[Mesa-dev] [PATCH] android: fix bison warning of conflicting outputs to file

Chad Versace chad.versace at linux.intel.com
Mon Dec 3 12:10:04 PST 2012


On 11/30/2012 09:58 AM, Ian Romanick wrote:
> On 11/29/2012 11:52 PM, groleo at gmail.com wrote:
>> From: Adrian Marius Negreanu <adrian.m.negreanu at intel.com>
>>
>> Bison -o parameter expects a .c file.
>> The corresponding .h filename is obtained
>> by removing the extension of the initial .c.
>>
>> This was breaking compilation on Ubuntu 12.04
>>
>> libmesa_dricore_intermediates/libmesa_dricore.a(program_parse.tab.o):
>> In function `_mesa_parse_arb_program':
>> external/mesa/src/mesa/program/program_parse.y:2682: multiple definition
>> of `_mesa_parse_arb_program'
>> libmesa_dricore_intermediates/libmesa_dricore.a(lex.yy.o):external/mesa/src/mesa/program/program_parse.y:2682:
>>
>> first defined here
>>
>> Signed-off-by: Adrian Marius Negreanu <adrian.m.negreanu at intel.com>
>> ---
>>   src/mesa/program/Android.mk | 3 ---
>>   1 file changed, 3 deletions(-)
>>
>> diff --git a/src/mesa/program/Android.mk b/src/mesa/program/Android.mk
>> index 5eb0439..7125061 100644
>> --- a/src/mesa/program/Android.mk
>> +++ b/src/mesa/program/Android.mk
>> @@ -65,9 +65,6 @@ LOCAL_GENERATED_SOURCES := \
>>   $(intermediates)/program/program_parse.tab.c: $(LOCAL_PATH)/program_parse.y
>>       $(mesa_local-y-to-c-and-h)
>>
>> -$(intermediates)/program/program_parse.tab.h: $(LOCAL_PATH)/program_parse.y
>> -    $(mesa_local-y-to-c-and-h)
>> -
> 
> Isn't it possible that removing this dependency will break parallel builds in
> some cases?

I can imagine a manufactured scenario in which the lack of a dependency of
program_parse.tag.h on program_parse.y would break a parallel build. But,
given how the android buildtree is separate from the source tree, it really
is just a manufactured scenario.

As it stands, that make rule is completely wrong, and so the build is broken.
Let's fix the build ASAP, and discuss how to correctly declare the dependency
later when the situation is no longer urgent.

Tested-and-reviewed-by: Chad Versace <chad.versace at linux.intel.com>



More information about the mesa-dev mailing list