[llvm-commits] [llvm] r67139 - in /llvm/trunk: lib/Transforms/IPO/DeadArgumentElimination.cpp test/Transforms/DeadArgElim/2009-03-17-MRE-Invoke.ll

[Bill Wendling]

On Wed, Mar 18, 2009 at 11:45 AM, Dale Johannesen <dalej at apple.com> wrote:
> \On Mar 18, 2009, at 11:26 AMPDT, Bill Wendling wrote:
>>
>> On Wed, Mar 18, 2009 at 11:22 AM, Chris Lattner <clattner at apple.com>
>> wrote:
>>>
>>> On Mar 18, 2009, at 10:46 AM, Bill Wendling wrote:
>>>
>>>> Chris,
>>>>
>>>> These were causing XPASSes (on Darwin) in these tests:
>>>>
>>>> Sending        test/FrontendC/2009-03-13-dbg.c
>>>> Sending        test/FrontendC++/2009-02-16-CtorNames-dbg.cpp
>>>> Sending        test/FrontendC++/2009-03-17-dbg.cpp
>>>>
>>>> Is this expected? I modified them to make them passing. Please check
>>>> to see if that's correct.
>>>
>>> Are you sure?  I can't imagine how these would be related.
>>>
>> Buildbot has them passing right after your patch:
>>
>> http://blamebot.apple.com:80/waterfall
>
> Hey, that's a nice display, but you gotta read it...
>
:-)

>> The debug info stuff is in flux. Maybe they have weird dependencies.
>>
>> Devang & Dale, do you know how these could be affected by Chris's r67139
>> patch?
>
> Looks to me like they passed until Evan reverted Devang's debug info patches
> last night, in 67141, and simultaneously (from the buildbot's point of view)
> XFAILed the tests in 67143.  I'm not sure why they didn't start failing,
> which is what I'd expect, but I don't think it's anything to do with Chris's
> patches.
>
Okay. Let's chalk it up to weirdness in our testing then (they should
have been failing in the *next* run, not this one...maybe the
buildbots know the future. ;-). Once Devang's patches are
re-committed, we can visit these tests then.

-bw