Merge lp:~matthew-gretton-dann/gcc-linaro/4.7-2013.05-merge into lp:gcc-linaro/4.7

[Matthew Gretton-Dann]

[Linaro Toolchain Builder (cbuild)]

cbuild has taken a snapshot of this branch at r115071 and queued it for build.

The diff against the ancestor r115070 is available at:
 http://cbuild.validation.linaro.org/snapshots/gcc-linaro-4.7+bzr115071~matthew-gretton-dann~4.7-2013.05-merge.diff

and will be built on the following builders:
 a9-builder a9hf-builder armv5-builder i686 x86_64 xaarch64 xaarch64_bare xcortexa15hf

You can track the build queue at:
 http://cbuild.validation.linaro.org/helpers/scheduler

cbuild-snapshot: gcc-linaro-4.7+bzr115071~matthew-gretton-dann~4.7-2013.05-merge
cbuild-ancestor: lp:gcc-linaro/4.7+bzr115070
cbuild-state: check

[Christophe Lyon (christophe-lyon)]

Diff looks OK (matches diffs between bzr 117031 and 117167.

[Christophe Lyon (christophe-lyon)]

Validations clean for x86, x86_64, cortex-a15 (2 FAILs are probably caused by qemu: 27_io/basic_ofstream/pthread2.cc and 30_threads/lock/2.cc), aarch64_bare.

For aarch64, there are several new tests that FAIL at execution: gcc.c-torture/execute/pr56866.c, gfortran.dg/fmt_cache_3.f9, gfortran.dg/namelist_77.f9, gfortran.dg/namelist_78.f90, gfortran.dg/namelist_79.f90, gfortran.dg/namelist_80.f90, gfortran.dg/namelist_81.f90). Is this expected?

Native ARM (a9, a9hf, armv5) have not run yet.

review: Needs Information

[Matthew Gretton-Dann (matthew-gretton-dann)]

Checking changes manually show no major differences:

http://cbuild.validation.linaro.org/helpers/buildlog/gcc-linaro-4.7+bzr115071

http://cbuild.validation.linaro.org/helpers/testcompare/gcc-linaro-4.7+bzr115071~matthew-gretton-dann~4.7-2013.05-merge/logs/armv7l-precise-cbuild454-calxeda02_18_00_precise_armel-cortexa9r1/gcc-testsuite.txt?base=gcc-linaro-4.7-2013.04

The only difference is a guality test where values have been optimised out.

Merge approved.