Restore deps/third_party/valgrind to its state at r48920

Restore deps/third_party/valgrind to its state at r48920. This backs out
r48921, r48923, and r49047.

BUG=none
TEST=none

[Mark Mentovai, 2010-06-08 18:14:31]

TBR (per your request)

[TVL, 2010-06-08 18:23:56]

tx

[willchan no longer on Chromium, 2010-06-08 18:34:16]

[ +timurrrr, +glider ]

Heads up for the people whose changes got reverted.  glider's change
broke the chromium os valgrind bots, which seems to be because the
test binaries were not built with USE_TCMALLOC (actually, we don't
seem to be defining that anywhere anymore.  so maybe
process_util_linux.cc is broken?).  Something's broken there.
Timurrrr's changes were reverted too because they had binary changes
to the same file which prevented a clean revert of glider's change.
Sorry.  Mark notes that the shell scripts should not be marked as
binary.

On Tue, Jun 8, 2010 at 11:14 AM,  <mark@chromium.org> wrote:
> Reviewers: willchan, TVL,
>
> Message:
> TBR (per your request)
>
> Description:
> Restore deps/third_party/valgrind to its state at r48920. This backs out
> r48921, r48923, and r49047.
>
> BUG=none
> TEST=none
>
> Please review this at http://codereview.chromium.org/2746004/show
>
> SVN Base: svn://svn.chromium.org/chrome/trunk/deps/third_party/valgrind/
>
> Affected files:
>  M     binaries/linux_x64/bin/no_op_client_for_valgrind
>  M     binaries/linux_x64/bin/valgrind
>  M     binaries/linux_x64/bin/valgrind-listener
>  M     binaries/linux_x64/bin/valgrind-tsan.sh
>  M     binaries/linux_x64/lib/valgrind/memcheck-amd64-linux
>  M     binaries/linux_x64/lib/valgrind/memcheck-x86-linux
>  M     binaries/linux_x64/lib/valgrind/vgpreload_core-amd64-linux.so
>  M     binaries/linux_x64/lib/valgrind/vgpreload_core-x86-linux.so
>  M     binaries/linux_x64/lib/valgrind/vgpreload_memcheck-amd64-linux.so
>  M     binaries/linux_x64/lib/valgrind/vgpreload_memcheck-x86-linux.so
>  M     binaries/linux_x86/bin/no_op_client_for_valgrind
>  M     binaries/linux_x86/bin/valgrind
>  M     binaries/linux_x86/bin/valgrind-listener
>  M     binaries/linux_x86/bin/valgrind-tsan.sh
>  M     binaries/linux_x86/lib/valgrind/memcheck-x86-linux
>  M     binaries/linux_x86/lib/valgrind/vgpreload_core-x86-linux.so
>  M     binaries/linux_x86/lib/valgrind/vgpreload_memcheck-x86-linux.so
>  binaries/mac/bin/ms_print
>  M     binaries/mac/bin/no_op_client_for_valgrind
>  M     binaries/mac/bin/valgrind
>  binaries/mac/bin/valgrind-listener
>  M     binaries/mac/bin/valgrind-tsan.sh
>  M     binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so
>  binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so.dSYM/Contents/Info.plist
>  M
>
binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so.dSYM/Contents/Resources/DWARF/libmpiwrap-x86-darwin.so
>  M     binaries/mac/lib/valgrind/memcheck-x86-darwin
>  binaries/mac/lib/valgrind/memcheck-x86-darwin.dSYM/Contents/Info.plist
>
 binaries/mac/lib/valgrind/memcheck-x86-darwin.dSYM/Contents/Resources/DWARF/memcheck-x86-darwin
>  M     binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so
>  M
>
binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so.dSYM/Contents/Info.plist
>  M
>
binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so.dSYM/Contents/Resources/DWARF/vgpreload_core-x86-darwin.so
>  M     binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so
>  M
>
binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so.dSYM/Contents/Info.plist
>  M
>
binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so.dSYM/Contents/Resources/DWARF/vgpreload_memcheck-x86-darwin.so
>  M     scripts/build-tsan-for-chromium.sh
>  M     scripts/common.sh
>  scripts/fork.patch
>  M     scripts/intercept_tcmalloc.patch
>
>
>

[Timur Iskhodzhanov, 2010-06-08 18:42:37]

Re: shell script - is it about valgrind-tsan.sh?
It is a binary, just open it using text editor and you'll see the
magic we're using to deploy TSan as a single file, so binary flag is
appropriate.

Please consider adding me and glider to the cc list of the CLs
regarding Valgrind binaries so at least we know they are reverted.
I'll un-revert TSan binaries now since they are in use only on TSan
bots and they don't have issues with TCMalloc.

Alexander, please investigate the reports so see if they are related
to your TCMalloc patch.

Timur Iskhodzhanov,
Google Russia


On Tue, Jun 8, 2010 at 10:33 PM, William Chan (陈智昌)
<willchan@chromium.org> wrote:
> [ +timurrrr, +glider ]
>
> Heads up for the people whose changes got reverted.  glider's change
> broke the chromium os valgrind bots, which seems to be because the
> test binaries were not built with USE_TCMALLOC (actually, we don't
> seem to be defining that anywhere anymore.  so maybe
> process_util_linux.cc is broken?).  Something's broken there.
> Timurrrr's changes were reverted too because they had binary changes
> to the same file which prevented a clean revert of glider's change.
> Sorry.  Mark notes that the shell scripts should not be marked as
> binary.
>
> On Tue, Jun 8, 2010 at 11:14 AM,  <mark@chromium.org> wrote:
>> Reviewers: willchan, TVL,
>>
>> Message:
>> TBR (per your request)
>>
>> Description:
>> Restore deps/third_party/valgrind to its state at r48920. This backs out
>> r48921, r48923, and r49047.
>>
>> BUG=none
>> TEST=none
>>
>> Please review this at http://codereview.chromium.org/2746004/show
>>
>> SVN Base: svn://svn.chromium.org/chrome/trunk/deps/third_party/valgrind/
>>
>> Affected files:
>>  M     binaries/linux_x64/bin/no_op_client_for_valgrind
>>  M     binaries/linux_x64/bin/valgrind
>>  M     binaries/linux_x64/bin/valgrind-listener
>>  M     binaries/linux_x64/bin/valgrind-tsan.sh
>>  M     binaries/linux_x64/lib/valgrind/memcheck-amd64-linux
>>  M     binaries/linux_x64/lib/valgrind/memcheck-x86-linux
>>  M     binaries/linux_x64/lib/valgrind/vgpreload_core-amd64-linux.so
>>  M     binaries/linux_x64/lib/valgrind/vgpreload_core-x86-linux.so
>>  M     binaries/linux_x64/lib/valgrind/vgpreload_memcheck-amd64-linux.so
>>  M     binaries/linux_x64/lib/valgrind/vgpreload_memcheck-x86-linux.so
>>  M     binaries/linux_x86/bin/no_op_client_for_valgrind
>>  M     binaries/linux_x86/bin/valgrind
>>  M     binaries/linux_x86/bin/valgrind-listener
>>  M     binaries/linux_x86/bin/valgrind-tsan.sh
>>  M     binaries/linux_x86/lib/valgrind/memcheck-x86-linux
>>  M     binaries/linux_x86/lib/valgrind/vgpreload_core-x86-linux.so
>>  M     binaries/linux_x86/lib/valgrind/vgpreload_memcheck-x86-linux.so
>>  binaries/mac/bin/ms_print
>>  M     binaries/mac/bin/no_op_client_for_valgrind
>>  M     binaries/mac/bin/valgrind
>>  binaries/mac/bin/valgrind-listener
>>  M     binaries/mac/bin/valgrind-tsan.sh
>>  M     binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so
>>  binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so.dSYM/Contents/Info.plist
>>  M
>>
binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so.dSYM/Contents/Resources/DWARF/libmpiwrap-x86-darwin.so
>>  M     binaries/mac/lib/valgrind/memcheck-x86-darwin
>>  binaries/mac/lib/valgrind/memcheck-x86-darwin.dSYM/Contents/Info.plist
>>
 binaries/mac/lib/valgrind/memcheck-x86-darwin.dSYM/Contents/Resources/DWARF/memcheck-x86-darwin
>>  M     binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so
>>  M
>>
binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so.dSYM/Contents/Info.plist
>>  M
>>
binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so.dSYM/Contents/Resources/DWARF/vgpreload_core-x86-darwin.so
>>  M     binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so
>>  M
>>
binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so.dSYM/Contents/Info.plist
>>  M
>>
binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so.dSYM/Contents/Resources/DWARF/vgpreload_memcheck-x86-darwin.so
>>  M     scripts/build-tsan-for-chromium.sh
>>  M     scripts/common.sh
>>  scripts/fork.patch
>>  M     scripts/intercept_tcmalloc.patch
>>
>>
>>
>

[willchan no longer on Chromium, 2010-06-08 19:16:23]

On Tue, Jun 8, 2010 at 11:42 AM, Timur Iskhodzhanov
<timurrrr@chromium.org> wrote:
> Re: shell script - is it about valgrind-tsan.sh?
> It is a binary, just open it using text editor and you'll see the
> magic we're using to deploy TSan as a single file, so binary flag is
> appropriate.

Ok, thanks for the explanation.

>
> Please consider adding me and glider to the cc list of the CLs
> regarding Valgrind binaries so at least we know they are reverted.

Sorry, Mark was probably unaware of what was going on.  thomasvl and I
asked him to help us do the revert, and as soon as it happened, I cc'd
you.  Maybe we can use a WATCHLIST or something in trunk/deps?  I
dunno.

> I'll un-revert TSan binaries now since they are in use only on TSan
> bots and they don't have issues with TCMalloc.

Thanks.

>
> Alexander, please investigate the reports so see if they are related
> to your TCMalloc patch.
>
> Timur Iskhodzhanov,
> Google Russia
>
>
> On Tue, Jun 8, 2010 at 10:33 PM, William Chan (陈智昌)
> <willchan@chromium.org> wrote:
>> [ +timurrrr, +glider ]
>>
>> Heads up for the people whose changes got reverted.  glider's change
>> broke the chromium os valgrind bots, which seems to be because the
>> test binaries were not built with USE_TCMALLOC (actually, we don't
>> seem to be defining that anywhere anymore.  so maybe
>> process_util_linux.cc is broken?).  Something's broken there.
>> Timurrrr's changes were reverted too because they had binary changes
>> to the same file which prevented a clean revert of glider's change.
>> Sorry.  Mark notes that the shell scripts should not be marked as
>> binary.
>>
>> On Tue, Jun 8, 2010 at 11:14 AM,  <mark@chromium.org> wrote:
>>> Reviewers: willchan, TVL,
>>>
>>> Message:
>>> TBR (per your request)
>>>
>>> Description:
>>> Restore deps/third_party/valgrind to its state at r48920. This backs out
>>> r48921, r48923, and r49047.
>>>
>>> BUG=none
>>> TEST=none
>>>
>>> Please review this at http://codereview.chromium.org/2746004/show
>>>
>>> SVN Base: svn://svn.chromium.org/chrome/trunk/deps/third_party/valgrind/
>>>
>>> Affected files:
>>>  M     binaries/linux_x64/bin/no_op_client_for_valgrind
>>>  M     binaries/linux_x64/bin/valgrind
>>>  M     binaries/linux_x64/bin/valgrind-listener
>>>  M     binaries/linux_x64/bin/valgrind-tsan.sh
>>>  M     binaries/linux_x64/lib/valgrind/memcheck-amd64-linux
>>>  M     binaries/linux_x64/lib/valgrind/memcheck-x86-linux
>>>  M     binaries/linux_x64/lib/valgrind/vgpreload_core-amd64-linux.so
>>>  M     binaries/linux_x64/lib/valgrind/vgpreload_core-x86-linux.so
>>>  M     binaries/linux_x64/lib/valgrind/vgpreload_memcheck-amd64-linux.so
>>>  M     binaries/linux_x64/lib/valgrind/vgpreload_memcheck-x86-linux.so
>>>  M     binaries/linux_x86/bin/no_op_client_for_valgrind
>>>  M     binaries/linux_x86/bin/valgrind
>>>  M     binaries/linux_x86/bin/valgrind-listener
>>>  M     binaries/linux_x86/bin/valgrind-tsan.sh
>>>  M     binaries/linux_x86/lib/valgrind/memcheck-x86-linux
>>>  M     binaries/linux_x86/lib/valgrind/vgpreload_core-x86-linux.so
>>>  M     binaries/linux_x86/lib/valgrind/vgpreload_memcheck-x86-linux.so
>>>  binaries/mac/bin/ms_print
>>>  M     binaries/mac/bin/no_op_client_for_valgrind
>>>  M     binaries/mac/bin/valgrind
>>>  binaries/mac/bin/valgrind-listener
>>>  M     binaries/mac/bin/valgrind-tsan.sh
>>>  M     binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so
>>>  binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so.dSYM/Contents/Info.plist
>>>  M
>>>
binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so.dSYM/Contents/Resources/DWARF/libmpiwrap-x86-darwin.so
>>>  M     binaries/mac/lib/valgrind/memcheck-x86-darwin
>>>  binaries/mac/lib/valgrind/memcheck-x86-darwin.dSYM/Contents/Info.plist
>>>
 binaries/mac/lib/valgrind/memcheck-x86-darwin.dSYM/Contents/Resources/DWARF/memcheck-x86-darwin
>>>  M     binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so
>>>  M
>>>
binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so.dSYM/Contents/Info.plist
>>>  M
>>>
binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so.dSYM/Contents/Resources/DWARF/vgpreload_core-x86-darwin.so
>>>  M     binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so
>>>  M
>>>
binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so.dSYM/Contents/Info.plist
>>>  M
>>>
binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so.dSYM/Contents/Resources/DWARF/vgpreload_memcheck-x86-darwin.so
>>>  M     scripts/build-tsan-for-chromium.sh
>>>  M     scripts/common.sh
>>>  scripts/fork.patch
>>>  M     scripts/intercept_tcmalloc.patch
>>>
>>>
>>>
>>
>

[Timur Iskhodzhanov, 2010-06-08 19:24:52]

On Tue, Jun 8, 2010 at 11:16 PM, William Chan (陈智昌)
<willchan@chromium.org> wrote:
> On Tue, Jun 8, 2010 at 11:42 AM, Timur Iskhodzhanov
> <timurrrr@chromium.org> wrote:
>> Re: shell script - is it about valgrind-tsan.sh?
>> It is a binary, just open it using text editor and you'll see the
>> magic we're using to deploy TSan as a single file, so binary flag is
>> appropriate.
>
> Ok, thanks for the explanation.
>
>>
>> Please consider adding me and glider to the cc list of the CLs
>> regarding Valgrind binaries so at least we know they are reverted.
>
> Sorry, Mark was probably unaware of what was going on.  thomasvl and I
> asked him to help us do the revert, and as soon as it happened, I cc'd
> you.  Maybe we can use a WATCHLIST or something in trunk/deps?  I
> dunno.
Ah, alright then. I'll see what we can do with WATCHLISTs, thanks for the tip.

>> I'll un-revert TSan binaries now since they are in use only on TSan
>> bots and they don't have issues with TCMalloc.
>
> Thanks.
>
>>
>> Alexander, please investigate the reports so see if they are related
>> to your TCMalloc patch.
>>
>> Timur Iskhodzhanov,
>> Google Russia
>>
>>
>> On Tue, Jun 8, 2010 at 10:33 PM, William Chan (陈智昌)
>> <willchan@chromium.org> wrote:
>>> [ +timurrrr, +glider ]
>>>
>>> Heads up for the people whose changes got reverted.  glider's change
>>> broke the chromium os valgrind bots, which seems to be because the
>>> test binaries were not built with USE_TCMALLOC (actually, we don't
>>> seem to be defining that anywhere anymore.  so maybe
>>> process_util_linux.cc is broken?).  Something's broken there.
>>> Timurrrr's changes were reverted too because they had binary changes
>>> to the same file which prevented a clean revert of glider's change.
>>> Sorry.  Mark notes that the shell scripts should not be marked as
>>> binary.
>>>
>>> On Tue, Jun 8, 2010 at 11:14 AM,  <mark@chromium.org> wrote:
>>>> Reviewers: willchan, TVL,
>>>>
>>>> Message:
>>>> TBR (per your request)
>>>>
>>>> Description:
>>>> Restore deps/third_party/valgrind to its state at r48920. This backs out
>>>> r48921, r48923, and r49047.
>>>>
>>>> BUG=none
>>>> TEST=none
>>>>
>>>> Please review this at http://codereview.chromium.org/2746004/show
>>>>
>>>> SVN Base: svn://svn.chromium.org/chrome/trunk/deps/third_party/valgrind/
>>>>
>>>> Affected files:
>>>>  M     binaries/linux_x64/bin/no_op_client_for_valgrind
>>>>  M     binaries/linux_x64/bin/valgrind
>>>>  M     binaries/linux_x64/bin/valgrind-listener
>>>>  M     binaries/linux_x64/bin/valgrind-tsan.sh
>>>>  M     binaries/linux_x64/lib/valgrind/memcheck-amd64-linux
>>>>  M     binaries/linux_x64/lib/valgrind/memcheck-x86-linux
>>>>  M     binaries/linux_x64/lib/valgrind/vgpreload_core-amd64-linux.so
>>>>  M     binaries/linux_x64/lib/valgrind/vgpreload_core-x86-linux.so
>>>>  M     binaries/linux_x64/lib/valgrind/vgpreload_memcheck-amd64-linux.so
>>>>  M     binaries/linux_x64/lib/valgrind/vgpreload_memcheck-x86-linux.so
>>>>  M     binaries/linux_x86/bin/no_op_client_for_valgrind
>>>>  M     binaries/linux_x86/bin/valgrind
>>>>  M     binaries/linux_x86/bin/valgrind-listener
>>>>  M     binaries/linux_x86/bin/valgrind-tsan.sh
>>>>  M     binaries/linux_x86/lib/valgrind/memcheck-x86-linux
>>>>  M     binaries/linux_x86/lib/valgrind/vgpreload_core-x86-linux.so
>>>>  M     binaries/linux_x86/lib/valgrind/vgpreload_memcheck-x86-linux.so
>>>>  binaries/mac/bin/ms_print
>>>>  M     binaries/mac/bin/no_op_client_for_valgrind
>>>>  M     binaries/mac/bin/valgrind
>>>>  binaries/mac/bin/valgrind-listener
>>>>  M     binaries/mac/bin/valgrind-tsan.sh
>>>>  M     binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so
>>>>
 binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so.dSYM/Contents/Info.plist
>>>>  M
>>>>
binaries/mac/lib/valgrind/libmpiwrap-x86-darwin.so.dSYM/Contents/Resources/DWARF/libmpiwrap-x86-darwin.so
>>>>  M     binaries/mac/lib/valgrind/memcheck-x86-darwin
>>>>  binaries/mac/lib/valgrind/memcheck-x86-darwin.dSYM/Contents/Info.plist
>>>>
 binaries/mac/lib/valgrind/memcheck-x86-darwin.dSYM/Contents/Resources/DWARF/memcheck-x86-darwin
>>>>  M     binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so
>>>>  M
>>>>
binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so.dSYM/Contents/Info.plist
>>>>  M
>>>>
binaries/mac/lib/valgrind/vgpreload_core-x86-darwin.so.dSYM/Contents/Resources/DWARF/vgpreload_core-x86-darwin.so
>>>>  M     binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so
>>>>  M
>>>>
binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so.dSYM/Contents/Info.plist
>>>>  M
>>>>
binaries/mac/lib/valgrind/vgpreload_memcheck-x86-darwin.so.dSYM/Contents/Resources/DWARF/vgpreload_memcheck-x86-darwin.so
>>>>  M     scripts/build-tsan-for-chromium.sh
>>>>  M     scripts/common.sh
>>>>  scripts/fork.patch
>>>>  M     scripts/intercept_tcmalloc.patch
>>>>
>>>>
>>>>
>>>
>>
>

[Mark Mentovai, 2010-06-08 19:29:20]

William Chan (陈智昌) wrote:
> Sorry, Mark was probably unaware of what was going on.  thomasvl and I
> asked him to help us do the revert, and as soon as it happened, I cc'd
> you.

This is correct. I was acting in my role as fast-svn-bot, when Will
couldn’t get drover to revert this for him.

Mark