OUT
T6SR
T6OE
Input
Clock
Interrupt
Request
Core Timer T6
Up/Down
MCB02046B
Figure 10-24 GPT2 Register CAPREL in Capture-And-Reload Mode
This combined mode can be used to detect consecutive external events which may
occur aperiodically, but where a finer resolution, that means, more ‘ticks’ within the time
between two external events is required.
User’s Manual
10-36
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
For this purpose, the time between the external events is measured using timer T5 and
the CAPREL register. Timer T5 runs in timer mode counting up with a frequency of e.g.
fCPU/32. The external events are applied to pin CAPIN. When an external event occurs,
the timer T5 contents are latched into register CAPREL, and timer T5 is cleared
(T5CLR = ‘1’). Thus, register CAPREL always contains the correct time between two
events, measured in timer T5 increments. Timer T6, which runs in timer mode counting
down with a frequency of e.g. fCPU / 4, uses the value in register CAPREL to perform a
reload on underflow. This means, the value in register CAPREL represents the time
between two underflows of timer T6, now measured in timer T6 increments. Since timer
T6 runs 8 times faster than timer T5, it will underflow 8 times within the time between two
external events. Thus, the underflow signal of timer T6 generates 8 ‘ticks’. Upon each
underflow, the interrupt request flag T6IR will be set and bit T6OTL will be toggled. The
state of T6OTL may be output on pin T6OUT. This signal has 8 times more transitions
than the signal which is applied to pin CAPIN.
The underflow signal of timer T6 can furthermore be used to clock one or more of the
timers of the CAPCOM units, which gives the user the possibility to set compare events
based on a finer resolution than that of the external events.
User’s Manual
10-37
V3.1, 2000-03
C167CR
Derivatives
The General Purpose Timer Units
10.2.3
Interrupt Control for GPT2 Timers and CAPREL
When a timer overflows from FFFF to 0000 (when counting up), or when it underflows
H
H
from 0000 to FFFF (when counting down), its interrupt request flag (T5IR or T6IR) in
H
H
register TxIC will be set. Whenever a transition according to the selection in bit field CI
is detected at pin CAPIN, interrupt request flag CRIR in register CRIC is set. Setting any
request flag will cause an interrupt to the respective timer or CAPREL interrupt vector
(T5INT, T6INT or CRINT) or trigger a PEC service, if the respective interrupt enable bit
(T5IE or T6IE in register TxIC, CRIE in register CRIC) is set. There is an interrupt control
register for each of the two timers and for the CAPREL register.
T5IC
Timer 5 Intr. Ctrl. Reg.
SFR (FF66 /B3 )
Reset value: - - 00
H
H
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
5
5
4
4
4
3
2
1
0
-
T5IR T5IE
ILVL
GLVL
RW
-
-
-
-
-
-
-
rwh rw
rw
T6IC
Timer 6 Intr. Ctrl. Reg.
SFR (FF68 /B4 )
Reset value: - - 00
H
H
15
14
13
12
11
10
-
9
8
7
6
3
2
1
0
-
T6IR T6IE
ILVL
GLVL
RW
-
-
-
-
-
-
-
rwh rw
rw
CRIC
CAPREL Intr. Ctrl. Reg.
SFR (FF6A /B5 )
Reset value: - - 00
H
H
15
14
13
12
11
10
-
9
8
7
6
3
2
1
0
-
CRIR CRIE
ILVL
GLVL
RW
-
-
-
-
-
-
-
rwh rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
User’s Manual
10-38
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
11
The Asynchronous/Synchronous Serial Interface
The Asynchronous/Synchronous Serial Interface ASC0 provides serial communication
between the C167CR and other microcontrollers, microprocessors or external
peripherals.
The ASC0 supports full-duplex asynchronous communication up to 781 KBaud/
1.03 MBaud and half-duplex synchronous communication up to 3.1/4.1 MBaud (@ 25/
33 MHz CPU clock). In synchronous mode, data are transmitted or received
synchronous to a shift clock which is generated by the C167CR. In asynchronous mode,
8- or 9-bit data transfer, parity generation, and the number of stop bits can be selected.
Parity, framing, and overrun error detection is provided to increase the reliability of data
transfers. Transmission and reception of data is double-buffered. For multiprocessor
communication, a mechanism to distinguish address from data bytes is included. Testing
is supported by a loop-back option. A 13-bit baud rate generator provides the ASC0 with
a separate serial clock signal.
Ports & Direction Control
Alternate Functions
Data Registers
Control Registers
Interrupt Control
E
ODP3
DP3
P3
S0BG
S0CON
S0TIC
S0RIC
S0EIC
S0TBUF
S0RBUF
E
S0TBIC
RXD0/P3.11
TXD0/P3.10
ODP3
DP3
Port 3 Open Drain Control Register
Port 3 Direction Control Register
P3
Port 3 Data Register
S0CON ASC0 Control Register
S0BG
ASC0 Baud Rate Generator/Reload Reg. S0RBUF ASC0 Receive Buffer Register (read
S0TBUF ASC0 Transmit Buffer Register
S0TIC ASC0 Transmit Interrupt Control Register S0RIC
S0TBIC ASC0 Transmit Buffer Interrupt Ctrl. Reg.
S0EIC
only)
ASC0 Receive Interrupt Control
Register
ASC0 Error Interrupt Control Register
MCA04376
Figure 11-1 SFRs and Port Pins Associated with ASC0
User’s Manual
11-1
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
The operating mode of the serial channel ASC0 is controlled by its bitaddressable control
register S0CON. This register contains control bits for mode and error check selection,
and status flags for error identification.
S0CON
ASC0 Control Register
SFR (FFB0 /D8 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
S0
S0 S0
S0
S0
S0
S0
S0
S0
S0
S0R
-
PEN
S0M
LB BRS ODD
OE FE PE OEN FEN
REN STP
rw
rw
rw
rw
-
rwh rwh rwh rw
rw
rw rwh rw
rw
Bit
Function
S0M
ASC0 Mode Control
000: 8-bit data
synchronous operation
async. operation
001: 8-bit data
010: Reserved. Do not use this combination!
011: 7-bit data + parity
100: 9-bit data
101: 8-bit data + wake up bit
110: Reserved. Do not use this combination!
111: 8-bit data + parity
async. operation
async. operation
async. operation
async. operation
async. operation
S0STP
Number of Stop Bits Selection
0:
1:
One stop bit
Two stop bits
S0REN
Receiver Enable Bit
0:
1:
Receiver disabled
Receiver enabled
(Reset by hardware after reception of byte in synchronous mode)
S0PEN
S0FEN
S0OEN
Parity Check Enable Bit async. operation
0:
1:
Ignore parity
Check parity
Framing Check Enable Bit
async. operation
0:
1:
Ignore framing errors
Check framing errors
Overrun Check Enable Bit
0:
1:
Ignore overrun errors
Check overrun errors
User’s Manual
11-2
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
Bit
Function
S0PE
Parity Error Flag
Set by hardware on a parity error (S0PEN = ‘1’). Must be reset by
software.
S0FE
S0OE
S0ODD
S0BRS
S0LB
S0R
Framing Error Flag
Set by hardware on a framing error (S0FEN = ‘1’). Must be reset by
software.
Overrun Error Flag
Set by hardware on an overrun error (S0OEN = ‘1’). Must be reset by
software.
Parity Selection Bit
0:
1:
Even parity (parity bit set on odd number of ‘1’s in data)
Odd parity (parity bit set on even number of ‘1’s in data)
Baudrate Selection Bit
0:
1:
Divide clock by reload-value + constant (depending on mode)
Additionally reduce serial clock to 2/3rd
LoopBack Mode Enable Bit
0:
1:
Standard transmit/receive mode
Loopback mode enabled
Baudrate Generator Run Bit
0:
1:
Baudrate generator disabled (ASC0 inactive)
Baudrate generator enabled
A transmission is started by writing to the Transmit Buffer register S0TBUF (via an
instruction or a PEC data transfer). Only the number of data bits which is determined by
the selected operating mode will actually be transmitted, i.e. bits written to positions 9
through 15 of register S0TBUF are always insignificant. After a transmission has been
completed, the transmit buffer register is cleared to 0000 .
H
Data transmission is double-buffered, so a new character may be written to the transmit
buffer register, before the transmission of the previous character is complete. This allows
the transmission of characters back-to-back without gaps.
Data reception is enabled by the Receiver Enable Bit S0REN. After reception of a
character has been completed, the received data and, if provided by the selected
operating mode, the received parity bit can be read from the (read-only) Receive Buffer
register S0RBUF. Bits in the upper half of S0RBUF which are not valid in the selected
operating mode will be read as zeros.
Data reception is double-buffered, so that reception of a second character may already
begin before the previously received character has been read out of the receive buffer
register. In all modes, receive buffer overrun error detection can be selected through bit
User’s Manual
11-3
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
S0OEN. When enabled, the overrun error status flag S0OE and the error interrupt
request flag S0EIR will be set when the receive buffer register has not been read by the
time reception of a second character is complete. The previously received character in
the receive buffer is overwritten.
The Loop-Back option (selected by bit S0LB) allows the data currently being
transmitted to be received simultaneously in the receive buffer. This may be used to test
serial communication routines at an early stage without having to provide an external
network. In loop-back mode the alternate input/output functions of the Port 3 pins are not
necessary.
Note: Serial data transmission or reception is only possible when the Baud Rate
Generator Run Bit S0R is set to ‘1’. Otherwise the serial interface is idle.
Do not program the mode control field S0M in register S0CON to one of the
reserved combinations to avoid unpredictable behavior of the serial interface.
User’s Manual
11-4
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
11.1
Asynchronous Operation
Asynchronous mode supports full-duplex communication, where both transmitter and
receiver use the same data frame format and the same baud rate. Data is transmitted on
pin TXD0 and received on pin RXD0. These signals are alternate port functions.
Reload Register
CPU
Clock
÷ 2
Baud Rate Timer
÷ 16
S0R
S0PE
S0FE S0OE
S0M S0STP
Receive Int.
Request
Clock
S0REN
S0FEN
S0PEN
S0OEN
S0LB
S0RIR
S0TIR
S0EIR
Transmit Int.
Request
Serial Port Control
Shift Clock
RXD0/P3.11
Error Int.
Request
0
Samp-
ling
Receive Shift
Register
Transmit Shift
Register
MUX
1
TXD0/P3.10
Receive Buffer Reg.
S0RBUF
Transmit Buffer Reg.
S0TBUF
Internal Bus
MCB02219
Figure 11-2 Asynchronous Mode of Serial Channel ASC0
Asynchronous Data Frames
8-bit data frames either consist of 8 data bits D7 … D0 (S0M = ‘001 ’), or of 7 data bits
B
D6 … D0 plus an automatically generated parity bit (S0M = ‘011 ’). Parity may be odd
B
or even, depending on bit S0ODD in register S0CON. An even parity bit will be set, if the
modulo-2-sum of the 7 data bits is ‘1’. An odd parity bit will be cleared in this case. Parity
checking is enabled via bit S0PEN (always OFF in 8-bit data mode). The parity error flag
User’s Manual
11-5
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
S0PE will be set along with the error interrupt request flag, if a wrong parity bit is
received. The parity bit itself will be stored in bit S0RBUF.7.
(1st) 2nd
Stop Stop
Start
Bit (LSB)
D0
D7 /
Parity
D1
D2
D3
D4
D5
D6
Bit
Bit
MCT04377
Figure 11-3 Asynchronous 8-bit Data Frames
9-bit data frames either consist of 9 data bits D8 … D0 (S0M = ‘100 ’), of 8 data bits
B
D7 … D0 plus an automatically generated parity bit (S0M = ‘111 ’) or of 8 data bits
B
D7 … D0 plus wake-up bit (S0M = ‘101 ’). Parity may be odd or even, depending on bit
B
S0ODD in register S0CON. An even parity bit will be set, if the modulo-2-sum of the 8
data bits is ‘1’. An odd parity bit will be cleared in this case. Parity checking is enabled
via bit S0PEN (always OFF in 9-bit data and wake-up mode). The parity error flag S0PE
will be set along with the error interrupt request flag, if a wrong parity bit is received. The
parity bit itself will be stored in bit S0RBUF.8.
In wake-up mode received frames are only transferred to the receive buffer register, if
the 9th bit (the wake-up bit) is ‘1’. If this bit is ‘0’, no receive interrupt request will be
activated and no data will be transferred.
This feature may be used to control communication in multi-processor system:
When the master processor wants to transmit a block of data to one of several slaves, it
first sends out an address byte which identifies the target slave. An address byte differs
from a data byte in that the additional 9th bit is a ‘1’ for an address byte and a ‘0’ for a
data byte, so no slave will be interrupted by a data ‘byte’. An address ‘byte’ will interrupt
all slaves (operating in 8-bit data + wake-up bit mode), so each slave can examine the 8
LSBs of the received character (the address). The addressed slave will switch to 9-bit
data mode (e.g. by clearing bit S0M.0), which enables it to also receive the data bytes
that will be coming (having the wake-up bit cleared). The slaves that were not being
addressed remain in 8-bit data + wake-up bit mode, ignoring the following data bytes.
(1st) 2nd
Stop Stop
Bit Bit
Start
Bit (LSB)
D0
9th
Bit
D1
D2
D3
D4
D5
D6
D7
Data Bit D8
Parity
Wake-up Bit
MCT04378
Figure 11-4 Asynchronous 9-bit Data Frames
User’s Manual
11-6
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
Asynchronous transmission begins at the next overflow of the divide-by-16 counter
(see Figure 11-4), provided that S0R is set and data has been loaded into S0TBUF. The
transmitted data frame consists of three basic elements:
• the start bit
• the data field (8 or 9 bits, LSB first, including a parity bit, if selected)
• the delimiter (1 or 2 stop bits)
Data transmission is double buffered. When the transmitter is idle, the transmit data
loaded into S0TBUF is immediately moved to the transmit shift register thus freeing
S0TBUF for the next data to be sent. This is indicated by the transmit buffer interrupt
request flag S0TBIR being set. S0TBUF may now be loaded with the next data, while
transmission of the previous one is still going on.
The transmit interrupt request flag S0TIR will be set before the last bit of a frame is
transmitted, i.e. before the first or the second stop bit is shifted out of the transmit shift
register.
The transmitter output pin TXD0 must be configured for alternate data output, i.e. the
respective port output latch and the direction latch must be ‘1’.
Asynchronous reception is initiated by a falling edge (1-to-0 transition) on pin RXD0,
provided that bits S0R and S0REN are set. The receive data input pin RXD0 is sampled
at 16 times the rate of the selected baud rate. A majority decision of the 7th, 8th and 9th
sample determines the effective bit value. This avoids erroneous results that may be
caused by noise.
If the detected value is not a ‘0’ when the start bit is sampled, the receive circuit is reset
and waits for the next 1-to-0 transition at pin RXD0. If the start bit proves valid, the
receive circuit continues sampling and shifts the incoming data frame into the receive
shift register.
When the last stop bit has been received, the content of the receive shift register is
transferred to the receive data buffer register S0RBUF. Simultaneously, the receive
interrupt request flag S0RIR is set after the 9th sample in the last stop bit time slot (as
programmed), regardless whether valid stop bits have been received or not. The receive
circuit then waits for the next start bit (1-to-0 transition) at the receive data input pin.
The receiver input pin RXD0 must be configured for input, i.e. the respective direction
latch must be ‘0’.
Asynchronous reception is stopped by clearing bit S0REN. A currently received frame is
completed including the generation of the receive interrupt request and an error interrupt
request, if appropriate. Start bits that follow this frame will not be recognized.
Note: In wake-up mode received frames are only transferred to the receive buffer
register, if the 9th bit (the wake-up bit) is ‘1’. If this bit is ‘0’, no receive interrupt
request will be activated and no data will be transferred.
User’s Manual
11-7
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
11.2
Synchronous Operation
Synchronous mode supports half-duplex communication, basically for simple IO
expansion via shift registers. Data is transmitted and received via pin RXD0, while pin
TXD0 outputs the shift clock. These signals are alternate port functions. Synchronous
mode is selected with S0M = ‘000 ’.
B
8 data bits are transmitted or received synchronous to a shift clock generated by the
internal baud rate generator. The shift clock is only active as long as data bits are
transmitted or received.
Reload Register
CPU
Clock
÷ 2
Baud Rate Timer
÷ 4
S0R
S0M = 000B
S0OE
Receive Int.
Request
Clock
S0RIR
S0TIR
S0EIR
S0REN
S0OEN
S0LB
Transmit Int.
Request
Serial Port Control
TXD0/P3.10
Error Int.
Request
Shift Clock
Receive
0
1
Receive Shift
Register
Transmit Shift
Register
MUX
RXD0/P3.11
Transmit
Receive Buffer Reg.
S0RBUF
Transmit Buffer Reg.
S0TBUF
Internal Bus
MCB02220
Figure 11-5 Synchronous Mode of Serial Channel ASC0
User’s Manual
11-8
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
Synchronous transmission begins within 4 state times after data has been loaded into
S0TBUF, provided that S0R is set and S0REN = ‘0’ (half-duplex, no reception). Data
transmission is double buffered. When the transmitter is idle, the transmit data loaded
into S0TBUF is immediately moved to the transmit shift register thus freeing S0TBUF for
the next data to be sent. This is indicated by the transmit buffer interrupt request flag
S0TBIR being set. S0TBUF may now be loaded with the next data, while transmission
of the previous one is still going on. The data bits are transmitted synchronous with the
shift clock. After the bit time for the 8th data bit, both pins TXD0 and RXD0 will go high,
the transmit interrupt request flag S0TIR is set, and serial data transmission stops.
Pin TXD0 must be configured for alternate data output, i.e. the respective port output
latch and the direction latch must be ‘1’, in order to provide the shift clock. Pin RXD0 must
also be configured for output (output/direction latch = ‘1’) during transmission.
Synchronous reception is initiated by setting bit S0REN = ‘1’. If bit S0R = ‘1’, the data
applied at pin RXD0 are clocked into the receive shift register synchronous to the clock
which is output at pin TXD0. After the 8th bit has been shifted in, the content of the
receive shift register is transferred to the receive data buffer S0RBUF, the receive
interrupt request flag S0RIR is set, the receiver enable bit S0REN is reset, and serial
data reception stops.
Pin TXD0 must be configured for alternate data output, i.e. the respective port output
latch and the direction latch must be ‘1’, in order to provide the shift clock. Pin RXD0 must
be configured as alternate data input, i.e. the respective direction latch must be ‘0’.
Synchronous reception is stopped by clearing bit S0REN. A currently received byte is
completed including the generation of the receive interrupt request and an error interrupt
request, if appropriate. Writing to the transmit buffer register while a reception is in
progress has no effect on reception and will not start a transmission.
If a previously received byte has not been read out of the receive buffer register at the
time the reception of the next byte is complete, both the error interrupt request flag
S0EIR and the overrun error status flag S0OE will be set, provided the overrun check
has been enabled by bit S0OEN.
User’s Manual
11-9
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
11.3
Hardware Error Detection Capabilities
To improve the safety of serial data exchange, the serial channel ASC0 provides an error
interrupt request flag, which indicates the presence of an error, and three (selectable)
error status flags in register S0CON, which indicate which error has been detected
during reception. Upon completion of a reception, the error interrupt request flag S0EIR
will be set simultaneously with the receive interrupt request flag S0RIR, if one or more of
the following conditions are met:
• If the framing error detection enable bit S0FEN is set and any of the expected stop bits
is not high, the framing error flag S0FE is set, indicating that the error interrupt request
is due to a framing error (Asynchronous mode only).
• If the parity error detection enable bit S0PEN is set in the modes where a parity bit is
received, and the parity check on the received data bits proves false, the parity error
flag S0PE is set, indicating that the error interrupt request is due to a parity error
(Asynchronous mode only).
• If the overrun error detection enable bit S0OEN is set and the last character received
was not read out of the receive buffer by software or PEC transfer at the time the
reception of a new frame is complete, the overrun error flag S0OE is set indicating that
the error interrupt request is due to an overrun error (Asynchronous and synchronous
mode).
User’s Manual
11-10
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
11.4
ASC0 Baud Rate Generation
The serial channel ASC0 has its own dedicated 13-bit baud rate generator with 13-bit
reload capability, allowing baud rate generation independent of the GPT timers.
The baud rate generator is clocked with the CPU clock divided by 2 (fCPU/2). The timer
is counting downwards and can be started or stopped through the Baud Rate Generator
Run Bit S0R in register S0CON. Each underflow of the timer provides one clock pulse to
the serial channel. The timer is reloaded with the value stored in its 13-bit reload register
each time it underflows. The resulting clock is again divided according to the operating
mode and controlled by the Baudrate Selection Bit S0BRS. If S0BRS = ‘1’, the clock
signal is additionally divided to 2/3rd of its frequency (see formulas and table). So the
baud rate of ASC0 is determined by the CPU clock, the reload value, the value of S0BRS
and the operating mode (asynchronous or synchronous).
Register S0BG is the dual-function Baud Rate Generator/Reload register. Reading
S0BG returns the content of the timer (bits 15 … 13 return zero), while writing to S0BG
always updates the reload register (bits 15 … 13 are insiginificant).
An auto-reload of the timer with the content of the reload register is performed each time
S0BG is written to. However, if S0R = ‘0’ at the time the write operation to S0BG is
performed, the timer will not be reloaded until the first instruction cycle after S0R = ‘1’.
Asynchronous Mode Baud Rates
For asynchronous operation, the baud rate generator provides a clock with 16 times the
rate of the established baud rate. Every received bit is sampled at the 7th, 8th and 9th
cycle of this clock. The baud rate for asynchronous operation of serial channel ASC0 and
the required reload value for a given baudrate can be determined by the following
formulas:
f
CPU
B
=
Async
16 × (2 + <S0BRS>) × (<S0BRL> + 1)
f
CPU
S0BRL = (
) – 1
16 × (2 + <S0BRS>) × B
Async
<S0BRL> represents the content of the reload register, taken as unsigned 13-bit integer,
<S0BRS> represents the value of bit S0BRS (i.e. ‘0’ or ‘1’), taken as integer.
The tables below list various commonly used baud rates together with the required
reload values and the deviation errors compared to the intended baudrate for a number
of CPU frequencies.
Note: The deviation errors given in the tables below are rounded. Using a baudrate
crystal (e.g. 18.432 MHz) will provide correct baudrates without deviation errors.
User’s Manual
11-11
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
Table 11-1 ASC0 Asynchronous Baudrate Generation for fCPU = 16 MHz
Baud Rate
S0BRS = ‘0’
S0BRS = ‘1’
Deviation Error Reload Value Deviation Error Reload Value
500
KBaud ± 0.0%
0000
–
–
H
19.2 KBaud + 0.2%/ – 3.5%
0019 /001A
+ 2.1%/ – 3.5%
+ 2.1%/ – 0.8%
+ 0.6%/ – 0.8%
+ 0.6%/ – 0.1%
+ 0.3%/ – 0.1%
+ 0.1%/ – 0.1%
+ 0.0%/ – 0.0%
+ 1.7%
0010 /0011
H
H
H
H
H
H
H
H
9600
4800
2400
1200
600
61
Baud + 0.2%/ – 1.7%
Baud + 0.2%/ – 0.8%
Baud + 0.2%/ – 0.3%
Baud + 0.4%/ – 0.1%
Baud + 0.0%/ – 0.1%
Baud + 0.1%
0033 /0034
0021 /0022
H
H
0067 /0068
0044 /0045
H
H
00CF /00D0
0089 /008A
H
H
H
H
H
019F /01A0
0114 /0115
H
H
H
0340 /0341
022A /022B
H H
H
H
1FFF
–
115B /115C
H H
H
40
Baud –
1FFF
H
Table 11-2 ASC0 Asynchronous Baudrate Generation for fCPU = 20 MHz
Baud Rate S0BRS = ‘0’ S0BRS = ‘1’
Deviation Error Reload Value Deviation Error Reload Value
625
KBaud ± 0.0%
0000
–
–
H
19.2 KBaud + 1.7%/ – 1.4%
001F /0020
+ 3.3%/ – 1.4%
+ 1.0%/ – 1.4%
+ 1.0%/ – 0.2%
+ 0.4%/ – 0.2%
+ 0.1%/ – 0.2%
+ 0.1%/ – 0.1%
+ 0.0%/ – 0.0%
+ 1.7%
0014 /0015
H H
H
H
H
H
H
H
H
9600
4800
2400
1200
600
75
Baud + 0.2%/ – 1.4%
Baud + 0.2%/ – 0.6%
Baud + 0.2%/ – 0.2%
Baud + 0.2%/ – 0.4%
Baud + 0.1%/ – 0.0%
Baud + 1.7%
0040 /0041
002A /002B
H H
H
0081 /0082
0055 /0056
H H
H
0103 /0104
00AC /00AD
H H
H
0207 /0208
015A /015B
H H
H
0410 /0411
02B5 /02B6
H
H
H
H
1FFF
–
15B2 /15B3
H
H
50
Baud –
1FFF
H
User’s Manual
11-12
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
Table 11-3 ASC0 Asynchronous Baudrate Generation for fCPU = 25 MHz
Baud Rate
S0BRS = ‘0’
S0BRS = ‘1’
Deviation Error Reload Value Deviation Error Reload Value
781
KBaud + 0.2%
0000
–
–
H
19.2 KBaud + 1.7%/ – 0.8%
0027 /0028
+ 0.5%/ – 3.1%
+ 0.5%/ – 1.4%
+ 0.5%/ – 0.5%
+ 0.0%/ – 0.5%
+ 0.0%/ – 0.2%
+ 0.0%/ – 0.1%
+ 0.0%/ – 0.0%
+ 1.0%
001A /001B
H H
H
H
H
9600
4800
2400
1200
600
95
Baud + 0.5%/ – 0.8%
Baud + 0.5%/ – 0.2%
Baud + 0.2%/ – 0.2%
Baud + 0.0%/ – 0.2%
Baud + 0.0%/ – 0.1%
Baud + 0.4%
0050 /0051
0035 /0036
H H
H
00A1 /00A2
006B /006C
H H
H
H
0145 /0146
00D8 /00D9
H
H
H
H
H
028A /028B
01B1 /01B2
H
H
H
0515 /0516
0363 /0364
H H
H
H
1FFF
–
1569 /156A
H
H
H
63
Baud –
1FFF
H
Table 11-4 ASC0 Asynchronous Baudrate Generation for fCPU = 33 MHz
Baud Rate S0BRS = ‘0’ S0BRS = ‘1’
Deviation Error Reload Value Deviation Error Reload Value
1.031 MBaud ± 0.0%
0000
–
–
H
19.2 KBaud + 1.3%/ – 0.5%
0034 /0035
+ 2.3%/ – 0.5%
+ 0.9%/ – 0.5%
+ 0.2%/ – 0.5%
+ 0.2%/ – 0.2%
+ 0.2%/ – 0.0%
+ 0.1%/ – 0.0%
± 0.0%
0022 /0023
H
H
H
H
H
9600
4800
2400
1200
600
Baud + 0.4%/ – 0.5%
Baud + 0.4%/ – 0.1%
Baud + 0.2%/ – 0.1%
Baud + 0.0%/ – 0.1%
Baud + 0.0%/ – 0.0%
Baud + 7.1%
006A /006B
0046 /0047
H
H
H
00D5 /00D6
008E /008F
H H
H
H
01AC /01AD
011D /011E
H
H
H
H
H
035A /035B
023B /023C
H
H
H
06B5 /06B6
0478 /0479
H H
H
H
125
1FFF
–
157C
1FFF
H
H
H
84
Baud –
– 0.9%
User’s Manual
11-13
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
Synchronous Mode Baud Rates
For synchronous operation, the baud rate generator provides a clock with 4 times the
rate of the established baud rate. The baud rate for synchronous operation of serial
channel ASC0 can be determined by the following formula:
f
CPU
S0BRL = (
) – 1
4 × (2 + <S0BRS>) × B
Sync
f
CPU
B
=
Sync
4 × (2 + <S0BRS>) × (<S0BRL> + 1)
<S0BRL> represents the content of the reload register, taken as unsigned 13-bit integers,
<S0BRS> represents the value of bit S0BRS (i.e. ‘0’ or ‘1’), taken as integer.
Table 11-5 gives the limit baudrates depending on the CPU clock frequency and bit
S0BRS.
Table 11-5 ASC0 Synchronous Baudrate Generation
CPU clock
S0BRS = ‘0’
S0BRS = ‘1’
fCPU
Min. Baudrate
244 Baud
Max. Baudrate Min. Baudrate Max. Baudrate
16 MHz
20 MHz
25 MHz
33 MHz
2.000 MBaud
2.500 MBaud
3.125 MBaud
4.125 MBaud
162 Baud
203 Baud
254 Baud
336 Baud
1.333 MBaud
1.666 MBaud
2.083 MBaud
2.750 MBaud
305 Baud
381 Baud
504 Baud
User’s Manual
11-14
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
11.5
ASC0 Interrupt Control
Four bit addressable interrupt control registers are provided for serial channel ASC0.
Register S0TIC controls the transmit interrupt, S0TBIC controls the transmit buffer
interrupt, S0RIC controls the receive interrupt and S0EIC controls the error interrupt of
serial channel ASC0. Each interrupt source also has its own dedicated interrupt vector.
S0TINT is the transmit interrupt vector, S0TBINT is the transmit buffer interrupt vector,
S0RINT is the receive interrupt vector, and S0EINT is the error interrupt vector.
The cause of an error interrupt request (framing, parity, overrun error) can be identified
by the error status flags in control register S0CON.
Note: In contrast to the error interrupt request flag S0EIR, the error status flags S0FE/
S0PE/S0OE are not reset automatically upon entry into the error interrupt service
routine, but must be cleared by software.
S0TIC
ASC0 Tx Intr. Ctrl. Reg.
SFR (FF6C /B6 )
Reset value: - - 00
H
H
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
5
5
4
4
4
3
2
1
0
S0
S0
-
ILVL
GLVL
rw
TIR TIE
-
-
-
-
-
-
-
rwh rw
rw
S0TBIC
ASC0 Tx Buf. Intr. Ctrl. Reg.
SFR (FF9C /CE )
Reset value: - - 00
H
H
15
14
13
12
11
10
9
8
7
6
3
2
1
0
S0
S0
-
ILVL
GLVL
rw
TBIR TBIE
-
-
-
-
-
-
-
-
rwh rw
rw
S0RIC
ASC0 Rx Intr. Ctrl. Reg.
SFR (FF6E /B7 )
Reset value: - - 00
H
H
15
14
13
12
11
10
-
9
8
7
6
3
2
1
0
S0
S0
-
ILVL
GLVL
rw
RIR RIE
-
-
-
-
-
-
-
rwh rw
rw
User’s Manual
11-15
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
S0EIC
ASC0 Error Intr. Ctrl. Reg.
SFR (FF70 /B8 )
Reset value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
S0
EIR EIE
S0
-
ILVL
GLVL
rw
-
-
-
-
-
-
-
rwh rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
Using the ASC0 Interrupts
For normal operation (i.e. besides the error interrupt) the ASC0 provides three interrupt
requests to control data exchange via this serial channel:
• S0TBIR is activated when data is moved from S0TBUF to the transmit shift register.
• S0TIR is activated before the last bit of an asynchronous frame is transmitted,
or after the last bit of a synchronous frame has been transmitted.
• S0RIR is activated when the received frame is moved to S0RBUF.
While the task of the receive interrupt handler is quite clear, the transmitter is serviced
by two interrupt handlers. This provides advantages for the servicing software.
For single transfers is sufficient to use the transmitter interrupt (S0TIR), which indicates
that the previously loaded data has been transmitted, except for the last bit of an
asynchronous frame.
For multiple back-to-back transfers it is necessary to load the following piece of data
at last until the time the last bit of the previous frame has been transmitted. In
asynchronous mode this leaves just one bit-time for the handler to respond to the
transmitter interrupt request, in synchronous mode it is impossible at all.
Using the transmit buffer interrupt (S0TBIR) to reload transmit data gives the time to
transmit a complete frame for the service routine, as S0TBUF may be reloaded while the
previous data is still being transmitted.
User’s Manual
11-16
V3.1, 2000-03
C167CR
Derivatives
The Asynchronous/Synchronous Serial Interface
S0TIR
S0TBIR
S0TIR
S0TBIR
S0TIR
S0TBIR
Idle
Idle
S0RIR
S0RIR
S0RIR
S0TIR
Asynchronous Mode
S0TIR
S0TIR
S0TBIR
S0TBIR
S0TBIR
Idle
Idle
S0RIR
S0RIR
S0RIR
Synchronous Mode
MCT04379
Figure 11-6 ASC0 Interrupt Generation
As shown in Figure 11-6, S0TBIR is an early trigger for the reload routine, while S0TIR
indicates the completed transmission. Software using handshake therefore should rely
on S0TIR at the end of a data block to make sure that all data has really been
transmitted.
User’s Manual
11-17
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
12
The High-speed Synchronous Serial Interface
The high-speed Synchronous Serial Interface SSC provides flexible high-speed serial
communication between the C167CR and other microcontrollers, microprocessors or
external peripherals.
The SSC supports full-duplex and half-duplex synchronous communication up to 6.25/
8.25 MBaud (@ 25/33 MHz CPU clock). The serial clock signal can be generated by the
SSC itself (master mode) or be received from an external master (slave mode). Data
width, shift direction, clock polarity and phase are programmable. This allows
communication with SPI-compatible devices. Transmission and reception of data is
double-buffered. A 16-bit baud rate generator provides the SSC with a separate serial
clock signal.
The high-speed synchronous serial interface can be configured in a very flexible way, so
it can be used with other synchronous serial interfaces (e.g. the ASC0 in synchronous
mode), serve for master/slave or multimaster interconnections or operate compatible
with the popular SPI interface. So it can be used to communicate with shift registers (IO
expansion), peripherals (e.g. EEPROMs etc.) or other controllers (networking). The SSC
supports half-duplex and full-duplex communication. Data is transmitted or received on
pins MTSR/P3.9 (Master Transmit/Slave Receive) and MRST/P3.8 (Master Receive/
Slave Transmit). The clock signal is output or input on pin SCLK/P3.13. These pins are
alternate functions of Port 3 pins.
Ports & Direction Control
Alternate Functions
Data Registers
Control Registers
SSCCON
Interrupt Control
E
ODP3
DP3
P3
SSCBR
SSCTB
SSCRB
E
SSCTIC
SSCRIC
SSCEIC
E
E
SCLK/P3.13
MTSR/P3.9
MRST/P3.8
ODP3
DP3
Port 3 Open Drain Control Register
Port 3 Direction Control Register
P3
Port 3 Data Register
SSCCON SSC Control Register
SSCBR SSC Baud Rate Generator/Reload Reg. SSCRB
SSCTB SSC Transmit Buffer Register SSCRIC
SSCTIC SSC Transmit Interrupt Control Register SSCEIC
SSC Receive Buffer Register
SSC Receive Interrupt Control Register
SSC Error Interrupt Control Register
MCA04380
Figure 12-1 SFRs and Port Pins Associated with the SSC
User’s Manual
12-1
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
Slave Clock
Clock
Control
CPU
Clock
Baud Rate
Generator
SCLK
Master Clock
Shift
Clock
Receive Int. Request
Transmit Int. Request
Error Int.Request
SSC Control Block
Status
Control
MTSR
MRST
Pin
Control
16-Bit Shift Register
Transmit Buffer
Register SSCTB
Receive Buffer
Register SSCRB
I n t e r n a l B u s
MCB01957
Figure 12-2 Synchronous Serial Channel SSC Block Diagram
The operating mode of the serial channel SSC is controlled by its bit-addressable control
register SSCCON. This register serves for two purposes:
• during programming (SSC disabled by SSCEN = ‘0’) it provides access to a set of
control bits,
• during operation (SSC enabled by SSCEN = ‘1’) it provides access to a set of status
flags.
Register SSCCON is shown below in each of the two modes.
SSCCON
SSC Control Reg. (Pr.M.)
SFR (FFB2 /D9 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
SSC
EN
= 0
SSC
AR
EN
SSC
MS
SSC SSC SSC SSC
BEN PEN REN TEN
SSC SSC SSC
PO PH HB
-
-
SSCBM
rw
rw
-
rw
rw
rw
rw
rw
rw
rw
rw
rw
rw
User’s Manual
12-2
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
Bit
Function (Programming Mode, SSCEN = ‘0’)
SSC Data Width Selection
SSCBM
0:
Reserved. Do not use this combination.
1…15: Transfer Data Width is 2 … 16 bit (<SSCBM> + 1)
SSCHB
SSC Heading Control Bit
0:
1:
Transmit/Receive LSB First
Transmit/Receive MSB First
SSCPH
SSC Clock Phase Control Bit
0:
1:
Shift transmit data on the leading clock edge, latch on trailing edge
Latch receive data on leading clock edge, shift on trailing edge
SSCPO
SSC Clock Polarity Control Bit
0:
1:
Idle clock line is low, leading clock edge is low-to-high transition
Idle clock line is high, leading clock edge is high-to-low transition
SSCTEN
SSCREN
SSCPEN
SSCBEN
SSC Transmit Error Enable Bit
0:
1:
Ignore transmit errors
Check transmit errors
SSC Receive Error Enable Bit
0:
1:
Ignore receive errors
Check receive errors
SSC Phase Error Enable Bit
0:
1:
Ignore phase errors
Check phase errors
SSC Baudrate Error Enable Bit
0:
1:
Ignore baudrate errors
Check baudrate errors
SSCAREN SSC Automatic Reset Enable Bit
0:
1:
No additional action upon a baudrate error
The SSC is automatically reset upon a baudrate error
SSCMS
SSCEN
SSC Master Select Bit
0:
1:
Slave Mode. Operate on shift clock received via SCLK.
Master Mode. Generate shift clock and output it via SCLK.
SSC Enable Bit = ‘0’
Transmission and reception disabled. Access to control bits.
User’s Manual
12-3
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
SSCCON
SSC Control Reg. (Op.M.)
SFR (FFB2 /D9 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
SSC
EN
= 1
SSC
MS
SSC SSC SSC SSC SSC
BSY BE PE RE TE
-
-
-
-
-
SSCBC
rw
rw
-
rw
rw
rw
rw
rw
-
-
-
-
r
Bit
Function (Operating Mode, SSCEN = ‘1’)
SSC Bit Count Field
Shift counter is updated with every shifted bit. Do not write to!!!
SSCBC
SSCTE
SSCRE
SSCPE
SSCBE
SSC Transmit Error Flag
1:
Transfer starts with the slave’s transmit buffer not being updated
SSC Receive Error Flag
1:
Reception completed before the receive buffer was read
SSC Phase Error Flag
1:
Received data changes around sampling clock edge
SSC Baudrate Error Flag
1:
More than factor 2 or less than factor 0.5 between Slave’s actual
and expected baudrate
SSCBSY
SSCMS
SSC Busy Flag
Set while a transfer is in progress. Do not write to!!!
SSC Master Select Bit
0:
1:
Slave Mode. Operate on shift clock received via SCLK.
Master Mode. Generate shift clock and output it via SCLK.
SSCEN
SSC Enable Bit = ‘1’
Transmission and reception enabled. Access to status flags and M/S
control.
Note: The target of an access to SSCCON (control bits or flags) is determined by the
state of SSCEN prior to the access, i.e. writing C057 to SSCCON in
H
programming mode (SSCEN = ‘0’) will initialize the SSC (SSCEN was ‘0’) and then
turn it on (SSCEN = ‘1’).
When writing to SSCCON, make sure that reserved locations receive zeros.
User’s Manual
12-4
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
The shift register of the SSC is connected to both the transmit pin and the receive pin via
the pin control logic (see block diagram). Transmission and reception of serial data is
synchronized and takes place at the same time, i.e. the number of transmitted bits is also
received.
The major steps of the state machine of the SSC are controlled by the shift clock signal
(see Figure 12-3).
In master mode (SSCMS = ‘1’) two clocks per bit-time are generated, each upon an
underflow of the baudrate counter.
In slave mode (SSCMS = ‘0’) one clock per bit-time is generated, when the latching
edge of the external SCLK signal has been detected.
Transmit data is written into the transmit buffer SSCTB. When the contents of the buffer
are moved to the shift register (immediately if no transfer is currently active) a transmit
interrupt request (SSCTIR) is generated indicating that SSCTB may be reloaded again.
The busy flag SSCBSY is set when the transfer starts (with the next following shift clock
in master mode, immediately in slave mode).
Note: If no data is written to SSCTB prior to a slave transfer, this transfer starts after the
first latching edge of the external SCLK signal is detected. No transmit interrupt is
generated in this case.
When the contents of the shift register are moved to the receive buffer SSCRB after the
programmed number of bits (2 … 16) have been transferred, i.e. after the last latching
edge of the current transfer, a receive interrupt request (SSCRIR) is generated.
The busy flag SSCBSY is cleared at the end of the current transfer (with the next
following shift clock in master mode, immediately in slave mode).
When the transmit buffer is not empty at that time (in the case of continuous transfers)
the busy flag is not cleared and the transfer goes on after moving data from the buffer to
the shift register.
Software should not modify SSCBSY, as this flag is hardware controlled.
Note: Only one SSC (etc.) can be master at a given time.
The transfer of serial data bits can be programmed in many respects:
• the data width can be chosen from 2 bits to 16 bits
• transfer may start with the LSB or the MSB
• the shift clock may be idle low or idle high
• data bits may be shifted with the leading or trailing edge of the clock signal
• the baudrate may be set within a wide range (see baudrate generation)
• the shift clock can be generated (master) or received (slave)
This allows the adaptation of the SSC to a wide range of applications, where serial data
transfer is required.
User’s Manual
12-5
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
The Data Width Selection supports the transfer of frames of any length, from 2-bit
“characters” up to 16-bit “characters”. Starting with the LSB (SSCHB = ‘0’) allows
communication e.g. with ASC0 devices in synchronous mode (C166 Family) or 8051 like
serial interfaces. Starting with the MSB (SSCHB = ‘1’) allows operation compatible with
the SPI interface.
Regardless which data width is selected and whether the MSB or the LSB is transmitted
first, the transfer data is always right aligned in registers SSCTB and SSCRB, with the
LSB of the transfer data in bit 0 of these registers. The data bits are rearranged for
transfer by the internal shift register logic. The unselected bits of SSCTB are ignored, the
unselected bits of SSCRB will be not valid and should be ignored by the receiver service
routine.
The Clock Control allows the adaptation of transmit and receive behavior of the SSC to
a variety of serial interfaces. A specific clock edge (rising or falling) is used to shift out
transmit data, while the other clock edge is used to latch in receive data. Bit SSCPH
selects the leading edge or the trailing edge for each function. Bit SSCPO selects the
level of the clock line in the idle state. So for an idle-high clock the leading edge is a
falling one, a 1-to-0 transition. Figure 12-3 is a summary.
Figure 12-3 Serial Clock Phase and Polarity Options
User’s Manual
12-6
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
12.1
Full-duplex Operation
The different devices are connected through three lines. The definition of these lines is
always determined by the master: The line connected to the master’s data output pin
MTSR is the transmit line, the receive line is connected to its data input line MRST, and
the clock line is connected to pin SCLK. Only the device selected for master operation
generates and outputs the serial clock on pin SCLK. All slaves receive this clock, so their
pin SCLK must be switched to input mode (DP3.13 = ‘0’). The output of the master’s shift
register is connected to the external transmit line, which in turn is connected to the
slaves’ shift register input. The output of the slaves’ shift register is connected to the
external receive line in order to enable the master to receive the data shifted out of the
slave. The external connections are hard-wired, the function and direction of these pins
is determined by the master or slave operation of the individual device.
Note: The shift direction shown in Figure 12-4 applies for MSB-first operation as well as
for LSB-first operation.
When initializing the devices in this configuration, select one device for master operation
(SSCMS = ‘1’), all others must be programmed for slave operation (SSCMS = ‘0’).
Initialization includes the operating mode of the device’s SSC and also the function of
the respective port lines (see “Port Control”).
Master
Device #1
MTSR
Device #2
Slave
Shift Register
Shift Register
Transmit
MTSR
MRST
Receive
Clock
MRST
SCLK
SCLK
Clock
Clock
Device #3
Slave
Shift Register
MTSR
MRST
SCLK
Clock
MCS01963
Figure 12-4 SSC Full-duplex Configuration
User’s Manual
12-7
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
The data output pins MRST of all slave devices are connected together onto the one
receive line in this configuration. During a transfer each slave shifts out data from its shift
register. There are two ways to avoid collisions on the receive line due to different slave
data:
Only one slave drives the line, i.e. enables the driver of its MRST pin. All the other
slaves have to program their MRST pins to input. So only one slave can put its data onto
the master’s receive line. Only receiving of data from the master is possible. The master
selects the slave device from which it expects data either by separate select lines, or by
sending a special command to this slave. The selected slave then switches its MRST line
to output, until it gets a deselection signal or command.
The slaves use open drain output on MRST. This forms a Wired-AND connection. The
receive line needs an external pullup in this case. Corruption of the data on the receive
line sent by the selected slave is avoided, when all slaves which are not selected for
transmission to the master only send ones (‘1’). Since this high level is not actively driven
onto the line, but only held through the pullup device, the selected slave can pull this line
actively to a low level when transmitting a zero bit. The master selects the slave device
from which it expects data either by separate select lines, or by sending a special
command to this slave.
After performing all necessary initializations of the SSC, the serial interfaces can be
enabled. For a master device, the alternate clock line will now go to its programmed
polarity. The alternate data line will go to either ‘0’ or ‘1’, until the first transfer will start.
When the serial interfaces are enabled, the master device can initiate the first data
transfer by writing the transmit data into register SSCTB. This value is copied into the
shift register (which is assumed to be empty at this time), and the selected first bit of the
transmit data will be placed onto the MTSR line on the next clock from the baudrate
generator (transmission only starts, if SSCEN = ‘1’). Depending on the selected clock
phase, also a clock pulse will be generated on the SCLK line. With the opposite clock
edge the master at the same time latches and shifts in the data detected at its input line
MRST. This “exchanges” the transmit data with the receive data. Since the clock line is
connected to all slaves, their shift registers will be shifted synchronously with the
master’s shift register, shifting out the data contained in the registers, and shifting in the
data detected at the input line. After the preprogrammed number of clock pulses (via the
data width selection) the data transmitted by the master is contained in all slaves’ shift
registers, while the master’s shift register holds the data of the selected slave. In the
master and all slaves the content of the shift register is copied into the receive buffer
SSCRB and the receive interrupt flag SSCRIR is set.
A slave device will immediately output the selected first bit (MSB or LSB of the transfer
data) at pin MRST, when the content of the transmit buffer is copied into the slave’s shift
register. It will not wait for the next clock from the baudrate generator, as the master
does. The reason for this is that, depending on the selected clock phase, the first clock
User’s Manual
12-8
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
edge generated by the master may be already used to clock in the first data bit. So the
slave’s first data bit must already be valid at this time.
Note: On the SSC always a transmission and a reception takes place at the same time,
regardless whether valid data has been transmitted or received. This is different
e.g. from asynchronous reception on ASC0.
The initialization of the SCLK pin on the master requires some attention in order to
avoid undesired clock transitions, which may disturb the other receivers. The state of the
internal alternate output lines is ‘1’ as long as the SSC is disabled. This alternate output
signal is ANDed with the respective port line output latch. Enabling the SSC with an idle-
low clock (SSCPO = ‘0’) will drive the alternate data output and (via the AND) the port
pin SCLK immediately low. To avoid this, use the following sequence:
• select the clock idle level (SSCPO = ‘x’)
• load the port output latch with the desired clock idle level (P3.13 = ‘x’)
• switch the pin to output (DP3.13 = ‘1’)
• enable the SSC (SSCEN = ‘1’)
• if SSCPO = ‘0’: enable alternate data output (P3.13 = ‘1’)
The same mechanism as for selecting a slave for transmission (separate select lines or
special commands) may also be used to move the role of the master to another device
in the network. In this case the previous master and the future master (previous slave)
will have to toggle their operating mode (SSCMS) and the direction of their port pins (see
description above).
User’s Manual
12-9
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
12.2
Half-duplex Operation
In a half duplex configuration only one data line is necessary for both receiving and
transmitting of data. The data exchange line is connected to both pins MTSR and MRST
of each device, the clock line is connected to the SCLK pin.
The master device controls the data transfer by generating the shift clock, while the slave
devices receive it. Due to the fact that all transmit and receive pins are connected to the
one data exchange line, serial data may be moved between arbitrary stations.
Similar to full duplex mode there are two ways to avoid collisions on the data
exchange line:
• only the transmitting device may enable its transmit pin driver
• the non-transmitting devices use open drain output and only send ones.
Since the data inputs and outputs are connected together, a transmitting device will clock
in its own data at the input pin (MRST for a master device, MTSR for a slave). By these
means any corruptions on the common data exchange line are detected, where the
received data is not equal to the transmitted data.
Master
Device #1
MTSR
Device #2
Slave
Shift Register
Shift Register
MTSR
MRST
MRST
SCLK
Clock
SCLK
Clock
Clock
Common
Transmit/
Receive
Line
Device #3
Slave
Shift Register
MTSR
MRST
SCLK
Clock
MCS01965
Figure 12-5 SSC Half-duplex Configuration
User’s Manual
12-10
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
12.3
Continuous Transfers
When the transmit interrupt request flag is set, it indicates that the transmit buffer SSCTB
is empty and ready to be loaded with the next transmit data. If SSCTB has been reloaded
by the time the current transmission is finished, the data is immediately transferred to the
shift register and the next transmission will start without any additional delay. On the data
line there is no gap between the two successive frames. E.g. two byte transfers would
look the same as one word transfer. This feature can be used to interface with devices
which can operate with or require more than 16 data bits per transfer. It is just a matter
of software, how long a total data frame length can be. This option can also be used e.g.
to interface to byte-wide and word-wide devices on the same serial bus.
Note: Of course, this can only happen in multiples of the selected basic data width, since
it would require disabling/enabling of the SSC to reprogram the basic data width
on-the-fly.
User’s Manual
12-11
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
12.4
Port Control
The SSC uses three pins of Port 3 to communicate with the external world. Pin P3.13/
SCLK serves as the clock line, while pins P3.8/MRST (Master Receive/Slave Transmit)
and P3.9/MTSR (Master Transmit/Slave Receive) serve as the serial data input/output
lines.The operation of these pins depends on the selected operating mode (master or
slave). In order to enable the alternate output functions of these pins instead of the
general purpose IO operation, the respective port latches have to be set to ‘1’, since the
port latch outputs and the alternate output lines are ANDed. When an alternate data
output line is not used (function disabled), it is held at a high level, allowing IO operations
via the port latch. The direction of the port lines depends on the operating mode. The
SSC will automatically use the correct alternate input or output line of the ports when
switching modes. The direction of the pins, however, must be programmed by the user,
as shown in the tables. Using the open drain output feature helps to avoid bus contention
problems and reduces the need for hardwired hand-shaking or slave select lines. In this
case it is not always necessary to switch the direction of a port pin. Table 12-1
summarizes the required values for the different modes and pins.
Table 12-1 SSC Port Control
Pin
Master Mode
Slave Mode
Function Port Latch Direction
Function Port Latch Direction
SCLK
Serial
Clock
Output
P3.13 = ‘1’ DP3.13 = ‘1’ Serial
P3.13 = ‘x’ DP3.13 = ‘0’
Clock
Input
MTSR Serial
Data
P3.9 = ‘1’
P3.8 = ‘x’
DP3.9 = ‘1’ Serial
P3.9 = ‘x’
P3.8 = ‘1’
DP3.9 = ‘0’
DP3.8 = ‘1’
Data
Input
Output
MRST Serial
Data
DP3.8 = ‘0’ Serial
Data
Input
Output
Note: In Table 12-1, an ‘x’ means that the actual value is irrelevant in the respective
mode, however, it is recommended to set these bits to ‘1’, so they are already in
the correct state when switching between master and slave mode.
User’s Manual
12-12
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
12.5
Baud Rate Generation
The serial channel SSC has its own dedicated 16-bit baud rate generator with 16-bit
reload capability, permitting baud rate generation independent from the timers.
The baud rate generator is clocked with the CPU clock divided by 2 (fCPU / 2). The timer
is counting downwards and can be started or stopped through the global enable bit
SSCEN in register SSCCON. Register SSCBR is the dual-function Baud Rate
Generator/Reload register. Reading SSCBR, while the SSC is enabled, returns the
content of the timer. Reading SSCBR, while the SSC is disabled, returns the
programmed reload value. In this mode the desired reload value can be written to
SSCBR.
Note: Never write to SSCBR, while the SSC is enabled.
The formulas below calculate either the resulting baud rate for a given reload value, or
the required reload value for a given baudrate:
fCPU
fCPU
,
SSCBR = (
) – 1
B
=
SSC
2 × Baudrate
2 × (<SSCBR> + 1)
SSC
<SSCBR> represents the content of the reload register, taken as an unsigned 16-bit
integer.
The tables below list some possible baud rates together with the required reload values
and the resulting bit times, for different CPU clock frequencies.
Table 12-2 SSC Bit-Time Calculation
Bit-time for fCPU = …
Reload Val.
(SSCBR)
16 MHz
20 MHz
25 MHz
33 MHz
Reserved. SSCBR must be > 0.
0000
0001
0002
0003
0004
0007
0009
H
H
H
H
H
H
H
250
375
500
625
1.00
1.25
10
ns
ns
ns
ns
µs
µs
µs
µs
µs
µs
200
300
400
500
800
1
ns
ns
ns
ns
ns
µs
µs
µs
µs
µs
160
240
320
400
640
800
6.4
8
ns
ns
ns
ns
ns
ns
µs
µs
µs
µs
121
182
242
303
485
606
4.8
ns
ns
ns
ns
ns
ns
µs
µs
µs
µs
8
004F
H
H
12.5
15.6
20.6
10
6.1
0063
12.5
16.5
10
7.6
007C
H
13.2
10
00A4
H
User’s Manual
12-13
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
Table 12-2 SSC Bit-Time Calculation (cont’d)
Bit-time for fCPU = …
Reload Val.
(SSCBR)
16 MHz
1
20 MHz
25 MHz
640
800
1
33 MHz
485
ms
ms
ms
ms
800
1
µs
µs
µs
µs
µs
ms
1F3F
H
1.25
1.56
8.2
ms
ms
ms
µs
606
270F
H
1.25
6.6
ms
ms
758
30D3
H
5.2
4.0
FFFF
H
Table 12-3 SSC Baudrate Calculation
Baud Rate for fCPU = …
20 MHz 25 MHz
Reload Val.
(SSCBR)
16 MHz
33 MHz
Reserved. SSCBR must be > 0.
0000
H
H
H
H
H
H
H
4.00
2.67
2.00
1.60
1.00
800
100
80
MBaud 5.00
MBaud 6.25
MBaud 8.25
MBaud 0001
MBaud 0002
MBaud 0003
MBaud 0004
MBaud 0007
MBaud 0009
MBaud
MBaud
MBaud
3.33
2.50
2.00
MBaud 4.17
MBaud 3.13
MBaud 2.50
MBaud 1.56
MBaud 1.25
MBaud 5.50
MBaud 4.13
MBaud 3.30
MBaud 2.06
MBaud 1.65
MBaud 1.25
KBaud
KBaud
KBaud
KBaud
KBaud
KBaud
Baud
1.0
125
100
80
KBaud
156
KBaud
KBaud
206
165
KBaud
KBaud
KBaud
004F
H
H
KBaud 125
0063
64
KBaud
KBaud
KBaud
100
KBaud 132
007C
H
H
H
H
48.5
1.0
60.6
1.25
1.0
75.8
1.56
KBaud
KBaud
KBaud
100
KBaud 00A4
2.06
1.65
KBaud
KBaud
KBaud
1F3F
270F
30D3
800
640
KBaud 1.25
Baud
800
Baud
1.0
KBaud 1.32
H
122.1 Baud
152.6 Baud
190.7 Baud
251.7 Baud
FFFF
H
User’s Manual
12-14
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
12.6
Error Detection Mechanisms
The SSC is able to detect four different error conditions. Receive Error and Phase Error
are detected in all modes, while Transmit Error and Baudrate Error only apply to slave
mode. When an error is detected, the respective error flag is set. When the
corresponding Error Enable Bit is set, also an error interrupt request will be generated
by setting SSCEIR (see Figure 12-6). The error interrupt handler may then check the
error flags to determine the cause of the error interrupt. The error flags are not reset
automatically (like SSCEIR), but rather must be cleared by software after servicing. This
allows servicing of some error conditions via interrupt, while the others may be polled by
software.
Note: The error interrupt handler must clear the associated (enabled) errorflag(s) to
prevent repeated interrupt requests.
A Receive Error (Master or Slave mode) is detected, when a new data frame is
completely received, but the previous data was not read out of the receive buffer register
SSCRB. This condition sets the error flag SSCRE and, when enabled via SSCREN, the
error interrupt request flag SSCEIR. The old data in the receive buffer SSCRB will be
overwritten with the new value and is unretrievably lost.
A Phase Error (Master or Slave mode) is detected, when the incoming data at pin MRST
(master mode) or MTSR (slave mode), sampled with the same frequency as the CPU
clock, changes between one sample before and two samples after the latching edge of
the clock signal (see “Clock Control”). This condition sets the error flag SSCPE and,
when enabled via SSCPEN, the error interrupt request flag SSCEIR.
A Baud Rate Error (Slave mode) is detected, when the incoming clock signal deviates
from the programmed baud rate by more than 100%, i.e. it either is more than double or
less than half the expected baud rate. This condition sets the error flag SSCBE and,
when enabled via SSCBEN, the error interrupt request flag SSCEIR. Using this error
detection capability requires that the slave’s baud rate generator is programmed to the
same baud rate as the master device. This feature detects false additional, or missing
pulses on the clock line (within a certain frame).
Note: If this error condition occurs and bit SSCAREN = ‘1’, an automatic reset of the SSC
will be performed in case of this error. This is done to reinitialize the SSC, if too
few or too many clock pulses have been detected.
User’s Manual
12-15
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
A Transmit Error (Slave mode) is detected, when a transfer was initiated by the master
(shift clock gets active), but the transmit buffer SSCTB of the slave was not updated
since the last transfer. This condition sets the error flag SSCTE and, when enabled via
SSCTEN, the error interrupt request flag SSCEIR. If a transfer starts while the transmit
buffer is not updated, the slave will shift out the ‘old’ contents of the shift register, which
normally is the data received during the last transfer.
This may lead to the corruption of the data on the transmit/receive line in half-duplex
mode (open drain configuration), if this slave is not selected for transmission. This mode
requires that slaves not selected for transmission only shift out ones, i.e. their transmit
buffers must be loaded with ‘FFFF ’ prior to any transfer.
H
Note: A slave with push/pull output drivers, which is not selected for transmission, will
normally have its output drivers switched. However, in order to avoid possible
conflicts or misinterpretations, it is recommended to always load the slave’s
transmit buffer prior to any transfer.
Register SSCCON
Register SSCEIC
SSCTEN
SSCTE
&
&
&
&
Transmit
Error
SSCREN
SSCRE
SSCEIE
&
Error
Interrupt
SSCEINT
Receive
Error
1
SSCEIR
SSCPEN
SSCPE
Phase
Error
SSCBEN
SSCBE
Baudrate
Error
MCA01968
Figure 12-6 SSC Error Interrupt Control
User’s Manual
12-16
V3.1, 2000-03
C167CR
Derivatives
The High-speed Synchronous Serial Interface
12.7
SSC Interrupt Control
Three bit addressable interrupt control registers are provided for serial channel SSC.
Register SSCTIC controls the transmit interrupt, SSCRIC controls the receive interrupt
and SSCEIC controls the error interrupt of serial channel SSC. Each interrupt source
also has its own dedicated interrupt vector. SCTINT is the transmit interrupt vector,
SCRINT is the receive interrupt vector, and SCEINT is the error interrupt vector.
The cause of an error interrupt request (receive, phase, baudrate, transmit error) can be
identified by the error status flags in control register SSCCON.
Note: In contrary to the error interrupt request flag SSCEIR, the error status flags SSCxE
are not reset automatically upon entry into the error interrupt service routine, but
must be cleared by software.
SSCTIC
SSC Transmit Intr. Ctrl. Reg.
SFR (FF72 /B9 )
Reset value: - - 00
H
H
H
H
H
15
14
13
12
11
10
9
8
7
6
5
5
5
4
4
4
3
2
1
0
SSC SSC
TIR TIE
ILVL
GLVL
rw
-
-
-
-
-
-
-
-
rw
rw
rw
SSCRIC
SSC Receive Intr. Ctrl. Reg.
SFR (FF74 /BA )
Reset value: - - 00
H
H
15
14
13
12
11
10
9
8
7
6
3
2
1
0
SSC SSC
RIR RIE
ILVL
GLVL
rw
-
-
-
-
-
-
-
-
rw
rw
rw
SSCEIC
SSC Error Intr. Ctrl. Reg.
SFR (FF76 /BB )
Reset value: - - 00
H
H
15
14
13
12
11
10
-
9
8
7
6
3
2
1
0
SSC SSC
EIR EIE
ILVL
GLVL
rw
-
-
-
-
-
-
-
rw
rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
User’s Manual
12-17
V3.1, 2000-03
C167CR
Derivatives
The Watchdog Timer (WDT)
13
The Watchdog Timer (WDT)
To allow recovery from software or hardware failure, the C167CR provides a Watchdog
Timer. If the software fails to service this timer before an overflow occurs, an internal
reset sequence will be initiated. This internal reset will also pull the RSTOUT pin low,
which also resets the peripheral hardware which might be the cause for the malfunction.
When the watchdog timer is enabled and the software has been designed to service it
regularly before it overflows, the watchdog timer will supervise the program execution as
it only will overflow if the program does not progress properly. The watchdog timer will
also time out if a software error was due to hardware related failures. This prevents the
controller from malfunctioning for longer than a user-specified time.
Note: When the bidirectional reset is enabled also pin RSTIN will be pulled low for the
duration of the internal reset sequence upon a software reset or a watchdog timer
reset.
The watchdog timer provides two registers:
• a read-only timer register that contains the current count, and
• a control register for initialization and reset source detection.
Reset Indication Pins
Data Registers
WDT
Control Registers
WDTCON
RSTOUT
(deactivated by EINIT)
RSTIN
(bidirectional reset only)
MCA04381
Figure 13-1 SFRs and Port Pins Associated with the Watchdog Timer
The watchdog timer is a 16-bit up counter which is clocked with the prescaled CPU clock
(fCPU). The prescaler divides the CPU clock:
• by 2 (WDTIN = ‘0’), or
• by 128 (WDTIN = ‘1’).
User’s Manual
13-1
V3.1, 2000-03
C167CR
Derivatives
The Watchdog Timer (WDT)
The 16-bit watchdog timer is realized as two concatenated 8-bit timers (see
Figure 13-2). The upper 8 bits of the watchdog timer can be preset to a
user-programmable value via a watchdog service access in order to vary the watchdog
expire time. The lower 8 bits are reset upon each service access.
÷ 2
fCPU
MUX
WDT Low Byte
Clear
WDT High Byte
WDTR
÷ 128
RSTOUT
WDTIN
Reset
WDT
Control
WDTREL
MCB02052
Figure 13-2 Watchdog Timer Block Diagram
User’s Manual
13-2
V3.1, 2000-03
C167CR
Derivatives
The Watchdog Timer (WDT)
13.1
Operation of the Watchdog Timer
The current count value of the Watchdog Timer is contained in the Watchdog Timer
Register WDT which is a non-bitaddressable read-only register. The operation of the
Watchdog Timer is controlled by its bitaddressable Watchdog Timer Control Register
WDTCON. This register specifies the reload value for the high byte of the timer, selects
the input clock prescaling factor and also provides flags that indicate the source of a
reset.
After any reset (except see note) the watchdog timer is enabled and starts counting up
from 0000 with the default frequency fWDT = fCPU / 2. The default input frequency may
H
be changed to another frequency (fWDT = fCPU / 128) by programming the prescaler (bit
WDTIN).
The watchdog timer can be disabled by executing the instruction DISWDT (Disable
Watchdog Timer). Instruction DISWDT is a protected 32-bit instruction which will ONLY
be executed during the time between a reset and execution of either the EINIT (End of
Initialization) or the SRVWDT (Service Watchdog Timer) instruction. Either one of these
instructions disables the execution of DISWDT.
Note: After a hardware reset that activates the Bootstrap Loader the watchdog timer will
be disabled. The software reset that terminates the BSL mode will then enable the
WDT.
When the watchdog timer is not disabled via instruction DISWDT it will continue counting
up, even during Idle Mode. If it is not serviced via the instruction SRVWDT by the time
the count reaches FFFF the watchdog timer will overflow and cause an internal reset.
H
This reset will pull the external reset indication pin RSTOUT low. The Watchdog Timer
Reset Indication Flag (WDTR) in register WDTCON will be set in this case.
In bidirectional reset mode also pin RSTIN will be pulled low for the duration of the
internal reset sequence and a long hardware reset will be indicated instead.
A watchdog reset will also complete a running external bus cycle before starting the
internal reset sequence if this bus cycle does not use READY or samples READY active
(low) after the programmed waitstates. Otherwise the external bus cycle will be aborted.
To prevent the watchdog timer from overflowing it must be serviced periodically by the
user software. The watchdog timer is serviced with the instruction SRVWDT which is a
protected 32-bit instruction. Servicing the watchdog timer clears the low byte and reloads
the high byte of the watchdog timer register WDT with the preset value from bitfield
WDTREL which is the high byte of register WDTCON. Servicing the watchdog timer will
also reset bit WDTR. After being serviced the watchdog timer continues counting up from
8
the value (<WDTREL> × 2 ).
User’s Manual
13-3
V3.1, 2000-03
C167CR
Derivatives
The Watchdog Timer (WDT)
Instruction SRVWDT has been encoded in such a way that the chance of unintentionally
servicing the watchdog timer (e.g. by fetching and executing a bit pattern from a wrong
location) is minimized. When instruction SRVWDT does not match the format for
protected instructions the Protection Fault Trap will be entered, rather than the
instruction be executed.
WDTCON
WDT Control Register
SFR (FFAE /D7 )
Reset value: 00XX
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
LHW SHW SW WDT WDT
WDTREL
-
-
-
R
R
R
R
IN
-
-
-
-
rh
rh
rh
rh
rw
Bit
Function
WDTIN
Watchdog Timer Input Frequency Select
Controls the input clock prescaler. See Table 13-1.
WDTR
Watchdog Timer Reset Indication Flag
Cleared by a hardware reset or by the SRVWDT instruction.
SWR
Software Reset Indication Flag
SHWR
LHWR
WDTREL
Short Hardware Reset Indication Flag
Long Hardware Reset Indication Flag
Watchdog Timer Reload Value (for the high byte of WDT)
Note: The reset value depends on the reset source (see description below).
The execution of EINIT clears the reset indication flags.
The time period for an overflow of the watchdog timer is programmable in two ways:
• The input frequency to the watchdog timer can be selected via a prescaler controlled
by bit WDTIN in register WDTCON to be
fCPU / 2 or fCPU / 128.
• The reload value WDTREL for the high byte of WDT can be programmed in register
WDTCON.
User’s Manual
13-4
V3.1, 2000-03
C167CR
Derivatives
The Watchdog Timer (WDT)
The period P
between servicing the watchdog timer and the next overflow can
WDT
therefore be determined by the following formula:
(1 + <WDTIN> × 6)
16
8
2
× (2 – <WDTREL> × 2 )
P
=
WDT
f
CPU
Table 13-1 marks the possible ranges (depending on the prescaler bit WDTIN) for the
watchdog time which can be achieved using a certain CPU clock.
Table 13-1 Watchdog Time Ranges
CPU clock
fCPU
Prescaler
WDT fWDT
Reload value in WDTREL
7F 00
FF
H
H
H
IN
0
1
0
1
0
1
0
1
0
1
fCPU / 2
42.67
2.73
µs 5.50
ms 352.3
µs 4.13
ms 264.2
µs 3.30
ms 211.4
µs 2.64
ms 169.1
µs 2.00
ms 128.1
ms 10.92
ms 699.1
ms 8.19
ms 524.3
ms 6.55
ms 419.4
ms 5.24
ms 335.5
ms 3.97
ms 254.2
ms
ms
ms
ms
ms
ms
ms
ms
ms
ms
12 MHz
16 MHz
20 MHz
25 MHz
33 MHz
fCPU / 128
fCPU / 2
32.00
2.05
fCPU / 128
fCPU / 2
25.60
1.64
fCPU / 128
fCPU / 2
20.48
1.31
fCPU / 128
fCPU / 2
15.52
0.99
fCPU / 128
Note: For safety reasons, the user is advised to rewrite WDTCON each time before the
watchdog timer is serviced.
User’s Manual
13-5
V3.1, 2000-03
C167CR
Derivatives
The Watchdog Timer (WDT)
13.2
Reset Source Indication
The reset indication flags in register WDTCON provide information on the source for the
last reset. As the C167CR starts executing from location 00’0000 after any possible
H
reset event the initialization software may check these flags in order to determine if the
recent reset event was triggered by an external hardware signal (via RSTIN), by software
itself or by an overflow of the watchdog timer. The initialization (and also the further
operation) of the microcontroller system can thus be adapted to the respective
circumstances, e.g. a special routine may verify the software integrity after a watchdog
timer reset.
The reset indication flags are not mutually exclusive, i.e. more than one flag may be set
after reset depending on its source. Table 13-2 summarizes the possible combinations:
Table 13-2 Reset Indication Flag Combinations
1)
Event
Reset Indication Flags
LHWR
SHWR
SWR
WDTR
Long Hardware Reset
Short Hardware Reset
Software Reset
1
1
1
1
1
1
1
0
–
0
0
–
1
–
0
Watchdog Timer Reset
EINIT instruction
0
–
0
–
SRVWDT instruction
1)
Description of table entries:
‘1’ = flag is set, ‘0’ = flag is cleared, ‘–’ = flag is not affected,
‘ ’ = flag is set in bidirectional reset mode, not affected otherwise.
Long Hardware Reset is indicated when the RSTIN input is still sampled low (active) at
the end of a hardware triggered internal reset sequence.
Short Hardware Reset is indicated when the RSTIN input is sampled high (inactive) at
the end of a hardware triggered internal reset sequence.
Software Reset is indicated after a reset triggered by the excution of instruction SRST.
Watchdog Timer Reset is indicated after a reset triggered by an overflow of the
watchdog timer.
Note: When bidirectional reset is enabled the RSTIN pin is pulled low for the duration of
the internal reset sequence upon any sort of reset.
Therefore always a long hardware reset (LHWR) will be recognized in any case.
User’s Manual
13-6
V3.1, 2000-03
C167CR
Derivatives
The Bootstrap Loader
14
The Bootstrap Loader
The built-in bootstrap loader of the C167CR provides a mechanism to load the startup
program, which is executed after reset, via the serial interface. In this case no external
memory or an internal ROM/OTP/Flash is required for the initialization code starting at
location 00’0000 .
H
The bootstrap loader moves code/data into the internal RAM, but it is also possible to
transfer data via the serial interface into an external RAM using a second level loader
routine. ROM memory (internal or external) is not necessary. However, it may be used
to provide lookup tables or may provide “core-code”, i.e. a set of general purpose
subroutines, e.g. for IO operations, number crunching, system initialization, etc.
RSTIN
P0L.4
1)
2)
4)
RxD0
TxD0
3)
5)
CSP:IP
32 bytes
User Software
6)
Int. Boot ROM BSL-routine
1) BSL initialization time, < 70/ fCPU µs, (fCPU in [MHZ]).
2) Zero byte (1 start bit, eight '0' data bits, 1 stop bit), sent by host.
3) Identification byte, sent by C167CR.
4) 32 bytes of code / data, sent by host.
5) Caution: TxD0 is only driven a certain time after reception of the zero byte (< 40/ fCPU µs,
f
CPU in [MHz]).
6) Internal Boot ROM.
MCT04382
Figure 14-1 Bootstrap Loader Sequence
The Bootstrap Loader may be used to load the complete application software into
ROMless systems, it may load temporary software into complete systems for testing or
calibration, it may also be used to load a programming routine for Flash devices.
The BSL mechanism may be used for standard system startup as well as only for special
occasions like system maintenance (firmware update) or end-of-line programming or testing.
User’s Manual
14-1
V3.1, 2000-03
C167CR
Derivatives
The Bootstrap Loader
Entering the Bootstrap Loader
The C167CR enters BSL mode if pin P0L.4 is sampled low at the end of a hardware
reset. In this case the built-in bootstrap loader is activated independent of the selected
bus mode. The bootstrap loader code is stored in a special Boot-ROM, no part of the
standard mask ROM, OTP or Flash memory area is required for this.
1)
After entering BSL mode and the respective initialization the C167CR scans the RXD0
line to receive a zero byte, i.e. one start bit, eight ‘0’ data bits and one stop bit. From the
duration of this zero byte it calculates the corresponding baudrate factor with respect to
the current CPU clock, initializes the serial interface ASC0 accordingly and switches pin
TxD0 to output. Using this baudrate, an identification byte is returned to the host that
provides the loaded data.
This identification byte identifies the device to be booted. The following codes are
defined:
55 :
8xC166.
H
A5 :
Previous versions of the C167 (obsolete).
Previous versions of the C165.
C167 derivatives.
H
B5 :
H
C5 :
H
D5 :
All devices equipped with identification registers.
H
Note: The identification byte D5 does not directly identify a specific derivative. This
H
information can in this case be obtained from the identification registers.
When the C167CR has entered BSL mode, the following configuration is automatically
set (values that deviate from the normal reset values, are marked):
Watchdog Timer:
Context Pointer CP: FA00
Stack Pointer SP:
Register S0CON:
Register S0BG:
Disabled
Register STKUN:
Register STKOV:
Register BUSCON0: acc. to startup config.
P3.10/TXD0:
DP3.10:
FA40
H
FA0C 0<–>C
H
H
H
FA40
8011
‘1’
‘1’
H
acc. to ‘00’ byte
Other than after a normal reset the watchdog timer is disabled, so the bootstrap loading
sequence is not time limited. Pin TXD0 is configured as output, so the C167CR can
return the identification byte.
Note: Even if the internal ROM/OTP/Flash is enabled, no code can be executed out of it.
The hardware that activates the BSL during reset may be a simple pull-down resistor on
P0L.4 for systems that use this feature upon every hardware reset. You may want to use
a switchable solution (via jumper or an external signal) for systems that only temporarily
use the bootstrap loader.
1)
The external host should not send the zero byte before the end of the BSL initialization time (see Figure 14-1)
to make sure that it is correctly received.
User’s Manual
14-2
V3.1, 2000-03
C167CR
Derivatives
The Bootstrap Loader
External Signal
Circuit_1
Circuit_2
P0L.4
P0L.4
Normal Boot
BSL
RP0L.4
RP0L.4
8 k
Ω
8 kΩ
MCA02261
Figure 14-2 Hardware Provisions to Activate the BSL
After sending the identification byte the ASC0 receiver is enabled and is ready to receive
the initial 32 Bytes from the host. A half duplex connection is therefore sufficient to feed
the BSL.
Note: In order to properly enter BSL mode it is not only required to pull P0L.4 low, but
also pins P0L.2, P0L.3, P0L.5 must receive defined levels.
This is described in Chapter 19.
User’s Manual
14-3
V3.1, 2000-03
C167CR
Derivatives
The Bootstrap Loader
Memory Configuration after Reset
The configuration (i.e. the accessibility) of the C167CR’s memory areas after reset in
Bootstrap-loader mode differs from the standard case. Pin EA is not evaluated when
BSL mode is selected, and accesses to the internal code memory are partly redirected,
while the C167CR is in BSL mode (see Table 14-1). All code fetches are made from the
special Boot-ROM, while data accesses read from the internal code memory. Data
accesses will return undefined values on ROMless devices.
Note: The code in the Boot-ROM is not an invariant feature of the C167CR. User
software should not try to execute code from the internal ROM area while the BSL
mode is still active, as these fetches will be redirected to the Boot-ROM.
The Boot-ROM will also “move” to segment 1, when the internal ROM area is
mapped to segment 1.
Table 14-1 BSL Memory Configurations
16 MBytes
16 MBytes
16 MBytes
Access to
external
bus
Access to
external
bus
Depends
on reset
config.
disabled
enabled
(EA, P0)
1
0
1
0
1
0
Int.
RAM
Int.
RAM
Int.
RAM
Access to
int. ROM
enabled
Access to
int. ROM
enabled
Depends
on reset
config.
MCA04383
MCA04384
MCA04385
BSL mode active
Yes (P0L.4 = ‘0’)
high
Yes (P0L.4 = ‘0’)
low
No (P0L.4 = ‘1’)
acc. to application
User ROM access
EA pin
Code fetch from
Boot-ROM access
Boot-ROM access
internal ROM area
Data fetch from
User ROM access
User ROM access
User ROM access
internal ROM area
User’s Manual
14-4
V3.1, 2000-03
C167CR
Derivatives
The Bootstrap Loader
Loading the Startup Code
After sending the identification byte the BSL enters a loop to receive 32 Bytes via ASC0.
These bytes are stored sequentially into locations 00’FA40 through 00’FA5F of the
H
H
internal RAM. So up to 16 instructions may be placed into the RAM area. To execute the
loaded code the BSL then jumps to location 00’FA40 , i.e. the first loaded instruction.
H
The bootstrap loading sequence is now terminated, the C167CR remains in BSL mode,
however. Most probably the initially loaded routine will load additional code or data, as
an average application is likely to require substantially more than 16 instructions. This
second receive loop may directly use the pre-initialized interface ASC0 to receive data
and store it to arbitrary user-defined locations.
This second level of loaded code may be the final application code. It may also be
another, more sophisticated, loader routine that adds a transmission protocol to enhance
the integrity of the loaded code or data. It may also contain a code sequence to change
the system configuration and enable the bus interface to store the received data into
external memory.
This process may go through several iterations or may directly execute the final
application. In all cases the C167CR will still run in BSL mode, i.e. with the watchdog
timer disabled and limited access to the internal code memory. All code fetches from the
internal ROM area (00’0000 … 00’7FFF or 01’0000 … 01’7FFF , if mapped to
H
H
H
H
segment 1) are redirected to the special Boot-ROM. Data fetches access will access the
internal code memory of the C167CR, if any is available, but will return undefined data
on ROMless devices.
Exiting Bootstrap Loader Mode
In order to execute a program in normal mode, the BSL mode must be terminated first.
The C167CR exits BSL mode upon a software reset (ignores the level on P0L.4) or a
hardware reset (P0L.4 must be high then!). After a reset the C167CR will start executing
from location 00’0000 of the internal ROM or the external memory, as programmed via
H
pin EA.
User’s Manual
14-5
V3.1, 2000-03
C167CR
Derivatives
The Bootstrap Loader
Choosing the Baudrate for the BSL
The calculation of the serial baudrate for ASC0 from the length of the first zero byte that
is received, allows the operation of the bootstrap loader of the C167CR with a wide range
of baudrates. However, the upper and lower limits have to be kept, in order to insure
proper data transfer.
fCPU
32 (S0BRL + 1)
B
= ---------------------------------------------
C167CR
The C167CR uses timer T6 to measure the length of the initial zero byte. The
quantization uncertainty of this measurement implies the first deviation from the real
baudrate, the next deviation is implied by the computation of the S0BRL reload value
from the timer contents. The formula below shows the association:
fCPU
-- -------------
T6 =
T6 – 36
S0BRL = -------------------
72
9
,
4 BHost
For a correct data transfer from the host to the C167CR the maximum deviation between
the internal initialized baudrate for ASC0 and the real baudrate of the host should be
below 2.5%. The deviation (F , in percent) between host baudrate and C167CR
B
baudrate can be calculated via the formula below:
BContr – BHost
------------------------------------
BContr
F
=
100%
FB ≤ 2,5%
,
B
Note: Function (F ) does not consider the tolerances of oscillators and other devices
B
supporting the serial communication.
This baudrate deviation is a nonlinear function depending on the CPU clock and the
baudrate of the host. The maxima of the function (F ) increase with the host baudrate
B
due to the smaller baudrate prescaler factors and the implied higher quantization error
(see Figure 14-3).
User’s Manual
14-6
V3.1, 2000-03
C167CR
Derivatives
The Bootstrap Loader
FB
Ι
2.5%
BLow
BHigh
BHost
ΙΙ
MCA02260
Figure 14-3 Baudrate Deviation between Host and C167CR
The minimum baudrate (B in Figure 14-3) is determined by the maximum count
Low
capacity of timer T6, when measuring the zero byte, i.e. it depends on the CPU clock.
16
The minimum baudrate is obtained by using the maximum T6 count 2 in the baudrate
formula. Baudrates below B
would cause T6 to overflow. In this case ASC0 cannot
Low
be initialized properly and the communication with the external host is likely to fail.
The maximum baudrate (B in Figure 14-3) is the highest baudrate where the
High
deviation still does not exceed the limit, i.e. all baudrates between B
and B
are
Low
High
below the deviation limit. B
marks the baudrate up to which communication with the
High
external host will work properly without additional tests or investigations.
Higher baudrates, however, may be used as long as the actual deviation does not
exceed the indicated limit. A certain baudrate (marked I) in Figure 14-3) may e.g. violate
the deviation limit, while an even higher baudrate (marked II) in Figure 14-3) stays very
well below it. Any baudrate can be used for the bootstrap loader provided that the
following three prerequisites are fulfilled:
• the baudrate is within the specified operating range for the ASC0
• the external host is able to use this baudrate
• the computed deviation error is below the limit.
Table 14-2 Bootstrap Loader Baudrate Ranges
f
[MHz]
10
312,500
12
375,000
16
500,000
20
625,000
25
33
CPU
B
B
B
B
781,250 1,031,250
MAX
9,600
600
19,200
600
19,200
600
19,200
1,200
687
38,400
1,200
859
38,400
1,200
1,133
High
STDmin
Low
344
412
550
User’s Manual
14-7
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
15
The Capture/Compare Units
The C167CR provides two almost identical Capture/Compare (CAPCOM) units which
only differ in the way they are connected to the C167CR’s IO pins. They provide
32 channels which interact with 4 timers. The CAPCOM units can capture the contents
of a timer on specific internal or external events, or they can compare a timer’s content
with given values and modify output signals in case of a match. With this mechanism
they support generation and control of timing sequences on up to 16 channels per unit
with a minimum of software intervention.
Ports & Direction Control
Alternate Functions
Data Registers
Control Registers
T01CON
Interrupt Control
DP1H
P1H
ODP2
DP2
P2
E
E
T0
T0IC
T1IC
T0REL
T1
T1REL
T7
E
E
E
E
T78CON
T7IC
T8IC
E
ODP3
DP3
P3
E
E
E
T7REL
T8
E
T8REL
CC0-3
CC4-7
CC8-11
CC12-15
CC16-19
CC20-23
CC24-27
CC28-31
ODP7
DP7
P7
CCM0
CCM1
CCM2
CCM3
CCM4
CCM5
CCM6
CCM7
CC0IC-3IC
CC4IC-7IC
CC8IC-11IC
CC12IC-15IC
CC16IC-19IC
CC20IC-23IC
CC24IC-27IC
CC28IC-31IC
ODP8
DP8
P8
E
E
E
E
CC0IO/P2.0..CC15IO/P2.15
CC16IO/P8.0..CC23IO/P8.7
CC28IO/P7.4..CC31IO/P7.7
CC24IO/P1H.4..CC27IO/
P1H.7
ODPx
DPx
Px
Port x Open Drain Control Register
Port x Direction Control Register
Port x Data Register
Tx
CAPCOM Timer x Register
CAPCOM1 Register 0..15
CAPCOM2 Register 16..31
CAPCOM1 Mode Control Register 0..3
CAPCOM2 Mode Control Register 4..7
CC0..15
CC16..31
CCM0..3
CCM4..7
T01CON CAPCOM1 Timers T0/T1 Control Register
T78CON CAPCOM2 Timers T7/T8 Control Register
T0IC/T1IC CAPCOM1 Timer 0/1 Interrupt Ctrl. Reg.
T7IC/T8IC CAPCOM2 Timer 7/8 Interrupt Ctrl. Reg.
CC0..15IC CAPCOM1 Interrupt Ctrl. Reg. 0..15
CC16..31IC CAPCOM2 Interrupt Ctrl. Reg. 16..31
TxREL
CAPCOM Timer x Reload Register
MCA04386
Figure 15-1 SFRs and Port Pins Associated with the CAPCOM Units
User’s Manual
15-1
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
From the programmer’s point of view, the term ‘CAPCOM unit’ refers to a set of SFRs
which are associated with this peripheral, including the port pins which may be used for
alternate input/output functions including their direction control bits.
A CAPCOM unit is typically used to handle high speed IO tasks such as pulse and
waveform generation, pulse width modulation, or recording of the time at which specific
events occur. It also allows the implementation of up to 16 software timers. The
maximum resolution of the CAPCOM units is 8 CPU clock cycles (=16 TCL).
Each CAPCOM unit consists of two 16-bit timers (T0/T1 in CAPCOM1, T7/T8 in
CAPCOM2), each with its own reload register (TxREL), and a bank of sixteen dual
purpose 16-bit capture/compare registers (CC0 through CC15 in CAPCOM1, CC16
through CC31 in CAPCOM2).
The input clock for the CAPCOM timers is programmable to several prescaled values of
the CPU clock, or it can be derived from an overflow/underflow of timer T6 in block GPT2.
T0 and T7 may also operate in counter mode (from an external input) where they can be
clocked by external events.
Each capture/compare register may be programmed individually for capture or compare
function, and each register may be allocated to either timer of the associated unit. All
capture/compare registers of each module have one port pin associated with it,
respectively, which serves as an input pin for the capture function or as an output pin for
the compare function. The capture function causes the current timer contents to be
latched into the respective capture/compare register triggered by an event (transition) on
its associated port pin. The compare function may cause an output signal transition on
that port pin whose associated capture/compare register matches the current timer
contents. Specific interrupt requests are generated upon each capture/compare event or
upon timer overflow.
Figure 15-2 shows the basic structure of the two CAPCOM units.
User’s Manual
15-2
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Reload Reg. TxREL
fCPU
TxIN
2n : 1
Tx
Input
Control
Interrupt
Request
CAPCOM Timer Tx
GPT2 Timer T6
Over/Underflow
CCxIO
Mode
Control
(Capture
or
16-Bit
16 Capture Inputs
16 Compare Outputs
Capture/
Compare
Registers
16 Capture/Compare
Interrupt Request
Compare)
CCxIO
fCPU
2n : 1
Ty
Interrupt
Request
Input
Control
CAPCOM Timer Ty
Reload Reg. TyREL
GPT2 Timer T6
Over/Underflow
MCB02143B
Figure 15-2 CAPCOM Unit Block Diagram
Table 15-1 CAPCOM Channel Port Connections
Unit
Channel
Port
Capture
Input
Compare
Output
Output
–
CAPCOM1
CAPCOM2
CC0IO … CC15IO
P2.0 … P2.15
CC16IO … CC23IO P8.0 … P8.7
Input
CC24IO … CC27IO P1H.4 … P1H.7 Input
CC28IO … CC31IO P7.4 … P7.7
Σ = 32 Σ = 32
Input
Output
Σ = 28
Σ = 32
User’s Manual
15-3
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
15.1
The CAPCOM Timers
The primary use of the timers T0/T1 and T7/T8 is to provide two independent time bases
(16 TCL maximum resolution) for the capture/compare registers of each unit, but they
may also be used independent of the capture/compare registers.
The basic structure of the four timers is identical, while the selection of input signals is
different for timers T0/T7 and timers T1/T8 (see figures below).
Reload Reg. TxREL
Txl
Input
Control
fCPU
2n : 1
GPT2 Timer T6
Over/Underflow
Interrupt
Request
CAPCOM Timer Tx
MUX
Edge
Select
TxR
TxI TxM
TxIN
TxI
MCB02013
x = 0, 7
Figure 15-3 Block Diagram of CAPCOM Timers T0 and T7
Reload Reg. TxREL
Txl
fCPU
2n : 1
Interrupt
Request
CAPCOM Timer Tx
MUX
TxM
GPT2 Timer T6
Over/Underflow
TxR
MCB02014
x = 1, 8
Figure 15-4 Block Diagram of CAPCOM Timers T1 and T8
Note: When an external input signal is connected to the input lines of both T0 and T7,
these timers count the input signal synchronously. Thus the two timers can be
regarded as one timer whose contents can be compared with 32 capture registers.
User’s Manual
15-4
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
The functions of the CAPCOM timers are controlled via the bitaddressable 16-bit control
registers T01CON and T78CON. The high-byte of T01CON controls T1, the low-byte of
T01CON controls T0, the high-byte of T78CON controls T8, the low-byte of T78CON
controls T7. The control options are identical for all four timers (except for external input).
T01CON
CAPCOM Timer 0/1 Ctrl. Reg. SFR (FF50 /A8 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
T1R
-
-
T1M
T1I
-
T0R
-
-
T0M
T0I
-
rw
-
-
rw
rw
-
rw
-
-
rw
rw
T78CON
CAPCOM Timer 7/8 Ctrl. Reg.
SFR (FF20 /90 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
T8R
-
-
T8M
T8I
-
T7R
-
-
T7M
T7I
-
rw
-
-
rw
rw
-
rw
-
-
rw
rw
Bit
Function
Timer/Counter x Input Selection
Timer Mode (TxM = ‘0’): Input Frequency = fCPU / 2
See also Table 15-2 - Table 15-4 for examples.
TxI
(<TxI> + 3)
Counter Mode (TxM = ‘1’): 000 Overflow/Underflow of GPT2 Timer 6
1)
001 Positive (rising) edge on pin T7IN
1)
010 Negative (falling) edge on pin T7IN
1)
011 Any edge (rising and falling) on pin T7IN
1XX Reserved
TxM
Timer/Counter x Mode Selection
0:
1:
Timer Mode (Input derived from internal clock)
Counter Mode (Input from External Input or T6)
TxR
Timer/Counter x Run Control
0:
1:
Timer/Counter x is disabled
Timer/Counter x is enabled
1)
This selection is available for timers T0 and T7. Timers T1 and T8 will stop at this selection!
User’s Manual
15-5
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
The timer run flags T0R, T1R, T7R, and T8R allow for enabling and disabling the timers.
The following description of the timer modes and operation always applies to the enabled
state of the timers, i.e. the respective run flag is assumed to be set to ‘1’.
In all modes, the timers are always counting upward. The current timer values are
accessible for the CPU in the timer registers Tx, which are non-bitaddressable SFRs.
When the CPU writes to a register Tx in the state immediately before the respective timer
increment or reload is to be performed, the CPU write operation has priority and the
increment or reload is disabled to guarantee correct timer operation.
Timer Mode
The bits TxM in SFRs T01CON and T78CON select between timer or counter mode for
the respective timer. In timer mode (TxM = ‘0’), the input clock for a timer is derived from
the internal CPU clock divided by a programmable prescaler. The different options for
the prescaler are selected separately for each timer by the bit fields TxI.
The input frequencies fTx for Tx are determined as a function of the CPU clock as follows,
where <TxI> represents the contents of the bit field TxI:
fCPU
(<TxI> + 3)
fTx
=
2
When a timer overflows from FFFF to 0000 it is reloaded with the value stored in its
H
H
respective reload register TxREL. The reload value determines the period P between
Tx
two consecutive overflows of Tx as follows:
16
(<TxI> + 3)
(2 – <TxREL>) × 2
P =
Tx
f
CPU
After a timer has been started by setting its run flag (TxR) to ‘1’, the first increment will
occur within the time interval which is defined by the selected timer resolution. All further
increments occur exactly after the time defined by the timer resolution.
When both timers of a CAPCOM unit are to be incremented or reloaded at the same time
T0 is always serviced one CPU clock before T1, T7 before T8, respectively.
The timer input frequencies, resolution and periods which result from the selected
prescaler option in TxI when using a certain CPU clock are listed in Table 15-2 -
Table 15-4. The numbers for the timer periods are based on a reload value of 0000 .
H
Note that some numbers may be rounded to 3 significant digits.
User’s Manual
15-6
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Table 15-2 Timer Input Frequencies, Resolution and Period @ 20 MHz
fCPU = 20 MHz
Timer Input Selection TxI
000
001
010
32
011
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler (1:N)
8
16
64
128
256
512
1024
Input Frequency 2.5
MHz
400
1.25
MHz
625
KHz
312.5 156.25 78.125 39.06 19.53
KHz
KHz
KHz
KHz
KHz
Resolution
Period
800
ns
1.6
µs
3.2
6.4
12.8
25.6
51.2
ns
µs
µs
µs
µs
µs
26
ms
52.5
ms
105
ms
210
ms
420
ms
840
ms
1.68
s
3.36
s
Table 15-3 Timer Input Frequencies, Resolution and Period @ 25 MHz
fCPU = 25 MHz
Timer Input Selection TxI
000
001
010
011
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler (1:N)
8
16
32
64
128
256
512
1024
Input Frequency 3.125 1.563 781.25 390.63 195.31 97.656 48.828 24.414
MHz
MHz
KHz
KHz
KHz
KHz
KHz
KHz
Resolution
Period
320
ns
640
ns
1.28
µs
2.56
µs
5.12
µs
10.24 20.48 40.96
µs
µs
µs
21
ms
42
ms
84
ms
168
ms
336
ms
672
ms
1.344 2.688
s
s
Table 15-4 Timer Input Frequencies, Resolution and Period @ 33 MHz
fCPU = 33 MHz
Timer Input Selection TxI
000
001
010
011
100
101
110
111
B
B
B
B
B
B
B
B
Prescaler (1:N)
8
16
32
64
128
256
512
1024
Input Frequency 4.125 2.063 1.031 515.63 257.81 128.91 64.453 32.227
MHz
MHz
MHz
KHz
KHz
KHz
KHz
15.52 31.03
µs µs
KHz
Resolution
Period
242
ns
485
ns
970
ns
1.94
µs
3.88
µs
7.76
µs
15.89 31.78 63.55 127.10 254.20 508.40 1.017 2.034
ms
ms
ms
ms
ms
ms
s
s
User’s Manual
15-7
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Counter Mode
The bits TxM in SFRs T01CON and T78CON select between timer or counter mode for
the respective timer. In Counter mode (TxM = ‘1’) the input clock for a timer can be
derived from the overflows/underflows of timer T6 in block GPT2. In addition, timers T0
and T7 can be clocked by external events. Either a positive, a negative, or both a positive
and a negative transition at pin T0IN or T7IN (alternate port input function), respectively,
can be selected to cause an increment of T0/T7.
When T1 or T8 is programmed to run in counter mode, bit field TxI is used to enable the
overflows/underflows of timer T6 as the count source. This is the only option for these
timers and it is selected by the combination TxI = 000 . When bit field TxI is programmed
B
to any other valid combination, the respective timer will stop.
When T0 or T7 is programmed to run in counter mode, bit field TxI is used to select the
count source and transition (if the source is the input pin) which should cause a count
trigger (see description of TxyCON for the possible selections).
Note: In order to use pin T0IN or T7IN as external count input pin, the respective port pin
must be configured as input, i.e. the corresponding direction control bit must be
cleared (DPx.y = ‘0’).
If the respective port pin is configured as output, the associated timer may be
clocked by modifying the port output latches Px.y via software, e.g. for testing
purposes.
The maximum external input frequency to T0 or T7 in counter mode is fCPU/16. To
ensure that a signal transition is properly recognized at the timer input, an external count
input signal should be held for at least 8 CPU clock cycles before it changes its level
again. The incremented count value appears in SFR T0/T7 within 8 CPU clock cycles
after the signal transition at pin TxIN.
Reload
A reload of a timer with the 16-bit value stored in its associated reload register in both
modes is performed each time a timer would overflow from FFFF to 0000 . In this case
H
H
the timer does not wrap around to 0000 , but rather is reloaded with the contents of the
H
respective reload register TxREL. The timer then resumes incrementing starting from the
reloaded value.
The reload registers TxREL are not bitaddressable.
User’s Manual
15-8
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
15.2
CAPCOM Unit Timer Interrupts
Upon a timer overflow the corresponding timer interrupt request flag TxIR for the
respective timer will be set. This flag can be used to generate an interrupt or trigger a
PEC service request, when enabled by the respective interrupt enable bit TxIE.
Each timer has its own bitaddressable interrupt control register (TxIC) and its own
interrupt vector (TxINT). The organization of the interrupt control registers TxIC is
identical with the other interrupt control registers.
T0IC
CAPCOM T0 Intr. Ctrl. Reg.
SFR (FF9C /CE )
Reset value: - - 00
H
H
H
H
H
H
15
14
13
12
11
10
9
8
7
6
5
5
5
5
4
4
4
4
3
2
1
0
T0IR T0IE
ILVL
GLVL
rw
-
-
-
-
-
-
-
-
rwh rw
rw
T1IC
CAPCOM T1 Intr. Ctrl. Reg.
SFR (FF9E /CF )
Reset value: - - 00
H
H
15
14
13
12
11
10
9
8
7
6
3
2
1
0
T1IR T1IE
ILVL
GLVL
rw
-
-
-
-
-
-
-
-
rwh rw
rw
T7IC
CAPCOM T7 Intr. Ctrl. Reg.
ESFR (F17A /BE )
Reset value: - - 00
H
H
15
14
13
12
11
10
9
8
7
6
3
2
1
0
T7IR T7IE
ILVL
GLVL
rw
-
-
-
-
-
-
-
-
rwh rw
rw
T8IC
CAPCOM T8 Intr. Ctrl. Reg.
ESFR (F17C /BF )
Reset value: - - 00
H
H
15
14
13
12
11
10
9
8
7
6
3
2
1
0
T8IR T8IE
ILVL
GLVL
rw
-
-
-
-
-
-
-
-
rwh rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
User’s Manual
15-9
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
15.3
Capture/Compare Registers
The 16-bit capture/compare registers CC0 through CC31 are used as data registers for
capture or compare operations with respect to timers T0/T1 and T7/T8. The capture/
compare registers are not bitaddressable.
The functions of the 32 capture/compare registers are controlled by 8 bitaddressable
16-bit mode control registers named CCM0 … CCM7 which are organized identically (see
description below). Each register contains bits for mode selection and timer allocation of
four capture/compare registers.
Capture/Compare Mode Registers for the CAPCOM1 Unit (CC0 … CC15)
CCM0
CAPCOM Mode Ctrl. Reg. 0
SFR (FF52 /A9 )
Reset value: 0000
H
H
H
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
4
4
4
3
2
1
0
ACC
3
ACC
2
ACC
1
ACC
0
CCMOD3
CCMOD2
CCMOD1
CCMOD0
rw
rw
rw
rw
rw
rw
rw
rw
CCM1
CAPCOM Mode Ctrl. Reg. 1
SFR (FF54 /AA )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
3
2
1
0
ACC
7
ACC
6
ACC
5
ACC
4
CCMOD7
CCMOD6
CCMOD5
CCMOD4
rw
rw
rw
rw
rw
rw
rw
rw
CCM2
CAPCOM Mode Ctrl. Reg. 2
SFR (FF56 /AB )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
3
2
1
0
ACC
11
ACC
10
ACC
9
ACC
8
CCMOD11
CCMOD10
CCMOD9
CCMOD8
rw
rw
rw
rw
rw
rw
rw
rw
CCM3
CAPCOM Mode Ctrl. Reg. 3
SFR (FF58 /AC )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
3
2
1
0
ACC
15
ACC
14
ACC
13
ACC
12
CCMOD15
CCMOD14
CCMOD13
CCMOD12
rw
rw
rw
rw
rw
rw
rw
rw
User’s Manual
15-10
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Capture/Compare Mode Registers for the CAPCOM2 Unit (CC16 … CC32)
CCM4
CAPCOM Mode Ctrl. Reg. 4
SFR (FF22 /91 )
Reset value: 0000
H
H
H
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
4
4
4
3
2
1
0
ACC
19
ACC
18
ACC
17
ACC
16
CCMOD19
CCMOD18
CCMOD17
CCMOD16
rw
rw
rw
rw
rw
rw
rw
rw
CCM5
CAPCOM Mode Ctrl. Reg. 5
SFR (FF24 /92 )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
3
2
1
0
ACC
23
ACC
22
ACC
21
ACC
20
CCMOD23
CCMOD22
CCMOD21
CCMOD20
rw
rw
rw
rw
rw
rw
rw
rw
CCM6
CAPCOM Mode Ctrl. Reg. 6
SFR (FF26 /93 )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
3
2
1
0
ACC
27
ACC
26
ACC
25
ACC
24
CCMOD27
CCMOD26
CCMOD25
CCMOD24
rw
rw
rw
rw
rw
rw
rw
rw
CCM7
CAPCOM Mode Ctrl. Reg. 7
SFR (FF28 /94 )
Reset value: 0000
H
H
15
14
13
12
11
10
9
8
7
6
5
3
2
1
0
ACC
31
ACC
30
ACC
29
ACC
28
CCMOD31
CCMOD30
CCMOD29
CCMOD28
rw
rw
rw
rw
rw
rw
rw
rw
Bit
Function
Mode Selection for Capture/Compare Register CCx
CCMODx
The available capture/compare modes are listed in Table 15-5.
ACCx
Allocation Bit for Capture/Compare Register CCx
0:
1:
CCx allocated to Timer T7
CCx allocated to Timer T8
User’s Manual
15-11
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Each of the registers CCx may be individually programmed for capture mode or one of
4 different compare modes (no output signal for CC24 … CC27), and may be allocated
individually to one of the two timers of the respective CAPCOM unit (T0 or T1, and T7 or
T8, respectively). A special combination of compare modes additionally allows the
implementation of a ‘double-register’ compare mode. When capture or compare
operation is disabled for one of the CCx registers, it may be used for general purpose
variable storage.
Table 15-5 Selection of Capture Modes and Compare Modes
CCMODx Selected Operating Mode
0 0 0
Disable Capture and Compare Modes
The respective CAPCOM register may be used for general variable storage.
0 0 1
0 1 0
0 1 1
1 0 0
Capture on Positive Transition (Rising Edge) at Pin CCxIO
Capture on Negative Transition (Falling Edge) at Pin CCxIO
Capture on Positive and Negative Transition (Both Edges) at Pin CCxIO
Compare Mode 0:
Interrupt Only
Several interrupts per timer period; Enables double-register compare mode
for registers CC8 … CC15 and CC24 … CC31.
1 0 1
Compare Mode 1:
Toggle Output Pin on each Match
Several compare events per timer period; This mode is required for double-
register compare mode for registers CC0 … CC7 and CC16 … CC23.
1 1 0
1 1 1
Compare Mode 2:
Only one interrupt per timer period.
Interrupt Only
Compare Mode 3:
Set Output Pin on each Match
Reset output pin on each timer overflow; Only one interrupt per timer period.
The detailed discussion of the capture and compare modes is valid for all the capture/
compare channels, so registers, bits and pins are only referenced by the placeholder ‘x’.
Note: Capture/compare channels 27 … 24 generate an interrupt request but do not
provide an output signal. The resulting exceptions are indicated in the following
subsections.
A capture or compare event on channel 31 may be used to trigger a channel
injection on the C167CR’s A/D converter if enabled.
User’s Manual
15-12
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
15.4
Capture Mode
In response to an external event the content of the associated timer (T0/T1 or T7/T8,
depending on the used CAPCOM unit and the state of the allocation control bit ACCx) is
latched into the respective capture register CCx. The external event causing a capture
can be programmed to be either a positive, a negative, or both a positive or a negative
transition at the respective external input pin CCxIO.
The triggering transition is selected by the mode bits CCMODx in the respective
CAPCOM mode control register. In any case, the event causing a capture will also set
the respective interrupt request flag CCxIR, which can cause an interrupt or a PEC
service request, when enabled.
Edge
Select
Capture Reg. CCx
Interrupt
Request
CCxIO
CCMODx
Input
Clock
Interrupt
Request
CAPCOM Timer Ty
x = 31...0
y = 0, 1, 7, 8
MCB02015
Figure 15-5 Capture Mode Block Diagram
In order to use the respective port pin as external capture input pin CCxIO for capture
register CCx, this port pin must be configured as input, i.e. the corresponding direction
control bit must be set to ‘0’. To ensure that a signal transition is properly recognized, an
external capture input signal should be held for at least 8 CPU clock cycles before it
changes its level.
During these 8 CPU clock cycles the capture input signals are scanned sequentially.
When a timer is modified or incremented during this process, the new timer contents will
already be captured for the remaining capture registers within the current scanning
sequence.
Note: When the timer modification can generate an overflow the capture interrupt routine
should check if the timer overflow was serviced during these 8 CPU clock cycles.
If pin CCxIO is configured as output, the capture function may be triggered by modifying
the corresponding port output latch via software, e.g. for testing purposes.
User’s Manual
15-13
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
15.5
Compare Modes
The compare modes allow triggering of events (interrupts and/or output signal
transitions) with minimum software overhead. In all compare modes, the 16-bit value
stored in compare register CCx (in the following also referred to as ‘compare value’) is
continuously compared with the contents of the allocated timer (T0/T1 or T7/T8). If the
current timer contents match the compare value, an appropriate output signal, which is
based on the selected compare mode, can be generated at the corresponding output pin
CCxIO (except for CC27IO … CC24IO) and the associated interrupt request flag CCxIR
is set, which can generate an interrupt request (if enabled).
As for capture mode, the compare registers are also processed sequentially during
compare mode. When any two compare registers are programmed to the same compare
value, their corresponding interrupt request flags will be set to ‘1’ and the selected output
signals will be generated within 8 CPU clock cycles after the allocated timer is
incremented to the compare value. Further compare events on the same compare value
are disabled until the timer is incremented again or written to by software. After a reset,
compare events for register CCx will only become enabled, if the allocated timer has
been incremented or written to by software and one of the compare modes described in
the following has been selected for this register.
The different compare modes which can be programmed for a given compare register
CCx are selected by the mode control field CCMODx in the associated capture/compare
mode control register. In the following, each of the compare modes, including the special
‘double-register’ mode, is discussed in detail.
Compare Mode 0
This is an interrupt-only mode which can be used for software timing purposes. Compare
mode 0 is selected for a given compare register CCx by setting bit field CCMODx of the
corresponding mode control register to ‘100 ’.
B
In this mode, the interrupt request flag CCxIR is set each time a match is detected
between the content of compare register CCx and the allocated timer. Several of these
compare events are possible within a single timer period, when the compare value in
register CCx is updated during the timer period. The corresponding port pin CCxIO is not
affected by compare events in this mode and can be used as general purpose IO pin.
If compare mode 0 is programmed for one of the registers CC8 … CC15 or
CC24 … CC31, the double-register compare mode becomes enabled for this register if
the corresponding bank 1 register is programmed to compare mode 1 (see “Double-
register Compare Mode” on Page 15-19).
User’s Manual
15-14
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Interrupt
Request
Compare Reg. CCx
Comparator
Port
CCxIO
Latch
Toggle
(Mode 1)
CCMODx
Input
Clock
Interrupt
Request
CAPCOM Timer Ty
not for CC27...CC24
x = 31...0
y = 0, 1, 7, 8
MCB02016
Figure 15-6 Compare Mode 0 and 1 Block Diagram
Note: The port latch and pin remain unaffected in compare mode 0.
In the example below, the compare value in register CCx is modified from cv1 to cv2 after
compare events #1 and #3, and from cv2 to cv1 after events #2 and #4, etc. This results
in periodic interrupt requests from timer Ty, and in interrupt requests from register CCx
which occur at the time specified by the user through cv1 and cv2.
Contents of Ty
FFFFH
Compare Value cv2
Compare Value cv1
Reload Value<TyREL>
0000H
Interrupt
Requests:
TyIR
CCxIR
CCxIR TyIR
CCxIR
CCxIR TyIR
Event #1 Event #2
CCx: = cv2 CCx: = cv1
Event #3 Event #4
CCx: = cv2 CCx: = cv1
t
Output pin CCxIO only effected in mode 1. No changes in mode 0.
x = 31...0
y = 0, 1, 7, 8
MCB02017
Figure 15-7 Timing Example for Compare Modes 0 and 1
User’s Manual
15-15
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Compare Mode 1
Compare mode 1 is selected for register CCx by setting bit field CCMODx of the
corresponding mode control register to ‘101 ’.
B
When a match between the content of the allocated timer and the compare value in
register CCx is detected in this mode, interrupt request flag CCxIR is set to ‘1’, and in
addition the corresponding output pin CCxIO (alternate port output function) is toggled.
For this purpose, the state of the respective port output latch (not the pin) is read,
inverted, and then written back to the output latch.
Compare mode 1 allows several compare events within a single timer period. An
overflow of the allocated timer has no effect on the output pin, nor does it disable or
enable further compare events.
In order to use the respective port pin as compare signal output pin CCxIO for compare
register CCx in compare mode 1, this port pin must be configured as output, i.e. the
corresponding direction control bit must be set to ‘1’. With this configuration, the initial
state of the output signal can be programmed or its state can be modified at any time by
writing to the port output latch.
In compare mode 1 the port latch is toggled upon each compare event (see Timing
Example above).
If compare mode 1 is programmed for one of the registers CC0 … CC7 or CC16 … CC23
the double-register compare mode becomes enabled for this register if the
corresponding bank 2 register is programmed to compare mode 0 (see “Double-
register Compare Mode” on Page 15-19).
Note: If the port output latch is written to by software at the same time it would be altered
by a compare event, the software write will have priority. In this case the hardware-
triggered change will not become effective.
On channels 27 … 24 compare mode 1 will generate an interrupt request but no
output function is provided.
Compare Mode 2
Compare mode 2 is an interrupt-only mode similar to compare mode 0, but only one
interrupt request per timer period will be generated. Compare mode 2 is selected for
register CCx by setting bit field CCMODx of the corresponding mode control register to
‘110 ’.
B
When a match is detected in compare mode 2 for the first time within a timer period, the
interrupt request flag CCxIR is set to ‘1’. The corresponding port pin is not affected and
can be used for general purpose IO. However, after the first match has been detected in
this mode, all further compare events within the same timer period are disabled for
compare register CCx until the allocated timer overflows. This means, that after the first
match, even when the compare register is reloaded with a value higher than the current
timer value, no compare event will occur until the next timer period.
User’s Manual
15-16
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
In the example below, the compare value in register CCx is modified from cv1 to cv2 after
compare event #1. Compare event #2, however, will not occur until the next period of
timer Ty.
Interrupt
Compare Reg. CCx
Request
Set
(Mode 3)
Reset
Comparator
Port
Latch
CCxIO
CCMODx
Input
Clock
Interrupt
Request
CAPCOM Timer Ty
not for CC27...CC24
x = 31...0
y = 0, 1, 7, 8
MCB02019
Figure 15-8 Compare Mode 2 and 3 Block Diagram
Note: The port latch and pin remain unaffected in compare mode 2.
Contents of Ty
FFFFH
Compare Value cv2
Compare Value cv1
Reload Value<TyREL>
0000H
Interrupt
Requests:
TyIR
CCxIR
TyIR
CCxIR TyIR
State
CCxIO:
1
0
time
Event #1
CCx: = cv2
Event #2
CCx: = cv1
Output pin CCxIO only effected in mode 3. No changes in mode 2.
x = 31...0
y = 0, 1, 7, 8
MCB02021
Figure 15-9 Timing Example for Compare Modes 2 and 3
User’s Manual
15-17
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Compare Mode 3
Compare mode 3 is selected for register CCx by setting bit field CCMODx of the
corresponding mode control register to ‘111 ’. In compare mode 3 only one compare
B
event will be generated per timer period.
When the first match within the timer period is detected the interrupt request flag CCxIR
is set to ‘1’ and also the output pin CCxIO (alternate port function) will be set to ‘1’. The
pin will be reset to ‘0’, when the allocated timer overflows.
If a match was found for register CCx in this mode, all further compare events during the
current timer period are disabled for CCx until the corresponding timer overflows. If, after
a match was detected, the compare register is reloaded with a new value, this value will
not become effective until the next timer period.
In order to use the respective port pin as compare signal output pin CCxIO for compare
register CCx in compare mode 3 this port pin must be configured as output, i.e. the
corresponding direction control bit must be set to ‘1’. With this configuration, the initial
state of the output signal can be programmed or its state can be modified at any time by
writing to the port output latch.
In compare mode 3 the port latch is set upon a compare event and cleared upon a timer
overflow (see Timing Example above).
However, when compare value and reload value for a channel are equal the respective
interrupt requests will be generated, only the output signal is not changed (set and clear
would coincide in this case).
Note: If the port output latch is written to by software at the same time it would be altered
by a compare event, the software write will have priority. In this case the hardware-
triggered change will not become effective.
On channels 27 … 24 compare mode 3 will generate an interrupt request but no
output function is provided.
User’s Manual
15-18
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Double-register Compare Mode
In double-register compare mode two compare registers work together to control one
output pin. This mode is selected by a special combination of modes for these two
registers.
For double-register mode the 16 capture/compare registers of each CAPCOM unit are
regarded as two banks of 8 registers each. Registers CC0 … CC7 and CC16 … CC23
form bank 1 while registers CC8 … CC15 and CC24 … CC31 form bank 2 (respectively).
For double-register mode a bank 1 register and a bank 2 register form a register pair.
Both registers of this register pair operate on the pin associated with the bank 1 register
(pins CC0IO … CC7IO and CC16IO … CC23IO).
The relationship between the bank 1 and bank 2 register of a pair and the effected output
pins for double-register compare mode is listed in Table 15-6.
Table 15-6 Register Pairs for Double-register Compare Mode
CAPCOM1 Unit
Register Pair
CAPCOM2 Unit
Register Pair
Associated
Output Pin
Associated
Output Pin
Bank 1
CC0
CC1
CC2
CC3
CC4
CC5
CC6
CC7
Bank 2
CC8
Bank 1
CC16
CC17
CC18
CC19
CC20
CC21
CC22
CC23
Bank 2
CC24
CC25
CC26
CC27
CC28
CC29
CC30
CC31
CC0IO
CC1IO
CC2IO
CC3IO
CC4IO
CC5IO
CC6IO
CC7IO
CC16IO
CC17IO
CC18IO
CC19IO
CC20IO
CC21IO
CC22IO
CC23IO
CC9
CC10
CC11
CC12
CC13
CC14
CC15
The double-register compare mode can be programmed individually for each register
pair. In order to enable double-register mode the respective bank 1 register (see
Table 15-6) must be programmed to compare mode 1 and the corresponding bank 2
register (see Table 15-6) must be programmed to compare mode 0.
If the respective bank 1 compare register is disabled or programmed for a mode other
than mode 1 the corresponding bank 2 register will operate in compare mode 0
(interrupt-only mode).
In the following, a bank 2 register (programmed to compare mode 0) will be referred to
as CCz while the corresponding bank 1 register (programmed to compare mode 1) will
be referred to as CCx.
User’s Manual
15-19
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
When a match is detected for one of the two registers in a register pair (CCx or CCz) the
associated interrupt request flag (CCxIR or CCzIR) is set to ‘1’ and pin CCxIO
corresponding to bank 1 register CCx is toggled. The generated interrupt always
corresponds to the register that caused the match.
Note: If a match occurs simultaneously for both register CCx and register CCz of the
register pair pin CCxIO will be toggled only once but two separate compare
interrupt requests will be generated, one for vector CCxINT and one for vector
CCzINT.
In order to use the respective port pin as compare signal output pin CCxIO for compare
register CCx in double-register compare mode, this port pin must be configured as
output, i.e. the corresponding direction control bit must be set to ‘1’. With this
configuration, the output pin has the same characteristics as in compare mode 1.
Interrupt
Request
Compare Reg. CCx
Comparator
Mode 1 (CCMODx)
Input
Clock
Port
Latch
1
CAPCOM Timer Ty
Comparator
CCxIO
Toggle
Mode 0 (CCMODz)
Compare Reg. CCz
Interrupt
Request
x = 23...16, 7...0
y = 0, 1, 7, 8
z = 31...24, 15...8
MCB02022
Figure 15-10 Double-Register Compare Mode Block Diagram
User’s Manual
15-20
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
In this configuration example, the same timer allocation was chosen for both compare
registers, but each register may also be individually allocated to one of the two timers of
the respective CAPCOM unit. In the timing example for this compare mode (below) the
compare values in registers CCx and CCz are not modified.
Note: The pins CCzIO (which do not serve for double-register compare mode) may be
used for general purpose IO.
Contents of Ty
FFFFH
Compare Value CCz
Compare Value CCx
Reload Value<TyREL>
0000H
Interrupt
Requests:
TyIR
CCxIR
CCxIR TyIR
CCxIR
CCxIR TyIR
State of CCxIO:
1
0
time
x = 23...16, 7...0
y = 0, 1, 7, 8
z = 31...24, 15...8
MCB02023
Figure 15-11 Timing Example for Double-register Compare Mode
User’s Manual
15-21
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
15.6
Capture/Compare Interrupts
Upon a capture or compare event, the interrupt request flag CCxIR for the respective
capture/compare register CCx is set to ‘1’. This flag can be used to generate an interrupt
or trigger a PEC service request when enabled by the interrupt enable bit CCxIE.
Capture interrupts can be regarded as external interrupt requests with the additional
feature of recording the time at which the triggering event occurred (see also
Section 5.7).
Each of the 32 capture/compare registers has its own bitaddressable interrupt control
register (CC0IC … CC31IC) and its own interrupt vector (CC0INT … CC31INT). These
registers are organized the same way as all other interrupt control registers. The basic
register layout is shown below, Table 15-7 lists the associated addresses.
CCxIC
CAPCOM Intr. Ctrl. Reg.
(E)SFR (See Table 15-7)
Reset value: - - 00
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
CCx CCx
IR IE
ILVL
GLVL
rw
-
-
-
-
-
-
-
-
rwh rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
User’s Manual
15-22
V3.1, 2000-03
C167CR
Derivatives
The Capture/Compare Units
Table 15-7 Register Pairs for Double-Register Compare Mode
CAPCOM1 Unit
Address
CAPCOM2 Unit
Address
Register
CC0IC
CC1IC
CC2IC
CC3IC
CC4IC
CC5IC
CC6IC
CC7IC
CC8IC
CC9IC
CC10IC
CC11IC
CC12IC
CC13IC
CC14IC
CC15IC
Reg. Space Register
Reg. Space
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
ESFR
FF78 /BC
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
SFR
CC16IC
CC17IC
CC18IC
CC19IC
CC20IC
CC21IC
CC22IC
CC23IC
CC24IC
CC25IC
CC26IC
CC27IC
CC28IC
CC29IC
CC30IC
CC31IC
F160 /B0
H
H
H
H
H
H
H
H
FF7A /BD
F162 /B1
H
H
H
H
H
H
H
H
H
H
FF7C /BE
F164 /B2
H
H
FF7E /BF
F166 /B3
H
H
FF80 /C0
F168 /B4
H
H
FF82 /C1
F16A /B5
H H
H
FF84 /C2
F16C /B6
H
H
H
H
H
H
FF86 /C3
F16E /B7
H
H
FF88 /C4
F170 /B8
H
H
FF8A /C5
F172 /B9
H
H
H
FF8C /C6
F174 /BA
H
H
H
H
H
H
H
H
FF8E /C7
F176 /BB
H
H
FF90 /C8
F178 /BC
H
H
H
H
FF92 /C9
F184 /C2
H
H
FF94 /CA
F18C /C6
H
H
H
H
H
H
FF96 /CB
F194 /CA
H
H
User’s Manual
15-23
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
16
The Pulse Width Modulation Module
The Pulse Width Modulation (PWM) Module of the C167CR allows the generation of up
to 4 independent PWM signals. These PWM signals can be generated within a wide
range of output frequencies, which depends on:
• the CPU clock frequency fCPU
• the selected counter resolution (fCPU / 1 or fCPU / 64)
• the operating mode (edge/center aligned)
• the required PWM resolution (1-bit … 16-bit)
The maximum PWM output frequency in a real application is primarily determined by the
PWM resolution which is required for that application.
Ports & Direction Control
Alternate Functions
Data Registers
Control Registers
Control Registers and
Interrupt Control
E
ODP7
DP7
P7
PP0
E
PT0
PT1
PT2
PT3
E
E
E
E
PWMCON0
PWMCON1
PW0
PP1
PW1
PP2
PW2
PP3
PW3
E
E
E
PWMIC
E
POUT0/P7.0
POUT1/P7.1
POUT2/P7.2
POUT3/P7.3
ODP7
DP7
P7
Port 7 Open Drain Control Register
Port 7 Direction Control Register
Port 7 Data Register
PPx
PWx
PTx
PWM Period Register x
PWM Pulse Width Register x
PWM Counter Register x
PWMIC PWM Interrupt Control Register
PWMCONx PWM Control Register 0/1
MCA04387
Figure 16-1 SFRs and Port Pins Associated with the PWM Module
The Pulse Width Modulation Module consists of 4 independent PWM channels. Each
channel has a 16-bit up/down counter PTx, a 16-bit period register PPx with a shadow
latch, a 16-bit pulse width register PWx with a shadow latch, two comparators, and the
necessary control logic.
The operation of all four channels is controlled by two common control registers,
PWMCON0 and PWMCON1, and the interrupt control and status is handled by one
interrupt control register PWMIC, which is also common for all channels.
User’s Manual
16-1
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
PPx Period Register
Shadow Register
Comparator
Write
Control
Match
Run
PTx
16-Bit Up/Down
Counter
Up/Down/
Clear
Control
fCPU
Input
Control
Match
Enable
Output
POUTx
Control
Comparator
Shadow Register
Write
Control
PWx Pulse Width Reg.
User Read-
& Writeable
MCB01948
Figure 16-2 PWM Channel Block Diagram
16.1
Operating Modes
The PWM module provides four different operating modes:
• Standard PWM generation (edge aligned PWM) available on all four channels
• Symmetrical PWM generation (center aligned PWM) available on all four channels
• Burst mode combines channels 0 and 1
• Single shot mode available on channels 2 and 3
Note: The output signals of the PWM module are XORed with the outputs of the
respective port output latches. After reset these latches are cleared, so the PWM
signals are directly driven to the port pins. By setting the respective port output
latch to ‘1’ the PWM signal may be inverted (XORed with ‘1’) before being driven
to the port pin. The descriptions below refer to the standard case after reset, i.e.
direct driving.
User’s Manual
16-2
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
Mode 0: Standard PWM Generation (Edge Aligned PWM)
Mode 0 is selected by clearing the respective bit PMx in register PWMCON1 to ‘0’. In this
mode the timer PTx of the respective PWM channel is always counting up until it reaches
the value in the associated period shadow register. Upon the next count pulse the timer
is reset to 0000 and continues counting up with subsequent count pulses. The PWM
H
output signal is switched to high level when the timer contents are equal to or greater
than the contents of the pulse width shadow register. The signal is switched back to low
level when the respective timer is reset to 0000 , i.e. below the pulse width shadow
H
register. The period of the resulting PWM signal is determined by the value of the
respective PPx shadow register plus 1, counted in units of the timer resolution.
PWM_Period
= [PPx] + 1
Mode0
The duty cycle of the PWM output signal is controlled by the value in the respective pulse
width shadow register. This mechanism allows the selection of duty cycles from 0% to
100% including the boundaries. For a value of 0000 the output will remain at a high
H
level, representing a duty cycle of 100%. For a value higher than the value in the period
register the output will remain at a low level, which corresponds to a duty cycle of 0%.
Figure 16-3 illustrates the operation and output waveforms of a PWM channel in mode
0 for different values in the pulse width register. This mode is referred to as Edge Aligned
PWM, because the value in the pulse width (shadow) register only effects the positive
edge of the output signal. The negative edge is always fixed and related to the clearing
of the timer.
User’s Manual
16-3
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
PPx
Period = 7
7
7
7
6
6
6
5
5
PTx Count
Value
4
4
3
3
2
2
1
1
1
0
0
0
Duty
Cycle
PWx Pulse
Width = 0
100%
87.5%
75%
PWx = 1
PWx = 2
PWx = 4
PWx = 6
PWx = 7
PWx = 8
50%
25%
12.5%
0%
LSR
LSR
LSR
Latch Shadow
Register;
Interrupt Request
MCA01949
Figure 16-3 Operation and Output Waveform in Mode 0
User’s Manual
16-4
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
Mode 1: Symmetrical PWM Generation (Center Aligned PWM)
Mode 1 is selected by setting the respective bit PMx in register PWMCON1 to ‘1’. In this
mode the timer PTx of the respective PWM channel is counting up until it reaches the
value in the associated period shadow register. Upon the next count pulse the count
direction is reversed and the timer starts counting down now with subsequent count
pulses until it reaches the value 0000 . Upon the next count pulse the count direction is
H
reversed again and the count cycle is repeated with the following count pulses.
The PWM output signal is switched to a high level when the timer contents are equal to
or greater than the contents of the pulse width shadow register while the timer is counting
up. The signal is switched back to a low level when the respective timer has counted
down to a value below the contents of the pulse width shadow register. So in mode 1 this
PWM value controls both edges of the output signal.
Note that in mode 1 the period of the PWM signal is twice the period of the timer:
PWM_Period
= 2 × ([PPx] + 1)
Mode1
Figure 16-4 illustrates the operation and output waveforms of a PWM channel in mode 1
for different values in the pulse width register. This mode is referred to as Center Aligned
PWM, because the value in the pulse width (shadow) register effects both edges of the
output signal symmetrically.
User’s Manual
16-5
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
PPx
Period = 7
7
7
6
6
5
4
3
2
5
PTx Count
Value
4
3
2
1
1
1
1
0
0
0
0
Duty
Cycle
PWx Pulse
Width = 0
100%
87.5%
75%
PWx = 1
PWx = 2
PWx = 4
PWx = 6
PWx = 7
PWx = 8
50%
25%
12.5%
0%
LSR
Latch Shadow
Register;
Change Count
Direction
LSR
MCA01950
Interrupt Request
Figure 16-4 Operation and Output Waveform in Mode 1
User’s Manual
16-6
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
Burst Mode
Burst mode is selected by setting bit PB01 in register PWMCON1 to ‘1’. This mode
combines the signals from PWM channels 0 and 1 onto the port pin of channel 0. The
output of channel 0 is replaced with the logical AND of channels 0 and 1. The output of
channel 1 can still be used at its associated output pin (if enabled).
Each of the two channels can either operate in mode 0 or 1.
Note: It is guaranteed by design, that no spurious spikes will occur at the output pin of
channel 0 in this mode. The output of the AND gate will be transferred to the output
pin synchronously to internal clocks.
XORing of the PWM signal and the port output latch value is done after the
ANDing of channel 0 and 1.
PP0
Period
Value
PT0
Count
Value
Channel 0
PP1
PT1
Count
Channel 1
Resulting
Output
POUT0
MCA01951
Figure 16-5 Operation and Output Waveform in Burst Mode
User’s Manual
16-7
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
Single Shot Mode
Single shot mode is selected by setting the respective bit PSx in register PWMCON1 to
‘1’. This mode is available for PWM channels 2 and 3.
In this mode the timer PTx of the respective PWM channel is started via software and is
counting up until it reaches the value in the associated period shadow register. Upon the
next count pulse the timer is cleared to 0000 and stopped via hardware, i.e. the
H
respective PTRx bit is cleared. The PWM output signal is switched to high level when the
timer contents are equal to or greater than the contents of the pulse width shadow
register. The signal is switched back to low level when the respective timer is cleared,
i.e. is below the pulse width shadow register. Thus starting a PWM timer in single shot
mode produces one single pulse on the respective port pin, provided that the pulse width
value is between 0000 and the period value. In order to generate a further pulse, the
H
timer has to be started again via software by setting bit PTRx.
After starting the timer (i.e. PTRx = ‘1’) the output pulse may be modified via software.
Writing to timer PTx changes the positive and/or negative edge of the output signal,
depending on whether the pulse has already started (i.e. the output is high) or not (i.e.
the output is still low). This (multiple) retriggering is always possible while the timer is
running, i.e. after the pulse has started and before the timer is stopped.
Loading counter PTx directly with the value in the respective PPx shadow register will
abort the current PWM pulse upon the next clock pulse (counter is cleared and stopped
by hardware).
By setting the period (PPx), the timer start value (PTx) and the pulse width value (PWx)
appropriately, the pulse width (tw) and the optional pulse delay (td) may be varied in a
wide range.
User’s Manual
16-8
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
PPx
Period = 7
7
7
6
6
5
5
PTx Count
Value
4
4
3
3
2
2
1
1
0
PWx Pulse
Width = 4
Set PTRx by
Software
LSR;
Set PTRx by
Software for
Next Pulse
LSR
PTRx Reset by
Hardware
PTx stopped
PPx
Period = 7
7
7
6
6
6
5
5
5
PTx Count
Value
4
4
4
3
2
1
1
0
0
tD
tD
PWx Pulse
Width = 4
Retrigger after Pulse
has started: Write PWx
value to PTx
Trigger before Pulse has
started: Write PWx value
to PTx;
Shortens Delay Time tD
MCA01952
Figure 16-6 Operation and Output Waveform in Single Shot Mode
User’s Manual
16-9
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
16.2
PWM Module Registers
The PWM module is controlled via two sets of registers. The waveforms are selected by
the channel specific registers PTx (timer), PPx (period) and PWx (pulse width). Three
common registers control the operating modes and the general functions (PWMCON0
and PWMCON1) of the PWM module as well as the interrupt behavior (PWMIC).
Up/Down Counters PTx
Each counter PTx of a PWM channel is clocked either directly with fCPU or with fCPU/64.
Bit PTIx in register PWMCON0 selects the respective clock source. A PWM counter
counts up or down (controlled by hardware), while its respective run control bit PTRx is
set. A timer is started (PTRx = ‘1’) via software and is stopped (PTRx = ‘0’) either via
hardware or software, depending on its operating mode. Control bit PTRx enables or
disables the clock input of counter PTx rather than controlling the PWM output signal.
Note: For the register locations please refer to Table 16-2 further below.
Table 16-1 summarizes the PWM frequencies that result from various combinations of
operating mode, counter resolution (input clock) and pulse width resolution.
Table 16-1 PWM Output Frequency and Resolution
Inp.Clk. (fCPU/x) PWM 8-bit PWM 10-bitPWM 12-bitPWM 14-bitPWM 16-bitPWM
(Counter resol.) Mode Resolution Resolution Resolution Resolution Resolution
16 MHz/1
(62.5 ns)
0
1
62.50 KHz 15.63 KHz 3.91 KHz 976.6 Hz 244.1 Hz
31.25 KHz 7.81 KHz 1.95 KHz 488.3 Hz 122.1 Hz
16 MHz/64
(4.0 µs)
0
1
976.6 Hz 244.1 Hz 61.04 Hz 15.29 Hz 3.81
488.3 Hz 122.1 Hz 30.52 Hz 7.63 Hz 1.91
Hz
Hz
20 MHz/1
(50 ns)
0
1
78.13 KHz 19.53 KHz 4.88 KHz 1.22 KHz 305.2 Hz
39.06 KHz 9.77 KHz 2.44 KHz 610.4 Hz 152.6 Hz
20 MHz/64
(3.2 µs)
0
1
1.22 KHz 305.2 Hz 76.29 Hz 19.07 Hz 4.77
610.4 Hz 152.6 Hz 38.15 Hz 9.54 Hz 2.38
Hz
Hz
25 MHz/1
(40 ns)
0
1
97.66 KHz 24.41 KHz 6.10 KHz 1.53 KHz 381.5 Hz
48.83 KHz 12.21 KHz 3.05 KHz 762.9 Hz 190.7 Hz
25 MHz/64
(2.56 µs)
0
1
1.53 KHz 381.5 Hz 95.37 Hz 23.84 Hz 5.96
762.9 Hz 190.7 Hz 47.68 Hz 11.92 Hz 2.98
Hz
Hz
33 MHz/1
(30.3 ns)
0
1
128.9 KHz 32.23 KHz 8.06 KHz 2.01 KHz 503.5 Hz
64.45 KHz 16.11 KHz 4.03 KHz 1.01 KHz 251.8 Hz
33 MHz/64
(1.94 µs)
0
1
2.01 KHz 503.5 Hz 125.9 Hz 31.47 Hz 7.87
1.01 KHz 251.8 Hz 62.94 Hz 15.74 Hz 3.93
Hz
Hz
User’s Manual
16-10
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
Period Registers PPx
The 16-bit period register PPx (see Table 16-2 for locations) of a PWM channel
determines the period of a PWM cycle, i.e. the frequency of the PWM signal. This
register is buffered with a shadow register. The shadow register is loaded from the
respective PPx register at the beginning of every new PWM cycle, or upon a write access
to PPx, while the timer is stopped. The CPU accesses the PPx register while the
hardware compares the contents of the shadow register with the contents of the
associated counter PTx. When a match is found between counter and PPx shadow
register, the counter is either reset to 0000 , or the count direction is switched from
H
counting up to counting down, depending on the selected operating mode of that PWM
channel.
Pulse Width Registers PWx
The 16-bit pulse width register PWx (see Table 16-2 for locations) of a PWM channel
holds the actual PWM pulse width value which corresponds to the duty cycle of the PWM
signal. This register is buffered with a shadow register. The CPU accesses the PWx
register while the hardware compares the contents of the shadow register with the
contents of the associated counter PTx. The shadow register is loaded from the
respective PWx register at the beginning of every new PWM cycle, or upon a write
access to PWx, while the timer is stopped.
Table 16-2 PWM Module Channel Specific Register Addresses
Register Address
Register Space Register Address
Register Space
ESFR
PW0
PW1
PW2
PW3
FE30 /18
SFR
SFR
SFR
SFR
PT0
PT1
PT2
PT3
PP0
PP1
PP2
PP3
F030 /18
H
H
H
H
H
H
FE32 /19
F032 /19
ESFR
H
H
FE34 /1A
F034 /1A
ESFR
H
H
H
H
H
H
FE36 /1B
F036 /1B
ESFR
H
H
F038 /1C
ESFR
H
H
F03A /1D
ESFR
H
H
H
F03C /1E
ESFR
H
Note: These registers are not
bitaddressable.
F03E /1F
ESFR
H
H
When the counter value is greater than or equal to the shadow register value, the PWM
signal is set, otherwise it is reset. The output of the comparators may be described by
the boolean formula:
PWM output signal = [PTx] ≥ [PWx shadow latch].
This type of comparison allows a flexible control of the PWM signal.
User’s Manual
16-11
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
PWM Control Register PWMCON0
Register PWMCON0 controls the function of the timers of the four PWM channels and
the channel specific interrupts. Having the control bits organized in functional groups
allows e.g. to start or stop all 4 PWM timers simultaneously with one bitfield instruction.
PWMCON0
PWM Control Register 0
SFR (FF30 /98 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
PIR PIR PIR PIR PIE PIE PIE PIE PTI PTI PTI PTI PTR PTR PTR PTR
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
rwh rwh rwh rwh rw
rw
rw
rw
rw
rw
rw
rw rwh rwh rw
rw
Bit
Function
PWM Timer x Run Control Bit
PTRx
PTIx
PIEx
PIRx
0:
1:
Timer PTx is disconnected from its input clock
Timer PTx is running
PWM Timer x Input Clock Selection
0:
1:
Timer PTx clocked with CLK
Timer PTx clocked with CLK
CPU
CPU
/64
PWM Channel x Interrupt Enable Flag
0:
1:
Interrupt from channel x disabled
Interrupt from channel x enabled
PWM Channel x Interrupt Request Flag
0:
1:
No interrupt request from channel x
Channel x interrupt pending (must be reset via software)
User’s Manual
16-12
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
PWM Control Register PWMCON1
Register PWMCON1 controls the operating modes and the outputs of the four PWM
channels. The basic operating mode for each channel (standard = edge aligned, or
symmetrical = center aligned PWM mode) is selected by the mode bits PMx. Burst mode
(channels 0 and 1) and single shot mode (channel 2 or 3) are selected by separate
control bits. The output signal of each PWM channel is individually enabled by bit PENx.
If the output is not enabled the respective pin can be used for general purpose IO and
the PWM channel can only be used to generate an interrupt request.
PWMCON1
PWM Control Register 1
SFR (FF32 /99 )
Reset value: 0000
H
H
H
15
PS3 PS2
rw rw
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
PB
01
PEN PEN PEN PEN
3
-
-
-
-
-
PM3 PM2 PM1 PM0
rw rw rw rw
2
1
0
-
rw
-
-
-
-
rw
rw
rw
rw
Bit
Function
PWM Channel x Output Enable Bit
PENx
0:
1:
Channel x output signal disabled, generate interrupt only
Channel x output signal enabled
PMx
PWM Channel x Mode Control Bit
0:
1:
Channel x operates in mode 0, i.e. edge aligned PWM
Channel x operates in mode 1, i.e. center aligned PWM
PB01
PWM Channel 0/1 Burst Mode Control Bit
0:
Channel 0 and channel 1 work independently in their respective
standard mode
1:
Outputs of channels 0 and 1 are ANDed to POUT0 in burst mode
PSx
PWM Channel x Single Shot Mode Control Bit
0:
1:
Channel x works in respective standard mode
Channel x operates in single shot mode
User’s Manual
16-13
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
16.3
Interrupt Request Generation
Each of the four channels of the PWM module can generate an individual interrupt
request. Each of these “channel interrupts” can activate the common “module interrupt”,
which actually interrupts the CPU. This common module interrupt is controlled by the
PWM Module Interrupt Control register PWMIC. The interrupt service routine can
determine the active channel interrupt(s) from the channel specific interrupt request flags
PIRx in register PWMCON0. The interrupt request flag PIRx of a channel is set at the
beginning of a new PWM cycle, i.e. upon latching the shadow registers (LSR). This
indicates that registers PPx and PWx are now ready to receive a new value. If a channel
interrupt is enabled via its respective PIEx bit, also the common interrupt request flag
PWMIR in register PWMIC is set, provided that it is enabled via the common interrupt
enable bit PWMIE.
Note: The channel interrupt request flags (PIRx in register PWMCON0) are not
automatically cleared by hardware upon entry into the interrupt service routine, so
they must be cleared via software. The module interrupt request flag PWMIR is
cleared by hardware upon entry into the service routine, regardless of how many
channel interrupts were active. However, it will be set again if during execution of
the service routine a new channel interrupt request is generated.
PWMIC
PWM Intr. Ctrl. Reg.
ESFR (F17E /BF )
Reset value: - - 00
H
H
H
15
14
13
12
11
-
10
-
9
8
7
6
5
4
3
2
1
0
PWM PWM
IR IE
ILVL
GLVL
rw
-
-
-
-
-
-
rwh rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
User’s Manual
16-14
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
16.4
PWM Output Signals
The output signals of the four PWM channels (POUT3 … POUT0) are alternate output
functions on Port 7 (P7.3 … P7.0). The output signal of each PWM channel is individually
enabled by control bit PENx in register PWMCON1.
The PWM signals are XORed with the respective port latch outputs before being driven
to the port pins. This allows driving the PWM signal directly to the port pin (P7.x = ‘0’) or
drive the inverted PWM signal (P7.x = ‘1’).
PWM 3
XOR
XOR
XOR
XOR
Pin P7.3
Pin P7.2
Pin P7.1
Latch P7.3
Latch P7.2
Latch P7.1
Latch P7.0
PWM 2
PWM 1
&
PWM 0
Pin P7.0
PENx
PB 01
MCA02277
Figure 16-7 PWM Output Signal Generation
Note: Using the open drain mode on Port 7 allows the combination of two or more PWM
outputs through a Wired-AND configuration, using an external pullup device. This
provides sort of a burst mode for any PWM channel.
User’s Manual
16-15
V3.1, 2000-03
C167CR
Derivatives
The Pulse Width Modulation Module
Software Control of the PWM Outputs
In an application the PWM output signals are generally controlled by the PWM module.
However, it may be necessary to influence the level of the PWM output pins via software
either to initialize the system or to react on some extraordinary condition, e.g. a system
fault or an emergency.
Clearing the timer run bit PTRx stops the associated counter and leaves the respective
output at its current level.
The individual PWM channel outputs are controlled by comparators according to the
formula:
PWM output signal = [PTx] ≥ [PWx shadow latch].
So whenever software changes registers PTx, the respective output will reflect the
condition after the change. Loading timer PTx with a value greater than or equal to the
value in PWx immediately sets the respective output, a PTx value below the PWx value
clears the respective output.
By clearing or setting the respective Port 7 output latch the PWM channel signal is driven
directly or inverted to the port pin.
Clearing the enable bit PENx disconnects the PWM channel and switches the respective
port pin to the value in the port output latch.
Note: To prevent further PWM pulses from occurring after such a software intervention
the respective counter must be stopped first.
User’s Manual
16-16
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
17
The Analog/Digital Converter
The C167CR provides an Analog/Digital Converter with 10-bit resolution and a sample
& hold circuit on-chip. A multiplexer selects between up to 16 analog input channels
(alternate functions of Port 5) either via software (fixed channel modes) or automatically
(auto scan modes).
To fulfill most requirements of embedded control applications the ADC supports the
following conversion modes:
• Fixed Channel Single Conversion
produces just one result from the selected channel
• Fixed Channel Continuous Conversion
repeatedly converts the selected channel
• Auto Scan Single Conversion
produces one result from each of a selected group of channels
• Auto Scan Continuous Conversion
repeatedly converts the selected group of channels
• Wait for ADDAT Read Mode
start a conversion automatically when the previous result was read
• Channel Injection Mode
insert the conversion of a specific channel into a group conversion (auto scan)
A set of SFRs and port pins provide access to control functions and results of the ADC.
Ports & Direction Control
Alternate Functions
Data Registers
Control Registers
ADCON
Interrupt Control
P5
ADDAT
ADCIC
ADEIC
P5DIDIS
ADDAT2
E
P5
Port 5 Analog Input Port:
AN0/P5.0...AN15/P5.15
ADCON A/D Converter Control Register
ADCIC A/D Converter Interrupt Control Register
(End of Conversion)
ADEIC A/D Converter Interrupt Control Register
(Overrun Error / Channel Injection)
P5DIDIS Port 5 Digital Input Disable Register
ADDAT A/D Converter Result Register
ADDAT2 A/D Conv. Channel Injection Result Reg.
MCA04388
Figure 17-1 SFRs and Port Pins Associated with the A/D Converter
User’s Manual
17-1
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
The external analog reference voltages VAREF and VAGND are fixed. The separate
supply for the ADC reduces the interference with other digital signals.
The sample time as well as the conversion time is programmable, so the ADC can be
adjusted to the internal resistances of the analog sources and/or the analog reference
voltage supply.
ADCON
ADCIR
ADEIR
Conversion
Control
Inerrupt
Requests
AN0
.
.
.
Result Reg. ADDAT
Result Reg. ADDAT2
10-Bit
Converter
MUX
S + H
AN15
VAREF VAGND
MCB04389
Figure 17-2 Analog/Digital Converter Block Diagram
17.1
Mode Selection and Operation
The analog input channels AN15 … AN0 are alternate functions of Port 5 which is an
input-only port. The Port 5 lines may either be used as analog or digital inputs. For pins
that shall be used as analog inputs it is recommended to disable the digital input stage
via register P5DIDIS. This avoids undesired cross currents and switching noise while the
(analog) input signal level is between VIL and VIH.
The functions of the A/D converter are controlled by the bit-addressable A/D Converter
Control Register ADCON. Its bitfields specify the analog channel to be acted upon, the
conversion mode, and also reflect the status of the converter.
User’s Manual
17-2
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
ADCON
ADC Control Register
SFR (FFA0 /D0 )
Reset value: 0000
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
AD AD AD AD AD
CRQ CIN WR BSY ST
ADCTC
ADSTC
-
ADM
rw
ADCH
rw
rw
rw
rwh rw
rw rwh rwh
-
Bit
Function
ADC Analog Channel Input Selection
Selects the (first) ADC channel which is to be converted.
Note: Valid channel numbers are 0 to 7 .
ADCH
H
H
ADM
ADC Mode Selection
00: Fixed Channel Single Conversion
01: Fixed Channel Continuous Conversion
10: Auto Scan Single Conversion
11: Auto Scan Continuous Conversion
ADST
ADC Start Bit
0:
1:
Stop a running conversion
Start conversion(s)
ADBSY
ADC Busy Flag
0:
1:
ADC is idle
A conversion is active.
ADWR
ADCIN
ADCRQ
ADSTC
ADC Wait for Read Control
ADC Channel Injection Enable
ADC Channel Injection Request Flag
ADC Sample Time Control (Defines the ADC sample time in a certain range)
00: tBC × 8
01: tBC × 16
10: tBC × 32
11: tBC × 64
ADCTC
ADC Conversion Time Control (Defines the ADC basic conversion clock fBC)
00: fBC = fCPU / 4
01: fBC = fCPU / 2
10: fBC = fCPU / 16
11: fBC = fCPU / 8
User’s Manual
17-3
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
Bitfield ADCH specifies the analog input channel which is to be converted (first channel
of a conversion sequence in auto scan modes). Bitfield ADM selects the operating mode
of the A/D converter. A conversion (or a sequence) is then started by setting bit ADST.
Clearing ADST stops the A/D converter after a certain operation which depends on the
selected operating mode.
The busy flag (read-only) ADBSY is set, as long as a conversion is in progress.
The result of a conversion is stored in the result register ADDAT, or in register ADDAT2
for an injected conversion.
Note: Bitfield CHNR of register ADDAT is loaded by the ADC to indicate, which channel
the result refers to.
Bitfield CHNR of register ADDAT2 is loaded by the CPU to select the analog
channel, which is to be injected.
ADDAT
ADC Result Register
SFR (FEA0 /50 )
Reset value: 0000
H
H
H
15
14
CHNR
rwh
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
ADRES
-
-
rwh
ADDAT2
ADC Chan. Inj. Result Reg.
ESFR (F0A0 /50 )
Reset value: 0000
H
H
H
15
14
CHNR
rw
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
ADRES
-
-
rwh
Bit
Function
A/D Conversion Result
The 10-bit digital result of the most recent conversion.
Channel Number (identifies the converted analog channel)
Note: Valid channel numbers are F to 0 .
ADRES
CHNR
H
H
User’s Manual
17-4
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
A conversion is started by setting bit ADST = ‘1’. The busy flag ADBSY will be set and
the converter then selects and samples the input channel, which is specified by the
channel selection field ADCH in register ADCON. The sampled level will then be held
internally during the conversion. When the conversion of this channel is complete, the
10-bit result together with the number of the converted channel is transferred into the
result register ADDAT and the interrupt request flag ADCIR is set. The conversion result
is placed into bitfield ADRES of register ADDAT.
If bit ADST is reset via software, while a conversion is in progress, the A/D converter will
stop after the current conversion (fixed channel modes) or after the current conversion
sequence (auto scan modes).
Setting bit ADST while a conversion is running, will abort this conversion and start a new
conversion with the parameters specified in ADCON.
Note: Abortion and restart (see above) are triggered by bit ADST changing from ‘0’ to ‘1’,
i.e. ADST must be ‘0’ before being set.
While a conversion is in progress, the mode selection field ADM and the channel
selection field ADCH may be changed. ADM will be evaluated after the current
conversion. ADCH will be evaluated after the current conversion (fixed channel modes)
or after the current conversion sequence (auto scan modes).
Fixed Channel Conversion Modes
These modes are selected by programming the mode selection bitfield ADM in register
ADCON to ‘00 ’ (single conversion) or to ‘01 ’ (continuous conversion). After starting the
B
B
converter through bit ADST the busy flag ADBSY will be set and the channel specified
in bit field ADCH will be converted. After the conversion is complete, the interrupt request
flag ADCIR will be set.
In Single Conversion Mode the converter will automatically stop and reset bits ADBSY
and ADST.
In Continuous Conversion Mode the converter will automatically start a new conversion
of the channel specified in ADCH. ADCIR will be set after each completed conversion.
When bit ADST is reset by software, while a conversion is in progress, the converter will
complete the current conversion and then stop and reset bit ADBSY.
User’s Manual
17-5
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
Auto Scan Conversion Modes
These modes are selected by programming the mode selection field ADM in register
ADCON to ‘10 ’ (single conversion) or to ‘11 ’ (continuous conversion). Auto Scan
B
B
modes automatically convert a sequence of analog channels, beginning with the channel
specified in bit field ADCH and ending with channel 0, without requiring software to
change the channel number.
After starting the converter through bit ADST, the busy flag ADBSY will be set and the
channel specified in bit field ADCH will be converted. After the conversion is complete,
the interrupt request flag ADCIR will be set and the converter will automatically start a
new conversion of the next lower channel. ADCIR will be set after each completed
conversion. After conversion of channel 0 the current sequence is complete.
In Single Conversion Mode the converter will automatically stop and reset bits
ADBSY and ADST.
In Continuous Conversion Mode the converter will automatically start a new sequence
beginning with the conversion of the channel specified in ADCH.
When bit ADST is reset by software, while a conversion is in progress, the converter will
complete the current sequence (including conversion of channel 0) and then stop and
reset bit ADBSY.
# 3
# 2
# 1
# 0
# 3
# 2
Conversion
of Channel..
Write ADDAT
ADDAT Full
# x
# 3
# 2
# 1
# 0
# 3
Generate Interrupt
Request
ADDAT Full;
Channnel 0
Result Lost
Read of ADDAT;
Result of Channel:
# x
# 3
# 2
# 1
# 3
Overrun Error
Interrupt Request
MCA02241
Figure 17-3 Auto Scan Conversion Mode Example
User’s Manual
17-6
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
Wait for ADDAT Read Mode
If in default mode of the ADC a previous conversion result has not been read out of
register ADDAT by the time a new conversion is complete, the previous result in register
ADDAT is lost because it is overwritten by the new value, and the A/D overrun error
interrupt request flag ADEIR will be set.
In order to avoid error interrupts and the loss of conversion results especially when using
continuous conversion modes, the ADC can be switched to “Wait for ADDAT Read
Mode” by setting bit ADWR in register ADCON.
If the value in ADDAT has not been read by the time the current conversion is complete,
the new result is stored in a temporary buffer and the next conversion is suspended
(ADST and ADBSY will remain set in the meantime, but no end-of-conversion interrupt
will be generated). After reading the previous value from ADDAT the temporary buffer is
copied into ADDAT (generating an ADCIR interrupt) and the suspended conversion is
started. This mechanism applies to both single and continuous conversion modes.
Note: While in standard mode continuous conversions are executed at a fixed rate
(determined by the conversion time), in “Wait for ADDAT Read Mode” there may
be delays due to suspended conversions. However, this only affects the
conversions, if the CPU (or PEC) cannot keep track with the conversion rate.
# 3
# 2
# 1
wait
# 0
# 3
Conversion
of Channel..
Write ADDAT
ADDAT Full
# x
# 3
# 2
# 1
# 0
# 3
Temp-Latch Full
1
Generate Interrupt
Request
Hold Result in
Temp-Latch
Read of ADDAT;
Result of Channel:
# x
# 3
# 2
# 1
# 0
MCA01970
Figure 17-4 Wait for Read Mode Example
User’s Manual
17-7
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
Channel Injection Mode
Channel Injection Mode allows the conversion of a specific analog channel (also while
the ADC is running in a continuous or auto scan mode) without changing the current
operating mode. After the conversion of this specific channel the ADC continues with the
original operating mode.
Channel Injection mode is enabled by setting bit ADCIN in register ADCON and requires
the Wait for ADDAT Read Mode (ADWR = ‘1’). The channel to be converted in this mode
is specified in bitfield CHNR of register ADDAT2.
Note: Bitfield CHNR in ADDAT2 is not modified by the A/D converter, but only the
ADRES bit field. Since the channel number for an injected conversion is not
buffered, bitfield CHNR of ADDAT2 must never be modified during the sample
phase of an injected conversion, otherwise the input multiplexer will switch to the
new channel. It is recommended to only change the channel number with no
injected conversion running.
# x
# ...
# x-1
# x-2
# x-3
# x-4
Conversion
of Channel..
Write ADDAT; # x+1
ADDAT Full
# x
# x-1
# x-2
# x-3
# x-4
Read ADDAT
# x+1
# x
# x-1
# x-2
# y
# x-3
# x-4
Injected
Conversion
of Channel # y
Channel Injection
Request
Write ADDAT2
ADDAT2 Full
Int. Request
ADEINT
Read ADDAT2
MCA01971
Figure 17-5 Channel Injection Example
User’s Manual
17-8
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
A channel injection can be triggered in two ways:
• setting of the Channel Injection Request bit ADCRQ via software
• a compare or a capture event of Capture/Compare register CC31 of the CAPCOM2
unit, which also sets bit ADCRQ.
The second method triggers a channel injection at a specific time, on the occurrence of
a predefined count value of the CAPCOM timers or on a capture event of register CC31.
This can be either the positive, the negative, or both the positive and the negative edge
of an external signal. In addition, this option allows recording the time of occurrence of
this signal.
Note: The channel injection request bit ADCRQ will be set on any interrupt request of
CAPCOM2 channel CC31, regardless whether the channel injection mode is
enabled or not. It is recommended to always clear bit ADCRQ before enabling the
channel injection mode.
After the completion of the current conversion (if any is in progress) the converter will
start (inject) the conversion of the specified channel. When the conversion of this
channel is complete, the result will be placed into the alternate result register
ADDAT2, and a Channel Injection Complete Interrupt request will be generated,
which uses the interrupt request flag ADEIR (for this reason the Wait for ADDAT
Read Mode is required).
Note: If the temporary data register used in Wait for ADDAT Read Mode is full, the
respective next conversion (standard or injected) will be suspended. The
temporary register can hold data for ADDAT (from a standard conversion) or for
ADDAT2 (from an injected conversion).
User’s Manual
17-9
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
# x
# x-1
# x-2
# x-3
# ...
Conversion
of Channel..
Wait until
ADDAT2 is
read
Write ADDAT # x+1
ADDAT Full
# x
# x-1
# x-2
# x-3
Read ADDAT
# x+1
# x
# x-1
# x-2
# x-3
Injected
Conversion
of Channel y
Channel Injection
Request
Write ADDAT2
# y
# z
ADDAT2 Full
Int. Request
ADEINT
# z
# y
Read ADDAT2
Temp-Latch
Full
# x
# x-1
# x-2
# x-3
# ...
Conversion
of Channel..
Write ADDAT # x+1
ADDAT Full
# x
# x-1
# x-2
# x-3
Read ADDAT
# x+1
# x
# x-1
# x-2
# x-3
Temp-Latch
Full
# y
Channel Injection
Request
Wait until
ADDAT2 is
read
Write ADDAT2
ADDAT2 Full
Int. Request
ADEINT
# y
Read ADDAT2
MCA01972
Figure 17-6 Channel Injection Example with Wait for Read
User’s Manual
17-10
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
Arbitration of Conversions
Conversion requests that are activated while the ADC is idle immediately trigger the
respective conversion. If a conversion is requested while another conversion is currently
in progress the operation of the A/D converter depends on the kind of the involved
conversions (standard or injected).
Note: A conversion request is activated if the respective control bit (ADST or ADCRQ)
is toggled from ‘0’ to ‘1’, i.e. the bit must have been zero before being set.
Table 17-1 summarizes the ADC operation in the possible situations.
Table 17-1 Conversion Arbitration
Conversion
in Progress
New Requested Conversion
Injected
Standard
Standard
Abort running conversion,
and start requested new
conversion.
Complete running conversion,
start requested conversion after that.
Injected
Complete running conversion,
Complete running conversion,
start requested conversion after start requested conversion after that.
that.
Bit ADCRQ will be ‘0’ for the second
conversion, however.
User’s Manual
17-11
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
17.2
Conversion Timing Control
When a conversion is started, first the capacitances of the converter are loaded via the
respective analog input pin to the current analog input voltage. The time to load the
capacitances is referred to as sample time. Next the sampled voltage is converted to a
digital value in successive steps, which correspond to the resolution of the ADC. During
these phases (except for the sample time) the internal capacitances are repeatedly
charged and discharged via pins VAREF and VAGND
.
The current that has to be drawn from the sources for sampling and changing charges
depends on the time that each respective step takes, because the capacitors must reach
their final voltage level within the given time, at least with a certain approximation. The
maximum current, however, that a source can deliver, depends on its internal resistance.
The time that the two different actions during conversion take (sampling, and converting)
can be programmed within a certain range in the C167CR relative to the CPU clock. The
absolute time that is consumed by the different conversion steps therefore is
independent from the general speed of the controller. This allows adjusting the A/D
converter of the C167CR to the properties of the system:
Fast Conversion can be achieved by programming the respective times to their
absolute possible minimum. This is preferable for scanning high frequency signals. The
internal resistance of analog source and analog supply must be sufficiently low,
however.
High Internal Resistance can be achieved by programming the respective times to a
higher value, or the possible maximum. This is preferable when using analog sources
and supply with a high internal resistance in order to keep the current as low as possible.
The conversion rate in this case may be considerably lower, however.
The conversion time is programmed via the upper two bits of register ADCON. Bitfield
ADCTC (conversion time control) selects the basic conversion clock (fBC), used for the
operation of the A/D converter. The sample time is derived from this conversion clock.
Table 17-2 lists the possible combinations. The timings refer to CPU clock cycles, where
tCPU = 1 / fCPU
The limit values for fBC (see data sheet) must not be exceeded when selecting ADCTC
and fCPU
.
.
Table 17-2 ADC Conversion Timing Control
ADCON.15|14
(ADCTC)
A/D Converter
Basic Clock fBC
ADCON.13|12
(ADSTC)
Sample Time tS
00
01
10
11
fCPU / 4
fCPU / 2
fCPU / 16
fCPU / 8
00
01
10
11
tBC × 8
tBC × 16
tBC × 32
tBC × 64
User’s Manual
17-12
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
The time for a complete conversion includes the sample time t , the conversion itself and
S
the time required to transfer the digital value to the result register (2 tCPU) as shown in
the example below.
Note: The non-linear decoding of bit field ADCTC provides compatibility with 80C166
designs for the default value (‘00’ after reset).
Converter Timing Example
Assumptions:
fCPU = 25 MHz (i.e. tCPU = 40 ns), ADCTC = ‘00’, ADSTC = ‘00’.
Basic clock
Sample time
fBC = fCPU / 4 = 6.25 MHz, i.e. tBC = 160 ns.
tS = tBC × 8 = 1280 ns.
Conversion time tC = tS + 40 tBC + 2 tCPU = (1280 + 6400 + 80) ns = 7.76 µs.
Note: For the exact specification please refer to the data sheet of the selected derivative.
User’s Manual
17-13
V3.1, 2000-03
C167CR
Derivatives
The Analog/Digital Converter
17.3
A/D Converter Interrupt Control
At the end of each conversion, interrupt request flag ADCIR in interrupt control register
ADCIC is set. This end-of-conversion interrupt request may cause an interrupt to vector
ADCINT, or it may trigger a PEC data transfer which reads the conversion result from
register ADDAT e.g. to store it into a table in the internal RAM for later evaluation.
The interrupt request flag ADEIR in register ADEIC will be set either, if a conversion
result overwrites a previous value in register ADDAT (error interrupt in standard mode),
or if the result of an injected conversion has been stored into ADDAT2 (end-of-injected-
conversion interrupt). This interrupt request may be used to cause an interrupt to vector
ADEINT, or it may trigger a PEC data transfer.
ADCIC
ADC Conversion Intr.Ctrl.Reg. SFR (FF98 /CC )
Reset value: - - 00
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ADC ADC
IR IE
-
ILVL
GLVL
rw
-
-
-
-
-
-
-
-
rwh rw
rw
ADEIC
ADC Error Intr.Ctrl.Reg.
SFR (FF9A /CD )
Reset value: - - 00
H
H
H
15
14
13
12
11
10
-
9
8
7
6
5
4
3
2
1
0
ADE ADE
IR IE
-
ILVL
GLVL
rw
-
-
-
-
-
-
-
rwh rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
User’s Manual
17-14
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
18
The On-chip CAN Interface
The Controller Area Network (CAN) bus with its associated protocol allows
communication between a number of stations which are connected to this bus with high
efficiency.
Efficiency in this context means:
• Transfer speed (Data rates of up to 1 Mbit/sec can be achieved)
• Data integrity (The CAN protocol provides several means for error checking)
• Host processor unloading (The controller here handles most of the tasks autonomously)
• Flexible and powerful message passing (The extended CAN protocol is supported)
The integrated CAN module handles the completely autonomous transmission and
reception of CAN frames in accordance with the CAN specification V2.0 part B (active),
i.e. the on-chip CAN module can receive and transmit:
• Standard frames with 11-bit identifiers, as well as
• Extended frames with 29-bit identifiers.
Note: The CAN module is an XBUS peripheral and therefore requires bit XPEN in
register SYSCON to be set in order to be operable.
Core Registers
SYSCON
Control Registers
(within each module)
Object Registers
(within each module)
Interrupt Control
CSR
X
X
X
X
X
X
MCRn
UARn
LARn
Data
X
X
X
X
XP0IC
E
IR
BTR
DP4
GMS
U/LGML
U/LMLM
SYSCON System Configuration Register
CSR
IR
BTR
GMS
Control/Status Register
Interrupt Register
Bit Timing Register
Global Mask Short
DP4
Port 4 Direction Control Register
CAN1 Interrupt Control Register
XP0IC
U/LGML Global Mask Long
U/LMLM Last Message Mask
MCRn
Configuration Register of Message n
U/LARn Arbitration Register of Message n
MCA04390
Figure 18-1 Registers Associated with the CAN Module
User’s Manual
18-1
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
The bit timing is derived from the XCLK and is programmable up to a data rate of
1 MBaud. The minimum CPU clock frequency to achieve 1 MBaud is fCPU ≥ 8/16 MHz,
depending on the activation of the CAN module’s clock prescaler.
The CAN module uses two pins of Port 4 to interface to a bus transceiver.
It provides Full CAN functionality on up to 15 full-sized message objects (8 data bytes
each). Message object 15 may be configured for Basic CAN functionality with a double-
buffered receive object.
Both modes provide separate masks for acceptance filtering which allows the
acceptance of a number of identifiers in Full CAN mode and also allows disregarding a
number of identifiers in Basic CAN mode.
All message objects can be updated independent from the other objects during operation
of the module and are equipped with buffers for the maximum message length of 8
Bytes.
18.1
Functional Blocks of the CAN Module
The CAN module combines several functional blocks (see Figure 18-2) that work in
parallel and contribute to the controller’s performance. These units and the functions
they provide are described below.
Each of the message objects has a unique identifier and its own set of control and status
bits. Each object can be configured with its direction as either transmit or receive, except
the last message which is only a double receive buffer with a special mask register.
An object with its direction set as transmit can be configured to be automatically sent
whenever a remote frame with a matching identifier (taking into account the respective
global mask register) is received over the CAN bus. By requesting the transmission of a
message with the direction set as receive, a remote frame can be sent to request that
the appropriate object be sent by some other node. Each object has separate transmit
and receive interrupts and status bits, giving the CPU full flexibility in detecting when a
remote/data frame has been sent or received.
For general purpose two masks for acceptance filtering can be programmed, one for
identifiers of 11 bits and one for identifiers of 29 bits. However the CPU must configure
bit XTD (Normal or Extended Frame Identifier) for each valid message to determine
whether a standard or extended frame will be accepted.
The last message object has its own programmable mask for acceptance filtering,
allowing a large number of infrequent objects to be handled by the system.
The object layer architecture of the CAN controller is designed to be as regular and
orthogonal as possible. This makes it easy to use.
User’s Manual
18-2
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
CAN_TxD
CAN_RxD
BTL - Configuration
BTL
CRC
Timing
Generator
Tx/Rx Shift Register
Messages
Clocks
(to all)
Control
Messages
Handlers
Intelligent Memory
Interrupt
Register
Status +
Control
BSP
Status
Register
EML
to XBUS
MCB04391
Figure 18-2 CAN Controller Block Diagram
User’s Manual
18-3
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Tx/Rx Shift Register
The Transmit/Receive Shift Register holds the destuffed bit stream from the bus line to
allow the parallel access to the whole data or remote frame for the acceptance match
test and the parallel transfer of the frame to and from the Intelligent Memory.
Bit Stream Processor
The Bit Stream Processor (BSP) is a sequencer controlling the sequential data stream
between the Tx/Rx Shift Register, the CRC Register, and the bus line. The BSP also
controls the EML and the parallel data stream between the Tx/Rx Shift Register and the
Intelligent Memory such that the processes of reception, arbitration, transmission, and
error signalling are performed according to the CAN protocol. Note that the automatic
retransmission of messages which have been corrupted by noise or other external error
conditions on the bus line is handled by the BSP.
Cyclic Redundancy Check Register
This register generates the Cyclic Redundancy Check (CRC) code to be transmitted
after the data bytes and checks the CRC code of incoming messages. This is done by
dividing the data stream by the code generator polynomial.
Error Management Logic
The Error Management Logic (EML) is responsible for the fault confinement of the CAN
device. Its counters, the Receive Error Counter and the Transmit Error Counter, are
incremented and decremented by commands from the Bit Stream Processor. According
to the values of the error counters, the CAN controller is set into the states error active,
error passive and busoff.
The CAN controller is error active, if both error counters are below the error passive limit of 128.
It is error passive, if at least one of the error counters equals or exceeds 128.
It goes busoff, if the Transmit Error Counter equals or exceeds the busoff limit of 256.
The device remains in this state, until the busoff recovery sequence is finished.
Additionally, there is the bit EWRN in the Status Register, which is set, if at least one of
the error counters equals or exceeds the error warning limit of 96. EWRN is reset, if both
error counters are less than the error warning limit.
User’s Manual
18-4
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Bit Timing Logic
This block (BTL) monitors the busline input CAN_RXD and handles the busline related
bit timing according to the CAN protocol.
The BTL synchronizes on a recessive to dominant busline transition at Start of Frame
(hard synchronization) and on any further recessive to dominant busline transition, if the
CAN controller itself does not transmit a dominant bit (resynchronization).
The BTL also provides programmable time segments to compensate for the propagation
delay time and for phase shifts and to define the position of the Sample Point in the bit
time. The programming of the BTL depends on the baudrate and on external physical
delay times.
Intelligent Memory
The Intelligent Memory (CAM/RAM Array) provides storage for up to 15 message objects
of maximum 8 data bytes length. Each of these objects has a unique identifier and its
own set of control and status bits. After the initial configuration, the Intelligent Memory
can handle the reception and transmission of data without further CPU actions.
Organization of Registers and Message Objects
All registers and message objects of the CAN controller are located in the special CAN
address area of 256 Bytes, which is mapped into segment 0 and uses addresses
00’EF00 through 00’EFFF . All registers are organized as 16-bit registers, located on
H
H
word addresses. However, all registers may be accessed bytewise in order to select
special actions without effecting other mechanisms.
Register Naming reflects the specific name of a register as well as a general module
indicator. This results in unique register names.
Example: module indicator is C1 (CAN module 1), specific name is Control/Status
Register (CSR), unique register name is C1CSR.
Note: The address map shown below lists the registers which are part of the CAN
controller. There are also C167CR specific registers that are associated with the
CAN module.
User’s Manual
18-5
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
EF00
EF10
EF20
EF30
EF40
EF50
EF60
EF70
EF80
EF90
EFA0
EFB0
EFC0
EFD0
EFE0
EFF0
General Registers
Message Object 1
Message Object 2
Message Object 3
Message Object 4
Message Object 5
Message Object 6
Message Object 7
Message Object 8
Message Object 9
Message Object 10
Message Object 11
Message Object 12
Message Object 13
Message Object 14
Message Object 15
EF00
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Control/Status
Register CSR
H
H
H
H
H
EF02
EF04
EF06
EF08
Interrupt Register
IR
Bit Timing Register
BTR
Global Mask Short
GMS
Global Mask Long
LGML
UGML
EF0C
Mask of Last
Message
H
LMLM
UMLM
CAN Address Area
General Registers
MCA04392
Figure 18-3 CAN Module Address Map
User’s Manual
18-6
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
18.2
General Functional Description
The Control/Status Register (CSR) accepts general control settings for the module and
provides general status information.
CSR
Control/Status Register
XReg (EF00 )
Reset value: XX01
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
B
E
RX TX
OK OK
-
LEC
TM CCE
rw rw
0
CPS EIE SIE IE INIT
rw rw rw rwh
OFF WRN
rh
rh
r
rwh rwh
rwh
r
r
Bit
Function (Control Bits)
Initialization
INIT
Starts the initialization of the CAN controller, when set.
INIT is set – after a reset
– when entering the busoff state
– by the application software
IE
Interrupt Enable
Enables or disables interrupt generation from the CAN module via the signal
XINTR. Does not affect status updates.
SIE
Status Change Interrupt Enable
Enables or disables interrupt generation when a message transfer
(reception or transmission) is successfully completed or a CAN bus error is
detected (and registered in the status partition).
EIE
Error Interrupt Enable
Enables or disables interrupt generation on a change of bit BOFF or EWARN
in the status partition).
CPS
Clock Prescaler Control Bit
0:
This control bit position is fixed to ‘0’.
This bit is reserved and will disable the 2:1 input clock prescaler in future
version of the CAN module..
CCE
TM
Configuration Change Enable
Allows or inhibits CPU access to the Bit Timing Register.
Test Mode (must be ‘0’)
Make sure that this bit is always cleared when writing to the Control Register,
as this bit controls a special test mode, that is used for production testing.
During normal operation, however, this test mode may lead to undesired
behaviour of the device.
User’s Manual
18-7
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Bit
Function (Control Bits)
Last Error Code
LEC
This field holds a code which indicates the type of the last error occurred on
the CAN bus. If a message has been transferred (reception or transmission)
without error, this field will be cleared.
0
1
No Error
Stuff Error: More than 5 equal bits in a sequence have occurred in a
part of a received message where this is not allowed.
Form Error: Wrong format in fixed format part of a received frame.
AckError: The message this CAN controller transmitted was not
acknowledged by another node.
2
3
4
Bit1Error: During the transmission of a message (with the exception
of the arbitration field), the device wanted to send a recessive level
(“1”), but the monitored bus value was dominant.
5
Bit0Error: During the transmission of a message (or acknowledge bit,
active error flag, or overload flag), the device wanted to send a
dominant level (“0”), but the monitored bus value was recessive.
During busoff recovery this status is set each time a sequence of 11
recessive bits has been monitored. This enables the CPU to monitor
the proceeding of the busoff recovery sequence (indicates that the bus
is not stuck at dominant or continously disturbed).
6
7
CRCError: The received CRC check sum was incorrect.
Unused code: may be written by the CPU to check for updates.
TXOK
RXOK
Transmitted Message Successfully
Indicates that a message has been transmitted successfully (error free and
acknowledged by at least one other node), since this bit was last reset by the
CPU (the CAN controller does not reset this bit!).
Received Message Successfully
This bit is set each time a message has been received successfully, since
this bit was last reset by the CPU (the CAN controller does not reset this bit!).
RXOK is also set when a message is received that is not accepted (i.e. stored).
EWRN
BOFF
Error Warning Status
Indicates that at least one of the error counters in the EML has reached the
error warning limit of 96.
Busoff Status
Indicates when the CAN controller is in busoff state (see EML).
Note: Reading the upper half of the Control Register (status partition) will clear the
Status Change Interrupt value in the Interrupt Register, if it is pending. Use byte
accesses to the lower half to avoid this.
User’s Manual
18-8
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
CAN Interrupt Handling
The on-chip CAN module has one interrupt output, which is connected (through a
synchronization stage) to a standard interrupt node in the C167CR in the same manner
as all other interrupts of the standard on-chip peripherals. With this configuration, the
user has all control options available for this interrupt, such as enabling/disabling, level
and group priority, and interrupt or PEC service (see note below). The on-chip CAN
module is connected to an XBUS interrupt control register.
As for all other interrupts, the node interrupt request flag is cleared automatically by
hardware when this interrupt is serviced (either by standard interrupt or PEC service).
Note: As a rule, CAN interrupt requests can be serviced by a PEC channel. However,
because PEC channels only can execute single predefined data transfers (there
are no conditional PEC transfers), PEC service can only be used, if the respective
request is known to be generated by one specific source, and that no other
interrupt request will be generated in between. In practice this seems to be a rare
case.
Since an interrupt request of the CAN module can be generated due to different
conditions, the appropriate CAN interrupt status register must be read in the service
routine to determine the cause of the interrupt request. The interrupt identifier INTID (a
number) in the Port Control/Interrupt Register (PCIR) indicates the cause of an interrupt.
When no interrupt is pending, the identifier will have the value 00 .
H
If the value in INTID is not 00 , then there is an interrupt pending. If bit IE in the control/
H
status register is set also the interrupt signal to the CPU is activated. The interrupt signal
(to the interrupt node) remains active until INTID gets 00 (i.e. all interrupt requests have
H
been serviced) or until interrupt generation is disabled (CSR.IE = ‘0’).
Note: The interrupt node is activated only upon a 0 → 1 transition of the CAN interrupt
signal. The CAN interrupt service routine should only be left after INTID has been
verified to be 00 .
H
The interrupt with the lowest number has the highest priority. If a higher priority interrupt
(lower number) occurs before the current interrupt is processed, INTID is updated and
the new interrupt overrides the last one.
INTID is also updated when the respective source request has been processed. This is
indicated by clearing the INTPND flag in the respective object’s message control register
(MCRn) or by reading the status partition of register CSR (in case of a status change
interrupt). The updating of INTID is done by the CAN state machine and takes up to 6
CAN clock cycles (1 CAN clock cycle = 1 or 2 CPU clock cycles, detrmined by the
prescaler bit CPS), depending on current state of the state machine.
Note: A worst case condition can occur when BRP = 00 AND the CAN controller is storing
H
a just received message AND the CPU is executing consecutive accesses to the CAN
module. In this rare case the maximum delay may be 26 CAN clock cycles.
The impact of this delay can be minimized by clearing bit INTPND at an early stage
User’s Manual
18-9
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
of interrupt processing, and (if required) restricting CPU accesses to the CAN
module until the anticipated updating is complete.
IR
Interrupt Register
XReg (EF02 )
Reset value: XXXX
H
H
15
14
13
12
11
10
-
9
-
8
7
6
5
4
3
2
1
0
- reserved -
INTID
rh
-
-
-
-
-
-
Bit
Function
Interrupt Identifier
This number indicates the cause of the interrupt (if pending).
INTID
00
01
Interrupt Idle: There is no interrupt request pending.
H
H
Status Change Interrupt: The CAN controller has updated (not
necessarily changed) the status in the Control Register. This can
refer to a change of the error status of the CAN controller (EIE is
set and BOFF or EWRN change) or to a CAN transfer incident
(SIE must be set), like reception or transmission of a message
(RXOK or TXOK is set) or the occurrence of a CAN bus error (LEC
is updated). The CPU may clear RXOK, TXOK, and LEC,
however, writing to the status partition of the Control Register can
never generate or reset an interrupt. To update the INTID value
the status partition of the Control Register must be read.
02
Message 15 Interrupt: Bit INTPND in the Message Control
Register of message object 15 (last message) has been set.
The last message object has the highest interrupt priority of all
H
1)
message objects.
(02 + N) Message N Interrupt: Bit INTPND in the Message Control
Register of message object ‘N’ has been set (N = 1 … 14). Note
that a message interrupt code is only displayed, if there is no other
1)
interrupt request with a higher priority.
Example: message 1: INTID = 03 , message 14: INTID = 10
H
H
1)
Bit INTPND of the corresponding message object has to be cleared to give messages with a lower priority the
possibility to update INTID or to reset INTID to “00 ” (idle state).
H
User’s Manual
18-10
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Configuration of the Bit Timing
According to the CAN protocol specification, a bit time is subdivided into four segments:
Sync segment, propagation time segment, phase buffer segment 1 and phase buffer
segment 2.
Each segment is a multiple of the time quantum tq, with
(1 – CPS)
tq = (BRP + 1) × 2
× tXCLK.
The Synchronization Segment (Sync Seg) is always 1 tq long. The Propagation Time
Segment and the Phase Buffer Segment 1 (combined to TSeg1) define the time before
the sample point, while Phase Buffer Segment 2 (TSeg2) defines the time after the
sample point. The length of these segments is programmable (except Sync-Seg) via the
Bit Timing Register (BTR).
Note: For exact definition of these segments please refer to the CAN Protocol
Specification.
1 Bit Time
Sync-
Seg
Sync-
Seg
TSeg1
TSeg2
1 time quantum
Sample Point
Transmit Point
(tq)
MCT04393
Figure 18-4 Bit Timing Definition
The bit time is determined by the XBUS clock period tXCLK, the Baud Rate Prescaler,
and the number of time quanta per bit:
bit time = tSync-Seg + tTSeg1 + tTSeg2
tSync-Seg = 1 × tq
[18.1]
tTSeg1
tTSeg2
= (TSEG1 + 1) × tq
= (TSEG2 + 1) × tq
(1-CPS)
tq
= (BRP + 1) × 2
× tXCLK
[18.2]
Note: TSEG1, TSEG2, and BRP are the programmed numerical values from the
respective fields of the Bit Timing Register.
User’s Manual
18-11
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
BTR
Bit Timing Register
XReg (EF04 )
Reset value: UUUU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0
TSEG2
TSEG1
SJW
BRP
r
rw
rw
rw
rw
Bit
Function
Baud Rate Prescaler
For generating the bit time quanta the CPU frequency fCPU is divided by
BRP
(1 – CPS)
2
× (BRP + 1). See also the prescaler control bit CPS in register
CSR.
SJW
(Re)Synchronization Jump Width
Adjust the bit time by maximum (SJW + 1) time quanta for
resynchronization.
TSEG1
TSEG2
Time Segment before sample point
There are (TSEG1 + 1) time quanta before the sample point.
Valid values for TSEG1 are “2 … 15”.
Time Segment after sample point
There are (TSEG2 + 1) time quanta after the sample point.
Valid values for TSEG2 are “1 … 7”.
Note: This register can only be written, if the config. change enable bit (CCE) is set.
Hard Synchronization and Resynchronization
To compensate phase shifts between clock oscillators of different CAN controllers, any
CAN controller has to synchronize on any edge from recessive to dominant bus level if
the edge lies between a Sample Point and the next Synchronization Segment, and on
any other edge if it itself does not send a dominant level. If the Hard Synchronization is
enabled (at the Start of Frame), the bit time is restarted at the Synchronization Segment,
otherwise the Resynchronization Jump Width (SJW) defines the maximum number of
time quanta by which a bit time may be shortened or lengthened during one
Resynchronization. The current bit time is adjusted by
tSJW = (SJW + 1) × tq
Note: SJW is the programmed numerical value from the respective field of the Bit Timing
Register.
User’s Manual
18-12
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Calculation of the Bit Time
The programming of the bit time according to the CAN Specification depends on the
desired baudrate, the XCLK frequency, and on the external physical delay times of the
bus driver, of the bus line and of the input comparator. These delay times are
summarized in the Propagation Time Segment tProp, where
tProp is two times the maximum of the sum of physical bus delay, the input comparator
delay, and the output driver delay rounded up to the nearest multiple of tq.
To fulfill the requirements of the CAN specification, the following conditions must be met:
tTSeg2 ≥ 2 × tq = Information Processing Time
tTSeg2 ≥ tSJW
tTSeg1 ≥ 3 × tq
tTSeg1 ≥ tSJW + tProp
Note: In order to achieve correct operation according to the CAN protocol the total bit
time should be at least 8 t , i.e. TSEG1 + TSEG2 ≥ 5.
q
So, to operate with a baudrate of 1 MBit/sec, the XCLK frequency has to be at
least 8/16 MHz (depending on the prescaler control bit CPS in register CSR).
The maximum tolerance df for XCLK depends on the Phase Buffer Segment 1 (PB1),
the Phase Buffer Segment 2 (PB2), and the Resynchronization Jump Width (SJW):
min(PB1, PB2)
2 × (13 × bit time – PB2)
---------------------------------------------------------------
df
≤
AND
tSJW
20 × bit time
-------------------------------
df
≤
The examples below show how the bit timing is to be calculated under specific
circumstances.
User’s Manual
18-13
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Bit Timing Example for High Baudrate
This example makes the following assumptions:
• XCLK frequency = 20 MHz
• BRP = 00, CPS = 0
• Baudrate = 1 Mbit/sec
tq
100 ns = 2 × tXCLK
bus driver delay
50 ns
receiver circuit delay
30 ns
bus line (40 m) delay 220 ns
tProp
tSJW
tTSeg1
tTSeg2
tSync
tBit
600 ns = 6 × tq
100 ns = 1 × tq
700 ns = tProp + tSJW
200 ns = Information Processing Time
100 ns = 1 × tq
1000 ns = tSync + tTSeg1 + tTSeg2
min(PB1, PB2)
2 × (13 × bit time – PB2)
---------------------------------------------------------------
tolerance for tXCLK
0.39%
=
=
0,1µs
----------------------------------------------------------
2 × (13 × 1µs – 0,2µs)
Bit Timing Example for Low Baudrate
This example makes the following assumptions:
• XCLK frequency = 4 MHz
• BRP = 01, CPS = 0
• Baudrate = 100 kbit/sec
tq
1 µs
= 4 × tXCLK
bus driver delay
receiver circuit delay
200 ns
80 ns
bus line (40 m) delay 220 ns
tProp
tSJW
tTSeg1
tTSeg2
tSync
tBit
1 µs
4 µs
5 µs
4 µs
1 µs
10 µs
= 1 × tq
= 4 × tq
= tProp + tSJW
= Information Processing Time + 2 × tq
= 1 × tq
= tSync + tTSeg1 + tTSeg2
min(PB1, PB2)
2 × (13 × bit time – PB2)
---------------------------------------------------------------
tolerance for fXCLK
1.58%
=
4µs
---------------------------------------------------------
=
2 × (13 × 10µs – 4µs)
User’s Manual
18-14
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Mask Registers
Messages can use standard or extended identifiers. Incoming frames are masked with
their appropriate global masks. Bit IDE of the incoming message determines, if the
standard 11-bit mask in Global Mask Short (GMS) is to be used, or the 29-bit extended
mask in Global Mask Long (UGML&LGML). Bits holding a “0” mean “don’t care”, i.e. do
not compare the message’s identifier in the respective bit position.
The last message object (15) has an additional individually programmable acceptance
mask (Mask of Last Message, UMLM&LMLM) for the complete arbitration field. This
allows classes of messages to be received in this object by masking some bits of the
identifier.
Note: The Mask of Last Message is ANDed with the Global Mask that corresponds to
the incoming message.
GMS
Global Mask Short
XReg (EF06 )
Reset value: UFUU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ID20 … 18
1
1
1
1
1
ID28 … 21
rw
r
r
r
r
r
rw
Bit
Function
Identifier (11-bit)
Mask to filter incoming messages with standard identifier.
ID28 … 18
User’s Manual
18-15
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
UGML
Upper Global Mask Long
XReg (EF08 )
Reset value: UUUU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ID20 … 13
ID28 … 21
rw
rw
LGML
Lower Global Mask Long
XReg (EF0A )
Reset value: UUUU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ID4 … 0
0
0
0
ID12 … 5
rw
r
r
r
rw
Bit
Function
Identifier (29-bit)
Mask to filter incoming messages with extended identifier.
ID28 … 0
User’s Manual
18-16
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
UMLM
Upper Mask of Last Message
XReg (EF0C )
Reset value: UUUU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ID20 … 18
ID17 … 13
ID28 … 21
rw
rw
rw
LMLM
Lower Mask of Last Message
XReg (EF0E )
Reset value: UUUU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ID4 … 0
0
0
0
ID12 … 5
rw
r
r
r
rw
Bit
Function
Identifier (29-bit)
ID28 … 0
Mask to filter the last incoming message (Nr. 15) with standard or
extended identifier (as configured).
User’s Manual
18-17
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
18.3
The Message Object
The message object is the primary means of communication between CPU and CAN
controller. Each of the 15 message objects uses 15 consecutive bytes (see map below)
and starts at an address that is a multiple of 16.
Note: All message objects must be initialized by the CPU, even those which are not
going to be used, before clearing the INIT bit.
Offset
Message Control (MCR)
Arbitration (UAR&LAR)
+0
Object Start Address (EFn0 )
H
+2
+4
Message Object 1: EF10
H
Data0
Msg. Config. (MCFG)
+6
Message Object 2: EF20
H
Data2
Data4
Data1
Data3
Data5
Data7
+8
...
+10
+12
+14
Message Object 14: EFE0
H
Data6
Message Object 15: EFF0
H
Reserved
MCA04394
Figure 18-5 Message Object Address Map
The general properties of a message object are defined via the Message Control
Register (MCR). There is a dedicated register MCRn for each message object n.
Each element of the Message Control Register is made of two complementary bits.This
special mechanism allows the selective setting or resetting of specific elements (leaving
others unchanged) without requiring read-modify-write cycles. None of these elements
will be affected by reset.
Table 18-1 shows how to use and interpret these 2-bit fields.
Table 18-1 MCR Bitfield Encoding
Value Function on Write
Meaning on Read
– reserved –
0 0
0 1
1 0
1 1
– reserved –
Reset element
Element is reset
Element is set
– reserved –
Set element
Leave element unchanged
User’s Manual
18-18
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
MCRn
Message Control Register
XReg (EFn0 )
Reset value: UUUU
H
H
15
14
13
TXRQ
rw
12
11
10
9
8
7
6
5
4
3
2
1
0
MSGLST
CPUUPD
RMTPND
NEWDAT MSGVAL
rw rw
TXIE
rw
RXIE
rw
INTPND
rw
rw
rw
Bit
Function
INTPND
Interrupt Pending
Indicates, if this message object has generated an interrupt request (see
TXIE and RXIE), since this bit was last reset by the CPU, or not.
RXIE
Receive Interrupt Enable
Defines, if bit INTPND is set after successful reception of a frame.
TXIE
Transmit Interrupt Enable
Defines, if bit INTPND is set after successful transmission of a frame.
1)
MSGVAL
Message Valid
Indicates, if the corresponding message object is valid or not. The CAN
controller only operates on valid objects. Message objects can be tagged
invalid, while they are changed, or if they are not used at all.
NEWDAT
MSGLST
CPUUPD
New Data
Indicates, if new data has been written into the data portion of this
message object by CPU (transmit-objects) or CAN controller (receive-
objects) since this bit was last reset, or not.
2)
Message Lost (This bit applies to receive-objects only!)
Indicates that the CAN controller has stored a new message into this
object, while NEWDAT was still set, i.e. the previously stored message
is lost.
CPU Update (This bit applies to transmit-objects only!)
Indicates that the corresponding message object may not be transmitted
now. The CPU sets this bit in order to inhibit the transmission of a
message that is currently updated, or to control the automatic response
to remote requests.
TXRQ
Transmit Request
Indicates that the transmission of this message object is requested by
the CPU or via a remote frame and is not yet done. TXRQ can be
1) 3)
disabled by CPUUPD.
User’s Manual
18-19
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Bit
Function
RMTPND
Remote Pending (Used for transmit-objects)
Indicates that the transmission of this message object has been
requested by a remote node, but the data has not yet been transmitted.
When RMTPND is set, the CAN controller also sets TXRQ. RMTPND
and TXRQ are cleared, when the message object has been successfully
transmitted.
1)
In message object 15 (last message) these bits are hardwired to “0” (inactive) in order to prevent transmission
of message 15.
2)
When the CAN controller writes new data into the message object, unused message bytes will be overwritten
by non specified values. Usually the CPU will clear this bit before working on the data, and verify that the bit is
still cleared once it has finished working to ensure that it has worked on a consistent set of data and not part
of an old message and part of the new message.
For transmit-objects the CPU will set this bit along with clearing bit CPUUPD. This will ensure that, if the
message is actually being transmitted during the time the message was being updated by the CPU, the CAN
controller will not reset bit TXRQ. In this way bit TXRQ is only reset once the actual data has been transferred.
3)
When the CPU requests the transmission of a receive-object, a remote frame will be sent instead of a data
frame to request a remote node to send the corresponding data frame. This bit will be cleared by the CAN
controller along with bit RMTPND when the message has been successfully transmitted, if bit NEWDAT has
not been set.
If there are several valid message objects with pending transmission request, the message with the lowest
message number is transmitted first. This arbitration is done when several objects are requested for
transmission by the CPU, or when operation is resumed after an error frame or after arbitration has been lost.
Arbitration Registers
The Arbitration Registers (UARn&LARn) are used for acceptance filtering of incoming
messages and to define the identifier of outgoing messages. A received message with
a matching identifier is accepted as a data frame (matching object has DIR = ‘0’) or as a
remote frame (matching object has DIR = ‘1’). For matching, the corresponding Global
Mask has to be considered (in case of message object 15 also the Mask of Last
Message). Extended frames (using Global Mask Long) can be stored only in message
objects with XTD = ‘1’, standard frames (using Global Mask Short) only in message
objects with XTD = ‘0’.
Message objects should have unique identifiers, i.e. if some bits are masked out by the
Global Mask Registers (i.e. “don’t care”), then the identifiers of the valid message objects
should differ in the remaining bits which are used for acceptance filtering.
If a received message (data frame or remote frame) matches with more than one valid
message object, it is associated with the object with the lowest message number. I.e. a
received data frame is stored in the “lowest” object, or the “lowest” object is sent in
response to a remote frame. The Global Mask is used for matching here.
User’s Manual
18-20
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
After a transmission (data frame or remote frame) the transmit request flag of the
matching object with the lowest message number is cleared. The Global Mask is not
used in this case.
When the CAN controller accepts a data frame, the complete message is stored into
the corresponding message object, including the identifier (also masked bits, standard
identifiers have bits ID17-0 filled with ‘0’), the data length code (DLC), and the data bytes
(valid bytes indicated by DLC). This is implemented to keep the data bytes connected
with the identifier, even if arbitration mask registers are used.
When the CAN controller accepts a remote frame, the corresponding transmit
message object (1 … 14) remains unchanged, except for bits TXRQ and RMTPND,
which are set, of course. In the last message object 15 (which cannot start a
transmission) the identifier bits corresponding to the “don’t care” bits of the Last Message
Mask are copied from the received frame. Bits corresponding to the “don’t care” bits of
the corresponding global mask are not copied (i.e. bits masked out by the global and the
last message mask cannot be retrieved from object 15).
UARn
Upper Arbitration Register
XReg (EFn2 )
Reset value: UUUU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ID20 … 18
ID17 … 13
ID28 … 21
rw
rw
LARn
Lower Arbitration Register
XReg (EFn4 )
Reset value: UUUU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ID4 … 0
0
0
0
ID12 … 5
rw
r
r
r
rw
Bit
Function
Identifier (29-bit)
ID28 … 0
Identifier of a standard message (ID28 … 18) or an extended message
(ID28 … 0). For standard identifiers bits ID17 … 0 are “don’t care”.
User’s Manual
18-21
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Message Configuration
The Message Configuration Register (low byte of MCFGn) holds a description of the
message within this object.
Note: There is no “don’t care” option for bits XTD and DIR. So incoming frames can only
match with corresponding message objects, either standard (XTD = 0) or
extended (XTD = 1). Data frames only match with receive-objects, remote frames
only match with transmit-objects.
When the CAN controller stores a data frame, it will write all the eight data bytes
into a message object. If the data length code was less than 8, the remaining bytes
of the message object will be overwritten by non specified values.
MCFGn
Message Configuration Reg.
XReg (EFn6 )
Reset value: - - UU
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Data Byte 0
DLC
DIR XTD
rw rw
0
0
rw
rw
r
r
Bit
Function
XTD
Extended Identifier
0:
Standard
This message object uses a standard 11-bit identifier.
Extended
1:
This message object uses an extended 29-bit identifier.
DIR
Message Direction
0:
Receive object.
On TXRQ, a remote frame with the identifier of this message
object is transmitted.
On reception of a data frame with matching identifier, that
message is stored in this message object.
Transmit object.
1:
On TXRQ, the respective message object is transmitted.
On reception of a remote frame with matching identifier, the TXRQ
and RMTPND bits of this message object are set.
DLC
Data Length Code
Defines the number of valid data bytes within the data area.
Valid values for the data length are 0 … 8.
Note: The first data byte occupies the upper half of the message configuration register.
User’s Manual
18-22
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Data Area
The data area occupies 8 successive byte positions after the Message Configuration
Register, i.e. the data area of message object n covers locations 00’EFn7 through
H
00’EFnE .
H
Location 00’EFnF is reserved.
H
Message data for message object 15 (last message) will be written into a two-message-
alternating buffer to avoid the loss of a message, if a second message has been
received, before the CPU has read the first one.
Handling of Message Objects
The following diagrams summarize the actions that have to be taken in order to transmit
and receive messages over the CAN bus. The actions taken by the CAN controller are
described as well as the actions that have to be taken by the CPU (i.e. the servicing
program).
The diagrams show:
• CAN controller handling of transmit objects
• CAN controller handling of receive objects
• CPU handling of transmit objects
• CPU handling of receive objects
• CPU handling of last message object
• Handling of the last message’s alternating buffer
User’s Manual
18-23
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
yes
no
Bus free?
no
TXRQ = 1
CPUUPD = 0
Received remote frame
with same identifier as
no
this message object?
yes
yes
NEWDAT := 0
Load message
into buffer
TXRQ := 1
RMTPND := 1
Send message
no
RXIE = 1?
yes
no
Transmission
successful?
INTPND := 1
yes
no
TXRQ := 0
RMTPND := 0
NEWDAT = 1?
yes
no
TXIE = 1?
yes
INTPND := 1
0: Reset
1: Set
MCA04395
Figure 18-6 CAN Controller Handling of Transmit Objects (DIR = ‘1’)
User’s Manual
18-24
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
yes
no
Bus idle?
no
TXRQ = 1?
CPUUPD = 0?
Received frame with
same identifier as this
no
message object?
yes
yes
NEWDAT := 0
Load identifier and
control into buffer
yes
NEWDAT = 1
no
Send remote frame
MSGLST := 1
no
Transmission
successful?
yes
Store message
NEWDAT := 1
TXRQ := 0
TXRQ := 0
RMTPND := 0
RMTPND := 0
no
no
TXIE = 1?
RXIE = 1?
yes
yes
INTPND := 1
INTPND := 1
0: Reset
1: Set
MCA04396
Figure 18-7 CAN Controller Handling of Receive Objects (DIR = ‘0’)
User’s Manual
18-25
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Power Up
(all bits undefined)
TXIE := (application cpecific)
RXIE := (application cpecific)
INTPND := 0
RMTPND := 0
TXRQ := 0
CPUUPD := 1
Identifier := (application cpecific)
NEWDAT := 0
Initialization
Direction := transmit
DLC := (application cpecific)
MSGVAL := 1
XTD := (application cpecific)
CPUUPD := 1
NEWDAT := 1
Update: Start
Update
Write / calculate message contents
CPUUPD := 0
Update: End
yes
Want to send?
no
TXRQ := 1
yes
no
Update
message?
0: Reset
1: Set
MCA04397
Figure 18-8 CPU Handling of Transmit Objects (DIR = ‘1’)
User’s Manual
18-26
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Power Up
(all bits undefined)
TXIE := (application cpecific)
RXIE := (application cpecific)
INTPNDd := 0
RMTPND := 0
TXRQ := 0
MSGLST := 0
Identifier := (application cpecific)
NEWDAT := 0
Initialization
Direction := receive
DLC := (value of DLC in transmitter)
MSGVAL := 1
XTD := (application cpecific)
Process: Start
NEWDAT := 0
Process message contents
Process
yes
Process: End
NEWDAT = 1?
Restart process
no
no
Request
Update?
yes
TXRQ := 1
0: Reset
1: Set
MCA04398
Figure 18-9 CPU Handling of Receive Objects (DIR = ‘0’)
User’s Manual
18-27
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Power Up
(all bits undefined)
RXIE := (application cpecific)
INTPND := 0
RMTPND := 0
MSGLST := 0
Identifier := (application cpecific)
NEWDAT := 0
Initialization
Direction := receive
DLC := (value of DLC in transmitter)
MSGVAL := 1
XTD := (application cpecific)
Process: Start
Process message contents
NEWDAT := 0
Process
yes
Process: End
NEWDAT = 1?
Restart process
no
0: Reset
1: Set
MCA04399
Figure 18-10 CPU Handling of the Last Message Object
User’s Manual
18-28
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Reset
CPU releases Buffer 2
CPU releases Buffer 1
Buffer 1 = Released
Buffer 2 = Released
CPU access to Buffer 2
Store received
message into Buffer 1
CPU allocates Buffer 2
Buffer 1 = Released
Buffer 2 = Allocated
CPU access to Buffer 2
Buffer 1 = Allocated
Buffer 2 = Released
CPU access to Buffer 1
Store received
message
Store received
message
into Buffer 1
into Buffer 2
Buffer 1 = Allocated
Buffer 2 = Allocated
CPU access to Buffer 2
Buffer 1 = Allocated
Buffer 2 = Allocated
CPU access to Buffer 1
Store received
message into
Buffer 1
CPU releases CPU releases
Buffer 2 Buffer 1
Store received
message into
Buffer 2
MSGLST is set
MSGLST is set
Allocated: NEWDAT = 1 or RMTPND = 1
Released: NEWDAT = 0 and RMTPND = 0
MCA04400
Figure 18-11 Handling of the Last Message Object’s Alternating Buffer
User’s Manual
18-29
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
18.4
Controlling the CAN Module
The CAN module is controlled by the C167CR via hardware signals (e.g. reset) and via
register accesses executed by software.
Accessing the On-chip CAN Module
The CAN module is implemented as an X-Peripheral and is therefore accessed like an
external memory or peripheral. That means that the registers of the CAN module can be
read and written using 16-bit or 8-bit direct or indirect MEM addressing modes. Also bit
handling is not supported via the XBUS. Since the XBUS, to which the CAN module is
connected, also represents the external bus, CAN accesses follow the same rules and
procedures as accesses to the external bus. CAN accesses cannot be executed in
parallel to external instruction fetches or data read/writes, but are arbitrated and inserted
into the external bus access stream.
Accesses to the CAN module use demultiplexed addresses, a 16-bit data bus (byte
accesses possible), two waitstates and no tristate waitstate.
The CAN address area starts at 00’EF00 and covers 256 Bytes. This area is decoded
H
internally, so none of the programmable address windows must be sacrificed in order to
access the on-chip CAN module.
The advantage of locating the CAN address area in segment 0 is that the CAN module
is accessible via data page 3, which is the ‘system’ data page, accessed usually through
the ‘system’ data page pointer DPP3. In this way, the internal addresses, such like SFRs,
internal RAM, and the CAN registers, are all located within the same data page and form
a contiguous address space.
Power Down Mode
If the C167CR enters Power Down mode, the XCLK signal will be turned off which will
stop the operation of the CAN module. Any message transfer is interrupted. In order to
ensure that the CAN controller is not stopped while sending a dominant level (‘0’) on the
CAN bus, the CPU should set bit INIT in the Control Register prior to entering Power
Down mode. The CPU can check if a transmission is in progress by reading bits TXRQ
and NEWDAT in the message objects and bit TXOK in the Control Register. After
returning from Power Down mode via hardware reset, the CAN module has to be
reconfigured.
User’s Manual
18-30
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
CAN Module Reset
The on-chip CAN module is connected to the XBUS Reset signal. This signal is
activated, when the C167CR’s reset input is activated, when a software reset is
executed, and in case of a watchdog reset. Activating the CAN module’s reset line
triggers a hardware reset.
This hardware reset:
• disconnects the CAN_TXD output from the port logic
• clears the error counters
• resets the busoff state
• switches the Control Register’s low byte to 01
H
• leaves the Control Register’s high byte and the Interrupt Register undefined
• does not change the other registers including the message objects (notified as UUUU)
Note: The first hardware reset after power-on leaves the unchanged registers in an
undefined state, of course.
The value 01 in the Control Register’s low byte prepares for the module
H
initialization.
CAN Module Activation
After a reset the CAN module is disabled. Before it can be used to receive or transmit
messages the application software must activate the CAN module.
Two actions are required for this purpose:
• General Module Enable globally activates the CAN module. This is done by setting
bit XPEN in register SYSCON.
• Module Initialization determines the functionality of the CAN module (baudrate,
active objects, etc.). This is the major part of the activation and is described in the
following.
User’s Manual
18-31
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Module Initialization
The module initialization is enabled by setting bit INIT in the control register CSR. This
can be done by the CPU via software, or automatically by the CAN controller on a
hardware reset, or if the EML switches to busoff state.
While INIT is set:
• all message transfer from and to the CAN bus is stopped
• the CAN transmit line CAN_TXD is “1” (recessive)
• the control bits NEWDAT and RMTPND of the last message object are reset
• the counters of the EML are left unchanged.
Setting bit CCE in addition, permits changing the configuration in the Bit Timing Register.
To initialize the CAN Controller, the following actions are required:
• configure the Bit Timing Register (CCE required)
• set the Global Mask Registers
• initialize each message object.
If a message object is not needed, it is sufficient to clear its message valid bit (MSGVAL),
i.e. to define it as not valid. Otherwise, the whole message object has to be initialized.
After the initialization sequence has been completed, the CPU clears bit INIT.
Now the BSP synchronizes itself to the data transfer on the CAN bus by waiting for the
occurrence of a sequence of 11 consecutive recessive bits (i.e. Bus Idle) before it can
take part in bus activities and start message transfers.
The initialization of the message objects is independent of the state of bit INIT and can
be done on the fly. The message objects should all be configured to particular identifiers
or set to “not valid” before the BSP starts the message transfer, however.
To change the configuration of a message object during normal operation, the CPU first
clears bit MSGVAL, which defines it as not valid. When the configuration is completed,
MSGVAL is set again.
User’s Manual
18-32
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Busoff Recovery Sequence
If the device goes busoff, it will set bit BOFF and also set bit INIT of its own accord,
stopping all bus activities. To have the CAN module take part in the CAN bus activities
again, the bus-off recovery sequence must be started by clearing the bit INIT (via
software). Once INIT has been cleared, the module will then wait for 129 occurrences of
Bus idle before resuming normal operation.
At the end of the busoff recovery sequence the Error Management Counters will be
reset. This will automatically clear bits BOFF and EWRN.
During the waiting time after the resetting of INIT each time a sequence of 11 recessive
bits has been monitored, a Bit0Error code is written to the Control Register, enabling
the CPU to check up whether the CAN bus is stuck at dominant or continously disturbed
and to monitor the proceeding of the busoff recovery sequence.
Note: An interrupt can be generated when entering the busoff state if bits IE and EIE are
set. The corresponding interrupt code in bitfield INTID is 01 .
H
The busoff recovery sequence cannot be shortened by setting or resetting INIT.
User’s Manual
18-33
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
18.5
Configuration Examples for Message Objects
The two examples below represent standard applications for using CAN messages. Both
examples assume that identifier and direction are already set up correctly.
The respective contents of the Message Control Register (MCR) are shown.
Configuration Example of a Transmission Object
This object shall be configured for transmission. It shall be transmitted automatically in
response to remote frames, but no receive interrupts shall be generated for this object.
MCR (Data bytes are not written completely → CPUUPD = ‘1’)
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0 1
RMTPND
0 1
TXRQ
1 0
0 1
1 0
0 1
TXIE
0 1
RXIE
0 1
INTPND
CPUUPD NEWDAT MSGVAL
MCR (Remote frame was received in the meantime → RMTPND = ‘1’, TXRQ = ‘1’)
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1 0
RMTPND
1 0
TXRQ
1 0
0 1
1 0
0 1
TXIE
0 1
RXIE
0 1
INTPND
CPUUPD NEWDAT MSGVAL
After updating the message the CPU should clear CPUUPD and set NEWDAT. The
previously received remote request will then be answered.
If the CPU wants to transmit the message actively it should also set TXRQ (which should
otherwise be left alone).
User’s Manual
18-34
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Configuration Example of a Reception Object
This object shall be configured for reception. A receive interrupt shall be generated each
time new data comes in. From time to time the CPU sends a remote request to trigger
the sending of this data from a remote node.
MCR (Message object is idle, i.e. waiting for a frame to be received)
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0 1
RMTPND
0 1
TXRQ
0 1
0 1
1 0
0 1
TXIE
1 0
RXIE
0 1
INTPND
MSGLST NEWDAT MSGVAL
.
MCR (A data frame was received → NEWDAT = ‘1’, INTPND = ‘1’)
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0 1
RMTPND
0 1
TXRQ
0 1
1 0
1 0
0 1
TXIE
1 0
1 0
INTPND
MSGLST NEWDAT MSGVAL
RXIE
To process the message the CPU should clear INTPND and NEWDAT, process the
data, and check that NEWDAT is still clear after that. If not, the processing should be
repeated.
To send a remote frame to request the data, simply bit TXRQ needs to be set. This bit
will be cleared by the CAN controller, once the remote frame has been sent or if the data
is received before the CAN controller could transmit the remote frame.
User’s Manual
18-35
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
18.6
The CAN Application Interface
The on-chip CAN module of the C167CR is connected to the (external) physical layer
(i.e. the CAN bus) via two signals:
Table 18-2 CAN Interface Signals
CAN Signal Port Pin
Function
CAN_RXD
CAN_TXD
P4.5
P4.6
Receive data from the physical layer of the CAN bus.
Transmit data to the physical layer of the CAN bus.
A logic low level (‘0’) is interpreted as the dominant CAN bus level, a logic high level (‘1’)
is interpreted as the recessive CAN bus level.
Connection to an External Transceiver
The CAN module of the C167CR can be connected to an external CAN bus via a CAN
transceiver.
Note: Basically it is also possible to connect several CAN modules directly (on-board)
without using CAN transceivers.
C167CR
CAN_RxD
CAN
Transceiver
CAN 1
Physical
Layer
CAN_TxD
MCS04401
Figure 18-12 Connection to a Single CAN Bus
User’s Manual
18-36
V3.1, 2000-03
C167CR
Derivatives
The On-chip CAN Interface
Port Control
The receive data line and the transmit data line of the CAN module are alternate port
functions. Make sure that the respective port pin for the receive line is switched to input
in order to enable proper reception. The respective port driver for the transmit will
automatically be switched ON.
This provides a standard pin configuration without additional software control and also
works in emulation mode where the port direction registers cannot be controlled.
Note: The CAN interface signals are only available on Port 4 if the respective pins are
not programmed to output segment address lines (alternate function of Port 4).
Select 0, 2, or 4 segment address lines if the CAN interface is to be used.
User’s Manual
18-37
V3.1, 2000-03
C167CR
Derivatives
System Reset
19
System Reset
The internal system reset function provides initialization of the C167CR into a defined
default state and is invoked either by asserting a hardware reset signal on pin RSTIN
(Hardware Reset Input), upon the execution of the SRST instruction (Software Reset) or
by an overflow of the watchdog timer.
Whenever one of these conditions occurs, the microcontroller is reset into its predefined
default state through an internal reset procedure. When a reset is initiated, pending
internal hold states are cancelled and the current internal access cycle (if any) is
completed. An external bus cycle is aborted, except for a watchdog reset (see
description). After that the bus pin drivers and the IO pin drivers are switched off
(tristate).
The internal reset procedure requires 516 CPU clock cycles in order to perform a
complete reset sequence. This 516 cycle reset sequence is started upon a watchdog
timer overflow, a SRST instruction or when the reset input signal RSTIN is latched low
(hardware reset). The internal reset condition is active at least for the duration of the
reset sequence and then until the RSTIN input is inactive and the PLL has locked (if the
PLL is selected for the basic clock generation). When this internal reset condition is
removed (reset sequence complete, RSTIN inactive, PLL locked) the reset configuration
is latched from PORT0. After that pins ALE, RD, and WR are driven to their inactive
levels.
Note: Bit ADP which selects the Adapt mode is latched with the rising edge of RSTIN.
After the internal reset condition is removed, the microcontroller will either start program
execution from external or internal memory, or enter boot mode.
C 167CR
RSTOUT
External
Hardware
RSTIN
VCC
External
Reset
a)
&
Reset
b)
Sources
+
a) Generated Warm Reset
b) Automatic Power-ON Reset
MCA02259
Figure 19-1 External Reset Circuitry
User’s Manual
19-1
V3.1, 2000-03
C167CR
Derivatives
System Reset
19.1
Reset Sources
Several sources (external or internal) can generate a reset for the C167CR. Software
can identify the respective reset source via the reset source indication flags in register
WDTCON. Generally any reset causes the same actions on the C167CR’s modules. The
differences are described in the following sections.
Hardware Reset
A hardware reset is triggered when the reset input signal RSTIN is latched low. To ensure
the recognition of the RSTIN signal (latching), it must be held low for at least 100 ns plus
2 CPU clock cycles (input filter plus synchronization). Also shorter RSTIN pulses may
trigger a hardware reset, if they coincide with the latch’s sample point. The actual
minimum duration for a reset pulse depends on the current CPU clock generation mode.
The worstcase is generating the CPU clock via the prescaler (f
= f
/ 2 in this case).
CPU
OSC
After the reset sequence has been completed, the RSTIN input is sampled again. When
the reset input signal is inactive at that time, the internal reset condition is terminated
(indicated as short hardware reset, SHWR). When the reset input signal is still active at
that time, the internal reset condition is prolonged until RSTIN gets inactive (indicated as
long hardware reset, LHWR).
During a hardware reset the inputs for the reset configuration (PORT0) need some time
to settle on the required levels, especially if the hardware reset aborts a read operation
from an external peripheral. During this settling time the configuration may intermittently
be wrong. For the duration of one internal reset sequence after a reset has been
recognized the configuration latches are not transparent, i.e. the (new) configuration
becomes valid earliest after the completion of one reset sequence. This usually covers
the required settling time.
When the basic clock is generated by the PLL the internal reset condition should be
extended until the on-chip PLL has locked (longer RSTIN signal).
The input RSTIN provides an internal pullup device equalling a resistor of 50 kΩ to
250 kΩ (the minimum reset time must be determined by the lowest value). Simply
connecting an external capacitor is sufficient for an automatic power-on reset (see b) in
Figure 19-1). RSTIN may also be connected to the output of other logic gates (see a) in
Figure 19-1). See also “Bidirectional Reset” on Page 19-4 in this case.
Note: A power-on reset requires an active time of two reset sequences (1036 CPU clock
cycles) after a stable clock signal is available (about 10 … 50 ms, depending on
the oscillator frequency, to allow the on-chip oscillator to stabilize).
User’s Manual
19-2
V3.1, 2000-03
C167CR
Derivatives
System Reset
Software Reset
The reset sequence can be triggered at any time via the protected instruction SRST
(Software Reset). This instruction can be executed deliberately within a program, e.g. to
leave bootstrap loader mode, or upon a hardware trap that reveals a system failure.
Note: A software reset only latches the configuration of the bus interface (SALSEL,
CSSEL, WRC, BUSTYP) from PORT0.
If bidirectional reset is enabled, a software reset is executed like a long hardware
reset.
Watchdog Timer Reset
When the watchdog timer is not disabled during the initialization or serviced regularly
during program execution it will overflow and trigger the reset sequence. Other than
hardware and software reset the watchdog reset completes a running external bus cycle
if this bus cycle either does not use READY at all, or if READY is sampled active (low)
after the programmed waitstates. When READY is sampled inactive (high) after the
programmed waitstates the running external bus cycle is aborted. Then the internal reset
sequence is started.
Note: A watchdog reset only latches the configuration of the bus interface (SALSEL,
CSSEL, WRC, BUSTYP) from PORT0.
If bidirectional reset is enabled a watchdog timer reset is executed like a long
hardware reset.
The watchdog reset cannot occur while the C167CR is in bootstrap loader mode!
User’s Manual
19-3
V3.1, 2000-03
C167CR
Derivatives
System Reset
Bidirectional Reset
In a special mode (bidirectional reset) the C167CR’s line RSTIN(normally an input) may
be driven active by the chip logic e.g. in order to support external equipment which is
required for startup (e.g. flash memory).
RSTIN
Internal Circuitry
&
Reset Sequence Active
BDRSTEN = '1'
MCS04403
Figure 19-2 Bidirectional Reset Operation
Bidirectional reset reflects internal reset sources (software, watchdog) also to the RSTIN
pin and converts short hardware reset pulses to a minimum duration of the internal reset
sequence. Bidirectional reset is enabled by setting bit BDRSTEN in register SYSCON
and changes RSTIN from a pure input to an open drain IO line. When an internal reset is
triggered by the SRST instruction or by a watchdog timer overflow or a low level is applied
to the RSTIN line, an internal driver pulls it low for the duration of the internal reset
sequence. After that it is released and is then controlled by the external circuitry alone.
The bidirectional reset function is useful in applications where external devices require
a defined reset signal but cannot be connected to the C167CR’s RSTOUT signal, e.g.
an external flash memory which must come out of reset and deliver code well before
RSTOUT can be deactivated via EINIT.
The following behavior differences must be observed when using the bidirectional reset
feature in an application:
• Bit BDRSTEN in register SYSCON cannot be changed after EINIT.
• After a reset bit BDRSTEN is cleared.
• The reset indication flags always indicate a long hardware reset.
• The PORT0 configuration is treated like on a hardware reset. Especially the bootstrap
loader may be activated when P0L.4 or RD is low.
• Pin RSTIN may only be connected to ext. reset devices with an open drain output driver.
• A short hardware reset is extended to the duration of the internal reset sequence.
User’s Manual
19-4
V3.1, 2000-03
C167CR
Derivatives
System Reset
19.2
Status After Reset
After a reset is completed most units of the C167CR enter a well-defined default status.
This ensures repeatable start conditions and avoids spurious activities after reset.
Watchdog Timer Operation after Reset
The watchdog timer starts running after the internal reset has completed. It will be
clocked with the internal system clock divided by 2 (f
/ 2), and its default reload value
CPU
16
is 00 , so a watchdog timer overflow will occur 131,072 CPU clock cycles (2 × 2 ) after
H
completion of the internal reset, unless it is disabled, serviced or reprogrammed
meanwhile. When the system reset was caused by a watchdog timer overflow, the
WDTR (Watchdog Timer Reset Indication) flag in register WDTCON will be set to ‘1’.
This indicates the cause of the internal reset to the software initialization routine. WDTR
is reset to ‘0’ by an external hardware reset, by servicing the watchdog timer or after
EINIT. After the internal reset has completed, the operation of the watchdog timer can
be disabled by the DISWDT (Disable Watchdog Timer) instruction. This instruction has
been implemented as a protected instruction. For further security, its execution is only
enabled in the time period after a reset until either the SRVWDT (Service Watchdog
Timer) or the EINIT instruction has been executed. Thereafter the DISWDT instruction
will have no effect.
Reset Values for the C167CR Registers
During the reset sequence the registers of the C167CR are preset with a default value.
Most SFRs, including system registers and peripheral control and data registers, are
cleared to zero, so all peripherals and the interrupt system are off or idle after reset. A
few exceptions to this rule provide a first pre-initialization, which is either fixed or
controlled by input pins.
DPP1:
0001 (points to data page 1)
H
DPP2:
0002 (points to data page 2)
H
DPP3:
0003 (points to data page 3)
H
CP:
FC00
FC00
FA00
FC00
H
H
H
STKUN:
STKOV:
SP:
H
WDTCON:
S0RBUF:
SSCRB:
SYSCON:
00XX (value depends on the reset source)
H
XX (undefined)
H
XXXX (undefined)
H
0XX0 (set according to reset configuration)
H
BUSCON0: 0XX0 (set according to reset configuration)
H
RP0H:
ONES:
XX (reset levels of P0H)
H
FFFF (fixed value)
H
User’s Manual
19-5
V3.1, 2000-03
C167CR
Derivatives
System Reset
The C167CR’s Pins after Reset
After the reset sequence the different groups of pins of the C167CR are activated in
different ways depending on their function. Bus and control signals are activated
immediately after the reset sequence according to the configuration latched from
PORT0, so either external accesses can takes place or the external control signals are
inactive. The general purpose IO pins remain in input mode (high impedance) until
reprogrammed via software (see Figure 19-3). The RSTOUT pin remains active (low)
until the end of the initialization routine (see description).
8)
6)
RSTIN
RD, WR
ALE
7)
Bus
1)
2)
2)
IO
4)
5)
3)
RSTOUT
Internal Reset Condition
6)
Initialization
3)
RSTIN
Internal Reset Condition
Initialization
When the internal reset condition is extended by RSTIN, the activation of the output signals is
delayed until the end of the internal reset condition.
1)
Current bus cycle is completed or aborted.
Switches asinchronously with RSTIN, sinchronously upon software or watchdog reset.
The reset condition ends here. The C 167CR starts program execution.
Activation of the IO pins is controlled by software.
Execution of the EINIT instruction.
The shaded area designates the internal reset sequence, which starts after synchronization of RSTIN.
A short hardware reset is extended until the end of the reset sequence in Bidirectional reset mode.
A software or WDT reset activates the RSTIN line in Bidirectional reset mode.
2)
3)
4)
5)
6)
7)
8)
MCS02258
Figure 19-3 Reset Input and Output Signals
User’s Manual
19-6
V3.1, 2000-03
C167CR
Derivatives
System Reset
Ports and External Bus Configuration during Reset
During the internal reset sequence all of the C167CR’s port pins are configured as inputs
by clearing the associated direction registers, and their pin drivers are switched to the
high impedance state. This ensures that the C167CR and external devices will not try to
drive the same pin to different levels. Pin ALE is held low through an internal pulldown,
and pins RD, WR and READY are held high through internal pullups. Also the pins that
can be configured for CS output will be pulled high.
The registers SYSCON and BUSCON0 are initialized according to the configuration
selected via PORT0.
When an external start is selected (pin EA = ‘0’):
• the Bus Type field (BTYP) in register BUSCON0 is initialized according to
P0L.7 and P0L.6
• bit BUSACT0 in register BUSCON0 is set to ‘1’
• bit ALECTL0 in register BUSCON0 is set to ‘1’
• bit ROMEN in register SYSCON will be cleared to ‘0’
• bit BYTDIS in register SYSCON is set according to the data bus width (set if 8-bit)
• bit WRCFG in register SYSCON is set according to pin P0H.0 (WRC)
When an internal start is selected (pin EA = ‘1’):
• register BUSCON0 is cleared to 0000
H
• bit ROMEN in register SYSCON will be set to ‘1’
• bit BYTDIS in register SYSCON is set, i.e. BHE/WRH is disabled
• bit WRCFG in register SYSCON is set according to pin P0H.0 (WRC)
The other bits of register BUSCON0, and the other BUSCON registers are cleared.
This default initialization selects the slowest possible external accesses using the
configured bus type.
When the internal reset has completed, the configuration of PORT0, PORT1, Port 4,
Port 6, and of the BHE signal (High Byte Enable, alternate function of P3.12) depends
on the bus type which was selected during reset. When any of the external bus modes
was selected during reset, PORT0 will operate in the selected bus mode. Port 4 will
output the selected number of segment address lines (all zero after reset). Port 6 will
drive the selected number of CS lines (CS0 will be ‘0’, while the other active CS lines
will be ‘1’). When no memory accesses above 64 K are to be performed, segmentation
may be disabled.
When the on-chip bootstrap loader was activated during reset, pin TxD0 (alternate port
function) will be switched to output mode after the reception of the zero byte.
All other pins remain in the high-impedance state until they are changed by software or
peripheral operation.
User’s Manual
19-7
V3.1, 2000-03
C167CR
Derivatives
System Reset
Reset Output Pin
The RSTOUT pin is dedicated to generate a reset signal for the system components
besides the controller itself. RSTOUT will be driven active (low) at the begin of any reset
sequence (triggered by hardware, the SRST instruction or a watchdog timer overflow).
RSTOUT stays active (low) beyond the end of the internal reset sequence until the
protected EINIT (End of Initialization) instruction is executed (see Figure 19-3). This
allows the complete configuration of the controller including its on-chip peripheral units
before releasing the reset signal for the external peripherals of the system.
Note: RSTOUT will float during emulation mode or adapt mode.
The Internal RAM after Reset
The contents of the internal RAM are not affected by a system reset. However, after a
power-on reset, the contents of the internal RAM are undefined. This implies that the
GPRs (R15 … R0) and the PEC source and destination pointers (SRCP7 … SRCP0,
DSTP7 … DSTP0) which are mapped into the internal RAM are also unchanged after a
warm reset, software reset or watchdog reset, but are undefined after a power-on reset.
The Extension RAM (XRAM) after Reset
The contents of the on-chip extension RAM are not affected by a system reset. However,
after a power-on reset, the contents of the XRAM are undefined.
Operation after Reset
After the internal reset condition is removed the C167CR fetches the first instruction from
the program memory (location 00’0000 for a standard start). As a rule, this first location
H
holds a branch instruction to the actual initialization routine that may be located
anywhere in the address space.
Note: When the Bootstrap Loader Mode was activated during a hardware reset the
C167CR does not fetch instructions from the program memory.
The standard bootstrap loader expects data via serial interface ASC0.
User’s Manual
19-8
V3.1, 2000-03
C167CR
Derivatives
System Reset
19.3
Application-specific Initialization Routine
After a reset the modules of the C167CR must be initialized to enable their operation on
a given application. This initialization depends on the task the C167CR is to fulfill in that
application and on some system properties like operating frequency, connected external
circuitry, etc.
The following initializations should typically be done, before the C167CR is prepared to
run the actual application software:
Memory Areas
The external bus interface can be reconfigured after an external reset because register
BUSCON0 is initialized to the slowest possible bus cycle configuration. The
programmable address windows can be enabled in order to adapt the bus cycle
characteristics to different memory areas or peripherals. Also after a single-chip mode
reset the external bus interface can be enabled and configured.
The internal program memory (if available) can be enabled and mapped after an
external reset in order to use the on-chip resources. After a single-chip mode reset the
internal program memory can be remapped or disabled at all in order to utilize external
memory (partly or completely).
Programmable program memory can be programmed, e.g. with data received over a
serial link.
Note: Initial Flash or OTP programming will rather be done in bootstrap loader mode.
System Stack
The default setup for the system stack (size, stackpointer, upper and lower limit
registers) can be adjusted to application-specific values. After reset, registers SP and
STKUN contain the same reset value 00’FC00 , while register STKOV contains
H
00’FA00 . With the default reset initialization, 256 words of system stack are available,
H
where the system stack selected by the SP grows downwards from 00’FBFE .
H
Note: The interrupt system, which is disabled upon completion of the internal reset,
should remain disabled until the SP is initialized.
Traps (incl. NMI) may occur, even though the interrupt system is still disabled.
Register Bank
The location of a register bank is defined by the context pointer (CP) and can be adjusted
to an application-specific bank before the general purpose registers (GPRs) are used.
After reset, register CP contains the value 00’FC00 , i.e. the register bank selected by
H
the CP grows upwards from 00’FC00 .
H
User’s Manual
19-9
V3.1, 2000-03
C167CR
Derivatives
System Reset
On-chip RAM
Based on the application, the user may wish to initialize portions of the internal writable
memory (IRAM/XRAM) before normal program operation. Once the register bank has
been selected by programming the CP register, the desired portions of the internal
memory can easily be initialized via indirect addressing.
Interrupt System
After reset the individual interrupt nodes and the global interrupt system are disabled. In
order to enable interrupt requests the nodes must be assigned to their respective
interrupt priority levels and be enabled. The vector locations must receive pointers to the
respective exception handlers. The interrupt system must globally be enabled by setting
bit IEN in register PSW. Care must be taken not to enable the interrupt system before
the initialization is complete in order to avoid e.g. the corruption of internal memory
locations by stack operations using an uninitialized stack pointer.
Watchdog Timer
After reset the watchdog timer is active and is counting its default period. If the watchdog
timer shall remain active the desired period should be programmed by selecting the
appropriate prescaler value and reload value. Otherwise the watchdog timer must be
disabled before EINIT.
Ports
Generally all ports of the C167CR are switched to input after reset. Some pins may be
automatically controlled, e.g. bus interface pins for an external start, TxD in Boot mode,
etc. Pins that shall be used for general purpose IO must be initialized via software. The
required mode (input/output, open drain/push pull, input threshold, etc.) depends on the
intended function for a given pin.
Peripherals
After reset the C167CR’s on-chip peripheral modules enter a defined default state (see
respective peripheral description) where it is disabled from operation. In order to use a
certain peripheral it must be initialized according to its intended operation in the application.
This includes selecting the operating mode (e.g. counter/timer), operating parameters
(e.g. baudrate), enabling interface pins (if required), assigning interrupt nodes to the
respective priority levels, etc.
After these standard initialization also application-specific actions may be required like
asserting certain levels to output pins, sending codes via interfaces, latching input levels, etc.
User’s Manual
19-10
V3.1, 2000-03
C167CR
Derivatives
System Reset
Termination of Initialization
The software initialization routine should be terminated with the EINIT instruction. This
instruction has been implemented as a protected instruction.
The execution of the EINIT instruction:
• disables the action of the DISWDT instruction,
• disables write accesses to reg. SYSCON (all configurations regarding reg. SYSCON
(enable CLKOUT, stacksize, etc.) must be selected before the execution of EINIT),
• clears the reset source detection bits in register WDTCON,
• causes the RSTOUT pin to go high
(this signal can be used to indicate the end of the initialization routine and the proper
operation of the microcontroller to external hardware).
User’s Manual
19-11
V3.1, 2000-03
C167CR
Derivatives
System Reset
19.4
System Startup Configuration
Although most of the programmable features of the C167CR are selected by software
either during the initialization phase or repeatedly during program execution, there are
some features that must be selected earlier, because they are used for the first access
of the program execution (e.g. internal or external start selected via EA).
These configurations are accomplished by latching the logic levels at a number of pins
at the end of the internal reset sequence. During reset internal pullup/pulldown devices
are active on those lines. They ensure inactive/default levels at pins which are not driven
externally. External pulldown/pullup devices may override the default levels in order to
select a specific configuration. Many configurations can therefore be coded with a
minimum of external circuitry.
Note: The load on those pins that shall be latched for configuration must be small
enough for the internal pullup/pulldown device to sustain the default level, or
external pullup/pulldown devices must ensure this level.
Those pins whose default level shall be overridden must be pulled low/high
externally.
Make sure that the valid target levels are reached until the end of the reset
sequence.
There is a specific application note to illustrate this.
User’s Manual
19-12
V3.1, 2000-03
C167CR
Derivatives
System Reset
19.4.1
System Startup Configuration upon an External Reset
For an external reset (EA = ‘0’) the startup configuration uses the pins of PORT0. The
value on the upper byte of PORT0 (P0H) is latched into register RP0H upon reset, the
value on the lower byte (P0L) directly influences the BUSCON0 register (bus mode) or
the internal control logic of the C167CR.
H.7 H.6 H.5 H.4 H.3 H.2 H.1 H.0 L.7 L.6 L.5 L.4 L.3 L.2 L.1 L.0
CLKCFG
CALSEL
CSSEL WRC BUSTYP
SMOD
ADP EMU
Internal Control Logic
(only on Hardware Reset)
Clock
Generator
Port 4
Logic
Port 6
Logic
SYSCON
BUSCON0
MCA04405
Figure 19-4 PORT0 Configuration during Reset
The pins that control the operation of the internal control logic, the clock configuration,
and the reserved pins are evaluated only during a hardware triggered reset sequence.
The pins that influence the configuration of the C167CR are evaluated during any reset
sequence, i.e. also during software and watchdog timer triggered resets.
The configuration via P0H is latched in register RP0H for subsequent evaluation by
software. Register RP0H is described in Chapter 9.
The following describes the different selections that are offered for reset configuration.
The default modes refer to pins at high level, i.e. without external pulldown devices
connected.
Please also consider the note above.
User’s Manual
19-13
V3.1, 2000-03
C167CR
Derivatives
System Reset
Emulation Mode
Pin P0L.0 (EMU) selects the Emulation Mode, when latched low at the end of reset. This
mode is used for special emulation and testing purposes and is of minor use for standard
C167CR applications, so P0L.0 should be held high.
Emulation mode provides access to integrated XBUS peripherals via the external bus
interface pins (direction reversed) of the C167CR. The CPU and the generic peripherals
are disabled, all modules connected via the XBUS are active.
Table 19-1 Emulation Mode Summary
Pin(s)
Function
Notes
Port 4,
PORT1
Address input
The segment address lines configured at reset
must be driven externally
PORT0
RD, WR
ALE
Data input/output
Control signal input
Unused input
Hold LOW
CLKOUT
RSTOUT
RSTIN
Port 6
CPU clock output
Reset input
Enabled automatically
Drive externally for an XBUS peripheral reset
Standard reset for complete device
Reset input
Interrupt output
Sends XBUS peripheral interrupt request e.g. to
the emulation system
Default: Emulation Mode is off.
Note: In emulation mode pin P0.15 (P0H.7) is inverted, i.e. the configuration ‘111’ would
select direct drive in emulation mode.
User’s Manual
19-14
V3.1, 2000-03
C167CR
Derivatives
System Reset
Adapt Mode
Pin P0L.1 (ADP) selects the Adapt Mode, when latched low at the end of reset. In this
mode the C167CR goes into a passive state, which is similar to its state during reset.
The pins of the C167CR float to tristate or are deactivated via internal pullup/pulldown
devices, as described for the reset state. In addition also the RSTOUT pin floats to
tristate rather than be driven low. The on-chip oscillator is disabled.
This mode allows switching a C167CR that is mounted to a board virtually off, so an
emulator may control the board’s circuitry, even though the original C167CR remains in
its place. The original C167CR also may resume to control the board after a reset
sequence with P0L.1 high. Please note that adapt mode overrides any other
configuration via PORT0.
Default: Adapt Mode is off.
Note: When XTAL1 is fed by an external clock generator (while XTAL2 is left open), this
clock signal may also be used to drive the emulator device.
However, if a crystal is used, the emulator device’s oscillator can use this crystal
only, if at least XTAL2 of the original device is disconnected from the circuitry (the
output XTAL2 will be driven high in Adapt Mode).
User’s Manual
19-15
V3.1, 2000-03
C167CR
Derivatives
System Reset
Special Operation Modes
Pins P0L.5 to P0L.2 (SMOD) select special operation modes of the C167CR during reset
(see Table 19-2). Make sure to only select valid configurations in order to ensure proper
operation of the C167CR.
Table 19-2 Definition of Special Modes for Reset Configuration
P0.5-2 (P0L.5-2) Special Mode
Notes
1 1 1 1
Normal Start
Default configuration.
Begin of execution as defined via pin EA.
1 1 1 0
1 1 0 1
1 1 0 0
1 0 1 1
Reserved
Reserved
Reserved
Do not select this configuration!
Do not select this configuration!
Do not select this configuration!
Standard Bootstrap Load an initial boot routine of 32 Bytes via
Loader
interface ASC0.
1 0 1 0
1 0 0 1
1 0 0 0
0 1 1 1
0 1 1 0
0 1 0 1
0 1 0 0
0 0 X X
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Do not select this configuration!
Do not select this configuration!
Do not select this configuration!
Do not select this configuration!
Do not select this configuration!
Do not select this configuration!
Do not select this configuration!
Do not select this configuration!
The On-chip Bootstrap Loader allows moving the start code into the internal RAM of
the C167CR via the serial interface ASC0. The C167CR will remain in bootstrap loader
mode until a hardware reset not selecting BSL mode or a software reset.
Default: The C167CR starts fetching code from location 00’0000 , the bootstrap loader
H
is off.
User’s Manual
19-16
V3.1, 2000-03
C167CR
Derivatives
System Reset
External Bus Type
Pins P0L.7 and P0L.6 (BUSTYP) select the external bus type during reset, if an external
start is selected via pin EA. This allows the configuration of the external bus interface of
the C167CR even for the first code fetch after reset. The two bits are copied into bit field
BTYP of register BUSCON0. P0L.7 controls the data bus width, while P0L.6 controls the
address output (multiplexed or demultiplexed). This bit field may be changed via
software after reset, if required.
Table 19-3 Configuration of External Bus Type
P0L.7-6 (BTYP)
Encoding
External Data Bus Width
External Address Bus Mode
0 0
0 1
1 0
1 1
8-bit Data
8-bit Data
16-bit Data
16-bit Data
Demultiplexed Addresses
Multiplexed Addresses
Demultiplexed Addresses
Multiplexed Addresses
PORT0 and PORT1 are automatically switched to the selected bus mode. In multiplexed
bus modes PORT0 drives both the 16-bit intra-segment address and the output data,
while PORT1 remains in high impedance state as long as no demultiplexed bus is
selected via one of the BUSCON registers. In demultiplexed bus modes PORT1 drives
the 16-bit intra-segment address, while PORT0 or P0L (according to the selected data
bus width) drives the output data.
For a 16-bit data bus BHE is automatically enabled, for an 8-bit data bus BHE is disabled
via bit BYTDIS in register SYSCON.
Default: 16-bit data bus with multiplexed addresses.
Note: If an internal start is selected via pin EA, these two pins are disregarded and bit
field BTYP of register BUSCON0 is cleared.
Write Configuration
Pin P0H.0 (WRC) selects the initial operation of the control pins WR and BHE during
reset. When high, this pin selects the standard function, i.e. WR control and BHE. When
low, it selects the alternate configuration, i.e. WRH and WRL. Thus even the first access
after a reset can go to a memory controlled via WRH and WRL. This bit is latched in
register RP0H and its inverted value is copied into bit WRCFG in register SYSCON.
Default: Standard function (WR control and BHE).
User’s Manual
19-17
V3.1, 2000-03
C167CR
Derivatives
System Reset
Chip Select Lines
Pins P0H.2 and P0H.1 (CSSEL) define the number of active chip select signals during
reset. This allows the selection which pins of Port 6 drive external CS signals and which
are used for general purpose IO. The two bits are latched in register RP0H.
Table 19-4 Configuration of Chip Select Lines
P0H.2-1 (CSSEL) Chip Select Lines
Note
1 1
1 0
0 1
0 0
Five: CS4 … CS0
None
Default without pull-downs
Two: CS1 … CS0
Three:CS2 … CS0
Default: All 5 chip select lines active (CS4 … CS0).
Note: The selected number of CS signals cannot be changed via software after reset.
Segment Address Lines
Pins P0H.4 and P0H.3 (SALSEL) define the number of active segment address lines
during reset. This allows the selection which pins of Port 4 drive address lines and which
are used for general purpose IO. The two bits are latched in register RP0H. Depending
on the system architecture the required address space is chosen and accessible right
from the start, so the initialization routine can directly access all locations without prior
programming. The required pins of Port 4 are automatically switched to address output
mode.
Table 19-5 Configuration of Segment Address Lines
P0H.4-3 (SALSEL) Segment Address Lines Directly accessible A. Space
1 1
Two: A17 … A16
256
KByte (Default without pull-
downs)
1 0
0 1
0 0
Eight: A23 … A16
None
16
64
1
MByte (Maximum)
KByte (Minimum)
MByte
Four: A19 … A16
Even if not all segment address lines are enabled on Port 4, the C167CR internally uses
its complete 24-bit addressing mechanism. This allows the restriction of the width of the
effective address bus, while still deriving CS signals from the complete addresses.
Default: 2-bit segment address (A17 … A16) allowing access to 256 KByte.
Note: The selected number of segment address lines cannot be changed via software
after reset.
User’s Manual
19-18
V3.1, 2000-03
C167CR
Derivatives
System Reset
Clock Generation Control
Pins P0H.7, P0H.6 and P0H.5 (CLKCFG) select the basic clock generation mode during
reset. The oscillator clock either directly feeds the CPU and peripherals (direct drive), it
is divided by 2 or it is fed to the on-chip PLL which then provides the CPU clock signal
(selectable multiple of the oscillator frequency, i.e. the input frequency). These bits are
latched in register RP0H.
Table 19-6 C167CR Clock Generation Modes
(P0H.7-5)
CPU Frequency External Clock
Notes
1)
(CLKCFG) fCPU = fOSC × F Input Range
1 1 1
1 1 0
1 0 1
1 0 0
0 1 1
0 1 0
0 0 1
fOSC × 4
fOSC × 3
fOSC × 2
fOSC × 5
fOSC × 1
fOSC × 1.5
fOSC / 2
2.5 to 8.25 MHz
3.33 to 11 MHz
5 to 16.5 MHz
2 to 6.6 MHz
1 to 33 MHz
Default configuration
2)
Direct drive
6.66 to 22 MHz
2 to 66 MHz
CPU clock via prescaler
0 0 0
fOSC × 2.5
4 to 13.2 MHz
1)
The external clock input range refers to a CPU clock range of 10 … 33 MHz.
2)
The maximum frequency depends on the duty cycle of the external clock signal.
In emulation mode pin P0.15 (P0H.7) is inverted, i.e. the configuration ‘111’ would select direct drive in
emulation mode.
Default: On-chip PLL is active with a factor of 1:4.
Watch the different requirements for frequency and duty cycle of the oscillator input clock
for the possible selections.
User’s Manual
19-19
V3.1, 2000-03
C167CR
Derivatives
System Reset
19.4.2
System Startup Configuration upon a Single-chip Mode Reset
For a single-chip mode reset (indicated by EA = ‘1’) the configuration via PORT0 is only
partly evaluated.
No bus type is stored into BUSCON0. The external bus interface remains inactive
(PORT0, PORT1, Port 4, Port 6).
The first instruction is fetched from internal program memory.
Note: Single-chip mode reset cannot be selected on ROMless devices. The attempt to
read the first instruction after reset will fail in such a case.
User’s Manual
19-20
V3.1, 2000-03
C167CR
Derivatives
Power Management
20
Power Management
For an increasing number of microcontroller based systems it is an important objective
to reduce the power consumption of the system as much as possible. A contradictory
objective is, however, to reach a certain level of system performance. Besides
optimization of design and technology a microcontroller’s power consumption can
generally be reduced by lowering its operating frequency and/or by reducing the circuitry
that is clocked. The architecture of the C167CR provides two operating modes to reduce
its power consumption (see Figure 20-1) under software control:
• In Idle mode the CPU is stopped, while the peripherals continue their operation. Idle
mode can be terminated by any reset or interrupt request
• In Power Down mode both the CPU and the peripherals are stopped. Power Down
mode can only be terminated by a hardware reset.
Note: All external bus actions are completed before Idle or Power Down mode is
entered. However, Idle or Power Down mode is not entered if READY is enabled,
but has not been activated (driven low) during the last bus access.
Intermittent operation (i.e. alternating phases of high performance and power saving) is
supported by cyclic interrupt generation of a timer.
Power
Active
Idle
Power Down
fCPU
MCA04406
Figure 20-1 Power Reduction Possibilities
User’s Manual
20-1
V3.1, 2000-03
C167CR
Derivatives
Power Management
20.1
Idle Mode
The power consumption of the C167CR microcontroller can be decreased by entering
Idle mode. In this mode all enabled peripherals, including the watchdog timer, continue
to operate normally, only the CPU operation is halted and the on-chip memory modules
are disabled.
Idle mode is entered after the IDLE instruction has been executed and the instruction
before the IDLE instruction has been completed. To prevent unintentional entry into Idle
mode, the IDLE instruction has been implemented as a protected 32-bit instruction.
Idle mode is terminated by interrupt requests from any enabled interrupt source whose
individual Interrupt Enable flag was set before the Idle mode was entered, regardless of
bit IEN.
For a request selected for CPU interrupt service the associated interrupt service routine
is entered if the priority level of the requesting source is higher than the current CPU
priority and the interrupt system is globally enabled. After the RETI (Return from
Interrupt) instruction of the interrupt service routine is executed the CPU continues
executing the program with the instruction following the IDLE instruction. Otherwise, if
the interrupt request cannot be serviced because of a too low priority or a globally
disabled interrupt system the CPU immediately resumes normal program execution with
the instruction following the IDLE instruction.
For a request which was programmed for PEC service a PEC data transfer is performed
if the priority level of this request is higher than the current CPU priority and the interrupt
system is globally enabled. After the PEC data transfer has been completed the CPU
remains in Idle mode. Otherwise, if the PEC request cannot be serviced because of a
too low priority or a globally disabled interrupt system the CPU does not remain in Idle
mode but continues program execution with the instruction following the IDLE
instruction.
Denied
CPU Interrupt Request
Accepted
Active
Mode
Idle
Mode
IDLE Instruction
Denied PEC Request
Executed
PEC Request
MCA04407
Figure 20-2 Transitions between Idle Mode and Active Mode
User’s Manual
20-2
V3.1, 2000-03
C167CR
Derivatives
Power Management
Idle mode can also be terminated by a Non-Maskable Interrupt, i.e. a high to low
transition on the NMI pin. After Idle mode has been terminated by an interrupt or NMI
request, the interrupt system performs a round of prioritization to determine the highest
priority request. In the case of an NMI request, the NMI trap will always be entered.
Any interrupt request whose individual Interrupt Enable flag was set before Idle mode
was entered will terminate Idle mode regardless of the current CPU priority. The CPU
will not go back into Idle mode when a CPU interrupt request is detected, even when the
interrupt was not serviced because of a higher CPU priority or a globally disabled
interrupt system (IEN = ‘0’). The CPU will only go back into Idle mode when the interrupt
system is globally enabled (IEN = ‘1’) and a PEC service on a priority level higher than
the current CPU level is requested and executed.
Note: An interrupt request which is individually enabled and assigned to priority level 0
will terminate Idle mode. The associated interrupt vector will not be accessed,
however.
The watchdog timer may be used to monitor the Idle mode: an internal reset will be
generated if no interrupt or NMI request occurs before the watchdog timer overflows. To
prevent the watchdog timer from overflowing during Idle mode it must be programmed
to a reasonable time interval before Idle mode is entered.
User’s Manual
20-3
V3.1, 2000-03
C167CR
Derivatives
Power Management
20.2
Power Down Mode
The microcontroller can be switched to Power Down mode which reduces the power
consumption to a minimum. Clocking of all internal blocks is stopped, the contents of the
internal RAM, however, are preserved through the voltage supplied via the VDD pins.
The watchdog timer is stopped in Power Down mode. This mode can only be terminated
by an external hardware reset, i.e. by asserting a low level on the RSTIN pin. This reset
will initialize all SFRs and ports to their default state, but will not change the contents of
the internal RAM.
There are two levels of protection against unintentionally entering Power Down mode.
First, the PWRDN (Power Down) instruction which is used to enter this mode has been
implemented as a protected 32-bit instruction. Second, this instruction is effective only
if the NMI (Non Maskable Interrupt) pin is externally pulled low while the PWRDN
instruction is executed. The microcontroller will enter Power Down mode after the
PWRDN instruction has completed.
This feature can be used in conjunction with an external power failure signal which pulls
the NMI pin low when a power failure is imminent. The microcontroller will enter the NMI
trap routine which can save the internal state into RAM. After the internal state has been
saved, the trap routine may then execute the PWRDN instruction. If the NMI pin is still
low at this time, Power Down mode will be entered, otherwise program execution
continues.
The initialization routine (executed upon reset) can check the reset identification flags in
register WDTCON to determine whether the controller was initially switched on, or
whether it was properly restarted from Power Down mode.
During power down the voltage at the VDD pins can be lowered to 2.5 V while the
contents of the internal RAM will still be preserved.
The total power consumption in Power Down mode depends on the current that flows
through the port drivers. To minimize the consumed current the pin drivers can be
disabled (pins switched to tristate) simply by configuring them for input.
The bus interface pins can be separately disabled by releasing the external bus (disable
all address windows by clearing the BUSACT bits) and switching the ports to input (if
necessary). Of course the required software in this case must be executed from internal
memory.
User’s Manual
20-4
V3.1, 2000-03
C167CR
Derivatives
Power Management
20.2.1
Status of Output Pins During Power Reduction Modes
During Idle mode the CPU clocks are turned off, while all peripherals continue their
operation in the normal way. Therefore all ports pins, which are configured as general
purpose output pins, output the last data value which was written to their port output
latches. If the alternate output function of a port pin is used by a peripheral, the state of
the pin is determined by the operation of the peripheral.
Port pins which are used for bus control functions go into that state which represents the
inactive state of the respective function (e.g. WR), or to a defined state which is based
on the last bus access (e.g. BHE). Port pins which are used as external address/data
bus hold the address/data which was output during the last external memory access
before entry into Idle mode under the following conditions:
P0H outputs the high byte of the last address if a multiplexed bus mode with 8-bit data
bus is used, otherwise P0H is floating. P0L is always floating in Idle mode.
PORT1 outputs the lower 16 bits of the last address if a demultiplexed bus mode is used,
otherwise the output pins of PORT1 represent the port latch data.
Port 4 outputs the segment address for the last access on those pins that were selected
during reset, otherwise the output pins of Port 4 represent the port latch data.
During power down mode the oscillator and the clocks to the CPU and to the
peripherals are turned off. Like in Idle mode, all port pins which are configured as general
purpose output pins output the last data value which was written to their port output
latches.
When the alternate output function of a port pin is used by a peripheral the state of this
pin is determined by the last action of the peripheral before the clocks were switched off.
Note: When the supply voltage is lowered in Power Down mode the high voltage of
output pins will decrease accordingly.
User’s Manual
20-5
V3.1, 2000-03
C167CR
Derivatives
Power Management
Table 20-1 State of C167CR Output Pins during Idle and Power Down mode.
C167CR
Output Pin(s)
External Bus Enabled
Idle Mode Power Down
Active (toggling) High
No External Bus
Idle Mode Power Down
CLKOUT
ALE
Active (toggling) High
Low
Low
RD, WR
P0L
High
High
Floating
Port Latch Data
Port Latch Data
Port Latch Data
Port Latch Data
Port Latch Data
Port Latch Data
1)
P0H
A15 … A8 /Float
2)
PORT1
Port 4
BHE
Last Address /Port Latch Data
Port Latch Data/Last segment
Last value
3)
CSx
Last value
RSTOUT
High if EINIT was executed before entering Idle or Power Down mode,
Low otherwise.
Other Port
Port Latch Data/Alternate Function
Output Pins
1)
For multiplexed buses with 8-bit data bus.
For demultiplexed buses.
2)
3)
The CS signal that corresponds to the last address remains active (low), all other enabled CS signals remain
inactive (high). By accessing an on-chip X-Periperal prior to entering a power save mode all external CS
signals can be deactivated.
User’s Manual
20-6
V3.1, 2000-03
C167CR
Derivatives
System Programming
21
System Programming
To aid in software development, a number of features has been incorporated into the
instruction set of the C167CR, including constructs for modularity, loops, and context
switching. In many cases commonly used instruction sequences have been simplified
while providing greater flexibility. The following programming features help to fully utilize
this instruction set.
Instructions Provided as Subsets of Instructions
In many cases, instructions found in other microcontrollers are provided as subsets of
more powerful instructions in the C167CR. This allows the same functionality to be
provided while decreasing the hardware required and decreasing decode complexity. In
order to aid assembly programming, these instructions, familiar from other
microcontrollers, can be built in macros, thus providing the same names.
Directly Substitutable Instructions are instructions known from other microcontrollers
that can be replaced by the following instructions of the C167CR:
Table 21-1 Substitution of Instructions
Substituted Instruction
C167CR Instruction
Function
CLR
Rn
Bit
Rn
Rn
Rn
AND
Rn, #0
Bit, Bit
Rn, #1
Rn, #1
Rn, #8
Clear register
H
CPLB
DEC
BMOVN
SUB
Complement bit
Decrement register
Increment register
Swap bytes within word
H
H
H
INC
ADD
SWAPB
ROR
Modification of System Flags is performed using bit set or bit clear instructions
(BSET, BCLR). All bit and word instructions can access the PSW register, so no
instructions like CLEAR CARRY or ENABLE INTERRUPTS are required.
External Memory Data Access does not require special instructions to load data
pointers or explicitly load and store external data. The C167CR provides a Von
Neumann memory architecture and its on-chip hardware automatically detects accesses
to internal RAM, GPRs, and SFRs.
Multiplication and Division
Multiplication and division of words and double words is provided through multiple cycle
instructions implementing a Booth algorithm. Each instruction implicitly uses the 32-bit
register MD (MDL = lower 16 bits, MDH = upper 16 bits). The MDRIU flag (Multiply or
Divide Register In Use) in register MDC is set whenever either half of this register is
written to or when a multiply/divide instruction is started. It is cleared whenever the MDL
User’s Manual
21-1
V3.1, 2000-03
C167CR
Derivatives
System Programming
register is read. Because an interrupt can be acknowledged before the contents of
register MD are saved, this flag is required to alert interrupt routines, which require the
use of the multiply/divide hardware, so they can preserve register MD. This register,
however, only needs to be saved when an interrupt routine requires use of the MD
register and a previous task has not saved the current result. This flag is easily tested by
the Jump-on-bit instructions.
Multiplication or division is simply performed by specifying the correct (signed or
unsigned) version of the multiply or divide instruction. The result is then stored in register
MD. The overflow flag (V) is set if the result from a multiply or divide instruction is greater
than 16 bits. This flag can be used to determine whether both word halfs must be
transferred from register MD. The high portion of register MD (MDH) must be moved into
the register file or memory first, in order to ensure that the MDRIU flag reflects the correct
state.
The following instruction sequence performs an unsigned 16 by 16-bit multiplication:
SAVE:
JNB
SCXT
MDRIU, START;Test if MD was in use.
MDC, #0010H ;Save and clear control register,
;leaving MDRIU set
;(only required for interrupted
;multiply/divide instructions)
BSET
PUSH
PUSH
START:
MULU
JMPR
MOV
SAVED
MDH
MDL
;Indicate the save operation
;Save previous MD contents …
;… on system stack
R1, R2
;Multiply 16·16 unsigned, Sets MDRIU
cc_NV, COPYL;Test for only 16-bit result
R3, MDH
R4, MDL
;Move high portion of MD
COPYL:
MOV
;Move low portion of MD, Clears MDRIU
RESTORE:
JNB
SAVED, DONE ;Test if MD registers were saved
POP
MDL
;Restore registers
POP
MDH
POP
MDC
BCLR
SAVED
;Multiplication is completed,
;program continues
DONE:
…
User’s Manual
21-2
V3.1, 2000-03
C167CR
Derivatives
System Programming
The above save sequence and the restore sequence after COPYL are only required if
the current routine could have interrupted a previous routine which contained a MUL or
DIV instruction. Register MDC is also saved because it is possible that a previous
routine’s Multiply or Divide instruction was interrupted while in progress. In this case the
information about how to restart the instruction is contained in this register. Register
MDC must be cleared to be correctly initialized for a subsequent multiplication or
division. The old MDC contents must be popped from the stack before the RETI
instruction is executed.
For a division the user must first move the dividend into the MD register. If a 16/16-bit
division is specified, only the low portion of register MD must be loaded. The result is also
stored into register MD. The low portion (MDL) contains the integer result of the division,
while the high portion (MDH) contains the remainder.
The following instruction sequence performs a 32 by 16-bit division:
MOV
MOV
DIV
JMPR
MOV
MOV
MDH, R1
MDL, R2
R3
;Move dividend to MD register. Sets MDRIU
;Move low portion to MD
;Divide 32/16 signed, R3 holds divisor
cc_V, ERROR ;Test for divide overflow
R3, MDH
R4, MDL
;Move remainder to R3
;Move integer result to R4. Clears MDRIU
Whenever a multiply or divide instruction is interrupted while in progress, the address of
the interrupted instruction is pushed onto the stack and the MULIP flag in the PSW of the
interrupting routine is set. When the interrupt routine is exited with the RETI instruction,
this bit is implicitly tested before the old PSW is popped from the stack. If MULIP = ‘1’
the multiply/divide instruction is re-read from the location popped from the stack (return
address) and will be completed after the RETI instruction has been executed.
Note: The MULIP flag is part of the context of the interrupted task. When the
interrupting routine does not return to the interrupted task (e.g. scheduler switches
to another task) the MULIP flag must be set or cleared according to the context of
the task that is switched to.
BCD Calculations
No direct support for BCD calculations is provided in the C167CR. BCD calculations are
performed by converting BCD data to binary data, performing the desired calculations
using standard data types, and converting the result back to BCD data. Due to the
enhanced performance of division instructions binary data is quickly converted to BCD
data through division by 10 . Conversion from BCD data to binary data is enhanced by
D
multiple bit shift instructions. This provides similar performance compared to instructions
directly supporting BCD data types, while no additional hardware is required.
User’s Manual
21-3
V3.1, 2000-03
C167CR
Derivatives
System Programming
21.1
Stack Operations
The C167CR supports two types of stacks. The system stack is used implicitly by the
controller and is located in the internal RAM. The user stack provides stack access to the
user in either the internal or external memory. Both stack types grow from high memory
addresses to low memory addresses.
Internal System Stack
A system stack is provided to store return vectors, segment pointers, and processor
status for procedures and interrupt routines. A system register, SP, points to the top of
the stack. This pointer is decremented when data is pushed onto the stack, and
incremented when data is popped.
The internal system stack can also be used to temporarily store data or pass it between
subroutines or tasks. Instructions are provided to push or pop registers on/from the
system stack. However, in most cases the register banking scheme provides the best
performance for passing data between multiple tasks.
Note: The system stack allows the storage of words only. Bytes must either be
converted to words or the respective other byte must be disregarded.
Register SP can only be loaded with even byte addresses (The LSB of SP is
always ‘0’).
Detection of stack overflow/underflow is supported by two registers, STKOV (Stack
Overflow Pointer) and STKUN (Stack Underflow Pointer). Specific system traps (Stack
Overflow trap, Stack Underflow trap) will be entered whenever the SP reaches either
boundary specified in these registers.
The contents of the stack pointer are compared to the contents of the overflow register,
whenever the SP is DECREMENTED either by a CALL, PUSH or SUB instruction. An
overflow trap will be entered, when the SP value is less than the value in the stack
overflow register.
The contents of the stack pointer are compared to the contents of the underflow register,
whenever the SP is INCREMENTED either by a RET, POP or ADD instruction. An
underflow trap will be entered, when the SP value is greater than the value in the stack
underflow register.
Note: When a value is MOVED into the stack pointer, NO check against the overflow/
underflow registers is performed.
In many cases the user will place a software reset instruction (SRST) into the stack
underflow and overflow trap service routines. This is an easy approach, which does not
require special programming. However, this approach assumes that the defined internal
stack is sufficient for the current software and that exceeding its upper or lower boundary
represents a fatal error.
User’s Manual
21-4
V3.1, 2000-03
C167CR
Derivatives
System Programming
It is also possible to use the stack underflow and stack overflow traps to cache portions
of a larger external stack. Only the portion of the system stack currently being used is
placed into the internal memory, thus allowing a greater portion of the internal RAM to
be used for program, data or register banking. This approach assumes no error but
requires a set of control routines (see below).
Circular (Virtual) Stack
This basic technique allows pushing until the overflow boundary of the internal stack is
reached. At this point a portion of the stacked data must be saved into external memory
to create space for further stack pushes. This is called “stack flushing”. When executing
a number of return or pop instructions, the upper boundary (since the stack empties
upward to higher memory locations) is reached. The entries that have been previously
saved in external memory must now be restored. This is called “stack filling”. Because
procedure call instructions do not continue to nest infinitely and call and return
instructions alternate, flushing and filling normally occurs very infrequently. If this is not
true for a given program environment, this technique should not be used because of the
overhead of flushing and filling.
The basic mechanism is the transformation of the addresses of a virtual stack area,
controlled via registers SP, STKOV and STKUN, to a defined physical stack area within
the internal RAM via hardware. This virtual stack area covers all possible locations that
SP can point to, i.e. 00’F000 through 00’FFFE . STKOV and STKUN accept the same
H
H
4 KByte address range.
The size of the physical stack area within the internal RAM that effectively is used for
standard stack operations is defined via bitfield STKSZ in register SYSCON (see below).
Table 21-2 Circular Stack Address Transformation
STKSZ Stack Size Internal RAM Addresses (Words)
Significant Bits
of Stack Ptr. SP
(Words)
256
128
64
of Physical Stack
0 0 0
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1
00’FBFE … 00’FA00 (Default after Reset)
SP.8 … SP.0
SP.7 … SP.0
SP.6 … SP.0
SP.5 … SP.0
B
B
B
B
B
B
H
H
H
H
00’FBFE … 00’FB00
H
00’FBFE … 00’FB80
H
32
00’FBFE … 00’FBC0
H H
512
–
00’FBFE … 00’F800 (not for 1KByte IRAM) SP.9 … SP.0
H H
Reserved. Do not use this combination.
Reserved. Do not use this combination.
–
–
–
B
1024
00’FDFE … 00’FX00 (Note: No circular stack) SP.11 … SP.0
H H
B
00’FX00 represents the lower IRAM limit, i.e.
H
1 KB: 00’FA00 , 2 KB: 00’F600 ,
H
H
H
3 KB: 00’F200
User’s Manual
21-5
V3.1, 2000-03
C167CR
Derivatives
System Programming
The virtual stack addresses are transformed to physical stack addresses by
concatenating the significant bits of the stack pointer register SP (see Table 21-2) with
the complementary most significant bits of the upper limit of the physical stack area
(00’FBFE ). This transformation is done via hardware (see Figure 21-1).
H
The reset values (STKOV = FA00 , STKUN = FC00 , SP = FC00 , STKSZ = 000 )
H
H
H
B
map the virtual stack area directly to the physical stack area and allow using the internal
system stack without any changes, provided that the 256 word area is not exceeded.
1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 0 1 1 1 0 0 0 0 0 0 0
1 1 1 1 1 0 1 1 1 0 0 0 0 0 0 0
1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 0 1 0 0 0 0 0 0 0 0 0
1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0
FBFE
FB80
FB80
FBFE
FA00
F800
H
H
H
H
H
H
Phys.A.
<SP>
After PUSH
After PUSH
1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
FBFE
FBFE
FB7E
FBFE
FBFE
F7FE
H
H
H
H
H
H
Phys.A.
<SP>
1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 0 1 1 0 1 1 1 1 1 1 0
64 words
Stack Size
256 words
MCA04408
Figure 21-1 Physical Stack Address Generation
The following example demonstrates the circular stack mechanism which is also an
effect of this virtual stack mapping: First, register R1 is pushed onto the lowest physical
stack location according to the selected maximum stack size. With the following
instruction, register R2 will be pushed onto the highest physical stack location although
the SP is decremented by 2 as for the previous push operation.
MOV
SP, #0F802H ;Set SP before last entry…
;… of physical stack of 256 words
…
;(SP)= F802H: Physical stack addr.= FA02H
PUSH
PUSH
R1
R2
;(SP)= F800H: Physical stack addr.= FA00H
;(SP)= F7FEH: Physical stack addr.= FBFEH
User’s Manual
21-6
V3.1, 2000-03
C167CR
Derivatives
System Programming
The effect of the address transformation is that the physical stack addresses wrap
around from the end of the defined area to its beginning. When flushing and filling the
internal stack, this circular stack mechanism only requires to move that portion of stack
data which is really to be re-used (i.e. the upper part of the defined stack area) instead
of the whole stack area. Stack data that remain in the lower part of the internal stack
need not be moved by the distance of the space being flushed or filled, as the stack
pointer automatically wraps around to the beginning of the freed part of the stack area.
Note: This circular stack technique is applicable for stack sizes of 32 to 512 words
(STKSZ = ‘000 ’ to ‘100 ’), it does not work with option STKSZ = ‘111 ’, which
B
B
B
uses the complete internal RAM for system stack.
In the latter case the address transformation mechanism is deactivated.
When a boundary is reached, the stack underflow or overflow trap is entered, where the
user moves a predetermined portion of the internal stack to or from the external stack.
The amount of data transferred is determined by the average stack space required by
routines and the frequency of calls, traps, interrupts and returns. In most cases this will
be approximately one quarter to one tenth the size of the internal stack. Once the transfer
is complete, the boundary pointers are updated to reflect the newly allocated space on
the internal stack. Thus, the user is free to write code without concern for the internal
stack limits. Only the execution time required by the trap routines affects user programs.
The following procedure initializes the controller for usage of the circular stack
mechanism:
• Specify the size of the physical system stack area within the internal RAM (bitfield
STKSZ in register SYSCON).
• Define two pointers, which specify the upper and lower boundary of the external stack.
These values are then tested in the stack underflow and overflow trap routines when
moving data.
• Set the stack overflow pointer (STKOV) to the limit of the defined internal stack area
plus six words (for the reserved space to store two interrupt entries).
The internal stack will now fill until the overflow pointer is reached. After entry into the
overflow trap procedure, the top of the stack will be copied to the external memory. The
internal pointers will then be modified to reflect the newly allocated space. After exiting
from the trap procedure, the internal stack will wrap around to the top of the internal
stack, and continue to grow until the new value of the stack overflow pointer is reached.
When the underflow pointer is reached while the stack is meptied the bottom of stack is
reloaded from the external memory and the internal pointers are adjusted accordingly.
User’s Manual
21-7
V3.1, 2000-03
C167CR
Derivatives
System Programming
Linear Stack
The C167CR also offers a linear stack option (STKSZ = ‘111 ’), where the system stack
B
may use the complete internal RAM area. This provides a large system stack without
requiring procedures to handle data transfers for a circular stack. However, this method
also leaves less RAM space for variables or code. The RAM area that may effectively be
consumed by the system stack is defined via the STKUN and STKOV pointers. The
underflow and overflow traps in this case serve for fatal error detection only.
For the linear stack option all modifiable bits of register SP are used to access the
physical stack. Although the stack pointer may cover addresses from 00’F000 up to
H
00’FFFE the (physical) system stack must be located within the internal RAM and
H
therefore may only use the address range 00’F600 to 00’FDFE . It is the user’s
H
H
responsibility to restrict the system stack to the internal RAM range.
Note: Avoid stack accesses below the IRAM area (ESFR space and reserved area) and
within address range 00’FE00 and 00’FFFE (SFR space).
H
H
Otherwise unpredictable results will occur.
User Stacks
User stacks provide the ability to create task specific data stacks and to off-load data
from the system stack. The user may push both bytes and words onto a user stack, but
is responsible for using the appropriate instructions when popping data from the specific
user stack. No hardware detection of overflow or underflow of a user stack is provided.
The following addressing modes allow implementation of user stacks:
[–Rw], Rb or [–Rw], Rw: Pre-decrement Indirect Addressing.
Used to push one byte or word onto a user stack. This mode is only available for MOV
instructions and can specify any GPR as the user stack pointer.
Rb, [Rw +] or Rw, [Rw +]: Post-increment Index Register Indirect Addressing.
i
i
Used to pop one byte or word from a user stack. This mode is available to most
instructions, but only GPRs R0-R3 can be specified as the user stack pointer.
Rb, [Rw+] or Rw, [Rw+]: Post-increment Indirect Addressing.
Used to pop one byte or word from a user stack. This mode is only available for MOV
instructions and can specify any GPR as the user stack pointer.
User’s Manual
21-8
V3.1, 2000-03
C167CR
Derivatives
System Programming
21.2
Register Banking
Register banking provides the user with an extremely fast method to switch user context.
A single machine cycle instruction saves the old bank and enters a new register bank.
Each register bank may assign up to 16 registers. Each register bank should be
allocated during coding based on the needs of each task. Once the internal memory has
been partitioned into a register bank space, internal stack space and a global internal
memory area, each bank pointer is then assigned. Thus, upon entry into a new task, the
appropriate bank pointer is used as the operand for the SCXT (switch context)
instruction. Upon exit from a task a simple POP instruction to the context pointer (CP)
restores the previous task’s register bank.
21.3
Procedure Call Entry and Exit
To support modular programming a procedure mechanism is provided to allow coding of
frequently used portions of code into subroutines. The CALL and RET instructions store
and restore the value of the instruction pointer (IP) on the system stack before and after
a subroutine is executed.
Procedures may be called conditionally with instructions CALLA or CALLI, or be called
unconditionally using instructions CALLR or CALLS.
Note: Any data pushed onto the system stack during execution of the subroutine must
be popped before the RET instruction is executed.
Passing Parameters on the System Stack
Parameters may be passed via the system stack through PUSH instructions before the
subroutine is called, and POP instructions during execution of the subroutine. Base plus
offset indirect addressing also permits access to parameters without popping these
parameters from the stack during execution of the subroutine. Indirect addressing
provides a mechanism of accessing data referenced by data pointers, which are passed
to the subroutine.
In addition, two instructions have been implemented to allow one parameter to be
passed on the system stack without additional software overhead.
The PCALL (push and call) instruction first pushes the ‘reg’ operand and the IP contents
onto the system stack and then passes control to the subroutine specified by the ‘caddr’
operand.
When exiting from the subroutine, the RETP (return and pop) instruction first pops the IP
and then the ‘reg’ operand from the system stack and returns to the calling program.
User’s Manual
21-9
V3.1, 2000-03
C167CR
Derivatives
System Programming
Cross Segment Subroutine Calls
Calls to subroutines in different segments require the use of the CALLS (call inter-
segment subroutine) instruction. This instruction preserves both the CSP (code segment
pointer) and IP on the system stack.
Upon return from the subroutine, a RETS (return from inter-segment subroutine)
instruction must be used to restore both the CSP and IP. This ensures that the next
instruction after the CALLS instruction is fetched from the correct segment.
Note: It is possible to use CALLS within the same segment, but still two words of the
stack are used to store both the IP and CSP.
Providing Local Registers for Subroutines
For subroutines which require local storage, the following methods are provided:
Alternate Bank of Registers: Upon entry into a subroutine, it is possible to specify a
new set of local registers by executing the SCXT (switch context) instruction. This
mechanism does not provide a method to recursively call a subroutine.
Saving and Restoring of Registers: To provide local registers, the contents of the
registers which are required for use by the subroutine can be pushed onto the stack and
the previous values be popped before returning to the calling routine. This is the most
common technique used today and it does provide a mechanism to support recursive
procedures. This method, however, requires two machine cycles per register stored on
the system stack (one cycle to PUSH the register, and one to POP the register).
Use of the System Stack for Local Registers: It is possible to use the SP and CP to
set up local subroutine register frames. This enables subroutines to dynamically allocate
local variables as needed within two machine cycles. A local frame is allocated by simply
subtracting the number of required local registers from the SP, and then moving the
value of the new SP to the CP.
This operation is supported through the SCXT (switch context) instruction with the
addressing mode ‘reg, mem’. Using this instruction saves the old contents of the CP on
the system stack and moves the value of the SP into CP (see example below). Each local
register is then accessed as if it was a normal register. Upon exit from the subroutine,
first the old CP must be restored by popping it from the stack and then the number of
used local registers must be added to the SP to restore the allocated local space back
to the system stack.
Note: The system stack is growing downwards, while the register bank is growing
upwards.
User’s Manual
21-10
V3.1, 2000-03
C167CR
Derivatives
System Programming
Old
Stack
Area
Old SP
R4
R3
Newly
Allocated
Register
Bank
R2
R1
R0
New CP
New SP
Old CPContents
New
Stack
Area
MCA04409
Figure 21-2 Local Registers
The software to provide the local register bank for the example above is very compact:
After entering the subroutine:
SUB
SCXT
SP, #10D
CP, SP
;Free 5 words in the current system stack
;Set the new register bank pointer
Before exiting the subroutine:
POP
ADD
CP
SP, #10D
;Restore the old register bank
;Release the 5 words …
;… of the current system stack
User’s Manual
21-11
V3.1, 2000-03
C167CR
Derivatives
System Programming
21.4
Table Searching
A number of features have been included to decrease the execution time required to
search tables. First, branch delays are eliminated by the branch target cache after the
first iteration of the loop. Second, in non-sequentially searched tables, the enhanced
performance of the ALU allows more complicated hash algorithms to be processed to
obtain better table distribution. For sequentially searched tables, the auto-increment
indirect addressing mode and the E (end of table) flag stored in the PSW decrease the
number of overhead instructions executed in the loop.
The two examples below illustrate searching ordered tables and non-ordered tables,
respectively:
MOV
R0, #BASE
;Move table base into R0
LOOP:
CMP
R1, [R0+]
;Compare target to table entry
JMPR
cc_SGT, LOOP;Test whether target has not been found
Note: The last entry in the table must be greater than the largest possible target.
MOV
R0, #BASE
;Move table base into R0
LOOP:
CMP
R1, [R0+]
;Compare target to table entry
JMPR
cc_NET, LOOP;Test whether target is not found AND …
;…the end of table has not been reached.
Note: The last entry in the table must be equal to the lowest signed integer (8000 ).
H
21.5
Floating Point Support
All floating point operations are performed using software. Standard multiple precision
instructions are used to perform calculations on data types that exceed the size of the
ALU. Multiple bit rotate and logic instructions allow easy masking and extracting of
portions of floating point numbers.
To decrease the time required to perform floating point operations, two hardware
features have been implemented in the CPU core. First, the PRIOR instruction aids in
normalizing floating point numbers by indicating the position of the first set bit in a GPR.
This result can the be used to rotate the floating point result accordingly. The second
feature aids in properly rounding the result of normalized floating point numbers through
the overflow (V) flag in the PSW. This flag is set when a one is shifted out of the carry bit
during shift right operations. The overflow flag and the carry flag are then used to round
the floating point result based on the desired rounding algorithm.
User’s Manual
21-12
V3.1, 2000-03
C167CR
Derivatives
System Programming
21.6
Peripheral Control and Interface
All communication between peripherals and the CPU is performed either by PEC
transfers to and from internal memory, or by explicitly addressing the SFRs associated
with the specific peripherals. After resetting the C167CR all peripherals (except the
watchdog timer) are disabled and initialized to default values. A desired configuration of
a specific peripheral is programmed using MOV instructions of either constants or
memory values to specific SFRs. Specific control flags may also be altered via bit
instructions.
Once in operation, the peripheral operates autonomously until an end condition is
reached at which time it requests a PEC transfer or requests CPU servicing through an
interrupt routine. Information may also be polled from peripherals through read accesses
to SFRs or bit operations including branch tests on specific control bits in SFRs. To
ensure proper allocation of peripherals among multiple tasks, a portion of the internal
memory has been made bit addressable to allow user semaphores. Instructions have
also been provided to lock out tasks via software by setting or clearing user specific bits
and conditionally branching based on these specific bits.
It is recommended that bit fields in control SFRs are updated using the BFLDH and
BFLDL instructions or a MOV instruction to avoid undesired intermediate modes of
operation which can occur, when BCLR/BSET or AND/OR instruction sequences are
used.
21.7
Trap/Interrupt Entry and Exit
Interrupt routines are entered when a requesting interrupt has a priority higher than the
current CPU priority level. Traps are entered regardless of the current CPU priority.
When either a trap or interrupt routine is entered, the state of the machine is preserved
on the system stack and a branch to the appropriate trap/interrupt vector is made.
All trap and interrupt routines require the use of the RETI (return from interrupt)
instruction to exit from the called routine. This instruction restores the system state from
the system stack and then branches back to the location where the trap or interrupt
occurred.
User’s Manual
21-13
V3.1, 2000-03
C167CR
Derivatives
System Programming
21.8
Unseparable Instruction Sequences
The instructions of the C167CR are very efficient (most instructions execute in one
machine cycle) and even the multiplication and division are interruptable in order to
minimize the response latency to interrupt requests (internal and external). In many
microcontroller applications this is vital.
Some special occasions, however, require certain code sequences (e.g. semaphore
handling) to be uninterruptable to function properly. This can be provided by inhibiting
interrupts during the respective code sequence by disabling and enabling them before
and after the sequence. The necessary overhead may be reduced by means of the
ATOMIC instruction which allows locking 1 … 4 instructions to an unseparable code
sequence, during which the interrupt system (standard interrupts and PEC requests)
and Class A Traps (NMI, stack overflow/underflow) are disabled. A Class B Trap
(illegal opcode, illegal bus access, etc.), however, will interrupt the atomic sequence,
since it indicates a severe hardware problem.
The interrupt inhibit caused by an ATOMIC instruction gets active immediately, i.e. no
other instruction will enter the pipeline except the one that follows the ATOMIC
instruction, and no interrupt request will be serviced in between. All instructions requiring
multiple cycles or hold states are regarded as one instruction in this sense (e.g. MUL is
one instruction). Any instruction type can be used within an unseparable code sequence.
ATOMIC #3
;The next 3 instr. are locked (No NOP requ.)
MOV
MOV
MUL
MOV
R0, #1234H ;Instr. 1 (no other instr. enters pipeline!)
R1, #5678H ;Instr. 2
R0, R1
;Instr. 3: MUL regarded as one instruction
;This instruction is out of the scope …
;… of the ATOMIC instruction sequence
R2, MDL
Note: As long as any Class B trap is pending (any of the class B trap flags in register
TFR is set) the ATOMIC instruction will not work. Clear the respective B trap flag
at the beginning of a B trap routine if ATOMIC shall be used within the routine.
21.9
Overriding the DPP Addressing Mechanism
The standard mechanism to access data locations uses one of the four data page
pointers (DPPx), which selects a 16-KByte data page, and a 14-bit offset within this data
page. The four DPPs allow immediate access to up to 64 KByte of data. In applications
with big data arrays, especially in HLL applications using large memory models, this may
require frequent reloading of the DPPs, even for single accesses.
User’s Manual
21-14
V3.1, 2000-03
C167CR
Derivatives
System Programming
The EXTP (extend page) instruction allows switching to an arbitrary data page for
1 … 4 instructions without having to change the current DPPs.
EXTP
MOV
MOV
R15, #1
R0, [R14]
R1, [R13]
;The override page number is stored in R15
;The (14-bit) page offset is stored in R14
;This instruction uses the std. DPP scheme!
The EXTS (extend segment) instruction allows switching to a 64 KByte segment
oriented data access scheme for 1 … 4 instructions without having to change the current
DPPs. In this case all 16 bits of the operand address are used as segment offset, with
the segment taken from the EXTS instruction. This greatly simplifies address calculation
with continuous data like huge arrays in “C”.
EXTS
MOV
MOV
#15, #1
R0, [R14]
R1, [R13]
;The override seg. is 15 (0F’0000H..0F’FFFFH)
;The (16-bit) segment offset is stored in R14
;This instruction uses the std. DPP scheme!
Note: Instructions EXTP and EXTS inhibit interrupts the same way as ATOMIC.
As long as any Class B trap is pending (any of the class B trap flags in register
TFR is set) the EXTend instructions will not work. Clear the respective B trap flag
at the beginning of a B trap routine if EXT* shall be used within the routine.
Short Addressing in the Extended SFR (ESFR) Space
The short addressing modes of the C167CR (REG or BITOFF) implicitly access the SFR
space. The additional ESFR space would have to be accessed via long addressing
modes (MEM or [Rw]). The EXTR (extend register) instruction redirects accesses in
short addressing modes to the ESFR space for 1 … 4 instructions, so the additional
registers can be accessed this way, too.
The EXTPR and EXTSR instructions combine the DPP override mechanism with the
redirection to the ESFR space using a single instruction.
Note: Instructions EXTR, EXTPR and EXTSR inhibit interrupts the same way as ATOMIC.
The switching to the ESFR area and data page overriding is checked by the
development tools or handled automatically.
Nested Locked Sequences
Each of the described extension instruction and the ATOMIC instruction starts an
internal “extension counter” counting the effected instructions. When another extension
or ATOMIC instruction is contained in the current locked sequence this counter is
restarted with the value of the new instruction. This allows the construction of locked
sequences longer than 4 instructions.
Note: Interrupt latencies may be increased when using locked code sequences.
PEC requests are not serviced during idle mode, if the IDLE instruction is part of
a locked sequence.
User’s Manual
21-15
V3.1, 2000-03
C167CR
Derivatives
System Programming
21.10
Handling the Internal Code Memory
The Mask-ROM/OTP/Flash versions of the C167CR provide on-chip code memory that
may store code as well as data. The lower 32 KByte of this code memory are referred to
as the „internal ROM area“. Access to this internal ROM area is controlled during the
reset configuration and via software. The ROM area may be mapped to segment 0, to
segment 1 or the code memory may be disabled at all.
Note: The internal ROM area always occupies an address area of 32 KByte, even if the
implemented mask ROM/OTP/Flash memory is smaller than that (e.g. 8 KByte).
Of course the total implemented memory may exceed 32 KBytes.
Code Memory Configuration During Reset
The control input pin EA (External Access) enables the user to define the address area
from which the first instructions after reset are fetched. When EA is low (‘0’) during reset,
the internal code memory is disabled and the first instructions are fetched from external
memory. When EA is high (‘1’) during reset, the internal code memory is globally enabled
and the first instructions are fetched from the internal memory.
Note: Be sure not to select internal memory access after reset on ROMless devices.
Mapping the Internal ROM Area
After reset the internal ROM area is mapped into segment 0, the “system segment”
(00’0000 … 00’7FFF ) as a default. This is necessary to allow the first instructions to
H
H
be fetched from locations 00’0000 ff. The ROM area may be mapped to segment 1
H
(01’0000 … 01’7FFF ) by setting bit ROMS1 in register SYSCON. The internal ROM
H
H
area may now be accessed through the lower half of segment 1, while accesses to
segment 0 will now be made to external memory. This adds flexibility to the system
software. The interrupt/trap vector table, which uses locations 00’0000 through
H
00’01FF , is now part of the external memory and may therefore be modified, i.e. the
H
system software may now change interrupt/trap handlers according to the current
condition of the system. The internal code memory can still be used for fixed software
routines like IO drivers, math libraries, application specific invariant routines, tables, etc.
This combines the advantage of an integrated non-volatile memory with the advantage
of a flexible, adaptable software system.
User’s Manual
21-16
V3.1, 2000-03
C167CR
Derivatives
System Programming
Enabling and Disabling the Internal Code Memory After Reset
If the internal code memory does not contain an appropriate startup code, the system
may be booted from external memory, while the internal memory is enabled afterwards
to provide access to library routines, tables, etc.
If the internal code memory only contains the startup code and/or test software, the
system may be booted from internal memory, which may then be disabled, after the
software has switched to executing from (e.g.) external memory, in order to free the
address space occupied by the internal code memory, which is now unnecessary.
User’s Manual
21-17
V3.1, 2000-03
C167CR
Derivatives
System Programming
21.11
Pits, Traps and Mines
Although handling the internal code memory provides powerful means to enhance the
overall performance and flexibility of a system, extreme care must be taken in order to
avoid a system crash. Instruction memory is the most crucial resource for the C167CR
and it must be made sure that it never runs out of it. The following precautions help to
take advantage of the methods mentioned above without jeopardizing system security.
Internal code memory access after reset: When the first instructions are to be fetched
from internal memory (EA = ‘1’), the device must contain code memory, and this must
contain a valid reset vector and valid code at its destination.
Mapping the internal ROM area to segment 1: Due to instruction pipelining, any new
ROM mapping will at the earliest become valid for the second instruction after the
instruction which has changed the ROM mapping. To enable accesses to the ROM area
after mapping a branch to the newly selected ROM area (JMPS) and reloading of all data
page pointers is required.
This also applies to re-mapping the internal ROM area to segment 0.
Enabling the internal code memory after reset: When enabling the internal code
memory after having booted the system from external memory, note that the C167CR
will then access the internal memory using the current segment offset, rather than
accessing external memory.
Disabling the internal code memory after reset: When disabling the internal code
memory after having booted the system from there, note that the C167CR will not access
external memory before a jump to segment 0 (in this case) is executed.
General Rules
When mapping the code memory no instruction or data accesses should be made to the
internal memory, otherwise unpredictable results may occur.
To avoid these problems, the instructions that configure the internal code memory
should be executed from external memory or from the on-chip RAM.
Whenever the internal code memory is disabled, enabled or remapped the DPPs must
be explicitly (re)loaded to enable correct data accesses to the internal and/or external
memory.
User’s Manual
21-18
V3.1, 2000-03
C167CR
Derivatives
The Register Set
22
The Register Set
This section summarizes all registers, which are implemented in the C167CR and
explains the description format which is used in the chapters describing the function and
layout of the SFRs.
For easy reference the registers are ordered according to two different keys (except for GPRs):
• Ordered by address, to check which register a given address references,
• Ordered by register name, to find the location of a specific register.
22.1
Register Description Format
In the respective chapters the function and the layout of the SFRs is described in a
specific format which provides a number of details about the described special function
register. The example below shows how to interpret these details.
REG_NAME
Name of Register
E/SFR (A16 /A8 )
Reset value: ****
H
H
H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
read/
std hw write
bit
read write
bit bit
<empty for byte registers>
bitfield
-
-
-
-
-
-
-
-
rw rwh rw
r
w
rw
Bit
Function
bit(field)name Explanation of bit(field)name
Description of the functions controlled by the different possible values
of this bit(field).
Elements
REG_NAME
A16/A8
Short name of this register
Long 16-bit address/Short 8-bit address
SFR/ESFR/XReg
(**) **
Register space (SFR, ESFR or External/XBUS Register)
Register contents after reset
0/1: defined value, ‘X’: undefined,
‘U’: unchanged (undefined (‘X’) after power up)
Access modes: can be read and/or write
Bits that are set/cleared by hardware are marked with
a shaded access box and an ‘h’ in it.
r/w
×h
User’s Manual
22-1
V3.1, 2000-03
C167CR
Derivatives
The Register Set
22.2
CPU General Purpose Registers (GPRs)
The GPRs form the register bank that the CPU works with. This register bank may be
located anywhere within the internal RAM via the Context Pointer (CP). Due to the
addressing mechanism, GPR banks can only reside within the internal RAM. All GPRs
are bit-addressable.
Table 22-1 General Purpose Word Registers
Name
Physical 8-bit
Address Address
Description
Reset
Value
R0
(CP) + 0
(CP) + 2
(CP) + 4
(CP) + 6
(CP) + 8
F0
F1
F2
F3
F4
CPU General Purpose (Word) Reg. R0
CPU General Purpose (Word) Reg. R1
CPU General Purpose (Word) Reg. R2
CPU General Purpose (Word) Reg. R3
CPU General Purpose (Word) Reg. R4
CPU General Purpose (Word) Reg. R5
CPU General Purpose (Word) Reg. R6
CPU General Purpose (Word) Reg. R7
CPU General Purpose (Word) Reg. R8
CPU General Purpose (Word) Reg. R9
CPU General Purpose (Word) Reg. R10
CPU General Purpose (Word) Reg. R11
CPU General Purpose (Word) Reg. R12
CPU General Purpose (Word) Reg. R13
CPU General Purpose (Word) Reg. R14
CPU General Purpose (Word) Reg. R15
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
R1
R2
R3
R4
R5
(CP) + 10 F5
(CP) + 12 F6
(CP) + 14 F7
(CP) + 16 F8
(CP) + 18 F9
R6
R7
R8
R9
R10
R11
R12
R13
R14
R15
(CP) + 20 FA
(CP) + 22 FB
H
H
(CP) + 24 FC
(CP) + 26 FD
H
H
H
H
(CP) + 28 FE
(CP) + 30 FF
User’s Manual
22-2
V3.1, 2000-03
C167CR
Derivatives
The Register Set
The first 8 GPRs (R7 … R0) may also be accessed bytewise. Other than with SFRs,
writing to a GPR byte does not affect the other byte of the respective GPR.
The respective halves of the byte-accessible registers receive special names:
Table 22-2 General Purpose Byte Registers
Name
Physical 8-bit
Address Address
Description
Reset
Value
RL0
RH0
RL1
RH1
RL2
RH2
RL3
RH3
RL4
RH4
RL5
RH5
RL6
RH6
RL7
RH7
(CP) + 0
(CP) + 1
(CP) + 2
(CP) + 3
(CP) + 4
(CP) + 5
(CP) + 6
(CP) + 7
(CP) + 8
(CP) + 9
F0
F1
F2
F3
F4
F5
F6
F7
F8
F9
CPU General Purpose (Byte) Reg. RL0
CPU General Purpose (Byte) Reg. RH0
CPU General Purpose (Byte) Reg. RL1
CPU General Purpose (Byte) Reg. RH1
CPU General Purpose (Byte) Reg. RL2
CPU General Purpose (Byte) Reg. RH2
CPU General Purpose (Byte) Reg. RL3
CPU General Purpose (Byte) Reg. RH3
CPU General Purpose (Byte) Reg. RL4
CPU General Purpose (Byte) Reg. RH4
CPU General Purpose (Byte) Reg. RL5
CPU General Purpose (Byte) Reg. RH5
CPU General Purpose (Byte) Reg. RL6
CPU General Purpose (Byte) Reg. RH6
CPU General Purpose (Byte) Reg. RL7
CPU General Purpose (Byte) Reg. RH7
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
(CP) + 10 FA
(CP) + 11 FB
H
H
(CP) + 12 FC
(CP) + 13 FD
H
H
H
H
(CP) + 14 FE
(CP) + 15 FF
User’s Manual
22-3
V3.1, 2000-03
C167CR
Derivatives
The Register Set
22.3
Special Function Registers Ordered by Name
Table 22-3 lists all SFRs which are implemented in the C167CR in alphabetical order.
Bit-addressable SFRs are marked with the letter “b” in column “Name”.
SFRs within the extended SFR-space (ESFRs) are marked with the letter “E” in column
“Physical Address”. Registers within on-chip X-Peripherals are marked with the letter “X”
in column “Physical Address”.
Table 22-3 C167CR Registers, Ordered by Name
Name
Physical 8-bit Description
Address Addr.
Reset
Value
ADCIC
b FF98
CC
A/D Converter End of Conversion Interrupt
Control Register
0000
H
H
H
ADCON
ADDAT
ADDAT2
b FFA0
D0
A/D Converter Control Register
A/D Converter Result Register
A/D Converter 2 Result Register
Address Select Register 1
Address Select Register 2
Address Select Register 3
Address Select Register 4
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
FEA0
50
H
H
F0A0 E 50
H
H
ADDRSEL1 FE18
0C
H
H
H
H
H
ADDRSEL2 FE1A
0D
H
ADDRSEL3 FE1C
0E
0F
H
H
H
ADDRSEL4 FE1E
ADEIC b FF9A
CD
A/D Converter Overrun Error Interrupt
Control Register
H
BUSCON0 b FF0C
86
Bus Configuration Register 0
Bus Configuration Register 1
Bus Configuration Register 2
Bus Configuration Register 3
Bus Configuration Register 4
CAN1 Bit Timing Register
CAN1 Control/Status Register
CAN1 Global Mask Short
CAN1 Interrupt Register
0000
0000
0000
0000
0000
UUUU
XX01
UFUU
XX
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
BUSCON1 b FF14
BUSCON2 b FF16
BUSCON3 b FF18
8A
8B
H
H
8C
8D
H
H
BUSCON4 b FF1A
H
C1BTR
EF04 X –
H
C1CSR
C1GMS
C1IR
EF00 X –
H
EF06 X –
H
EF02 X –
H
C1LARn
C1LGML
C1LMLM
C1MCFGn
EFn4 X –
CAN1 Lower Arbitration Register (msg. n) UUUU
H
EF0A X –
CAN1 Lower Global Mask Long
UUUU
UUUU
UU
H
EF0E X –
CAN1 Lower Mask of Last Message
H
EFn6 X –
CAN1 Message Configuration Register
H
(msg. n)
User’s Manual
22-4
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-3 C167CR Registers, Ordered by Name (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
C1MCRn
C1UARn
C1UGML
C1UMLM
CAPREL
CC0
EFn0 X –
CAN1 Message Ctrl. Reg. (msg. n)
CAN1 Upper Arbitration Reg. (msg. n)
CAN1 Upper Global Mask Long
CAN1 Upper Mask of Last Message
GPT2 Capture/Reload Register
CAPCOM Register 0
UUUU
UUUU
UUUU
UUUU
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
EFn2 X –
H
EF08 X –
H
EF0C X –
H
FE4A
25
40
H
H
H
H
H
H
H
FE80
CC0IC
CC1
b FF78
FE82
BC
CAPCOM Register 0 Interrupt Ctrl. Reg.
CAPCOM Register 1
H
41
H
CC10
FE94
4A
CAPCOM Register 10
H
CC10IC
CC11
b FF8C
C6
CAPCOM Register 10 Interrupt Ctrl. Reg.
CAPCOM Register 11
H
H
H
H
H
H
FE96
4B
H
CC11IC
CC12
b FF8E
C7
4C
C8
4D
C9
CAPCOM Register 11 Interrupt Ctrl. Reg.
CAPCOM Register 12
H
H
H
H
H
H
FE98
b FF90
FE9A
CC12IC
CC13
CAPCOM Register 12 Interrupt Ctrl. Reg.
CAPCOM Register 13
H
CC13IC
CC14
b FF92
CAPCOM Register 13 Interrupt Ctrl. Reg.
CAPCOM Register 14
H
FE9C
4E
CA
H
H
H
CC14IC
CC15
b FF94
CAPCOM Register 14 Interrupt Ctrl. Reg.
CAPCOM Register 15
H
H
FE9E
b FF96
4F
H
CC15IC
CC16
CB
30
CAPCOM Register 15 Interrupt Ctrl. Reg.
CAPCOM Register 16
H
H
FE60
H
CC16IC
CC17
b F160 E B0
CAPCOM Register 16 Interrupt Ctrl. Reg.
CAPCOM Register 17
H
H
FE62
31
H
H
CC17IC
CC18
b F162 E B1
CAPCOM Register 17 Interrupt Ctrl. Reg.
CAPCOM Register 18
H
H
H
FE64
32
H
CC18IC
CC19
b F164 E B2
CAPCOM Register 18 Interrupt Ctrl. Reg.
CAPCOM Register 19
H
H
H
FE66
33
H
CC19IC
CC1IC
CC2
b F166 E B3
CAPCOM Register 19 Interrupt Ctrl. Reg.
CAPCOM Register 1 Interrupt Ctrl. Reg.
CAPCOM Register 2
H
H
b FF7A
BD
H
H
H
FE84
42
H
User’s Manual
22-5
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-3 C167CR Registers, Ordered by Name (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
CC20
FE68
34
CAPCOM Register 20
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CC20IC
CC21
b F168 E B4
CAPCOM Register 20 Interrupt Ctrl. Reg.
CAPCOM Register 21
H
H
H
FE6A
35
H
CC21IC
CC22
b F16A E B5
CAPCOM Register 21 Interrupt Ctrl. Reg.
CAPCOM Register 22
H
H
H
FE6C
36
H
CC22IC
CC23
b F16C E B6
CAPCOM Register 22 Interrupt Ctrl. Reg.
CAPCOM Register 23
H
H
FE6E
37
H
H
CC23IC
CC24
b F16E E B7
CAPCOM Register 23 Interrupt Ctrl. Reg.
CAPCOM Register 24
H
H
FE70
38
H
H
CC24IC
CC25
b F170 E B8
CAPCOM Register 24 Interrupt Ctrl. Reg.
CAPCOM Register 25
H
H
FE72
39
H
H
CC25IC
CC26
b F172 E B9
CAPCOM Register 25 Interrupt Ctrl. Reg.
CAPCOM Register 26
H
H
H
FE74
3A
H
CC26IC
CC27
b F174 E BA
CAPCOM Register 26 Interrupt Ctrl. Reg.
CAPCOM Register 27
H
H
H
FE76
3B
H
CC27IC
CC28
b F176 E BB
CAPCOM Register 27 Interrupt Ctrl. Reg.
CAPCOM Register 28
H
H
H
FE78
3C
H
CC28IC
CC29
b F178 E BC
CAPCOM Register 28 Interrupt Ctrl. Reg.
CAPCOM Register 29
H
H
H
H
FE7A
3D
H
CC29IC
CC2IC
CC3
b F184 E C2
CAPCOM Register 29 Interrupt Ctrl. Reg.
CAPCOM Register 2 Interrupt Ctrl. Reg.
CAPCOM Register 3
H
b FF7C
BE
H
H
H
FE86
43
H
CC30
FE7C
3E
CAPCOM Register 30
H
H
CC30IC
CC31
b F18C E C6
CAPCOM Register 30 Interrupt Ctrl. Reg.
CAPCOM Register 31
H
H
H
FE7E
3F
H
CC31IC
CC3IC
CC4
b F194 E CA
CAPCOM Register 31 Interrupt Ctrl. Reg.
CAPCOM Register 3 Interrupt Ctrl. Reg.
CAPCOM Register 4
H
H
H
b FF7E
BF
44
H
H
H
FE88
b FF80
H
CC4IC
CC5
C0
CAPCOM Register 4 Interrupt Ctrl. Reg.
CAPCOM Register 5
H
FE8A
45
H
H
User’s Manual
22-6
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-3 C167CR Registers, Ordered by Name (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
CC5IC
CC6
b FF82
C1
CAPCOM Register 5 Interrupt Ctrl. Reg.
CAPCOM Register 6
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
FC00
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
FE8C
46
H
H
CC6IC
CC7
b FF84
C2
47
CAPCOM Register 6 Interrupt Ctrl. Reg.
CAPCOM Register 7
H
FE8E
b FF86
H
H
CC7IC
CC8
C3
48
CAPCOM Register 7 Interrupt Ctrl. Reg.
CAPCOM Register 8
H
FE90
H
H
H
H
CC8IC
CC9
b FF88
C4
49
CAPCOM Register 8 Interrupt Ctrl. Reg.
CAPCOM Register 9
FE92
H
CC9IC
CCM0
CCM1
CCM2
CCM3
CCM4
CCM5
CCM6
CCM7
CP
b FF8A
C5
CAPCOM Register 9 Interrupt Ctrl. Reg.
CAPCOM Mode Control Register 0
CAPCOM Mode Control Register 1
CAPCOM Mode Control Register 2
CAPCOM Mode Control Register 3
CAPCOM Mode Control Register 4
CAPCOM Mode Control Register 5
CAPCOM Mode Control Register 6
CAPCOM Mode Control Register 7
CPU Context Pointer Register
H
H
H
H
H
H
H
H
H
H
H
b FF52
b FF54
b FF56
b FF58
b FF22
b FF24
b FF26
b FF28
A9
AA
AB
H
H
AC
H
91
92
93
94
08
H
H
H
H
H
FE10
H
CRIC
CSP
b FF6A
B5
GPT2 CAPREL Interrupt Control Register
H
H
H
H
FE08
04
CPU Code Segment Pointer Register
(8 bits, not directly writeable)
DP0H
DP0L
DP1H
DP1L
DP2
b F102 E 81
P0H Direction Control Register
P0L Direction Control Register
P1H Direction Control Register
P1L Direction Control Register
Port 2 Direction Control Register
Port 3 Direction Control Register
Port 4 Direction Control Register
Port 6 Direction Control Register
Port 7 Direction Control Register
00
00
H
H
H
H
H
H
H
H
H
H
H
H
H
H
b F100 E 80
H
b F106 E 83
00
H
b F104 E 82
00
H
b FFC2
b FFC6
E1
0000
0000
00
H
H
H
H
H
H
H
DP3
E3
E5
E7
E9
DP4
b FFCA
b FFCE
H
H
H
DP6
00
DP7
b FFD2
00
User’s Manual
22-7
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-3 C167CR Registers, Ordered by Name (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
DP8
b FFD6
EB
Port 8 Direction Control Register
00
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
DPP0
DPP1
DPP2
DPP3
FE00
FE02
FE04
FE06
00
01
02
03
CPU Data Page Pointer 0 Register (10 bits) 0000
CPU Data Page Pointer 1 Register (10 bits) 0001
CPU Data Page Pointer 2 Register (10 bits) 0002
CPU Data Page Pointer 3 Register (10 bits) 0003
H
H
H
H
EXICON b F1C0 E E0
External Interrupt Control Register
CPU Multiply Divide Control Register
CPU Multiply Divide Register – High Word
CPU Multiply Divide Register – Low Word
Port 2 Open Drain Control Register
Port 3 Open Drain Control Register
Port 6 Open Drain Control Register
Port 7 Open Drain Control Register
Port 8 Open Drain Control Register
Constant Value 1’s Register (read only)
Port 0 High Register (Upper half of PORT0)
Port 0 Low Register (Lower half of PORT0)
Port 1 High Register (Upper half of PORT1)
Port 1 Low Register (Lower half of PORT1)
Port 2 Register
0000
0000
0000
0000
0000
0000
00
H
H
H
H
H
MDC
MDH
MDL
ODP2
ODP3
ODP6
ODP7
ODP8
ONES
P0H
P0L
b FF0E
87
06
07
H
FE0C
H
FE0E
H
b F1C2 E E1
H
H
H
H
H
b F1C6 E E3
H
b F1CE E E7
H
b F1D2 E E9
00
H
b F1D6 E EB
00
H
H
b FF1E
8F
FFFF
00
H
H
H
H
H
H
b FF02
b FF00
b FF06
b FF04
81
80
83
82
H
H
H
H
00
P1H
P1L
00
00
P2
b FFC0
b FFC4
b FFC8
E0
E2
E4
0000
0000
00
H
H
H
H
H
H
H
H
P3
Port 3 Register
P4
Port 4 Register (7 bits)
P5
b FFA2
D1
D2
Port 5 Register (read only)
XXXX
0000
00
H
H
H
H
P5DIDIS b FFA4
Port 5 Digital Input Disable Register
Port 6 Register (8 bits)
P6
b FFCC
E6
E8
H
P7
b FFD0
b FFD4
Port 7 Register (8 bits)
00
H
P8
EA
Port 8 Register (8 bits)
00
H
H
PDCR
PECC0
PECC1
F0AA E 55
Port Driver Control Register
0000
0000
0000
H
H
FEC0
FEC2
60
61
PEC Channel 0 Control Register
PEC Channel 1 Control Register
H
H
H
H
User’s Manual
22-8
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-3 C167CR Registers, Ordered by Name (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
PECC2
PECC3
PECC4
PECC5
PECC6
PECC7
PICON
PP0
FEC4
FEC6
FEC8
62
63
64
65
66
67
PEC Channel 2 Control Register
PEC Channel 3 Control Register
PEC Channel 4 Control Register
PEC Channel 5 Control Register
PEC Channel 6 Control Register
PEC Channel 7 Control Register
Port Input Threshold Control Register
PWM Module Period Register 0
PWM Module Period Register 1
PWM Module Period Register 2
PWM Module Period Register 3
CPU Program Status Word
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
XX
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
FECA
H
FECC
H
H
FECE
F1C4 E E2
H
H
F038 E 1C
H
H
H
H
H
H
H
H
PP1
F03A E 1D
H
PP2
F03C E 1E
H
PP3
F03E E 1F
H
PSW
PT0
b FF10
88
H
F030 E 18
PWM Module Up/Down Counter 0
PWM Module Up/Down Counter 1
PWM Module Up/Down Counter 2
PWM Module Up/Down Counter 3
PWM Module Pulse Width Register 0
PWM Module Pulse Width Register 1
PWM Module Pulse Width Register 2
PWM Module Pulse Width Register 3
PWM Module Control Register 0
PWM Module Control Register 1
PWM Module Interrupt Control Register
H
PT1
F032 E 19
H
PT2
F034 E 1A
H
H
H
H
H
PT3
F036 E 1B
H
PW0
FE30
FE32
FE34
FE36
18
19
H
H
H
H
H
H
PW1
PW2
1A
1B
H
H
H
H
PW3
PWMCON0b FF30
PWMCON1b FF32
98
99
PWMIC
RP0H
b F17E E BF
H H
b F108 E 84
System Startup Configuration Register
(read only)
H
H
S0BG
FEB4
5A
Serial Channel 0 Baud Rate Generator
Reload Register
0000
H
H
H
S0CON
S0EIC
b FFB0
D8
Serial Channel 0 Control Register
0000
0000
H
H
H
H
b FF70
B8
Serial Channel 0 Error Interrupt Control
Register
H
H
User’s Manual
22-9
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-3 C167CR Registers, Ordered by Name (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
S0RBUF
S0RIC
FEB2
59
Serial Channel 0 Receive Buffer Register
(read only)
XXXX
0000
0000
H
H
H
H
H
b FF6E
B7
Serial Channel 0 Receive Interrupt Control
Register
H
H
S0TBIC
b F19C E CE
Serial Channel 0 Transmit Buffer Interrupt
Control Register
H
H
S0TBUF
S0TIC
FEB0
58
Serial Channel 0 Transmit Buffer Register
0000
0000
H
H
H
H
H
b FF6C
B6
Serial Channel 0 Transmit Interrupt Control
Register
H
SP
FE12
09
CPU System Stack Pointer Register
SSC Baudrate Register
FC00
0000
0000
0000
XXXX
0000
0000
0000
FA00
FC00
H
H
H
H
H
H
H
H
H
H
H
H
SSCBR
F0B4 E 5A
H H
SSCCON b FFB2
D9
SSC Control Register
H
H
H
SSCEIC
SSCRB
b FF76
BB
SSC Error Interrupt Control Register
SSC Receive Buffer
H
F0B2 E 59
H
H
SSCRIC b FF74
BA
SSC Receive Interrupt Control Register
SSC Transmit Buffer
H
H
SSCTB
SSCTIC
STKOV
STKUN
F0B0 E 58
H H
b FF72
B9
0A
0B
SSC Transmit Interrupt Control Register
CPU Stack Overflow Pointer Register
CPU Stack Underflow Pointer Register
CPU System Configuration Register
CAPCOM Timer 0 Register
H
H
H
H
H
H
FE14
FE16
H
H
H
1)
SYSCON b FF12
89
28
0xx0
H
T0
FE50
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
T01CON b FF50
A8
CAPCOM Timer 0 and Timer 1 Ctrl. Reg.
CAPCOM Timer 0 Interrupt Ctrl. Reg.
CAPCOM Timer 0 Reload Register
CAPCOM Timer 1 Register
H
H
T0IC
T0REL
T1
b FF9C
CE
H
H
H
H
H
H
H
H
H
H
FE54
FE52
2A
H
29
H
T1IC
T1REL
T2
b FF9E
CF
CAPCOM Timer 1 Interrupt Ctrl. Reg.
CAPCOM Timer 1 Reload Register
GPT1 Timer 2 Register
H
FE56
FE40
2B
H
20
H
T2CON
T2IC
T3
b FF40
b FF60
FE42
A0
B0
GPT1 Timer 2 Control Register
GPT1 Timer 2 Interrupt Control Register
GPT1 Timer 3 Register
H
H
H
21
User’s Manual
22-10
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-3 C167CR Registers, Ordered by Name (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
T3CON
T3IC
T4
b FF42
b FF62
A1
B1
GPT1 Timer 3 Control Register
GPT1 Timer 3 Interrupt Control Register
GPT1 Timer 4 Register
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
FE44
22
H
H
H
T4CON
T4IC
T5
b FF44
b FF64
A2
B2
GPT1 Timer 4 Control Register
GPT1 Timer 4 Interrupt Control Register
GPT2 Timer 5 Register
H
H
H
FE46
23
H
H
H
T5CON
T5IC
T6
b FF46
b FF66
A3
B3
GPT2 Timer 5 Control Register
GPT2 Timer 5 Interrupt Control Register
GPT2 Timer 6 Register
H
H
H
FE48
24
H
H
H
T6CON
T6IC
T7
b FF48
b FF68
A4
B4
GPT2 Timer 6 Control Register
GPT2 Timer 6 Interrupt Control Register
CAPCOM Timer 7 Register
H
H
H
H
F050 E 28
H
T78CON b FF20
90
CAPCOM Timer 7 and 8 Control Register
CAPCOM Timer 7 Interrupt Ctrl. Reg.
CAPCOM Timer 7 Reload Register
CAPCOM Timer 8 Register
H
T7IC
T7REL
T8
b F17A E BD
H H
F054 E 2A
H
H
F052 E 29
H
H
T8IC
T8REL
TFR
b F17C E BE
CAPCOM Timer 8 Interrupt Ctrl. Reg.
CAPCOM Timer 8 Reload Register
Trap Flag Register
H
H
F056 E 2B
H
H
b FFAC
D6
H
H
H
H
WDT
FEAE
57
Watchdog Timer Register (read only)
Watchdog Timer Control Register
CAN1 Interrupt Control Register
Unassigned Interrupt Control Register
Unassigned Interrupt Control Register
PLL/OWD Interrupt Control Register
Constant Value 0’s Register (read only)
H
2)
WDTCON b FFAE
D7
00xx
H
H
H
H
XP0IC
XP1IC
XP2IC
XP3IC
b F186 E C3
0000
0000
0000
0000
0000
H
H
H
H
H
H
b F18E E C7
H
b F196 E CB
H
H
H
H
b F19E E CF
H
ZEROS
b FF1C
8E
H
1)
The system configuration is selected during reset.
2)
The reset value depends on the indicated reset source.
User’s Manual
22-11
V3.1, 2000-03
C167CR
Derivatives
The Register Set
22.4
Special Function Registers Ordered by Address
Table 22-4 lists all SFRs which are implemented in the C167CR ordered by their
physical address. Bit-addressable SFRs are marked with the letter “b” in column
“Name”.
SFRs within the extended SFR-space (ESFRs) are marked with the letter “E” in column
“Physical Address”. Registers within on-chip X-Peripherals are marked with the letter “X”
in column “Physical Address”.
Table 22-4 C167CR Registers, Ordered by Address
Name
Physical 8-bit Description
Address Addr.
Reset
Value
C1CSR
EF00 X –
CAN1 Control/Status Register
CAN1 Interrupt Register
XX01
XX
H
H
H
H
H
H
H
H
H
H
H
H
H
C1IR
EF02 X –
H
C1BTR
EF04 X –
CAN1 Bit Timing Register
UUUU
UFUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UUUU
UU
H
C1GMS
C1UGML
C1LGML
C1UMLM
C1LMLM
C1MCRn
C1UARn
C1LARn
C1MCFGn
EF06 X –
CAN1 Global Mask Short
H
EF08 X –
CAN1 Upper Global Mask Long
CAN1 Lower Global Mask Long
CAN1 Upper Mask of Last Message
CAN1 Lower Mask of Last Message
CAN1 Message Ctrl. Reg. (msg. n)
CAN1 Upper Arbitration Reg. (msg. n)
CAN1 Lower Arbitration Reg. (msg. n)
H
EF0A X –
H
EF0C X –
H
EF0E X –
H
EFn0 X –
H
EFn2 X –
H
EFn4 X –
H
EFn6 X –
CAN1 Message Configuration Register
H
(msg. n)
PT0
PT1
PT2
PT3
PP0
PP1
PP2
PP3
T7
F030 E 18
PWM Module Up/Down Counter 0
PWM Module Up/Down Counter 1
PWM Module Up/Down Counter 2
PWM Module Up/Down Counter 3
PWM Module Period Register 0
PWM Module Period Register 1
PWM Module Period Register 2
PWM Module Period Register 3
CAPCOM Timer 7 Register
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
F032 E 19
H
F034 E 1A
H
H
H
F036 E 1B
H
F038 E 1C
H
H
H
H
H
H
H
F03A E 1D
H
F03C E 1E
H
F03E E 1F
H
F050 E 28
H
T8
F052 E 29
CAPCOM Timer 8 Register
H
User’s Manual
22-12
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-4 C167CR Registers, Ordered by Address (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
T7REL
T8REL
ADDAT2
PDCR
SSCTB
SSCRB
SSCBR
DP0L
F054 E 2A
CAPCOM Timer 7 Reload Register
CAPCOM Timer 8 Reload Register
A/D Converter 2 Result Register
Port Driver Control Register
SSC Transmit Buffer
0000
0000
0000
0000
0000
XXXX
0000
00
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
F056 E 2B
H
F0A0 E 50
H
F0AA E 55
H
F0B0 E 58
H
F0B2 E 59
SSC Receive Buffer
H
F0B4 E 5A
SSC Baudrate Register
H
H
H
H
H
H
H
b F100 E 80
P0L Direction Control Register
P0H Direction Control Register
P1L Direction Control Register
P1H Direction Control Register
H
DP0H
b F102 E 81
00
H
DP1L
b F104 E 82
00
H
DP1H
b F106 E 83
00
H
RP0H
b F108 E 84
System Startup Configuration Register
(read only)
XX
H
CC16IC
CC17IC
CC18IC
CC19IC
CC20IC
CC21IC
CC22IC
CC23IC
CC24IC
CC25IC
CC26IC
CC27IC
CC28IC
T7IC
b F160 E B0
CAPCOM Register 16 Interrupt Ctrl. Reg.
CAPCOM Register 17 Interrupt Ctrl. Reg.
CAPCOM Register 18 Interrupt Ctrl. Reg.
CAPCOM Register 19 Interrupt Ctrl. Reg.
CAPCOM Register 20 Interrupt Ctrl. Reg.
CAPCOM Register 21 Interrupt Ctrl. Reg.
CAPCOM Register 22 Interrupt Ctrl. Reg.
CAPCOM Register 23 Interrupt Ctrl. Reg.
CAPCOM Register 24 Interrupt Ctrl. Reg.
CAPCOM Register 25 Interrupt Ctrl. Reg.
CAPCOM Register 26 Interrupt Ctrl. Reg.
CAPCOM Register 27 Interrupt Ctrl. Reg.
CAPCOM Register 28 Interrupt Ctrl. Reg.
CAPCOM Timer 7 Interrupt Ctrl. Reg.
CAPCOM Timer 8 Interrupt Ctrl. Reg.
PWM Module Interrupt Control Register
CAPCOM Register 29 Interrupt Ctrl. Reg.
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
b F162 E B1
H
b F164 E B2
H
b F166 E B3
H
b F168 E B4
H
b F16A E B5
H
b F16C E B6
H
b F16E E B7
H
b F170 E B8
H
b F172 E B9
H
b F174 E BA
H
H
H
b F176 E BB
H
b F178 E BC
H
H
H
H
H
H
b F17A E BD
H
T8IC
b F17C E BE
H
PWMIC
CC29IC
b F17E E BF
H
b F184 E C2
H
User’s Manual
22-13
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-4 C167CR Registers, Ordered by Address (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
XP0IC
b F186 E C3
CAN1 Interrupt Control Register
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
CC30IC
XP1IC
b F18C E C6
CAPCOM Register 30 Interrupt Ctrl. Reg.
Unassigned Interrupt Control Register
CAPCOM Register 31 Interrupt Ctrl. Reg.
Unassigned Interrupt Control Register
H
b F18E E C7
H
CC31IC
XP2IC
b F194 E CA
H
H
H
H
b F196 E CB
H
S0TBIC
b F19C E CE
Serial Channel 0 Transmit Buffer Interrupt
Control Register
H
XP3IC
b F19E E CF
PLL/OWD Interrupt Control Register
External Interrupt Control Register
Port 2 Open Drain Control Register
Port Input Threshold Control Register
Port 3 Open Drain Control Register
Port 6 Open Drain Control Register
Port 7 Open Drain Control Register
Port 8 Open Drain Control Register
0000
0000
0000
0000
0000
00
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
EXICON b F1C0 E E0
H
ODP2
PICON
ODP3
ODP6
ODP7
ODP8
DPP0
DPP1
DPP2
DPP3
CSP
b F1C2 E E1
H
F1C4 E E2
H
b F1C6 E E3
H
b F1CE E E7
H
b F1D2 E E9
00
H
b F1D6 E EB
00
H
H
FE00
FE02
FE04
FE06
FE08
00
01
02
03
04
CPU Data Page Pointer 0 Register (10 bits) 0000
CPU Data Page Pointer 1 Register (10 bits) 0001
CPU Data Page Pointer 2 Register (10 bits) 0002
CPU Data Page Pointer 3 Register (10 bits) 0003
H
H
H
H
H
H
H
H
H
H
CPU Code Segment Pointer Register
(8 bits, not directly writeable)
0000
MDH
MDL
CP
FE0C
06
07
08
09
CPU Multiply Divide Register – High Word
CPU Multiply Divide Register – Low Word
CPU Context Pointer Register
0000
0000
FC00
FC00
FA00
FC00
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
FE0E
FE10
FE12
FE14
FE16
SP
CPU System Stack Pointer Register
CPU Stack Overflow Pointer Register
CPU Stack Underflow Pointer Register
Address Select Register 1
STKOV
STKUN
0A
0B
H
H
ADDRSEL1 FE18
ADDRSEL2 FE1A
0C
0D
H
H
H
Address Select Register 2
H
ADDRSEL3 FE1C
0E
Address Select Register 3
H
User’s Manual
22-14
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-4 C167CR Registers, Ordered by Address (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
ADDRSEL4 FE1E
0F
Address Select Register 4
PWM Module Pulse Width Register 0
PWM Module Pulse Width Register 1
PWM Module Pulse Width Register 2
PWM Module Pulse Width Register 3
GPT1 Timer 2 Register
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
PW0
PW1
PW2
PW3
T2
FE30
FE32
FE34
FE36
FE40
FE42
FE44
FE46
FE48
18
19
1A
1B
H
H
H
H
H
H
H
H
H
H
20
21
22
23
24
25
28
29
T3
GPT1 Timer 3 Register
T4
GPT1 Timer 4 Register
T5
GPT2 Timer 5 Register
T6
GPT2 Timer 6 Register
CAPREL
T0
FE4A
GPT2 Capture/Reload Register
CAPCOM Timer 0 Register
CAPCOM Timer 1 Register
CAPCOM Timer 0 Reload Register
CAPCOM Timer 1 Reload Register
CAPCOM Register 16
H
H
H
H
H
H
H
H
H
H
FE50
FE52
FE54
FE56
FE60
FE62
FE64
FE66
FE68
T1
T0REL
T1REL
CC16
CC17
CC18
CC19
CC20
CC21
CC22
CC23
CC24
CC25
CC26
CC27
CC28
CC29
CC30
2A
2B
H
H
H
H
H
H
H
H
H
H
H
H
30
31
32
33
34
35
36
37
38
39
CAPCOM Register 17
CAPCOM Register 18
CAPCOM Register 19
CAPCOM Register 20
FE6A
CAPCOM Register 21
H
FE6C
CAPCOM Register 22
H
H
H
H
H
H
H
FE6E
CAPCOM Register 23
FE70
FE72
FE74
FE76
FE78
CAPCOM Register 24
CAPCOM Register 25
3A
3B
CAPCOM Register 26
H
H
CAPCOM Register 27
3C
3D
CAPCOM Register 28
H
H
H
FE7A
CAPCOM Register 29
H
FE7C
3E
CAPCOM Register 30
H
User’s Manual
22-15
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-4 C167CR Registers, Ordered by Address (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
CC31
CC0
FE7E
3F
CAPCOM Register 31
CAPCOM Register 0
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
XXXX
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
FE80
FE82
FE84
FE86
FE88
40
41
42
43
44
45
46
47
48
49
CC1
CAPCOM Register 1
CC2
CAPCOM Register 2
CC3
CAPCOM Register 3
CC4
CAPCOM Register 4
CC5
FE8A
CAPCOM Register 5
H
CC6
FE8C
CAPCOM Register 6
H
H
H
H
H
H
H
CC7
FE8E
CAPCOM Register 7
CC8
FE90
FE92
FE94
FE96
FE98
CAPCOM Register 8
CC9
CAPCOM Register 9
CC10
CC11
CC12
CC13
CC14
CC15
ADDAT
WDT
S0TBUF
S0RBUF
4A
4B
CAPCOM Register 10
CAPCOM Register 11
CAPCOM Register 12
CAPCOM Register 13
CAPCOM Register 14
CAPCOM Register 15
A/D Converter Result Register
Watchdog Timer Register (read only)
Serial Channel 0 Transmit Buffer Register
H
H
4C
4D
H
H
H
H
H
H
H
H
FE9A
H
FE9C
4E
4F
H
H
H
FE9E
FEA0
50
57
58
59
FEAE
H
H
H
FEB0
FEB2
Serial Channel 0 Receive Buffer Register
(read only)
S0BG
FEB4
5A
Serial Channel 0 Baud Rate Generator
Reload Register
0000
H
H
H
PECC0
PECC1
PECC2
PECC3
PECC4
PECC5
FEC0
FEC2
FEC4
FEC6
FEC8
60
61
62
63
64
65
PEC Channel 0 Control Register
PEC Channel 1 Control Register
PEC Channel 2 Control Register
PEC Channel 3 Control Register
PEC Channel 4 Control Register
PEC Channel 5 Control Register
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
FECA
H
User’s Manual
22-16
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-4 C167CR Registers, Ordered by Address (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
PECC6
PECC7
P0L
FECC
66
67
80
81
82
83
86
87
88
89
PEC Channel 6 Control Register
PEC Channel 7 Control Register
Port 0 Low Register (Lower half of PORT0)
Port 0 High Register (Upper half of PORT0)
Port 1 Low Register (Lower half of PORT1)
Port 1 High Register (Upper half of PORT1)
Bus Configuration Register 0
0000
0000
00
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
FECE
b FF00
b FF02
b FF04
b FF06
H
P0H
00
H
H
H
P1L
00
P1H
00
BUSCON0 b FF0C
0000
0000
0000
H
H
H
H
H
H
H
MDC
PSW
b FF0E
b FF10
CPU Multiply Divide Control Register
CPU Program Status Word
1)
SYSCON b FF12
BUSCON1 b FF14
BUSCON2 b FF16
BUSCON3 b FF18
CPU System Configuration Register
Bus Configuration Register 1
0xx0
H
8A
8B
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Bus Configuration Register 2
8C
8D
Bus Configuration Register 3
H
H
H
H
H
H
H
H
H
H
H
BUSCON4 b FF1A
Bus Configuration Register 4
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
ZEROS
ONES
b FF1C
b FF1E
8E
8F
Constant Value 0’s Register (read only)
Constant Value 1’s Register (read only)
CAPCOM Timer 7 and 8 Control Register
CAPCOM Mode Control Register 4
CAPCOM Mode Control Register 5
CAPCOM Mode Control Register 6
CAPCOM Mode Control Register 7
PWM Module Control Register 0
PWM Module Control Register 1
GPT1 Timer 2 Control Register
FFFF
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
T78CON b FF20
90
91
92
93
94
98
99
CCM4
CCM5
CCM6
CCM7
b FF22
b FF24
b FF26
b FF28
PWMCON0b FF30
PWMCON1b FF32
T2CON
T3CON
T4CON
T5CON
T6CON
b FF40
b FF42
b FF44
b FF46
b FF48
A0
A1
A2
A3
A4
A8
A9
H
H
H
H
H
H
H
GPT1 Timer 3 Control Register
GPT1 Timer 4 Control Register
GPT2 Timer 5 Control Register
GPT2 Timer 6 Control Register
T01CON b FF50
CCM0 b FF52
CAPCOM Timer 0 and Timer 1 Ctrl. Reg.
CAPCOM Mode Control Register 0
User’s Manual
22-17
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-4 C167CR Registers, Ordered by Address (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
CCM1
CCM2
CCM3
T2IC
b FF54
b FF56
b FF58
b FF60
b FF62
b FF64
b FF66
b FF68
AA
AB
CAPCOM Mode Control Register 1
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CAPCOM Mode Control Register 2
AC
CAPCOM Mode Control Register 3
H
B0
B1
B2
B3
B4
B5
B6
GPT1 Timer 2 Interrupt Control Register
GPT1 Timer 3 Interrupt Control Register
GPT1 Timer 4 Interrupt Control Register
GPT2 Timer 5 Interrupt Control Register
GPT2 Timer 6 Interrupt Control Register
GPT2 CAPREL Interrupt Control Register
H
T3IC
H
H
H
H
H
H
T4IC
T5IC
T6IC
CRIC
S0TIC
b FF6A
b FF6C
H
H
Serial Channel 0 Transmit Interrupt Control
Register
S0RIC
b FF6E
B7
Serial Channel 0 Receive Interrupt Control
Register
0000
H
H
H
S0EIC
b FF70
b FF72
B8
B9
Serial Channel 0 Error Interrupt Ctrl. Reg.
SSC Transmit Interrupt Control Register
SSC Receive Interrupt Control Register
SSC Error Interrupt Control Register
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
SSCTIC
SSCRIC b FF74
BA
BB
H
H
SSCEIC
CC0IC
CC1IC
CC2IC
CC3IC
CC4IC
CC5IC
CC6IC
CC7IC
CC8IC
CC9IC
CC10IC
CC11IC
CC12IC
b FF76
b FF78
BC
BD
CAPCOM Register 0 Interrupt Ctrl. Reg.
CAPCOM Register 1 Interrupt Ctrl. Reg.
CAPCOM Register 2 Interrupt Ctrl. Reg.
CAPCOM Register 3 Interrupt Ctrl. Reg.
CAPCOM Register 4 Interrupt Ctrl. Reg.
CAPCOM Register 5 Interrupt Ctrl. Reg.
CAPCOM Register 6Interrupt Ctrl. Reg.
CAPCOM Register 7 Interrupt Ctrl. Reg.
CAPCOM Register 8 Interrupt Ctrl. Reg.
CAPCOM Register 9 Interrupt Ctrl. Reg.
CAPCOM Register 10 Interrupt Ctrl. Reg.
CAPCOM Register 11 Interrupt Ctrl. Reg.
CAPCOM Register 12 Interrupt Ctrl. Reg.
H
H
H
H
H
H
H
H
H
H
H
H
H
b FF7A
b FF7C
b FF7E
H
H
H
H
H
H
H
H
BE
BF
C0
C1
C2
C3
C4
C5
C6
C7
C8
b FF80
b FF82
b FF84
b FF86
b FF88
b FF8A
b FF8C
b FF8E
H
H
H
H
b FF90
User’s Manual
22-18
V3.1, 2000-03
C167CR
Derivatives
The Register Set
Table 22-4 C167CR Registers, Ordered by Address (cont’d)
Name
Physical 8-bit Description
Address Addr.
Reset
Value
CC13IC
CC14IC
CC15IC
ADCIC
b FF92
b FF94
b FF96
b FF98
C9
CAPCOM Register 13 Interrupt Ctrl. Reg.
CAPCOM Register 14 Interrupt Ctrl. Reg.
CAPCOM Register 15 Interrupt Ctrl. Reg.
0000
0000
0000
0000
H
H
H
H
H
H
H
H
H
CA
CB
H
H
CC
A/D Converter End of Conversion Interrupt
Control Register
H
ADEIC
b FF9A
CD
A/D Converter Overrun Error Interrupt
Control Register
0000
H
H
H
T0IC
T1IC
ADCON
P5
b FF9C
b FF9E
b FFA0
b FFA2
CE
CAPCOM Timer 0 Interrupt Ctrl. Reg.
CAPCOM Timer 1 Interrupt Ctrl. Reg.
A/D Converter Control Register
Port 5 Register (read only)
Port 5 Digital Input Disable Register
Trap Flag Register
0000
0000
0000
XXXX
0000
0000
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CF
D0
D1
D2
D6
D7
D8
D9
P5DIDIS b FFA4
TFR
b FFAC
H
H
H
H
2)
WDTCON b FFAE
Watchdog Timer Control Register
Serial Channel 0 Control Register
SSC Control Register
00xx
H
S0CON
b FFB0
0000
0000
0000
0000
0000
0000
00
H
H
H
H
H
H
H
H
H
H
H
H
H
H
SSCCON b FFB2
P2
b FFC0
E0
E1
E2
E3
E4
E5
E6
E7
E8
E9
Port 2 Register
H
H
H
H
H
DP2
P3
b FFC2
b FFC4
b FFC6
b FFC8
Port 2 Direction Control Register
Port 3 Register
DP3
P4
Port 3 Direction Control Register
Port 4 Register (7 bits)
DP4
P6
b FFCA
Port 4 Direction Control Register
Port 6 Register (8 bits)
00
H
b FFCC
00
H
H
H
H
H
H
DP6
P7
b FFCE
Port 6 Direction Control Register
Port 7 Register (8 bits)
00
b FFD0
b FFD2
b FFD4
b FFD6
00
DP7
P8
Port 7 Direction Control Register
Port 8 Register (8 bits)
00
EA
EB
00
H
H
DP8
Port 8 Direction Control Register
00
1)
The system configuration is selected during reset.
2)
The reset value depends on the indicated reset source.
User’s Manual
22-19
V3.1, 2000-03
C167CR
Derivatives
The Register Set
22.5
Special Notes
PEC Pointer Registers
The source and destination pointers for the peripheral event controller are mapped to a
special area within the internal RAM. Pointers that are not occupied by the PEC may
therefore be used like normal RAM. During Power Down mode or any warm reset the
PEC pointers are preserved.
The PEC and its registers are described in Chapter 5.
GPR Access in the ESFR Area
The locations 00’F000 … 00’F01E within the ESFR area are reserved and allow to
H
H
access the current register bank via short register addressing modes. The GPRs are
mirrored to the ESFR area which allows access to the current register bank even after
switching register spaces (see example below).
MOV
R5, DP3
;GPR access via SFR area
EXTR #1
MOV
R5, ODP3
;GPR access via ESFR area
Writing Bytes to SFRs
All special function registers may be accessed wordwise or bytewise (some of them even
bitwise). Reading bytes from word SFRs is a non-critical operation. However, when
writing bytes to word SFRs the complementary byte of the respective SFR is cleared with
the write operation.
User’s Manual
22-20
V3.1, 2000-03
C167CR
Derivatives
Instruction Set Summary
23
Instruction Set Summary
This chapter briefly summarizes the C167CR’s instructions ordered by instruction
classes. This provides a basic understanding of the C167CR’s instruction set, the power
and versatility of the instructions and their general usage.
A detailed description of each single instruction, including its operand data type,
condition flag settings, addressing modes, length (number of bytes) and object code
format is provided in the “Instruction Set Manual” for the C166 Family. This manual
also provides tables ordering the instructions according to various criteria, to allow quick
references.
Summary of Instruction Classes
Grouping the various instruction into classes aids in identifying similar instructions (e.g.
SHR, ROR) and variations of certain instructions (e.g. ADD, ADDB). This provides an
easy access to the possibilities and the power of the instructions of the C167CR.
Note: The used mnemonics refer to the detailed description.
Arithmetic Instructions
• Addition of two words or bytes:
• Addition with Carry of two words or bytes:
• Subtraction of two words or bytes:
• Subtraction with Carry of two words or bytes:
• 16 ×16 bit signed or unsigned multiplication:
• 16 / 16 bit signed or unsigned division:
• 32 / 16 bit signed or unsigned division:
• 1’s complement of a word or byte:
ADD
ADDC
SUB
SUBC
MUL
DIV
ADDB
ADDCB
SUBB
SUBCB
MULU
DIVU
DIVL
CPL
DIVLU
CPLB
• 2’s complement (negation) of a word or byte:
NEG
NEGB
Logical Instructions
• Bitwise ANDing of two words or bytes:
• Bitwise ORing of two words or bytes:
• Bitwise XORing of two words or bytes:
AND
OR
XOR
ANDB
ORB
XORB
Compare and Loop Control Instructions
• Comparison of two words or bytes:
• Comparison of two words with post-increment
by either 1 or 2:
• Comparison of two words with post-decrement
by either 1 or 2:
CMP
CMPB
CMPI2
CMPD2
CMPI1
CMPD1
User’s Manual
23-1
V3.1, 2000-03
C167CR
Derivatives
Instruction Set Summary
Boolean Bit Manipulation Instructions
• Manipulation of a maskable bit field
in either the high or the low byte of a word:
• Setting a single bit (to ‘1’):
• Clearing a single bit (to ‘0’):
• Movement of a single bit:
• Movement of a negated bit:
• ANDing of two bits:
BFLDH
BFLDL
BSET
BCLR
BMOV
BMOVN
BAND
BOR
• ORing of two bits:
• XORing of two bits:
• Comparison of two bits:
BXOR
BCMP
Shift and Rotate Instructions
• Shifting right of a word:
• Shifting left of a word:
• Rotating right of a word:
• Rotating left of a word:
SHR
SHL
ROR
ROL
• Arithmetic shifting right of a word (sign bit shifting): ASHR
Prioritize Instruction
• Determination of the number of shift cycles required
to normalize a word operand (floating point support): PRIOR
Data Movement Instructions
• Standard data movement of a word or byte:
• Data movement of a byte to a word location
with either sign or zero byte extension:
MOV
MOVB
MOVBS
MOVBZ
Note: The data movement instructions can be used with a big number of different
addressing modes including indirect addressing and automatic pointer in-/
decrementing.
System Stack Instructions
• Pushing of a word onto the system stack:
• Popping of a word from the system stack:
• Saving of a word on the system stack,
and then updating the old word with a new value
(provided for register bank switching):
PUSH
POP
SCXT
User’s Manual
23-2
V3.1, 2000-03
C167CR
Derivatives
Instruction Set Summary
Jump Instructions
• Conditional jumping to an either absolutely,
indirectly, or relatively addressed target instruction
within the current code segment:
JMPA
JMPS
JMPI
JNB
JMPR
• Unconditional jumping to an absolutely addressed
target instruction within any code segment:
• Conditional jumping to a relatively addressed
target instruction within the current code segment
depending on the state of a selectable bit:
• Conditional jumping to a relatively addressed
target instruction within the current code segment
depending on the state of a selectable bit
with a post-inversion of the tested bit
JB
in case of jump taken (semaphore support):
JBC
JNBS
Call Instructions
• Conditional calling of an either absolutely
or indirectly addressed subroutine within the current
code segment:
• Unconditional calling of a relatively addressed
subroutine within the current code segment:
• Unconditional calling of an absolutely addressed
subroutine within any code segment:
CALLA
CALLR
CALLS
CALLI
• Unconditional calling of an absolutely addressed
subroutine within the current code segment plus
an additional pushing of a selectable register onto
the system stack:
• Unconditional branching to the interrupt or
trap vector jump table in code segment 0:
PCALL
TRAP
Return Instructions
• Returning from a subroutine
within the current code segment:
RET
• Returning from a subroutine
within any code segment:
RETS
• Returning from a subroutine within the current
code segment plus an additional popping of a
selectable register from the system stack:
• Returning from an interrupt service routine:
RETP
RETI
User’s Manual
23-3
V3.1, 2000-03
C167CR
Derivatives
Instruction Set Summary
System Control Instructions
• Resetting the C167CR via software:
• Entering the Idle mode:
SRST
IDLE
• Entering the Power Down mode:
• Servicing the Watchdog Timer:
• Disabling the Watchdog Timer:
PWRDN
SRVWDT
DISWDT
• Signifying the end of the initialization routine
(pulls pin RSTOUT high, and disables the effect of
any later execution of a DISWDT instruction):
EINIT
Miscellaneous
• Null operation which requires 2 Bytes of
storage and the minimum time for execution:
NOP
• Definition of an unseparable instruction sequence: ATOMIC
• Switch ‘reg’, ‘bitoff’ and ‘bitaddr’ addressing modes
to the Extended SFR space:
EXTR
EXTP
EXTS
• Override the DPP addressing scheme
using a specific data page instead of the DPPs,
and optionally switch to ESFR space:
• Override the DPP addressing scheme
using a specific segment instead of the DPPs,
and optionally switch to ESFR space:
EXTPR
EXTSR
Note: The ATOMIC and EXT* instructions provide support for uninterruptable code
sequences e.g. for semaphore operations. They also support data addressing
beyond the limits of the current DPPs (except ATOMIC), which is advantageous
for bigger memory models in high level languages. Refer to Chapter 21 for
examples.
Protected Instructions
Some instructions of the C167CR which are critical for the functionality of the controller
are implemented as so-called Protected Instructions. These protected instructions use
the maximum instruction format of 32 bits for decoding, while the regular instructions
only use a part of it (e.g. the lower 8 bits) with the other bits providing additional
information like involved registers. Decoding all 32 bits of a protected doubleword
instruction increases the security in cases of data distortion during instruction fetching.
Critical operations like a software reset are therefore only executed if the complete
instruction is decoded without an error. This enhances the safety and reliability of a
microcontroller system.
User’s Manual
23-4
V3.1, 2000-03
C167CR
Derivatives
Device Specification
24
Device Specification
The device specification describes the electrical parameters of the device. It lists DC
characteristics like input, output or supply voltages or currents, and AC characteristics
like timing characteristics and requirements.
Other than the architecture, the instruction set or the basic functions of the C167CR core
and its peripherals, these DC and AC characteristics are subject to changes due to
device improvements or specific derivatives of the standard device.
Therefore these characteristics are not contained in this manual, but rather provided in
a separate Data Sheet, which can be updated more frequently.
Please refer to the current version of the Data Sheet of the respective device for all
electrical parameters.
Note: In any case the specific characteristics of a device should be verified, before a new
design is started. This ensures that the used information is up to date.
Figure 24-1 shows the pin diagram of the C167CR. It shows the location of the different
supply and IO pins. A detailed description of all the pins is also found in the Data Sheet.
Note: Not all alternate functions shown in the figure below are supported by all
derivatives.
Please refer to the corresponding descriptions in the data sheets.
User’s Manual
24-1
V3.1, 2000-03
C167CR
Derivatives
Device Specification
P6.0/CS0
P6.1/CS1
P6.2/CS2
P6.3/CS3
P6.4/CS4
1
2
3
4
5
6
7
8
9
108
107
106
105
104
103
102
101
100
99
98
97
96
95
94
93
P0L.0/AD8
P0L.7/AD7
P0L.6/AD6
P0L.5/AD5
P0L.4/AD4
P0L.3/AD3
P0L.2/AD2
P0L.1/AD1
P0L.0/AD0
EA
ALE
READY
WR/WRL
RD
VSS
P6.5/HOLD
P6.6/HLDA
P6.7/BREQ
P8.0/CC16IO
P8.1/CC17IO 10
P8.2/CC18IO 11
P8.3/CC19IO 12
P8.4/CC20IO 13
P8.5/CC21IO 14
P8.6/CC22IO 15
P8.7/CC23IO 16
VDD
17
18
19
92
91
90
89
88
87
86
85
84
83
82
81
80
79
78
77
P4.7/A23
P4.6/A22/CAN1_TxD
P4.5/A21/CAN1_RxD
P4.4/A20
P4.3/A19
P4.2/A18
P4.1/A17
P4.0/A16
OWE
VDD
VSS
P7.0/POUT0
C 167CR
P7.1/POUT1 20
P7.2/POUT2 21
P7.3/POUT3 22
P7.4/CC28IO 23
P7.5/CC29IO 24
P7.6/CC30IO 25
P7.7/CC31IO 26
P5.0/AN0 27
P5.1/AN1 28
P5.2/AN2 29
P5.3/AN3 30
P5.4/AN4 31
P5.5/AN5 32
P5.6/AN6 33
P5.7/AN7 34
P5.8/AN8 35
P5.9/AN9 36
VSS
VDD
P3.15/CLKOUT
P3.13/SCLK
P3.12/BHE/WRH
P3.111/RxD0
P3.10/TxD0
P3.9/MTSR
P3.8/MRST
P3.7/T2IN
P3.6/T3IN
76
75
74
73
MCP04410
Figure 24-1 Pin Configuration P-MQFP-144 Package (top view)
Note: Signals CAN1_TxD and CAN1_RxD are not available in all derivatives of the
C167CR.
User’s Manual
24-2
V3.1, 2000-03
C167CR
Derivatives
Keyword Index
25
Keyword Index
This section lists a number of keywords which refer to specific details of the C167CR in
terms of its architecture, its functional units or functions. This helps to quickly find the
answer to specific questions about the C167CR.
ASC0 11-11
Bootstrap Loader 14-6
SSC 12-13
A
Access to X-Peripherals 9-37
Acronyms 1-7
Adapt Mode 19-15
ADC 2-15, 17-1
ADCIC, ADEIC 17-14
BHE 7-26, 9-9
Bidirectional reset 19-4
Bit
addressable memory 3-4
ADCON 17-3
Handling 4-10
Manipulation Instructions 23-2
ADDAT, ADDAT2 17-5
Address
protected 2-18, 4-10
Arbitration 9-25
reserved 2-12
Area Definition 9-24
Bootstrap Loader 14-1, 19-16
Boundaries 3-12
Boundaries 3-12
BTR 18-12
Segment 9-9, 19-18
ADDRSELx 9-23, 9-25
ALE length 9-13
Alternate signals 7-7
ALU 4-16
Analog/Digital Converter 2-15, 17-1
Arbitration
Burst mode (PWM) 16-7
Bus
Arbitration 9-29
CAN 2-14, 18-1, 18-36
Demultiplexed 9-5
Idle State 9-28
Address 9-25
Mode Configuration 9-3, 19-17
External Bus 9-29
Multiplexed 9-4
Master/Slave mode 9-33
Signals 9-30
BUSCONx 9-21, 9-25
ASC0 11-1
C
Asynchronous mode 11-5
CAN Interface 2-14, 18-1
Baudrate 11-11
Error Detection 11-10
Interrupts 11-15
activation 18-31
port control 18-37
CAPCOM 2-17
Synchronous mode 11-8
interrupt 15-22
timer 15-4
unit 15-1
Asynchronous Serial Interface (->ASC0)
11-1
Auto Scan conversion 17-6
Capture Mode
CAPCOM 15-13
GPT1 10-20
GPT2 (CAPREL) 10-34
B
Baudrate
User’s Manual
25-1
V3.1, 2000-03
C167CR
Derivatives
Keyword Index
Capture/Compare unit 15-1
CCM0, CCM1, CCM2, CCM3 15-10
CCM4, CCM5, CCM6, CCM7 15-11
CCxIC 15-22
Demultiplexed Bus 9-5
Development Support 1-6
Direct Drive 6-5
Direction
Center aligned PWM 16-5
Chip Select
count 10-4, 10-25
Disable
Configuration 9-10, 19-18
Latched/Early 9-11
Clock
Interrupt 5-16
Segmentation 4-15
Division 4-30, 21-1
distribution 6-1
Double-Register compare 15-19
DP0L, DP0H 7-10
DP1L, DP1H 7-14
DP2 7-18
DP3 7-23
DP4 7-28, 7-36, 7-42
DP8 7-47
generator modes 6-7, 19-19
Code memory handling 21-16
Compare modes 15-14
double register 15-19
Concatenation of Timers 10-16, 10-33
Configuration
Address 9-9, 19-18
Bus Mode 9-3, 19-17
Chip Select 9-10, 19-18
PLL 6-7, 19-19
DPP 4-22, 21-14
E
Early chip select 9-11
Edge aligned PWM 16-3
Edge characteristic (ports) 7-5
Emulation Mode 19-14
Enable
Reset 19-7, 19-12
special modes 19-16
Write Control 19-17
Context
Pointer 4-25
Interrupt 5-16
Context Switching 5-19
Conversion
Segmentation 4-15
XBUS peripherals 9-36
Error Detection
analog/digital 17-1
Auto Scan 17-6
ASC0 11-10
timing control 17-12
Count direction 10-4, 10-25
Counter 10-8, 10-14, 10-29, 10-31, 16-10
CP 4-25
CAN 18-4
SSC 12-15
EXICON 5-27
External
CPU 2-2, 4-1
Bus 2-10
CRIC 10-38
CSP 4-20
CSR 18-7
Bus Characteristics 9-12–9-18
Bus Idle State 9-28
Bus Modes 9-3–9-8
Fast interrupts 5-27
Interrupts 5-25
D
Data Page 4-22, 21-14
boundaries 3-12
Delay
startup configuration 19-13
F
Read/Write 9-16
Fast external interrupts 5-27
User’s Manual
25-2
V3.1, 2000-03
C167CR
Derivatives
Keyword Index
Flags 4-16–4-19
Fast external 5-27
Full Duplex 12-7
Handling CAN 18-9
Priority 5-8
G
Processing 5-1, 5-6
Response Times 5-20
Sources 5-2
System 2-7, 5-2
Vectors 5-2
GMS 18-15
GPR 3-6, 4-25, 22-2
GPT 2-16
GPT1 10-1
GPT2 10-22
IP 4-19
IRAM 3-4
status after reset 19-8
H
Half Duplex 12-10
Hardware
Reset 19-2
Traps 5-28
Hold State
Entry 9-31
Exit 9-32
L
LARn 18-21
Latched chip select 9-11
LGML 18-16
LMLM 18-17
M
I
Management
Power 20-1
Master mode
External bus 9-33
MCFGn 18-22
MCRn 18-19
MDC 4-32
Idle
Mode 20-2
State (Bus) 9-28
Incremental Interface 10-9
Indication of reset source 13-6
Input threshold 7-2
Instruction 21-1, 23-1
Bit Manipulation 23-2
Branch 4-4
MDH 4-30
MDL 4-31
Memory 2-8
bit-addressable 3-4
Pipeline 4-3
protected 23-4
Code memory handling 21-16
Timing 4-11
unseparable 21-14
Interface
External 3-11
RAM/SFR 3-4
ROM area 3-3
CAN 2-14, 18-1
Tri-state time 9-15
XRAM 3-9
Memory Cycle Time 9-14
Multiplexed Bus 9-4
Multiplication 4-30, 21-1
External Bus 9-1
serial async. (->ASC0) 11-1
serial sync. (->SSC) 12-1
Internal RAM 3-4
Interrupt
N
CAPCOM 15-22
Enable/Disable 5-16
External 5-25
NMI 5-1, 5-31
User’s Manual
25-3
V3.1, 2000-03
C167CR
Derivatives
Keyword Index
Bits 2-18, 4-10
instruction 23-4
PSW 4-16, 5-10
Pulse Width Modulation 2-18
PWM 2-18
O
ODP2 7-19, 7-37, 7-43
ODP3 7-24
ODP8 7-48
ONES 4-33
PWM Module 16-1
PWMCON0 16-12
PWMCON1 16-13
PWMIC 16-14
Open Drain Mode 7-4
Oscillator
circuitry 6-2
Watchdog 6-8
R
RAM
P
P0L, P0H 7-9
P1L, P1H 7-13
P2 7-18
extension 3-9
internal 3-4
Read/Write Delay 9-16
READY 9-17
P3 7-23
P4 7-28, 7-36, 7-42
P5 7-32
P5DIDIS 7-34
P8 7-47
PCIR 18-10
PDCR 7-6
Registers 22-1
sorted by address 22-12
sorted by name 22-4
Reserved bits 2-12
Reset 19-1
Bidirectional 19-4
Configuration 19-7, 19-12
Hardware 19-2
Output 19-8
Software 19-3
Source indication 13-6
Values 19-5
PEC 2-8, 3-7, 5-12
Response Times 5-23
PECCx 5-12
Peripheral
enable on XBUS 9-36
Summary 2-11
Phase Locked Loop (->PLL) 6-1
PICON 7-2
Watchdog Timer 19-3
Pins 8-1, 24-2
S
in Idle and Power Down mode
20-6
Pipeline 4-3
S0BG 11-11
S0CON 11-2
S0EIC, S0RIC, S0TIC, S0TBIC 11-15
S0RBUF 11-7, 11-9
S0TBUF 11-7, 11-9
Segment
Address 9-9, 19-18
boundaries 3-12
Segmentation 4-20
Enable/Disable 4-15
Serial Interface 2-13, 11-1
Asynchronous 11-5
Effects 4-6
PLL 6-1, 19-19
Port 2-14
edge characteristic 7-5
input threshold 7-2
Power Down Mode 20-4
Power Management 20-1
Prescaler 6-5
Protected
User’s Manual
25-4
V3.1, 2000-03
C167CR
Derivatives
Keyword Index
CAN 2-14, 18-1
Synchronous 11-8, 12-1
SFR 3-8, 22-4, 22-12
Sharing X-Peripherals 9-37
Single Chip Mode 9-2
T5CON 10-30
T5IC, T6IC 10-38
T6CON 10-24
T78CON 15-5
T7IC 15-9
startup configuration 19-20
T8IC 15-9
Single shot mode (PWM) 16-8
Slave mode
External bus 9-33
Software
TFR 5-30
Threshold 7-2
Timer 2-16, 10-1, 10-22
Auxiliary Timer 10-12, 10-30
CAPCOM 15-4
Reset 19-3
Traps 5-28
Concatenation 10-16, 10-33
Core Timer 10-3, 10-24
Source
Interrupt 5-2
Tools 1-6
Reset 13-6
Traps 5-28
SP 4-27
Tri-State Time 9-15
Special operation modes (config.) 19-16
SSC 12-1
U
Baudrate generation 12-13
Error Detection 12-15
Full Duplex 12-7
UARn 18-21
UGML 18-16
UMLM 18-17
Half Duplex 12-10
SSCBR 12-13
Unseparable instructions 21-14
V
SSCCON 12-2, 12-4
SSCEIC, SSCRIC, SSCTIC 12-17
SSCRB, SSCTB 12-8
Stack 3-5, 4-27, 21-4
Startup Configuration 19-7, 19-12
external reset 19-13
single-chip 19-20
STKOV 4-28
Visible mode 9-36
W
Waitstate
Memory Cycle 9-14
Tri-State 9-15
XBUS peripheral 9-36
Watchdog 2-18, 13-1
after reset 19-5
Oscillator 6-8
Reset 19-3
STKUN 4-29
Subroutine 21-10
Synchronous Serial Interface (->SSC)
12-1
SYSCON 4-13, 9-19
WDT 13-2
WDTCON 13-4
T
X
T01CON 15-5
T2CON 10-12
T2IC, T3IC, T4IC 10-21
T3CON 10-3
T4CON 10-12
XBUS 2-10, 9-35
enable peripherals 9-36
external access 9-37
waitstates 9-36
User’s Manual
25-5
V3.1, 2000-03
C167CR
Derivatives
Keyword Index
XPER-Share mode 9-37
XRAM
on-chip 3-9
status after reset 19-8
xxIC 5-7
Z
ZEROS 4-33
User’s Manual
25-6
V3.1, 2000-03
Infineon goes for Business Excellence
“Business excellence means intelligent approaches and clearly
defined processes, which are both constantly under review and
ultimately lead to good operating results.
Better operating results and business excellence mean less
idleness and wastefulness for all of us, more professional
success, more accurate information, a better overview and,
thereby, less frustration and more satisfaction.”
Dr. Ulrich Schumacher
h t t p : / / w w w . i n f i n e o n . c o m
Published by Infineon Technologies AG
相关型号:
©2020 ICPDF网 联系我们和版权申明