win 2015: disable allocator shim

win 2015: disable allocator shim

The allocator shimming doesn't work on the 2015 CRT and we are hoping
to be able to remove it if an additional feature lands in the 2015 CRT
API. For now, don't shim and revisit once VS2015 is RTM.

R=wfh@chromium.org
BUG=481611
Committed: https://crrev.com/58b00a59bff3c44ef6e41f77d2afba4408f4564a
Cr-Commit-Position: refs/heads/master@{#327297}

[scottmg, 2015-04-27 20:36:47]

er, wait. that's not right, one sec.

[scottmg, 2015-04-27 20:39:53]

Patchset #2 (id:20001) has been deleted

[scottmg, 2015-04-27 20:42:02]

ok, let's go with that. ptal.

[scottmg, 2015-04-27 20:52:52]

oh feh, that doesn't actually work because it doesn't stop the prep_libc script
from running. ignore me for a while again.

sorry.

[Will Harris, 2015-04-27 21:45:06]

On 2015/04/27 20:52:52, scottmg wrote:
> oh feh, that doesn't actually work because it doesn't stop the prep_libc
script
> from running. ignore me for a while again.
> 
> sorry.

wondering if this should be done inside allocator.gyp:allocator target instead?

[scottmg, 2015-04-27 21:47:28]

On 2015/04/27 21:45:06, Will Harris wrote:
> On 2015/04/27 20:52:52, scottmg wrote:
> > oh feh, that doesn't actually work because it doesn't stop the prep_libc
> script
> > from running. ignore me for a while again.
> > 
> > sorry.
> 
> wondering if this should be done inside allocator.gyp:allocator target
instead?

Which part? It needs to not have ALLOCATOR_SHIM defined too?

[scottmg, 2015-04-27 22:29:06]

On 2015/04/27 21:47:28, scottmg wrote:
> On 2015/04/27 21:45:06, Will Harris wrote:
> > On 2015/04/27 20:52:52, scottmg wrote:
> > > oh feh, that doesn't actually work because it doesn't stop the prep_libc
> > script
> > > from running. ignore me for a while again.
> > > 
> > > sorry.
> > 
> > wondering if this should be done inside allocator.gyp:allocator target
> instead?
> 
> Which part? It needs to not have ALLOCATOR_SHIM defined too?

Oh, you meant the prep_libc.py script part? Yeah, that makes sense. Done.

[Will Harris, 2015-04-27 22:42:29]

hmm if win_use_allocator_shim is 0 then no targets should be pulling in the
allocator.gyp:allocator dependency (which in turn pulls in the libcmt
dependency).  What dependency chain was causing prep_libc.py to run when you
tried patch set 2?

[scottmg, 2015-04-27 23:09:07]

On 2015/04/27 22:42:29, Will Harris wrote:
> hmm if win_use_allocator_shim is 0 then no targets should be pulling in the
> allocator.gyp:allocator dependency (which in turn pulls in the libcmt
> dependency).  What dependency chain was causing prep_libc.py to run when you
> tried patch set 2?

Up until ps4 I had the wrong condition in common.gypi

<(MSVS_VERSION) == "2015" is wrong (silently, natch), MSVS_VERSION == "2015" is
right.

After that, hmm... I was testing with `ninja -C out\Debug libcmt` so I didn't
notice that it wasn't actually referenced.

So, yeah. This has been a productive 3 hours. I blame gyp.

[Will Harris, 2015-04-28 04:28:05]

lgtm

[Will Harris, 2015-04-28 04:35:54]

one random thought - does component get set to "static_library" or empty by
default? - just verifying - since that if statement changes behavior if
component can be something other than "shared_library" or "static_library"

[scottmg, 2015-04-28 15:11:27]

On 2015/04/28 04:35:54, Will Harris wrote:
> one random thought - does component get set to "static_library" or empty by
> default? - just verifying - since that if statement changes behavior if
> component can be something other than "shared_library" or "static_library"

I think that would break the world. It's set to static by default, yes:
https://code.google.com/p/chromium/codesearch#chromium/src/build/common.gypi&...

[scottmg, 2015-04-28 15:11:43]

The CQ bit was checked by scottmg@chromium.org

[commit-bot: I haz the power, 2015-04-28 15:11:51]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1105213002/100001

[commit-bot: I haz the power, 2015-04-28 16:14:48]

Committed patchset #5 (id:100001)

[commit-bot: I haz the power, 2015-04-28 16:15:37]

Patchset 5 (id:??) landed as
https://crrev.com/58b00a59bff3c44ef6e41f77d2afba4408f4564a
Cr-Commit-Position: refs/heads/master@{#327297}