|
|
Chromium Code Reviews
DescriptionFix componene debug build failure in chromium
The error log is as follows:
../../third_party/skia/include/core/SkSpinlock.h:24: error: undefined reference to 'SkPODSpinlock::contendedAcquire()'
collect2: error: ld returned 1 exit status
[mtklein added below here]
Despite the presence in include/ and the added SK_API, this file is not part of Skia's public API... it's just used by files which are.
TBR=reed@google.com
Committed: https://skia.googlesource.com/skia/+/5ce8b438da24b7a2f2de771d9276a6a6789236a2
Patch Set 1 #Patch Set 2 : alternative way #Messages
Total messages: 17 (3 generated)
dongseong.hwang@intel.com changed reviewers: + bsalomon@google.com, reed@google.com
could you review this build fix?
On 2015/07/10 14:29:29, dshwang wrote: > could you review this build fix? alternative way is to add SK_API to SkPODSpinlock class. I think inlining is better
On 2015/07/10 14:31:46, dshwang wrote: > On 2015/07/10 14:29:29, dshwang wrote: > > could you review this build fix? > > alternative way is to add SK_API to SkPODSpinlock class. I think inlining is > better Let's add SK_API. This is intentionally out-of-line.
On 2015/07/10 14:43:12, mtklein wrote: > On 2015/07/10 14:31:46, dshwang wrote: > > On 2015/07/10 14:29:29, dshwang wrote: > > > could you review this build fix? > > > > alternative way is to add SK_API to SkPODSpinlock class. I think inlining is > > better > > Let's add SK_API. This is intentionally out-of-line. Can you send me a link to the whole failing build?
On 2015/07/10 14:44:03, mtklein wrote: > On 2015/07/10 14:43:12, mtklein wrote: > > On 2015/07/10 14:31:46, dshwang wrote: > > > On 2015/07/10 14:29:29, dshwang wrote: > > > > could you review this build fix? > > > > > > alternative way is to add SK_API to SkPODSpinlock class. I think inlining is > > > better > > > > Let's add SK_API. This is intentionally out-of-line. > > Can you send me a link to the whole failing build? Oh, it looks like there are no failing Chrome bots? If not, that's fine, let's just add SK_API if that makes it build for you.
On 2015/07/10 14:44:03, mtklein wrote:
> On 2015/07/10 14:43:12, mtklein wrote:
> > On 2015/07/10 14:31:46, dshwang wrote:
> > > On 2015/07/10 14:29:29, dshwang wrote:
> > > > could you review this build fix?
> > >
> > > alternative way is to add SK_API to SkPODSpinlock class. I think inlining
is
> > > better
> >
> > Let's add SK_API. This is intentionally out-of-line.
>
> Can you send me a link to the whole failing build?
Done. added SK_API instead.
I found it during local build. The log is as follows,
ninja: Entering directory `out_ozone_cros/Debug/'
[181/9697] SOLINK lib/libmedia.so
FAILED: if [ ! -e lib/libmedia.so -o ! -e lib/libmedia.so.TOC ]; then c++
-shared -Wl,-z,now -Wl,-z,relro -Wl,--fatal-warnings -Wl,-z,defs -pthread
-Wl,-z,noexecstack -fPIC -fuse-ld=gold
-B/home/dshwang/chromium/src/third_party/binutils/Linux_x64/Release/bin
-Wl,--disable-new-dtags -m64 -Wl,--icf=all -o lib/libmedia.so
-Wl,-soname=libmedia.so @lib/libmedia.so.rsp && { readelf -d lib/libmedia.so |
grep SONAME ; nm -gD -f p lib/libmedia.so | cut -f1-2 -d' '; } >
lib/libmedia.so.TOC; else c++ -shared -Wl,-z,now -Wl,-z,relro
-Wl,--fatal-warnings -Wl,-z,defs -pthread -Wl,-z,noexecstack -fPIC -fuse-ld=gold
-B/home/dshwang/chromium/src/third_party/binutils/Linux_x64/Release/bin
-Wl,--disable-new-dtags -m64 -Wl,--icf=all -o lib/libmedia.so
-Wl,-soname=libmedia.so @lib/libmedia.so.rsp && { readelf -d lib/libmedia.so |
grep SONAME ; nm -gD -f p lib/libmedia.so | cut -f1-2 -d' '; } >
lib/libmedia.so.tmp && if ! cmp -s lib/libmedia.so.tmp lib/libmedia.so.TOC; then
mv lib/libmedia.so.tmp lib/libmedia.so.TOC ; fi; fi
../../third_party/skia/include/core/SkSpinlock.h:24: error: undefined reference
to 'SkPODSpinlock::contendedAcquire()'
collect2: error: ld returned 1 exit status
[181/9697] SOLINK lib/libblink_platform.so
FAILED: if [ ! -e lib/libblink_platform.so -o ! -e lib/libblink_platform.so.TOC
]; then c++ -shared -Wl,-z,now -Wl,-z,relro -Wl,--fatal-warnings -Wl,-z,defs
-pthread -Wl,-z,noexecstack -fPIC -fuse-ld=gold
-B/home/dshwang/chromium/src/third_party/binutils/Linux_x64/Release/bin
-Wl,--disable-new-dtags -m64 -Wl,--icf=all -o lib/libblink_platform.so
-Wl,-soname=libblink_platform.so @lib/libblink_platform.so.rsp && { readelf -d
lib/libblink_platform.so | grep SONAME ; nm -gD -f p lib/libblink_platform.so |
cut -f1-2 -d' '; } > lib/libblink_platform.so.TOC; else c++ -shared -Wl,-z,now
-Wl,-z,relro -Wl,--fatal-warnings -Wl,-z,defs -pthread -Wl,-z,noexecstack -fPIC
-fuse-ld=gold
-B/home/dshwang/chromium/src/third_party/binutils/Linux_x64/Release/bin
-Wl,--disable-new-dtags -m64 -Wl,--icf=all -o lib/libblink_platform.so
-Wl,-soname=libblink_platform.so @lib/libblink_platform.so.rsp && { readelf -d
lib/libblink_platform.so | grep SONAME ; nm -gD -f p lib/libblink_platform.so |
cut -f1-2 -d' '; } > lib/libblink_platform.so.tmp && if ! cmp -s
lib/libblink_platform.so.tmp lib/libblink_platform.so.TOC; then mv
lib/libblink_platform.so.tmp lib/libblink_platform.so.TOC ; fi; fi
../../third_party/skia/include/core/SkSpinlock.h:24: error: undefined reference
to 'SkPODSpinlock::contendedAcquire()'
collect2: error: ld returned 1 exit status
[181/9697] CXX
obj/third_party/WebKit/Source/platform/weborigin/blink_platform_unittests.DatabaseIdentifierTest.o
ninja: build stopped: subcommand failed.
mtklein@google.com changed reviewers: + mtklein@google.com
lgtm at ps2
The CQ bit was checked by mtklein@google.com
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1229003004/20001
The author dongseong.hwang@intel.com has not signed Google Contributor License Agreement. Please visit https://cla.developers.google.com to sign and manage CLA.
On 2015/07/10 15:25:11, commit-bot: I haz the power wrote: > The author mailto:dongseong.hwang@intel.com has not signed Google Contributor License > Agreement. Please visit https://cla.developers.google.com to sign and manage > CLA. Hrm. Think you can do that? Seems like this has become a hard requirement lately.
The author dongseong.hwang@intel.com has not signed Google Contributor License Agreement. Please visit https://cla.developers.google.com to sign and manage CLA.
Message was sent while issue was closed.
Committed patchset #2 (id:20001) as https://skia.googlesource.com/skia/+/5ce8b438da24b7a2f2de771d9276a6a6789236a2
Message was sent while issue was closed.
On 2015/07/10 15:32:27, commit-bot: I haz the power wrote: > Committed patchset #2 (id:20001) as > https://skia.googlesource.com/skia/+/5ce8b438da24b7a2f2de771d9276a6a6789236a2 Well, I guess that was just a warning for now! At some point that'll become mandatory I'm told.
Message was sent while issue was closed.
On 2015/07/10 15:25:59, mtklein wrote: > On 2015/07/10 15:25:11, commit-bot: I haz the power wrote: > > The author mailto:dongseong.hwang@intel.com has not signed Google Contributor > License > > Agreement. Please visit https://cla.developers.google.com to sign and manage > > CLA. > > Hrm. Think you can do that? Seems like this has become a hard requirement > lately. thx for reviewing. I signed CLA and am chromium committer. Let me ask it to Intel support team. |
