[Mesa-dev] [PATCH] Handle removal of LLVMAddTargetData in SVN revision 260919

[Roland Scheidegger]

Am 16.02.2016 um 15:45 schrieb Matthew Dawson:
> On Tuesday, February 16, 2016 4:19:49 PM EST Michel Dänzer wrote:
>> On 16.02.2016 15:25, Matthew Dawson wrote:
>>> LLVM removed LLVMAddTargetData for the 3.9 release in r260919.  For the
>>> two
>>> places in mesa where this is called, only enable the lines when compiling
>>> for less then 3.9.
>>>
>>> For the radeon driver, I'm not sure how to check if any other LLVM calls
>>> need to be adjusted.  I think since the target data used is extracted
>>> from the LLVMModule, it isn't necessary to pass it back to LLVM again.
>>>
>>> The code does compile, and at least for radeonsi does run OpenGL games.
>>
>> BTW, I recommend getting familiar with piglit so that you can make sure
>> your changes don't cause any piglit regressions. I did so for this change.
> Thanks.  I've tried running piglit in the past, but I always get some 
> failures.  I'm never sure what the baseline is for a given mesa version.  Is 
> there a good place to find what should be failing?  And about how long should 
> a piglit run take?
> 
> I know some tests are skipped, and I don't worry about them.  It's just the 
> failures.
> 
> Thanks for taking my patch,
> 

There's always some failures. The way to handle this is to just run it
without your change and then with them, the report stuff can generate
diffs (of course, sometimes I'm lazy and don't always rerun the baseline
if not too much time has passed, you just have to be sure if some other
differences _really_ couldn't have been caused by your new code...).

Roland