On 2016-05-25 10:15 AM, Richard Purdie wrote:
On Wed, 2016-05-25 at 09:31 -0400, Bruce Ashfield wrote:
On 2016-05-25 8:32 AM, Bruce Ashfield wrote:
On 2016-05-25 2:39 AM, Richard Purdie wrote:
Thanks to the patches from Bruce we're looking better, linux
-yocto 4.1
works, just 4.4 has a couple of issues. I also removed a patch
which
removed the xdg-utils failure. That leaves us with:
qt4-x11-free:
arm, mips, ppc, world-lsb, x86, x86_64:
https://autobuilder.yoctoproject.org/main/builders/nightly-arm-ls
b/buil
ds/767/steps/BuildImages/logs/stdio
qt4-embedded:
https://autobuilder.yoctoproject.org/main/builders/nightly-world-
lsb/bu
ilds/520/steps/BuildImages/logs/stdio
linux-yocto-4.4 mips:
https://autobuilder.yoctoproject.org/main/builders/nightly-mips/b
uilds/
795/steps/BuildImages_1/logs/stdio
linux-yocto-4.4 ppc:
https://autobuilder.yoctoproject.org/main/builders/nightly-ppc/bu
ilds/8
07/steps/BuildImages_1/logs/stdio
diffutils - non-gplv3:
https://autobuilder.yoctoproject.org/main/builders/nightly-non-gp
l3/bui
lds/797
Bruce: Looks like 4.4 might also need a patch you merged to 4.1?
I didn't see anything obvious missing from 4.4, and the existing
patches
posted for the build issues are part of:
-----------------
commit 574e9f4b4334807faad40a2d23c350f266d6febb
Author: Bruce Ashfield <bruce.ashfi...@windriver.com>
Date: Fri May 13 09:46:53 2016 -0700
linux-yocto/4.4: gcc6 build fixes (powerpc and mips)
Khem provided fixes to fix gcc6 build issues, these are safe
for
all gcc versions, so we integrate them directly.
(From OE-Core rev: f1c75b93a4e11425e595c5ce043fbb0276a41931)
Signed-off-by: Khem Raj <raj.k...@gmail.com>
Signed-off-by: Bruce Ashfield <bruce.ashfi...@windriver.com>
Signed-off-by: Richard Purdie <
richard.pur...@linuxfoundation.org>
---------------
So I can't say why the two build failures that that commit
addresses
are popping up.
I checked and the later 4.4-stable commit update I did has the
right
SRCREVs and encompasses that fix .. unless something ended up
getting
clobbered in the code itself.
What's the right incantation to start a gcc6 based build on my own
machine ? Just bump the preferred version of the compiler ? (I'll
do
that now, since I see the gcc6 recipe .. and that seems like the
obvious thing).
Hmm. Are we somehow building different SRCREVs ? I'm able to build
the
ppc kernel with gcc6:
yow-bashfiel-d4 [/home/bruc...dard-build]> ls -alF vmlinux
-rwxr-xr-x 1 bruce bruce 115353128 May 25 09:12 vmlinux*
yow-bashfiel-d4 [/home/bruc...dard-build]> pwd
/home/bruce/poky/build/tmp/work/qemuppc-poky-linux/linux
-yocto/4.4.11+gitAUTOINC+6ec93aaa70_628bf62756-r0/linux-qemuppc
-standard-build
yow-bashfiel-d4 [/home/bruc...dard-build]> ls -alF vmlinux
-rwxr-xr-x 1 bruce bruce 115353128 May 25 09:12 vmlinux*
yow-bashfiel-d4 [/home/bruc...dard-build]> strings vmlinux | grep
"gcc
version"
%s version %s (bruce@yow-bashfiel-d4) (gcc version 6.1.1 20160511
(GCC) ) %s
Linux version 4.4.11-yocto-standard (bruce@yow-bashfiel-d4) (gcc
version
6.1.1 20160511 (GCC) ) #1 PREEMPT Wed May 25 09:12:02 EDT 2016
Just realised the failure is in mpc8315e_rdb so its meta-yocto-bsp
bumps we're missing?
Indeed. I can prepare a compile-only SRCREV bump, and if it happens
to introduce a runtime issue .. we'll deal with that separately.
Bruce
Cheers,
Richard
--
_______________________________________________
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core
