Small Device C Compiler / Mailing Lists

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

From: Borut Razem <borut.razem@si...> - 2008-03-23 06:59:18

It happened in the svn revision #5064, rev. #5063 was OK.

Raphael, can you please take a look ASAP?

Borut

Jim Paris wrote:
> Hi,
>
> It seems pic14 is broken in 2.8.0-rc1:
>
> $ cd /tmp
> $ echo '#include <pic/pic14regs.h>' > test.c
> $ echo 'void main() {while(!RA2);}' >> test.c
> $ sdcc/bin/sdcc --version
> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.8.0 #5082 (Mar  9 2008) (UNIX)
> $ sdcc/bin/sdcc -mpic14 -p16f628a -S test.c
> $ grep -A 3 while test.asm 
> ;       .line   2; "test.c"     void main() {while(!RA2);}
>         BTFSS   INDF,2
>         GOTO    _00105_DS_
>         RETURN
>
> It should be testing PORTA, not INDF.
> This code was OK in 2.7.0:
>
> $ sdcc --version
> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.7.0 #4818 (Feb 14 2008) (UNIX) 
> $ sdcc -mpic14 -p16f628a -S test.c
> $ grep -A 3 while test.asm 
> ;       .line   2; "test.c"     void main() {while(!RA2);}
>         BANKSEL _PORTA_bits
>         BTFSS   _PORTA_bits,2
>         GOTO    _00105_DS_
>
> -jim
>

Thread view

[sdcc-devel] pic14 regression in 2.8.0-rc1

From: Jim Paris <jim@jt...> - 2008-03-23 05:56:14

Hi,

It seems pic14 is broken in 2.8.0-rc1:

$ cd /tmp
$ echo '#include <pic/pic14regs.h>' > test.c
$ echo 'void main() {while(!RA2);}' >> test.c
$ sdcc/bin/sdcc --version
SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.8.0 #5082 (Mar  9 2008) (UNIX)
$ sdcc/bin/sdcc -mpic14 -p16f628a -S test.c
$ grep -A 3 while test.asm 
;       .line   2; "test.c"     void main() {while(!RA2);}
        BTFSS   INDF,2
        GOTO    _00105_DS_
        RETURN

It should be testing PORTA, not INDF.
This code was OK in 2.7.0:

$ sdcc --version
SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.7.0 #4818 (Feb 14 2008) (UNIX) 
$ sdcc -mpic14 -p16f628a -S test.c
$ grep -A 3 while test.asm 
;       .line   2; "test.c"     void main() {while(!RA2);}
        BANKSEL _PORTA_bits
        BTFSS   _PORTA_bits,2
        GOTO    _00105_DS_

-jim

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

From: Borut Razem <borut.razem@si...> - 2008-03-23 06:59:18

It happened in the svn revision #5064, rev. #5063 was OK.

Raphael, can you please take a look ASAP?

Borut

Jim Paris wrote:
> Hi,
>
> It seems pic14 is broken in 2.8.0-rc1:
>
> $ cd /tmp
> $ echo '#include <pic/pic14regs.h>' > test.c
> $ echo 'void main() {while(!RA2);}' >> test.c
> $ sdcc/bin/sdcc --version
> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.8.0 #5082 (Mar  9 2008) (UNIX)
> $ sdcc/bin/sdcc -mpic14 -p16f628a -S test.c
> $ grep -A 3 while test.asm 
> ;       .line   2; "test.c"     void main() {while(!RA2);}
>         BTFSS   INDF,2
>         GOTO    _00105_DS_
>         RETURN
>
> It should be testing PORTA, not INDF.
> This code was OK in 2.7.0:
>
> $ sdcc --version
> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.7.0 #4818 (Feb 14 2008) (UNIX) 
> $ sdcc -mpic14 -p16f628a -S test.c
> $ grep -A 3 while test.asm 
> ;       .line   2; "test.c"     void main() {while(!RA2);}
>         BANKSEL _PORTA_bits
>         BTFSS   _PORTA_bits,2
>         GOTO    _00105_DS_
>
> -jim
>

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

From: Borut Razem <borut.razem@si...> - 2008-03-23 06:59:18

It happened in the svn revision #5064, rev. #5063 was OK.

Raphael, can you please take a look ASAP?

Borut

Jim Paris wrote:
> Hi,
>
> It seems pic14 is broken in 2.8.0-rc1:
>
> $ cd /tmp
> $ echo '#include <pic/pic14regs.h>' > test.c
> $ echo 'void main() {while(!RA2);}' >> test.c
> $ sdcc/bin/sdcc --version
> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.8.0 #5082 (Mar  9 2008) (UNIX)
> $ sdcc/bin/sdcc -mpic14 -p16f628a -S test.c
> $ grep -A 3 while test.asm 
> ;       .line   2; "test.c"     void main() {while(!RA2);}
>         BTFSS   INDF,2
>         GOTO    _00105_DS_
>         RETURN
>
> It should be testing PORTA, not INDF.
> This code was OK in 2.7.0:
>
> $ sdcc --version
> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.7.0 #4818 (Feb 14 2008) (UNIX) 
> $ sdcc -mpic14 -p16f628a -S test.c
> $ grep -A 3 while test.asm 
> ;       .line   2; "test.c"     void main() {while(!RA2);}
>         BANKSEL _PORTA_bits
>         BTFSS   _PORTA_bits,2
>         GOTO    _00105_DS_
>
> -jim
>

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

From: Borut Razem <borut.razem@si...> - 2008-03-23 07:55:26

I found a Q&D solution: if AOP_PCODE is set to 13 in gen.h, everything 
is OK.
It seems that a constant is used somewhere in the code :-( .
I'll commit the change for the RC2.

Borut


Borut Razem wrote:
> It happened in the svn revision #5064, rev. #5063 was OK.
>
> Raphael, can you please take a look ASAP?
>
> Borut
>
> Jim Paris wrote:
>   
>> Hi,
>>
>> It seems pic14 is broken in 2.8.0-rc1:
>>
>> $ cd /tmp
>> $ echo '#include <pic/pic14regs.h>' > test.c
>> $ echo 'void main() {while(!RA2);}' >> test.c
>> $ sdcc/bin/sdcc --version
>> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.8.0 #5082 (Mar  9 2008) (UNIX)
>> $ sdcc/bin/sdcc -mpic14 -p16f628a -S test.c
>> $ grep -A 3 while test.asm 
>> ;       .line   2; "test.c"     void main() {while(!RA2);}
>>         BTFSS   INDF,2
>>         GOTO    _00105_DS_
>>         RETURN
>>
>> It should be testing PORTA, not INDF.
>> This code was OK in 2.7.0:
>>
>> $ sdcc --version
>> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.7.0 #4818 (Feb 14 2008) (UNIX) 
>> $ sdcc -mpic14 -p16f628a -S test.c
>> $ grep -A 3 while test.asm 
>> ;       .line   2; "test.c"     void main() {while(!RA2);}
>>         BANKSEL _PORTA_bits
>>         BTFSS   _PORTA_bits,2
>>         GOTO    _00105_DS_
>>
>> -jim
>>

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

From: Borut Razem <borut.razem@si...> - 2008-03-23 07:59:37

Jim Paris,

can you please file the bug to the sdcc bug tracker?

Borut


Borut Razem wrote:
> I found a Q&D solution: if AOP_PCODE is set to 13 in gen.h, everything 
> is OK.
> It seems that a constant is used somewhere in the code :-( .
> I'll commit the change for the RC2.
>
> Borut
>
>
> Borut Razem wrote:
>   
>> It happened in the svn revision #5064, rev. #5063 was OK.
>>
>> Raphael, can you please take a look ASAP?
>>
>> Borut
>>
>> Jim Paris wrote:
>>   
>>     
>>> Hi,
>>>
>>> It seems pic14 is broken in 2.8.0-rc1:
>>>
>>> $ cd /tmp
>>> $ echo '#include <pic/pic14regs.h>' > test.c
>>> $ echo 'void main() {while(!RA2);}' >> test.c
>>> $ sdcc/bin/sdcc --version
>>> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.8.0 #5082 (Mar  9 2008) (UNIX)
>>> $ sdcc/bin/sdcc -mpic14 -p16f628a -S test.c
>>> $ grep -A 3 while test.asm 
>>> ;       .line   2; "test.c"     void main() {while(!RA2);}
>>>         BTFSS   INDF,2
>>>         GOTO    _00105_DS_
>>>         RETURN
>>>
>>> It should be testing PORTA, not INDF.
>>> This code was OK in 2.7.0:
>>>
>>> $ sdcc --version
>>> SDCC : mcs51/gbz80/z80/avr/ds390/pic16/pic14/TININative/xa51/ds400/hc08 2.7.0 #4818 (Feb 14 2008) (UNIX) 
>>> $ sdcc -mpic14 -p16f628a -S test.c
>>> $ grep -A 3 while test.asm 
>>> ;       .line   2; "test.c"     void main() {while(!RA2);}
>>>         BANKSEL _PORTA_bits
>>>         BTFSS   _PORTA_bits,2
>>>         GOTO    _00105_DS_
>>>
>>> -jim
>>>     
>>>       
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> sdcc-devel mailing list
> sdcc-devel@...
> https://lists.sourceforge.net/lists/listinfo/sdcc-devel
>
>
>

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

From: Raphael Neider <rneider@we...> - 2008-03-23 08:13:04

Hi,

the bug has gone away since r5106 and does not appear in the current svn 
revision r5114.

> can you please file the bug to the sdcc bug tracker?

No need for it, it's already fixed.

>> I found a Q&D solution: if AOP_PCODE is set to 13 in gen.h, everything 
>> is OK.
>> It seems that a constant is used somewhere in the code :-( .
>> I'll commit the change for the RC2.

This is ugly and wrong; r5106 has the correct fix.

I'd leave the source as it was.

Regards,
Raphael

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

From: Borut Razem <borut.razem@si...> - 2008-03-23 08:37:44

Raphael,

you are right, and I'm stupid :-[
I checked with older versions, not with the latest ones.
I'll revert the change.

Thank you,
Borut


Raphael Neider wrote:
> Hi,
>
> the bug has gone away since r5106 and does not appear in the current svn 
> revision r5114.
>
>   
>> can you please file the bug to the sdcc bug tracker?
>>     
>
> No need for it, it's already fixed.
>
>   
>>> I found a Q&D solution: if AOP_PCODE is set to 13 in gen.h, everything 
>>> is OK.
>>> It seems that a constant is used somewhere in the code :-( .
>>> I'll commit the change for the RC2.
>>>       
>
> This is ugly and wrong; r5106 has the correct fix.
>
> I'd leave the source as it was.
>
> Regards,
> Raphael
>

Thanks for helping keep SourceForge clean.

Small Device C Compiler

Brought to you by: benshi, bernhardheld, drdani, epetrich, and 8 others

Re: [sdcc-devel] pic14 regression in 2.8.0-rc1

Thread view