Re: [PATCH] gcc5: Add PR65779 patch to fix powerpc compile issues

[Richard Purdie <richard.purdie@linuxfoundation.org>]

On Mon, 2015-05-11 at 17:24 -0700, Khem Raj wrote:
> On Mon, May 11, 2015 at 4:20 PM, akuster808 <akuster808@gmail.com> wrote:
> >
> >
> > On 05/11/2015 09:54 AM, Richard Purdie wrote:
> >>
> >> This fixes compile issues on powerpc with gcc 5 which show up with
> >> errors like:
> >>
> >> | make[2]: Entering directory
> >> '/media/build1/poky/build/tmp/work/ppc7400-poky-linux/xprop/1_1.2.2-r0/build'
> >> | powerpc-poky-linux-gcc  -m32 -mhard-float -mcpu=7400
> >> --sysroot=/media/build1/poky/build/tmp/sysroots/qemuppc -Wall
> >> -Wpointer-arith -Wmissing-declarations -Wformat=2 -Wstrict-prototypes
> >> -Wmissing-prototypes -Wnested-externs -Wbad-function-cast
> >> -Wold-style-definition -Wdeclaration-after-statement -Wunused
> >> -Wuninitialized -Wshadow -Wmissing-noreturn -Wmissing-format-attribute
> >> -Wredundant-decls -Wlogical-op -Werror=implicit -Werror=nonnull
> >> -Werror=init-self -Werror=main -Werror=missing-braces -Werror=sequence-point
> >> -Werror=return-type -Werror=trigraphs -Werror=array-bounds
> >> -Werror=write-strings -Werror=address -Werror=int-to-pointer-cast
> >> -Werror=pointer-to-int-cast -fno-strict-aliasing  -O2 -pipe -g
> >> -feliminate-unused-debug-types  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed
> >> -o xprop dsimple.o clientwin.o xprop.o -lX11
> >> | /media/build1/poky/build/tmp/sysroots/qemuppc/usr/lib/../lib/libX11.so:
> >> undefined reference to `.LCL2'
> >> | collect2: error: ld returned 1 exit status
> >>
> >> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
> >
> >
> > Acked-by: Armin Kuster <Akuster808@gmail.com>
> >
> >
> > this fixes all but one package(lzop) failure on qemuppc base yocto world
> > build.
> >
> 
> I am fine witj this patch for now, since we are mostly build testing,
> we have to keep in mind
> its a codegen bug and can bite us at runtime. I am just hoping for an
> update to this patch and that will eliminate
> that doubt.

Agreed, an update upstream would obviously help. FWIW the automated
runtime tests did look reasonable on the autobuilder apart from:

core-image-sato-sdk has a problem with the C++ toolchain not finding
limits (all arches).

qemuarm is not booting (minimal, sato or sato-sdk).

However these issues were present before the patch so I don't think its
related.

Since the failures list is better with the patch I'll merge it until
something better comes along.

Cheers,

Richard