Hello, I'm working on the XC164X µc and I've a question concerning the storage of registers in stack when interrupt occur.
The function is written in c language
void ItFunc (void) interrupt 14 { funcCode (); }
The pre-processor code is as follow:
ItFunc PROC INTERRUPT = 14 GLOBAL ItFunc ; FUNCTION ItFunc (BEGIN RMASK = @0x3FFF)
SCXT DPP3,#3
PUSH DPP0
SCXT MDC,#16
PUSH MDH
PUSH MDL
MOV [-R0],R1
MOV [-R0],R2
MOV [-R0],R3
MOV [-R0],R4
MOV [-R0],R5
MOV [-R0],R6
MOV [-R0],R7
MOV [-R0],R8
MOV [-R0],R9
MOV [-R0],R10
MOV [-R0],R11
MOV [-R0],R12
; line 234: funcCode ();
CALL funcCode
?C0005:
MOV R12,[R0+]
MOV R11,[R0+]
MOV R10,[R0+]
MOV R9,[R0+]
MOV R8,[R0+]
MOV R7,[R0+]
MOV R6,[R0+]
MOV R5,[R0+]
MOV R4,[R0+]
MOV R3,[R0+]
MOV R2,[R0+]
MOV R1,[R0+]
POP MDL
POP MDH
POP MDC
POP DPP0
POP DPP3
RETI
; FUNCTION ItFunc (END RMASK = @0x3FFF)
ItFunc ENDP
?PR?BS_TST ENDS
As shown in this pre-processor code, we use the register R0 to save the registers R1—R12. I'd like to store the registers using the instruction PUSH R1. How can I do this? Which directive could I use that the compiler store the registers in stack using the instruction "PUSH R1" and not using "MOV [-R0], R1"
Thanks
I give you an example
In the document that joost gives to me « http://www.keil.com/support/man/docs/c166/c166_noframe.htm"
We find this example:
1 int i1, i2, i3;
2 3 void intr_func1 (void) interrupt 0x21 {
4 1 i1 = i2 * i3;
5 1 } 6
7 #pragma NOFRAME
8 void intr_func2 (void) interrupt 0x22 {
9 1 i1 = i2 * i3;
10 1 }
ASSEMBLY LISTING OF GENERATED OBJECT CODE
; FUNCTION intr_func1 (BEGIN RMASK = @0x2030)
; SOURCE LINE # 3
0000 C6871000 SCXT MDC,#010H
0004 EC06 PUSH MDH
0006 EC07 PUSH MDL
0008 ECF4 PUSH R4
000A ECF5 PUSH R5
; SOURCE LINE # 4
000C F2F50000 R MOV R5,i3
0010 F2F40200 R MOV R4,i2
0014 0B45 MUL R4,R5
0016 F2F40EFE MOV R4,MDL
001A F6070400 R MOV i1,MDL
; SOURCE LINE # 5
001E FCF5 POP R5
0020 FCF4 POP R4
0022 FC07 POP MDL
0024 FC06 POP MDH
0026 FC87 POP MDC
0028 FB88 RETI
; FUNCTION intr_func1 (END RMASK = @0x2030)
; FUNCTION intr_func2 (BEGIN RMASK = @0x2030)
; SOURCE LINE # 8
; SOURCE LINE # 9
002A F2F50000 R MOV R5,i3
002E F2F40200 R MOV R4,i2
0032 0B45 MUL R4,R5
0034 F2F40EFE MOV R4,MDL
0038 F6070400 R MOV i1,MDL
; SOURCE LINE # 10
003C FB88 RETI
; FUNCTION intr_func2 (END RMASK = @0x2030)
; ; ; ; ; ; ;
Let's concentrate on function intr_func1
; ; ;
2
3 void intr_func1 (void) interrupt 0x21 {
5 1 }
; ; ; ; ;
; ;
We see in that the ASSEMBLY LISTING OF GENERATED OBJECT CODE stores the register R4 and R5 using PUSH R4 and PUSH R5.
In my case when I've written the same function intr_func1, that the ASSEMBLY LISTING OF GENERATED OBJECT CODE stores the register R4 using MOV [-R0],R4
; ; ; ;
intr_func1 PROC INTERRUPT = 33
GLOBAL intr_func1
; line 997:
; line 998:
; line 999: i1 = i2 * i3;
MOV R5,WORD i3
MOV R4,WORD i2
MUL R4,R5
MOV WORD i1,MDL
; line 1000: }
intr_func1 ENDP
?PR?OS_TEST ENDS
So my question is, I'd like to know how in the first case the compiler store R4 in stack using PUSH, and why in my case he store R4 using MOV [-R0],R4
I hope now is more clear
Please post code in the correct way - the above text is unreadable.
There are many, many thousands of decisions the compiler makes when it generates it's code. However, that is not something you should care about, just as you shouldn't care about the algorithm for controlling the ignition system in your car.
Tou are working at the wrong conceptual level, and it doesn't matter how many questions you ask - and how many answers you receive. There will still be thousands of other issues where the compiler decides what to do, and when.
the code exist in « http://www.keil.com/support/man/docs/c166/c166_noframe.htm"
just I've made a copy of this code.
the compiler don't decides what to do, we use directives and configuration to assisit compiler what to do.
so please, the question is clear, why turn arround??
No, the compiler really do decide what to do. We write source code to inform it what we want solved, and we use directives to give it hints about how to solve it.
If it was you who decided what to do, then you wouldn't need to buy the compiler, since you would not have any need for it. The way you decide what to do, is by writing the program in assembler.
Per Westermark is refering to using the pre tags to post code on the forum.
I understand what your trying to do; give a hint to the compiler to use PUSH (system stack) instead of MOV (user stack). But it's a hint so the compiler can still decide to do otherwise.
Could you answer me/us this: -why- do you want the compiler to use PUSH instead of MOV? I don't see why it's important in the first place.
Is not important if we store registers using push or mov R0, just i'm curious to know why there are two possibility to stores registers in stack. so I think there's an option to check off to tell compiler when using push or mov R0.
thanks for advising me to use pre> to post code, it's the first time that I post in this forum.
There is checkbox in the Options for target/C166
"Save temporary variables on User Stack" If it is checked, then all registers are placed in User Stack, i.e commands like mov [-R0], Rx are used. If unchecked, then push/pop commands are used instead.
Thanks Andrey!
For those people who do not use the IDE :-) it's the SAVESYS and SAVEUSR compiler directive. I completly missed those...
See http://www.keil.com/support/man/docs/c166/c166_cm_directives.htm
Thanks MOMO, I learned something new because of your post.
-- Joost
yes, finally
thanks for you