Check if internal_format is valid for GenerateMipmap().

Check if internal_format is valid for GenerateMipmap().

BUG=429053
TEST=gpu_unittests,webgl2_conformance
R=piman@chromium.org

Committed: https://crrev.com/33040da405658793aa7b248e8ea6e292fad46193
Cr-Commit-Position: refs/heads/master@{#370562}

[Zhenyao Mo, 2016-01-20 22:22:38]

piman: PTAL

Note that I only enable the affected test on Linux. That's because I only tested
on Linux NVIDIA.  If it turns out to work on Mac/Win, I'll update in a separate
CL accordingly.

[Zhenyao Mo, 2016-01-20 22:22:43]

The CQ bit was checked by zmo@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-20 22:25:21]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1605323004/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1605323004/1

[piman, 2016-01-20 23:03:51]

lgtm

https://codereview.chromium.org/1605323004/diff/1/gpu/command_buffer/service/...
File gpu/command_buffer/service/texture_manager.cc (right):

https://codereview.chromium.org/1605323004/diff/1/gpu/command_buffer/service/...
gpu/command_buffer/service/texture_manager.cc:599: if
(feature_info->IsES3Enabled()) {
nit: I think you should go through this code even if
!feature_info->IsES3Enabled(), for the cases where, say, floating point formats
are enabled even though full ES3 support isn't there.
In a sense, ES3 shouldn't be more restrictive than ES2 (and this code only adds
restrictions).

[Zhenyao Mo, 2016-01-20 23:25:38]

https://codereview.chromium.org/1605323004/diff/1/gpu/command_buffer/service/...
File gpu/command_buffer/service/texture_manager.cc (right):

https://codereview.chromium.org/1605323004/diff/1/gpu/command_buffer/service/...
gpu/command_buffer/service/texture_manager.cc:599: if
(feature_info->IsES3Enabled()) {
On 2016/01/20 23:03:51, piman (Slow to review) wrote:
> nit: I think you should go through this code even if
> !feature_info->IsES3Enabled(), for the cases where, say, floating point
formats
> are enabled even though full ES3 support isn't there.
> In a sense, ES3 shouldn't be more restrictive than ES2 (and this code only
adds
> restrictions).

Done.

[Zhenyao Mo, 2016-01-20 23:25:53]

The CQ bit was checked by zmo@chromium.org

[Zhenyao Mo, 2016-01-20 23:25:53]

The patchset sent to the CQ was uploaded after l-g-t-m from piman@chromium.org
Link to the patchset: https://codereview.chromium.org/1605323004/#ps20001
(title: " ")

[commit-bot: I haz the power, 2016-01-20 23:27:25]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1605323004/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1605323004/20001

[commit-bot: I haz the power, 2016-01-21 02:13:48]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-01-21 02:14:59]

Description was changed from

==========
Check if internal_format is valid for GenerateMipmap().

BUG=429053
TEST=gpu_unittests,webgl2_conformance
R=piman@chromium.org
==========

to

==========
Check if internal_format is valid for GenerateMipmap().

BUG=429053
TEST=gpu_unittests,webgl2_conformance
R=piman@chromium.org

Committed: https://crrev.com/33040da405658793aa7b248e8ea6e292fad46193
Cr-Commit-Position: refs/heads/master@{#370562}
==========

[commit-bot: I haz the power, 2016-01-21 02:15:00]

Patchset 2 (id:??) landed as
https://crrev.com/33040da405658793aa7b248e8ea6e292fad46193
Cr-Commit-Position: refs/heads/master@{#370562}

[alancutter (OOO until 2018), 2016-01-21 04:04:40]

A revert of this CL (patchset #2 id:20001) has been created in
https://codereview.chromium.org/1619453002/ by alancutter@chromium.org.

The reason for reverting is: Broke the following webkit_tests on all platforms:
virtual/gpu/fast/canvas/canvas-imageSmoothingQuality.html
virtual/gpu/fast/canvas/canvas-imageSmoothingQuality-pixel.html.

[Zhenyao Mo, 2016-01-21 22:02:38]

On 2016/01/21 04:04:40, alancutter wrote:
> A revert of this CL (patchset #2 id:20001) has been created in
> https://codereview.chromium.org/1619453002/ by mailto:alancutter@chromium.org.
> 
> The reason for reverting is: Broke the following webkit_tests on all
platforms:
> virtual/gpu/fast/canvas/canvas-imageSmoothingQuality.html
> virtual/gpu/fast/canvas/canvas-imageSmoothingQuality-pixel.html.

After some digging, it is the BGRA (!) format that I didn't include in the
texture_unsized_internal_format validator.