[llvm-testresults] buildbot failure in llvm on dragonegg-i386-linux

[Eric Christopher]

On Jan 26, 2011, at 12:54 PM, Eric Christopher wrote:

> 
> On Jan 26, 2011, at 12:43 PM, Duncan Sands wrote:
> 
>> Hi Eric,
>> 
>>>>> Non-determinism! The dragonegg selfhost compared object files it built
>>>>> at different stages, and they weren't the same. I guess this might be
>>>>> due to echristo's inline cost changes.
>>>> 
>>>> ... Those were Sunday though. No costs changed yesterday, might be some
>>>> caching effects though. I'll take a look.
>>> 
>>> I've reverted for now to see if it fixes it, but a quick question. From looking
>>> at the output of the script it looks like it is comparing the output of
>>> stage 0 vs the output of stage 1? Or is that just an artifact of the script?
>> 
>> thanks for reverting your patch.  The stages are not exactly the same as with a
>> gcc bootstrap.  In stage 0, LLVM and dragonegg are built with gcc, but dragonegg
>> is then immediately rebuilt with itself, creating the object files that are used
>> for the comparison.  So these object files are analogous to the object files
>> produced by stage1 of a gcc bootstrap.
> 
> OK, that's a bit better. I'd like some sort of test case if you can arrange it
> though. I didn't see anything in the patch that could cause this sort of issue.

A preprocessed version of the miscomparing file (or even better, bitcode and
some optimization options) would be helpful at least.

-eric

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-testresults/attachments/20110126/e633e5bf/attachment.html>