| Submitter | Richard Purdie |
|---|---|
| Date | April 15, 2012, 5:05 p.m. |
| Message ID | <1334509553.16992.34.camel@ted> |
| Download | mbox | patch |
| Permalink | /patch/25877/ |
| State | Accepted |
| Commit | e7135f14f3ef37f93f5c28b319464f3d6dd9b8da |
| Headers | show |
Comments
On Sun, Apr 15, 2012 at 10:05 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> gcc_cv_collect2_libs
is it the /usr/lib/cmplrs/cc2.11 access issue ?
in that case the fix it fine. However if its something in LTO
then we need to dig deeper
On Sun, 2012-04-15 at 16:42 -0700, Khem Raj wrote: > On Sun, Apr 15, 2012 at 10:05 AM, Richard Purdie > <richard.purdie@linuxfoundation.org> wrote: > > gcc_cv_collect2_libs > > is it the /usr/lib/cmplrs/cc2.11 access issue ? It is. > in that case the fix it fine. However if its something in LTO > then we need to dig deeper Its not lto related. Cheers, Richard
Patch
diff --git a/meta/recipes-devtools/gcc/gcc-configure-common.inc b/meta/recipes-devtools/gcc/gcc-configure-common.inc index 7a96e91..9542dc9 100644 --- a/meta/recipes-devtools/gcc/gcc-configure-common.inc +++ b/meta/recipes-devtools/gcc/gcc-configure-common.inc @@ -107,6 +107,10 @@ do_configure () { export CXXFLAGS_FOR_BUILD="${BUILD_CXXFLAGS}" export LDFLAGS_FOR_BUILD="${BUILD_LDFLAGS}" export ARCH_FLAGS_FOR_TARGET="${ARCH_FLAGS_FOR_TARGET}" + # We need to set this else there is cross-compilation badness + # in the config.log files (which might not get generated until do_compile + # hence being missed by the insane do_configure check). + export gcc_cv_collect2_libs="none required" (cd ${S} && gnu-configize) || die "failure running gnu-configize" oe_runconf
There were puzzling failures when you make a force recompile of any gcc component. The error was in do_configure with cross-compilation badness being detected in config.log files. gcc is different in that many of the config.log files are generated during the do_compile phase. This means this host contamination issue has always been present but only shows up on a rebuild. The fix is to force the appropriate configuration variable to "none required" then gcc won't look in the bad locations. [YOCTO #2279] Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> ---