Revert of tcmalloc: Use C++11 atomics where appropriate.

Revert of tcmalloc: Use C++11 atomics where appropriate. (patchset #9 id:160001 of https://codereview.chromium.org/1549913002/ )

Reason for revert:
Doesn't build on 32-bit: https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.linux%2FLinux_Builder__dbg__32_%2F66018%2F%2B%2Frecipes%2Fsteps%2Fcompile%2F0%2Fstdout

../../build/linux/debian_jessie_i386-sysroot/usr/lib/gcc/i586-linux-gnu/4.8/../../../../include/c++/4.8/bits/atomic_base.h:506: error: undefined reference to '__atomic_load_8'
../../build/linux/debian_jessie_i386-sysroot/usr/lib/gcc/i586-linux-gnu/4.8/../../../../include/c++/4.8/bits/atomic_base.h:486: error: undefined reference to '__atomic_store_8'

Original issue's description:
> tcmalloc: Use C++11 atomics where appropriate.
>
> Reland now that we have a newer sysroot.
>
> Ports these CLs to tcmalloc:
> https://codereview.chromium.org/636783002/
> https://codereview.chromium.org/1466833002/ (except mac)
>
> No intended behavior change, but it should remove
> the static initializer in atomicops_internals_x86_gcc.h
> on Linux.  It's also less code.
>
> BUG=94925, 559247, 572525
>
> Committed: https://crrev.com/e13537fe418eff11d3cab9077f6a647d7c74f103
> Cr-Original-Commit-Position: refs/heads/master@{#366904}
> Review-Url: https://codereview.chromium.org/1549913002
> Cr-Commit-Position: refs/heads/master@{#464440}
> Committed: https://chromium.googlesource.com/chromium/src/+/2f6d8f01d9087e8bebab5b7d2d25b28d657dbbb7

TBR=jfb@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=94925, 559247, 572525
Review-Url: https://codereview.chromium.org/2818713003
Cr-Commit-Position: refs/heads/master@{#464445}
Committed: https://chromium.googlesource.com/chromium/src/+/b20672f0a5080fe4318e049bd408db1a01105904

[Nico, 2017-04-13 16:59:36]

The CQ bit was checked by thakis@chromium.org

[Nico, 2017-04-13 16:59:36]

Created Revert of tcmalloc: Use C++11 atomics where appropriate.

[commit-bot: I haz the power, 2017-04-13 17:00:19]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-04-13 17:01:11]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1492102776574170, "parent_rev":
"c08796e46810403645b6cf2cc4f8890c78a361de", "commit_rev":
"b20672f0a5080fe4318e049bd408db1a01105904"}

[commit-bot: I haz the power, 2017-04-13 17:02:55]

Description was changed from

==========
Revert of tcmalloc: Use C++11 atomics where appropriate. (patchset #9 id:160001
of https://codereview.chromium.org/1549913002/ )

Reason for revert:
Doesn't build on 32-bit:
https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.linux%2FLinux_B...

../../build/linux/debian_jessie_i386-sysroot/usr/lib/gcc/i586-linux-gnu/4.8/../../../../include/c++/4.8/bits/atomic_base.h:506:
error: undefined reference to '__atomic_load_8'
../../build/linux/debian_jessie_i386-sysroot/usr/lib/gcc/i586-linux-gnu/4.8/../../../../include/c++/4.8/bits/atomic_base.h:486:
error: undefined reference to '__atomic_store_8'


Original issue's description:
> tcmalloc: Use C++11 atomics where appropriate.
> 
> Reland now that we have a newer sysroot.
> 
> Ports these CLs to tcmalloc:
> https://codereview.chromium.org/636783002/
> https://codereview.chromium.org/1466833002/ (except mac)
> 
> No intended behavior change, but it should remove
> the static initializer in atomicops_internals_x86_gcc.h
> on Linux.  It's also less code.
> 
> BUG=94925,559247,572525
> 
> Committed: https://crrev.com/e13537fe418eff11d3cab9077f6a647d7c74f103
> Cr-Original-Commit-Position: refs/heads/master@{#366904}
> Review-Url: https://codereview.chromium.org/1549913002
> Cr-Commit-Position: refs/heads/master@{#464440}
> Committed:
https://chromium.googlesource.com/chromium/src/+/2f6d8f01d9087e8bebab5b7d2d25...

TBR=jfb@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=94925,559247,572525
==========

to

==========
Revert of tcmalloc: Use C++11 atomics where appropriate. (patchset #9 id:160001
of https://codereview.chromium.org/1549913002/ )

Reason for revert:
Doesn't build on 32-bit:
https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.linux%2FLinux_B...

../../build/linux/debian_jessie_i386-sysroot/usr/lib/gcc/i586-linux-gnu/4.8/../../../../include/c++/4.8/bits/atomic_base.h:506:
error: undefined reference to '__atomic_load_8'
../../build/linux/debian_jessie_i386-sysroot/usr/lib/gcc/i586-linux-gnu/4.8/../../../../include/c++/4.8/bits/atomic_base.h:486:
error: undefined reference to '__atomic_store_8'

Original issue's description:
> tcmalloc: Use C++11 atomics where appropriate.
>
> Reland now that we have a newer sysroot.
>
> Ports these CLs to tcmalloc:
> https://codereview.chromium.org/636783002/
> https://codereview.chromium.org/1466833002/ (except mac)
>
> No intended behavior change, but it should remove
> the static initializer in atomicops_internals_x86_gcc.h
> on Linux.  It's also less code.
>
> BUG=94925,559247,572525
>
> Committed: https://crrev.com/e13537fe418eff11d3cab9077f6a647d7c74f103
> Cr-Original-Commit-Position: refs/heads/master@{#366904}
> Review-Url: https://codereview.chromium.org/1549913002
> Cr-Commit-Position: refs/heads/master@{#464440}
> Committed:
https://chromium.googlesource.com/chromium/src/+/2f6d8f01d9087e8bebab5b7d2d25...

TBR=jfb@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=94925,559247,572525

Review-Url: https://codereview.chromium.org/2818713003
Cr-Commit-Position: refs/heads/master@{#464445}
Committed:
https://chromium.googlesource.com/chromium/src/+/b20672f0a5080fe4318e049bd408...
==========

[commit-bot: I haz the power, 2017-04-13 17:02:57]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/b20672f0a5080fe4318e049bd408...

[JF, 2017-04-13 17:07:10]

This is caused by the toolchain conservatively assuming that CMPXCHG8B isn't
available on all x86 machines.

Is Chrome supported on those older CPUs?

A while ago I ran a query against all crash reports to see which CPUs we
actually saw in the wild, to determine support. Maybe it would help quantify
this question.