Chromium Code Reviews
|
|
Chromium Code Reviews
Issue
1903833002:
Put r7 back into clobber registers in sys_clone() for thumb2. (Closed)
|
Created:
4 years, 8 months ago by Ted Mielczarek Modified:
3 years, 8 months ago CC:
chromium-reviews, Markus (顧孟勤) Base URL:
https://chromium.googlesource.com/linux-syscall-support/@master Target Ref:
refs/heads/master Visibility:
Public. |
DescriptionPut r7 back into clobber registers in sys_clone() for thumb2.
This change reverts this previous change: https://chromium.googlesource.com/linux-syscall-support/+/77ebebeffe70f6c41493f239e0a1bc449cc6b009%5E!/
The compiler bug causing the workaround is no longer present in modern GCC,
and the workaround was busted if the compiler happened to assign r7
to the __res variable.
More discussion at https://bugzilla.mozilla.org/show_bug.cgi?id=1247399#c12
R=markus@chromium.org
BUG=https://bugs.chromium.org/p/linux-syscall-support/issues/detail?id=13
Patch contributed by Ted Mielczarek (ted.mielczarek@gmail.com).
Committed: https://chromium.googlesource.com/linux-syscall-support/+/1549d20f6d3e7d66bb4e687c0ab9da42c2bff2ac
Patch Set 1 #
Messages
Total messages: 16 (4 generated)
On 20 April 2016 at 11:24, ted.mielczarek@gmail.com via codereview.chromium.org <reply@chromiumcodereview-hr.appspotmail.com> wrote: > Reviewers: Markus (顧孟勤) > CL: https://codereview.chromium.org/1903833002/ > > Description: > Put r7 back into clobber registers in sys_clone() for thumb2. > > The compiler bug causing the workaround is no longer present in modern GCC, > and the workaround was busted if the compiler happened to assign r7 > to the __res variable. > > More discussion at > https://bugzilla.mozilla.org/show_bug.cgi?id=1247399#c12 > > R=markus@chromium.org > BUG=https://bugs.chromium.org/p/linux-syscall-support/issues/detail?id=13 > So this is reverting this change: https://chromium.googlesource.com/linux-syscall-support/+/77ebebeffe70f6c4149... ? Can you link to that explicitly in the commit message, please, in order to make it clearer what GCC workaround you are referring to? Cheers, Mark -- You received this message because you are subscribed to the Google Groups "Chromium-reviews" group. To unsubscribe from this group and stop receiving emails from it, send an email to chromium-reviews+unsubscribe@chromium.org.
On 2016/04/26 19:11:08, Mark Seaborn wrote: > So this is reverting this change: > https://chromium.googlesource.com/linux-syscall-support/+/77ebebeffe70f6c4149... > ? Yes, that's correct. > Can you link to that explicitly in the commit message, please, in order to > make it clearer what GCC workaround you are referring to? OK, will do.
Description was changed from ========== Put r7 back into clobber registers in sys_clone() for thumb2. The compiler bug causing the workaround is no longer present in modern GCC, and the workaround was busted if the compiler happened to assign r7 to the __res variable. More discussion at https://bugzilla.mozilla.org/show_bug.cgi?id=1247399#c12 R=markus@chromium.org BUG=https://bugs.chromium.org/p/linux-syscall-support/issues/detail?id=13 ========== to ========== Put r7 back into clobber registers in sys_clone() for thumb2. This change reverts this previous change: https://chromium.googlesource.com/linux-syscall-support/+/77ebebeffe70f6c4149... The compiler bug causing the workaround is no longer present in modern GCC, and the workaround was busted if the compiler happened to assign r7 to the __res variable. More discussion at https://bugzilla.mozilla.org/show_bug.cgi?id=1247399#c12 R=markus@chromium.org BUG=https://bugs.chromium.org/p/linux-syscall-support/issues/detail?id=13 ==========
Sorry, lost track of this, is that commit message descriptive enough?
On 2016/06/10 15:40:56, Ted Mielczarek wrote: > Sorry, lost track of this, is that commit message descriptive enough? Thanks, LGTM.
On 2016/06/14 17:46:28, Mark Seaborn wrote: > On 2016/06/10 15:40:56, Ted Mielczarek wrote: > > Sorry, lost track of this, is that commit message descriptive enough? > > Thanks, LGTM. I didn't try, but I suspect I don't have push rights to this repository. Can you land this for me?
On 2016/06/14 21:29:31, Ted Mielczarek wrote: > I didn't try, but I suspect I don't have push rights to this repository. Can you > land this for me? Ping?
Description was changed from ========== Put r7 back into clobber registers in sys_clone() for thumb2. This change reverts this previous change: https://chromium.googlesource.com/linux-syscall-support/+/77ebebeffe70f6c4149... The compiler bug causing the workaround is no longer present in modern GCC, and the workaround was busted if the compiler happened to assign r7 to the __res variable. More discussion at https://bugzilla.mozilla.org/show_bug.cgi?id=1247399#c12 R=markus@chromium.org BUG=https://bugs.chromium.org/p/linux-syscall-support/issues/detail?id=13 ========== to ========== Put r7 back into clobber registers in sys_clone() for thumb2. This change reverts this previous change: https://chromium.googlesource.com/linux-syscall-support/+/77ebebeffe70f6c4149... The compiler bug causing the workaround is no longer present in modern GCC, and the workaround was busted if the compiler happened to assign r7 to the __res variable. More discussion at https://bugzilla.mozilla.org/show_bug.cgi?id=1247399#c12 R=markus@chromium.org BUG=https://bugs.chromium.org/p/linux-syscall-support/issues/detail?id=13 Patch contributed by Ted Mielczarek (ted.mielczarek@gmail.com). ==========
Description was changed from ========== Put r7 back into clobber registers in sys_clone() for thumb2. This change reverts this previous change: https://chromium.googlesource.com/linux-syscall-support/+/77ebebeffe70f6c4149... The compiler bug causing the workaround is no longer present in modern GCC, and the workaround was busted if the compiler happened to assign r7 to the __res variable. More discussion at https://bugzilla.mozilla.org/show_bug.cgi?id=1247399#c12 R=markus@chromium.org BUG=https://bugs.chromium.org/p/linux-syscall-support/issues/detail?id=13 Patch contributed by Ted Mielczarek (ted.mielczarek@gmail.com). ========== to ========== Put r7 back into clobber registers in sys_clone() for thumb2. This change reverts this previous change: https://chromium.googlesource.com/linux-syscall-support/+/77ebebeffe70f6c4149... The compiler bug causing the workaround is no longer present in modern GCC, and the workaround was busted if the compiler happened to assign r7 to the __res variable. More discussion at https://bugzilla.mozilla.org/show_bug.cgi?id=1247399#c12 R=markus@chromium.org BUG=https://bugs.chromium.org/p/linux-syscall-support/issues/detail?id=13 Patch contributed by Ted Mielczarek (ted.mielczarek@gmail.com). Committed: https://chromium.googlesource.com/linux-syscall-support/+/1549d20f6d3e7d66bb4... ==========
Message was sent while issue was closed.
Committed patchset #1 (id:1) manually as 1549d20f6d3e7d66bb4e687c0ab9da42c2bff2ac (presubmit successful).
Message was sent while issue was closed.
thakis@chromium.org changed reviewers: + thakis@chromium.org
Message was sent while issue was closed.
This doesn't build: https://codereview.chromium.org/2787843002/ https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Ftryserver.chromium.andro... ../../breakpad/src/client/linux/handler/exception_handler.cc: In member function 'bool google_breakpad::ExceptionHandler::GenerateDump(google_breakpad::ExceptionHandler::CrashContext*)': ../../breakpad/src/client/linux/handler/exception_handler.cc:554:1: error: r7 cannot be used in asm here } ^ ../../breakpad/src/client/linux/handler/exception_handler.cc:554:1: error: r7 cannot be used in asm here (error from this call: https://cs.chromium.org/chromium/src/breakpad/src/client/linux/handler/except... Can we revert this?
Message was sent while issue was closed.
On 2017/03/30 15:39:28, Nico (afk until Tue Apr 4) wrote: > This doesn't build: > > https://codereview.chromium.org/2787843002/ > https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Ftryserver.chromium.andro... > > ../../breakpad/src/client/linux/handler/exception_handler.cc: In member function > 'bool > google_breakpad::ExceptionHandler::GenerateDump(google_breakpad::ExceptionHandler::CrashContext*)': > ../../breakpad/src/client/linux/handler/exception_handler.cc:554:1: error: r7 > cannot be used in asm here > } > ^ > ../../breakpad/src/client/linux/handler/exception_handler.cc:554:1: error: r7 > cannot be used in asm here > > (error from this call: > https://cs.chromium.org/chromium/src/breakpad/src/client/linux/handler/except... > > > Can we revert this? (This is with the newest SDK available in the Android NDK.)
Message was sent while issue was closed.
On 2017/03/30 16:00:21, Nico (afk until Tue Apr 4) wrote: > On 2017/03/30 15:39:28, Nico (afk until Tue Apr 4) wrote: > > This doesn't build: > > > > https://codereview.chromium.org/2787843002/ > > > https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Ftryserver.chromium.andro... > > > > ../../breakpad/src/client/linux/handler/exception_handler.cc: In member > function > > 'bool > > > google_breakpad::ExceptionHandler::GenerateDump(google_breakpad::ExceptionHandler::CrashContext*)': > > ../../breakpad/src/client/linux/handler/exception_handler.cc:554:1: error: r7 > > cannot be used in asm here > > } > > ^ > > ../../breakpad/src/client/linux/handler/exception_handler.cc:554:1: error: r7 > > cannot be used in asm here > > > > (error from this call: > > > https://cs.chromium.org/chromium/src/breakpad/src/client/linux/handler/except... > > > > > > Can we revert this? > > (This is with the newest SDK available in the Android NDK.) *the newest gcc
Message was sent while issue was closed.
Revert: https://chromium-review.googlesource.com/c/463646/ |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
