[llvm-commits] [llvm] r67049 - in /llvm/trunk: lib/CodeGen/PHIElimination.cpp test/CodeGen/X86/2009-03-13-PHIElimBug.ll test/CodeGen/X86/2009-03-16-PHIElimInLPad.ll

[Bill Wendling]

On Tue, Mar 17, 2009 at 12:34 AM, Duncan Sands <baldrick at free.fr> wrote:
> Hi Bill,
>
>> FAIL: /Volumes/Sandbox/Buildbot/llvm/full-llvm/build/llvm.src/test/CodeGen/X86/2009-03-13-PHIElimBug.ll
>> for PR3784
>> Failed with exit(1) at line 1
>> while running:  llvm-as <
>> /Volumes/Sandbox/Buildbot/llvm/full-llvm/build/llvm.src/test/CodeGen/X86/2009-03-13-PHIElimBug.ll
>> |  llc -march=x86 | /usr/bin/grep -A 2 {call  f} | /usr/bin/grep movl
>
> what is the output from llc and does grep -A work on your machine (should
> produce 2 lines of output after the matching output)?
>
I don't have the output from llc. I don't really have time to recreate
this failure either...

It looks like 'grep -A' works:

$ grep -A 2 This llvmCore-2064/README.txt
This directory and its subdirectories contain source code for the Low Level
Virtual Machine, a toolkit for the construction of highly optimized compilers,
optimizers, and runtime environments.

> PS: I'd rather you just xfailed the test rather than reverting the patch.
> The test tests that the patch fixed what it thinks it fixed.  The
> problem is more likely to be in the testcase than the patch itself.
>
Because the tests was there before your patch, it's not an option to
XFAIL it's obvious that there's something wrong with the test, in
which case you should fix it. :-) I don't know how important that test
is, and regressing in our tests is really bad. XFAIL just covers these
things up.

-bw