Add note about using the system linker with icecc

Add note about using the system linker with icecc

Committed: https://crrev.com/f6847d5442430a35270e6c0583bc1da85d1471bd
Cr-Commit-Position: refs/heads/master@{#369177}

[davve, 2016-01-11 10:04:40]

davve@opera.com changed reviewers:
+ andybons@chromium.org, thakis@chromium.org

[davve, 2016-01-11 10:04:40]

Since a recent Debian upgrade (testing) this seems to be required to be able to
use icecc with the other flags above.

I have the feeling this won't work in the long run, tacking on variables with
poor architecture support. But this holds it together for now. At least for me
and philipj.

Sample error messages:

FAILED: ccache cc -Wl,-z,now -Wl,-z,relro -Wl,--fatal-warnings -Wl,-z,defs
-pthread -Wl,-z,noexecstack -fPIC -fuse-ld=gold
-B/home/davve/src/blink/src/third_party/binutils/Linux_x64/Release/bin
-Wl,--disable-new-dtags -m64 -Wl,--detect-odr-violations -Wl,--icf=all -o
genmacro -Wl,--start-group
obj/third_party/yasm/source/patched-yasm/tools/genmacro/genmacro.genmacro.o
-Wl,--end-group  
/home/davve/src/blink/src/third_party/binutils/Linux_x64/Release/bin/ld.gold:
error: /usr/lib/gcc/x86_64-linux-gnu/4.9/../../../x86_64-linux-gnu/crti.o:
unsupported reloc 42 against global symbol __gmon_start__
/home/davve/src/blink/src/third_party/binutils/Linux_x64/Release/bin/ld.gold:
error:
obj/third_party/yasm/source/patched-yasm/tools/genmacro/genmacro.genmacro.o:
unsupported reloc 42 against global symbol stderr
/home/davve/src/blink/src/third_party/binutils/Linux_x64/Release/bin/ld.gold:
error:
obj/third_party/yasm/source/patched-yasm/tools/genmacro/genmacro.genmacro.o:
unsupported reloc 42 against global symbol stderr
/home/davve/src/blink/src/third_party/binutils/Linux_x64/Release/bin/ld.gold:
error:
obj/third_party/yasm/source/patched-yasm/tools/genmacro/genmacro.genmacro.o:
unsupported reloc 42 against global symbol stderr
/home/davve/src/blink/src/third_party/binutils/Linux_x64/Release/bin/ld.gold:
error:
obj/third_party/yasm/source/patched-yasm/tools/genmacro/genmacro.genmacro.o:
unsupported reloc 42 against global symbol stderr
/usr/lib/gcc/x86_64-linux-gnu/4.9/../../../x86_64-linux-gnu/crti.o(.init+0x7):
error: unsupported reloc 42

[davve, 2016-01-13 09:46:12]

ping.

[Bons, 2016-01-13 13:38:03]

lgtm

[davve, 2016-01-13 15:03:03]

The CQ bit was checked by davve@opera.com

[commit-bot: I haz the power, 2016-01-13 15:04:13]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1575843002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1575843002/1

[commit-bot: I haz the power, 2016-01-13 15:35:51]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-01-13 15:36:40]

Description was changed from

==========
Add note about using the system linker with icecc
==========

to

==========
Add note about using the system linker with icecc

Committed: https://crrev.com/f6847d5442430a35270e6c0583bc1da85d1471bd
Cr-Commit-Position: refs/heads/master@{#369177}
==========

[commit-bot: I haz the power, 2016-01-13 15:36:41]

Patchset 1 (id:??) landed as
https://crrev.com/f6847d5442430a35270e6c0583bc1da85d1471bd
Cr-Commit-Position: refs/heads/master@{#369177}

[Nico, 2016-01-13 20:47:16]

https://codereview.chromium.org/1575843002/diff/1/docs/linux_faster_builds.md
File docs/linux_faster_builds.md (right):

https://codereview.chromium.org/1575843002/diff/1/docs/linux_faster_builds.md...
docs/linux_faster_builds.md:49: Using the system linker is necessary since a
recent binutils upgrade.
What's "recent"? What will this mean in 4 months? Is there an upstream bug for
this somewhere?

[davve, 2016-01-14 08:13:03]

https://codereview.chromium.org/1575843002/diff/1/docs/linux_faster_builds.md
File docs/linux_faster_builds.md (right):

https://codereview.chromium.org/1575843002/diff/1/docs/linux_faster_builds.md...
docs/linux_faster_builds.md:49: Using the system linker is necessary since a
recent binutils upgrade.
On 2016/01/13 20:47:16, Nico wrote:
> What's "recent"? What will this mean in 4 months? Is there an upstream bug for
> this somewhere?

I get your point. :)

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=808181

is an upstream bug I've found. From that it looks like it may have been not the
binutils but the libc upgrade to 2.21-4 that broke this.

New, more specific CL coming up. Thanks for your prodding!