buildbot failure in LLVM on llvm-mips-linux (probably r206769)

[Daniel Sanders]

I can reproduce it on the same machine the MIPS builder used.

The problem seems to be related to this hunk:
--- RuntimeDyld.cpp     2014-04-22 14:23:24.243823000 +0000
+++ RuntimeDyld.r206769.cpp     2014-04-22 14:10:56.729443000 +0000
@@ -324,13 +318,12 @@
   // necessary section allocation size in loadObject by walking all the sections
   // once.
   unsigned StubBufSize = 0;
-  for (section_iterator SI = Obj.begin_sections(), SE = Obj.end_sections();
-       SI != SE; ++SI) {
-    section_iterator RelSecI = SI->getRelocatedSection();
+  for (const SectionRef &Section : Obj.sections()) {
+    section_iterator RelSecI = Section.getRelocatedSection();
     if (!(RelSecI == Section))
       continue;

-    for (const RelocationRef &Reloc : SI->relocations()) {
+    for (const RelocationRef &Reloc : Section.relocations()) {
       (void)Reloc;
       StubBufSize += StubSize;
     }

The hunk by itself is enough to make the tests fail. Applying all hunks except that one passes.

> -----Original Message-----
> From: llvm-commits-bounces at cs.uiuc.edu [mailto:llvm-commits-
> bounces at cs.uiuc.edu] On Behalf Of Hal Finkel
> Sent: 21 April 2014 21:08
> To: Jim Grosbach
> Cc: llvm-commits; gkistanova at gmail.com; llvm buildmaster; Reed Kotler
> Subject: Re: buildbot failure in LLVM on llvm-mips-linux (probably r206769)
> 
> ----- Original Message -----
> > From: "Jim Grosbach" <grosbach at apple.com>
> > To: "llvm buildmaster" <llvm.buildmaster at lab.llvm.org>, "llvm-commits"
> <llvm-commits at cs.uiuc.edu>, "reed kotler"
> > <rkotler at mips.com>, "Hal Finkel" <hfinkel at anl.gov>
> > Cc: "Duncan P. N. Exon Smith" <dexonsmith at apple.com>,
> > gkistanova at gmail.com
> > Sent: Monday, April 21, 2014 2:42:53 PM
> > Subject: Re: buildbot failure in LLVM on llvm-mips-linux (probably
> > r206769)
> >
> > Hi all,
> >
> > I suspect this failure is related to my patch, but it doesn’t
> > reproduce for me. Can anyone help verify that it’s a real failure and
> > if so get a repro recipe to figure out what’s going on? Looks like
> > it’s also happening on ppc bots.
> 
> I can reproduce this problem locally on a ppc64/Linux system with both a gcc-
> compiled Release+Asserts build and a self-hosted Release build. Both of
> these use libstdc++ (4.7 and 4.8 respectively).
> 
>  -Hal
> 
> >
> > I reverted the change in r206773 in the meantime.
> >
> > -Jim
> >
> > On Apr 21, 2014, at 12:15 PM, llvm.buildmaster at lab.llvm.org wrote:
> >
> > > The Buildbot has detected a new failure on builder llvm-mips-linux
> > > while building llvm.
> > > Full details are available at:
> > > http://lab.llvm.org:8011/builders/llvm-mips-linux/builds/8305
> > >
> > > Buildbot URL: http://lab.llvm.org:8011/
> > >
> > > Buildslave for this Build: mipsswbrd002
> > >
> > > Build Reason: scheduler
> > > Build Source Stamp: [branch trunk] 206769
> > > Blamelist: dexonsmith,grosbach
> > >
> > > BUILD FAILED: failed test-llvm
> > >
> > > sincerely,
> > > -The Buildbot
> > >
> > >
> > >
> >
> >
> 
> --
> Hal Finkel
> Assistant Computational Scientist
> Leadership Computing Facility
> Argonne National Laboratory
> 
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits