Fix GLHelperReadbackSupport::SupportsFormat crash

Fix GLHelperReadbackSupport::SupportsFormat crash

GetString(GL_EXTENSIONS) can return NULL if the context was lost. Handle
gracefully.

BUG=421608
Committed: https://crrev.com/679d31a211c8844eda3ec6b51efadb54e8fe1d0b
Cr-Commit-Position: refs/heads/master@{#298759}

[piman, 2014-10-08 21:22:03]

piman@chromium.org changed reviewers:
+ sievers@chromium.org

[no sievers, 2014-10-08 21:27:13]

lgtm

https://codereview.chromium.org/639073003/diff/1/content/common/gpu/client/gl...
File content/common/gpu/client/gl_helper_readback_support.cc (right):

https://codereview.chromium.org/639073003/diff/1/content/common/gpu/client/gl...
content/common/gpu/client/gl_helper_readback_support.cc:98:
format_cache_.push_back(entry);
Should we also not do this if format_out and type_out are 0?

[piman, 2014-10-08 21:45:31]

On 2014/10/08 21:27:13, sievers wrote:
> lgtm
> 
>
https://codereview.chromium.org/639073003/diff/1/content/common/gpu/client/gl...
> File content/common/gpu/client/gl_helper_readback_support.cc (right):
> 
>
https://codereview.chromium.org/639073003/diff/1/content/common/gpu/client/gl...
> content/common/gpu/client/gl_helper_readback_support.cc:98:
> format_cache_.push_back(entry);
> Should we also not do this if format_out and type_out are 0?

Doesn't matter either way. Leaving 0,0 in the cache is essentially caching the
"fail" result, which is valid (the whole GLHelperReadbackSupport hence the cache
is recreated for each new context / GLHelper).

A valid question is whether GetAdditionalFormat clients handle 0,0 but they all
do, unless I missed something. Fundamentally you need to understand the enum
results to be able to correctly allocate buffers and convert data, so you
necessarily have explicit tests against "well known" enum values, and 0,0 would
fail these tests.

[piman, 2014-10-08 21:45:35]

The CQ bit was checked by piman@chromium.org

[commit-bot: I haz the power, 2014-10-08 21:50:46]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/639073003/1

[commit-bot: I haz the power, 2014-10-09 04:32:06]

Committed patchset #1 (id:1) as 227fc9ae75293e22505b4bc01f6d5955b8ff2166

[commit-bot: I haz the power, 2014-10-09 04:33:15]

Patchset 1 (id:??) landed as
https://crrev.com/679d31a211c8844eda3ec6b51efadb54e8fe1d0b
Cr-Commit-Position: refs/heads/master@{#298759}