NaCl: Update revision in DEPS, r7866 -> r7919

NaCl: Update revision in DEPS, r7866 -> r7919

This pulls in the following Native Client changes:

r7867: (sehr) SelLdrLauncher refactoring.
r7868: (bradchen) Remove redundant defn and use of asm_HasCPUID().
r7869: (mcgrathr) Add test case for sleep function
r7870: (pdox) Move the sandboxed translator build to its own archive.
r7871: (pdox) Roll PNaCl toolchain DEPS to 7870, and commit second part of CL:
r7873: (pdox) Whitespace change to trigger rebuild
r7874: (pdox) Re-enable browser tests on the PNaCl Windows toolchain bots.
r7875: (jvoung) Halt pnacl test example animations when loaded
r7876: (dschuff) For sandboxed translation, use LLC-returned DT_NEEDED values instead of using
r7877: (thakis) Use whitespace instead of ',' to separate declaration modifiers.
r7878: (mcgrathr) Update glibc revision to 32d9fc1
r7879: (mseaborn) Exception handling: Add test for crashes inside trusted code
r7880: (mseaborn) Exception handling: Ensure exception_stack and exception_flag are uint32_t
r7881: (jvoung) Set 'chromium_code' to true in gyp files to get more compiler warnings (when building the nacl plugin in chrome).
r7882: (pasko) validator_ragel: cleanup: Move *.C -> *.cc
r7883: (pasko) validator_ragel: Fix easiest presubmit warnings in gen-decoder
r7884: (dschuff) Fix handle count assertion in NaClMainForChromium
r7885: (dschuff) Remove addition of unused IMC handles
r7886: (mcgrathr) Update gcc revision to c69a5b7
r7887: (mseaborn) Exception handling: Reorganise Windows implementation
r7888: (pdox) PNaCl: Remove library sym-links from the translator directories.
r7889: (jvoung) Bump pnacl binutils rev for tiny commandline change.
r7890: (pasko) validator_ragel: reporting x86-32 error offsets as in x86-64
r7891: (jvoung) Revert "Set 'chromium_code' to true in gyp files to get more compiler warnings
r7892: (kschimpf) Change NCPrintInst to NCPrintInstWithHex to conform with function
r7893: (mcgrathr) Update toolchains
r7895: (sehr) Correct the socket number that the startup information is transmitted on.
r7896: (bradchen) Revert r7893, which breaks appears to break SPEC perlbmk.
r7898: (bradchen) Fix 7896 to properly revert 7893
r7900: (pdox) Roll toolchain DEPS to 7898, and patch perlbmk run scripts to provide a symlink to the "lenums" datafile.
r7902: (bradchen) Re-enable sleep test.
r7903: (kschimpf) Move private test enuminsts into native client.
r7904: (mseaborn) Exception handling: Add support for x86-64 Windows
r7906: (mseaborn) Exception handling: Rename "debug_exception_handler_32.c" to remove "_32" suffix
r7907: (mseaborn) ARM: Remove mallopt() hack now that the bottom 1GB is reserved properly
r7908: (mseaborn) Exception handling: Remove code that exposes &nacl_user and &nap->exception_handler
r7909: (mseaborn) Exception handling: Move exception_stack/exception_flag to NaClAppThread
r7910: (pasko) validator_ragel: 32bit tests borrowed from the old validator
r7911: (jvoung) Bump chrome deps and change scons to use "pnacl-translate" for pnacl bitcode to match. Also use the pnacl-nmf generator by default to get hashes.
r7913: (mcgrathr) Build as PIE in Gyp build.
r7914: (mseaborn) Revert SelLdrLauncher refactoring: r7867, r7884, r7885, r7895
r7915: (mseaborn) Cleanup: Add a header for sel_main_chrome.c
r7916: (kschimpf) Don't report pre/post conditions if instruction wouldn't be valid anyway.
r7917: (kschimpf) Fix API calls to NaClGetExpKidIndex and NaClGetExpParentIndex to properly
r7918: (pdox) PNaCl Driver: Disable logging and multi-threaded I/O by default.
r7919: (mseaborn) Revert r7913: "Build as PIE in Gyp build."

BUG=none
TEST=nacl_integration

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=124566

[Mark Seaborn, 2012-03-02 01:19:59]

The bots are flaky, but:
 * nacl_integration passes consistently.
 * the ui_tests haven't failed the same way twice, so I assume this is due to
Chromium-side changes and not the NaCl rev update.

[jvoung - send to chromium..., 2012-03-02 01:28:36]

https://chromiumcodereview.appspot.com/9564018/diff/1/DEPS
File DEPS (right):

https://chromiumcodereview.appspot.com/9564018/diff/1/DEPS#newcode37
DEPS:37: "nacl_toolchain_revision": "7799",
Should the toolchain revs be updated too to match the revs in NaCl at 7919?

[Mark Seaborn, 2012-03-02 01:35:01]

On 1 March 2012 17:28, <jvoung@google.com> wrote:

>
>
https://chromiumcodereview.**appspot.com/9564018/diff/1/**DEPS<https://chromi...
> File DEPS (right):
>
>
https://chromiumcodereview.**appspot.com/9564018/diff/1/**DEPS#newcode37<http...
> DEPS:37: "nacl_toolchain_revision": "7799",
> Should the toolchain revs be updated too to match the revs in NaCl at
> 7919?
>

It's not a necessity, which is why it's optional in nacl_deps_bump.py.
Toolchain changes rarely affect NaCl-in-Chromium, and pulling them in just
slows the bots and developers down unnecessarily as new tarballs are
downloaded.

nacl_revision in Chromium is a week behind, so I'd rather get this update
in soon than have to wait for another Chromium trybot run.

Mark

-- 
You received this message because you are subscribed to the Google Groups
"Native-Client-Reviews" group.
To post to this group, send email to native-client-reviews@googlegroups.com.
To unsubscribe from this group, send email to
native-client-reviews+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/native-client-reviews?hl=en.

[Mark Seaborn, 2012-03-02 01:35:03]

On 1 March 2012 17:28, <jvoung@google.com> wrote:

>
>
https://chromiumcodereview.**appspot.com/9564018/diff/1/**DEPS<https://chromi...
> File DEPS (right):
>
>
https://chromiumcodereview.**appspot.com/9564018/diff/1/**DEPS#newcode37<http...
> DEPS:37: "nacl_toolchain_revision": "7799",
> Should the toolchain revs be updated too to match the revs in NaCl at
> 7919?
>

It's not a necessity, which is why it's optional in nacl_deps_bump.py.
Toolchain changes rarely affect NaCl-in-Chromium, and pulling them in just
slows the bots and developers down unnecessarily as new tarballs are
downloaded.

nacl_revision in Chromium is a week behind, so I'd rather get this update
in soon than have to wait for another Chromium trybot run.

Mark

[bradn, 2012-03-02 01:40:08]

lgtm

[jvoung - send to chromium..., 2012-03-02 01:40:39]

On 2012/03/02 01:35:03, Mark Seaborn wrote:
> On 1 March 2012 17:28, <mailto:jvoung@google.com> wrote:
> 
> >
> >
>
https://chromiumcodereview.**appspot.com/9564018/diff/1/**DEPS%3Chttps://chro...>
> > File DEPS (right):
> >
> >
>
https://chromiumcodereview.**appspot.com/9564018/diff/1/**DEPS#newcode37%3Cht...>
> > DEPS:37: "nacl_toolchain_revision": "7799",
> > Should the toolchain revs be updated too to match the revs in NaCl at
> > 7919?
> >
> 
> It's not a necessity, which is why it's optional in nacl_deps_bump.py.
> Toolchain changes rarely affect NaCl-in-Chromium, and pulling them in just
> slows the bots and developers down unnecessarily as new tarballs are
> downloaded.
> 
> nacl_revision in Chromium is a week behind, so I'd rather get this update
> in soon than have to wait for another Chromium trybot run.
> 
> Mark

Okay.  I guess occasionally we would need to update it to pull in the new
toolchain to update the SDK tarballs, but that can be done another time.

[Mark Seaborn, 2012-03-05 16:18:44]

On 1 March 2012 17:40, <jvoung@google.com> wrote:

> On 2012/03/02 01:35:03, Mark Seaborn wrote:
>
>> On 1 March 2012 17:28, <mailto:jvoung@google.com> wrote:
>>
>
>  > DEPS:37: "nacl_toolchain_revision": "7799",
>> > Should the toolchain revs be updated too to match the revs in NaCl at
>> > 7919?
>> >
>>
>
>  It's not a necessity, which is why it's optional in nacl_deps_bump.py.
>> Toolchain changes rarely affect NaCl-in-Chromium, and pulling them in just
>> slows the bots and developers down unnecessarily as new tarballs are
>> downloaded.
>>
>
>  nacl_revision in Chromium is a week behind, so I'd rather get this update
>> in soon than have to wait for another Chromium trybot run.
>>
>
>  Mark
>>
>
> Okay.  I guess occasionally we would need to update it to pull in the new
> toolchain to update the SDK tarballs, but that can be done another time.
>

That's a good point.  I hadn't thought about the SDK.  I will change the
default in nacl_deps_bump.py to be to update the toolchain.

Mark

-- 
You received this message because you are subscribed to the Google Groups
"Native-Client-Reviews" group.
To post to this group, send email to native-client-reviews@googlegroups.com.
To unsubscribe from this group, send email to
native-client-reviews+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/native-client-reviews?hl=en.

[Mark Seaborn, 2012-03-05 16:18:51]

On 1 March 2012 17:40, <jvoung@google.com> wrote:

> On 2012/03/02 01:35:03, Mark Seaborn wrote:
>
>> On 1 March 2012 17:28, <mailto:jvoung@google.com> wrote:
>>
>
>  > DEPS:37: "nacl_toolchain_revision": "7799",
>> > Should the toolchain revs be updated too to match the revs in NaCl at
>> > 7919?
>> >
>>
>
>  It's not a necessity, which is why it's optional in nacl_deps_bump.py.
>> Toolchain changes rarely affect NaCl-in-Chromium, and pulling them in just
>> slows the bots and developers down unnecessarily as new tarballs are
>> downloaded.
>>
>
>  nacl_revision in Chromium is a week behind, so I'd rather get this update
>> in soon than have to wait for another Chromium trybot run.
>>
>
>  Mark
>>
>
> Okay.  I guess occasionally we would need to update it to pull in the new
> toolchain to update the SDK tarballs, but that can be done another time.
>

That's a good point.  I hadn't thought about the SDK.  I will change the
default in nacl_deps_bump.py to be to update the toolchain.

Mark

[Nico, 2012-03-05 16:25:34]

On Mon, Mar 5, 2012 at 8:18 AM, Mark Seaborn <mseaborn@chromium.org> wrote:
> On 1 March 2012 17:40, <jvoung@google.com> wrote:
>>
>> On 2012/03/02 01:35:03, Mark Seaborn wrote:
>>>
>>> On 1 March 2012 17:28, <mailto:jvoung@google.com> wrote:
>>
>>
>>> > DEPS:37: "nacl_toolchain_revision": "7799",
>>> > Should the toolchain revs be updated too to match the revs in NaCl at
>>> > 7919?
>>> >
>>
>>
>>> It's not a necessity, which is why it's optional in nacl_deps_bump.py.
>>> Toolchain changes rarely affect NaCl-in-Chromium, and pulling them in
>>> just
>>> slows the bots and developers down unnecessarily as new tarballs are
>>> downloaded.
>>
>>
>>> nacl_revision in Chromium is a week behind, so I'd rather get this update
>>> in soon than have to wait for another Chromium trybot run.
>>
>>
>>> Mark
>>
>>
>> Okay.  I guess occasionally we would need to update it to pull in the new
>> toolchain to update the SDK tarballs, but that can be done another time.
>
>
> That's a good point.  I hadn't thought about the SDK.  I will change the
> default in nacl_deps_bump.py to be to update the toolchain.

What happens when the SDK has out of date tarballs?

It would be nice if chromium devs wouldn't have to download 600MB (I
think that's what the toolchain binary size is) every week.

-- 
You received this message because you are subscribed to the Google Groups
"Native-Client-Reviews" group.
To post to this group, send email to native-client-reviews@googlegroups.com.
To unsubscribe from this group, send email to
native-client-reviews+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/native-client-reviews?hl=en.

[Nico, 2012-03-05 16:25:38]

On Mon, Mar 5, 2012 at 8:18 AM, Mark Seaborn <mseaborn@chromium.org> wrote:
> On 1 March 2012 17:40, <jvoung@google.com> wrote:
>>
>> On 2012/03/02 01:35:03, Mark Seaborn wrote:
>>>
>>> On 1 March 2012 17:28, <mailto:jvoung@google.com> wrote:
>>
>>
>>> > DEPS:37: "nacl_toolchain_revision": "7799",
>>> > Should the toolchain revs be updated too to match the revs in NaCl at
>>> > 7919?
>>> >
>>
>>
>>> It's not a necessity, which is why it's optional in nacl_deps_bump.py.
>>> Toolchain changes rarely affect NaCl-in-Chromium, and pulling them in
>>> just
>>> slows the bots and developers down unnecessarily as new tarballs are
>>> downloaded.
>>
>>
>>> nacl_revision in Chromium is a week behind, so I'd rather get this update
>>> in soon than have to wait for another Chromium trybot run.
>>
>>
>>> Mark
>>
>>
>> Okay.  I guess occasionally we would need to update it to pull in the new
>> toolchain to update the SDK tarballs, but that can be done another time.
>
>
> That's a good point.  I hadn't thought about the SDK.  I will change the
> default in nacl_deps_bump.py to be to update the toolchain.

What happens when the SDK has out of date tarballs?

It would be nice if chromium devs wouldn't have to download 600MB (I
think that's what the toolchain binary size is) every week.