Commits · edcc6bc5267c6409ea78d1c857cfa141cf4cb24b · Chen Yisong / angle

24 Nov, 2014 3 commits

Make device lost flag a RendererD3D member. · edcc6bc5

authored Nov 24, 2014

This flag we used in both D3D renderers, and hence we can move it
to the base class to share a bit of common code.

BUG=angle:795

Change-Id: I8bd60796bfeb9590bdf798f9d044f8e20c7ad779
Reviewed-on: https://chromium-review.googlesource.com/228912Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

edcc6bc5

Remove notify parameter from testDeviceLost. · 4c76feac

authored Nov 24, 2014

We can get rid of this parameter now that we call notify
directly on the display in cases where we need to.

BUG=angle:795

Change-Id: I2024b70d0d725e755f7aa742ba221c2d0179d8b6
Reviewed-on: https://chromium-review.googlesource.com/228911Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

4c76feac

Do not use notify feature of testDeviceLost. · 9dd0cf0a

authored Nov 24, 2014

This is part one of a simplification of testDeviceLost. This
patch changes the instances where we would use the notify
parameter to notify the Display directly. We can do this
by using the Display attached to the Renderer in some cases.

BUG=angle:795

Change-Id: I24fd827989d47b0b2cefef7afb99fa62581ddc1b
Reviewed-on: https://chromium-review.googlesource.com/228910Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

9dd0cf0a

21 Nov, 2014 9 commits

Fixed Mac and Linux build breakage. · b23deb26

authored Nov 21, 2014

Introduced in 1edac3b1 .

BUG=none

Change-Id: Ia5c5e38b794e9c991ec8624de34681dcf13a651f
Reviewed-on: https://chromium-review.googlesource.com/231441Tested-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Brandon Jones <bajones@chromium.org>

b23deb26

Rename and slightly reorganize test targets. · 7bff24c1

authored Nov 20, 2014

Use "angle_" prefix for all targets to disambiguate them from any
potential Chromium-side targets.

Combine compiler and preprocessor tests into angle_compiler_tests.
(Chromium's "angle_unittests" target contains both.) Deleted
preprocessor_test_main.cpp.

BUG=angleproject:833

Change-Id: I7acc5a76f49c15d0064c16d2ace256f9cc86e9e8
Reviewed-on: https://chromium-review.googlesource.com/231185Reviewed-by: Kenneth Russell <kbr@chromium.org>
Tested-by: Kenneth Russell <kbr@chromium.org>

7bff24c1

Fixed Visual Studio 2013 Update 4 build failures for WinRT · 1edac3b1

authored Nov 20, 2014

Change-Id: I393e9800d1c2e902f60728e1f7d0debca66b76a0
Reviewed-on: https://chromium-review.googlesource.com/231140Tested-by: Cooper Partin <coopp@microsoft.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

1edac3b1

Update the README to match the one on the google code page. · af224bf0

authored Nov 21, 2014

Change-Id: I05b61d9a7c80a6e53df1b12a99b0c107b570496c
Reviewed-on: https://chromium-review.googlesource.com/231223Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

af224bf0

Update the DISALLOW_COPY_AND_ASSIGN macro to use c++ operator deletion. · e371f1d7

authored Nov 21, 2014

Also remove the TSymbol copy constructor which is not used.

BUG=angle:836

Change-Id: I4f35f554c90d01dc2132d244a1974d9a8dc89bbd
Reviewed-on: https://chromium-review.googlesource.com/231231Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shannon Woods <shannonwoods@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

e371f1d7

Fix releasing textures after we kill D3D11. · cc4cd292

authored Nov 21, 2014

This bug was crashing the IncompleteTextureTest in Release mode.
Fix this by releasing the textures before we free the D3D11
libraries.

BUG=angle:838

Change-Id: I771e4cc6876e3529c5d26f36711dea9181e631f0
Reviewed-on: https://chromium-review.googlesource.com/231247Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shannon Woods <shannonwoods@chromium.org>

cc4cd292

Merge libGLESv2 and libEGL classes into libANGLE. · 2b5420c0

authored Nov 19, 2014

BUG=angle:733

Change-Id: Ic491c971411fe82c56cd97c5c8325ac14ec218df
Reviewed-on: https://chromium-review.googlesource.com/230830Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

2b5420c0

Disallow copying of the MemoryBuffer class. · 746d0ea7

authored Nov 21, 2014

BUG=angle:836

Change-Id: Ie3f674ef6c321770209a37b1af06ea21d788f476
Reviewed-on: https://chromium-review.googlesource.com/231232Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

746d0ea7

Fix leaked ID3D11Resource object. · aad0cd7f

authored Nov 21, 2014

BUG=427245

Change-Id: If4833df5dc405a2a8d87a577bbd47b954613e321
Reviewed-on: https://chromium-review.googlesource.com/231246Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

aad0cd7f

20 Nov, 2014 3 commits

Refactor gmock and gtest dependencies for ANGLE's tests. · 1027aa62

authored Nov 20, 2014

This change allows these targets to be referenced from Chromium's
build projects.

BUG=angleproject:833

Change-Id: I14694f610921d28ddcd8664887086ca30375b7a2
Reviewed-on: https://chromium-review.googlesource.com/231072Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Tested-by: Kenneth Russell <kbr@chromium.org>

1027aa62

Fix bugs in 2D Array Texture support. · f8fccb39

authored Nov 12, 2014

This fixes assertion failures and other errors, reproducible
by running the dEQP Texture Format tests.

BUG=angle:813

Change-Id: I3b97f89323f9656b45f617211fb4579a24013951
Reviewed-on: https://chromium-review.googlesource.com/229351Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shannon Woods <shannonwoods@chromium.org>

f8fccb39

Fix failure initting 2D Array textures. · e5685e08

authored Nov 12, 2014

Our code currently failed in retrieving a NULL Image internally,
when we relied on getting texture dimensions on an unitialized
texture.

BUG=angle:813

Change-Id: Iaf6791ee291e45daae9c80d6452c4d9154227fdb
Reviewed-on: https://chromium-review.googlesource.com/229350Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shannon Woods <shannonwoods@chromium.org>

e5685e08

19 Nov, 2014 8 commits

Adds a file to track known failures in the WebGL CTS · 970d2613

authored Nov 18, 2014

Because ANGLE does not have a framework for testing directly against
the WebGL CTS, we're unable to suppress failing tests at runtime. Instead,
we'll document them in the repository for the reference of contributors.

Change-Id: Ieea68a9aaa0b7bfebda2900265e3c4a857a0ce64
Reviewed-on: https://chromium-review.googlesource.com/230582Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Shannon Woods <shannonwoods@chromium.org>

970d2613

Trivial change to allow translator to be compiled via Emscripten. · 0c219b81

authored Nov 19, 2014

BUG=none

Change-Id: Ifdb02d4b7c061f403f50d2e1fbea4ab117462475
Reviewed-on: https://chromium-review.googlesource.com/230813Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Kenneth Russell <kbr@chromium.org>

0c219b81

Remove pre-generated projects. · 5312b5d8

authored Nov 11, 2014

BUG=angle:829

Change-Id: I220ffb0929cf20f2834e7e0ea163105b45a9e03b
Reviewed-on: https://chromium-review.googlesource.com/229065Tested-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

5312b5d8

Remove includes of libGLESv2/main.h. · acf5d5be

authored Nov 19, 2014

It was mostly being included for gl::error or the debug.h include.

BUG=angle:733

Change-Id: If40d511c5422ac9265fbd809f647a91bcc802d41
Reviewed-on: https://chromium-review.googlesource.com/230551Reviewed-by: Brandon Jones <bajones@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

acf5d5be

Remove includes of libEGL/main.h. · 9c308c8d

authored Nov 19, 2014

It was mostly being included for querying the current state and EGL includes.

BUG=angle:733

Change-Id: Ia985dbc9de5739f6ea7fa5bb6eb2e187cc1d60f3
Reviewed-on: https://chromium-review.googlesource.com/230780Reviewed-by: Brandon Jones <bajones@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

9c308c8d

Create a DefaultAttachment type and an implementation for it. · 6a1e6b97

authored Nov 06, 2014

This allows for dynamically sized default attachments instead of calling
Context::makeCurrent each time the surface changes size.

BUG=angle:824

Change-Id: Ic39c0d7dc4269db53a34c01c4d915cb1a3cfbd08
Reviewed-on: https://chromium-review.googlesource.com/228180Tested-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

6a1e6b97

Fix memory leak when default framebuffer has 0 depth bits and 0 stencil bits · 44f4d74c

authored Nov 11, 2014

Change-Id: If9dbcfc7eac3295b2523dd41dc87198d1b1e2d76
Reviewed-on: https://chromium-review.googlesource.com/229022Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shannon Woods <shannonwoods@chromium.org>
Tested-by: Austin Kinross <aukinros@microsoft.com>

44f4d74c

Limit D3D11 Feature Level 9 to GL ES 2.0, and say it's not conformant · aed9f94e

authored Oct 23, 2014

Change-Id: I2ce88217c9b78e83bac6df5975d9edfbbf90e557
Reviewed-on: https://chromium-review.googlesource.com/225251Reviewed-by: Shannon Woods <shannonwoods@chromium.org>
Tested-by: Shannon Woods <shannonwoods@chromium.org>

aed9f94e

18 Nov, 2014 1 commit

Change the component type of STENCIL_INDEX8 to UNSIGNED_NORMALIZED · e0902644

authored Nov 04, 2014

That way RenderbufferStorageMultisample accepts STENCIL_INDEX8 as
an internal format for a multisampled buffer (samples > 0) as it
isn't concerned by the restriction on integer component types anymore.
This is consistent with the component type returned by the NVIDIA driver.

BUG=angle:812

Change-Id: Ic03f502ffa082b1011e8127213a5c1df0617ef43
Reviewed-on: https://chromium-review.googlesource.com/227470Reviewed-by: Shannon Woods <shannonwoods@chromium.org>
Tested-by: Shannon Woods <shannonwoods@chromium.org>

e0902644

17 Nov, 2014 2 commits

Copy the D3D compiler dll in the GN build. · ea0c84d1

authored Nov 11, 2014

Change-Id: I773a411d07991d9ae160267eba8325c327146554
Reviewed-on: https://chromium-review.googlesource.com/229007Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: John Abd-El-Malek <jam@chromium.org>

ea0c84d1

Unset SRVs which resource are being bound to an RTV · ac6ff938

authored Nov 12, 2014

This is the extension of a previously debug-only workaround for
D3D doing the unsetting by itself (with a warning). It made our
state tracking of the bound SRVs out of sync with D3D's and caused
some rendering to not take place.

BUG=427245

Change-Id: Id378030b64fddafbb3d406872a7dfd2c84793221
Reviewed-on: https://chromium-review.googlesource.com/229066Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

ac6ff938

15 Nov, 2014 1 commit

Revert "Revert "Switch to 8.1 SDK and d3dcompiler_47.dll"" · 7e81c9d5

authored Nov 15, 2014

This reverts commit 3cbdc34d.

This reverts back to using 8.1 and d3dcompiler_47, as chrome is now rolled forward.

Change-Id: I7ef4ebdbc32146b9ba6f551e376ac03ac71131ba
Reviewed-on: https://chromium-review.googlesource.com/229990Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Reviewed-by: Scott Graham <scottmg@chromium.org>
Commit-Queue: Scott Graham <scottmg@chromium.org>
Tested-by: Scott Graham <scottmg@chromium.org>

7e81c9d5

14 Nov, 2014 4 commits

Revert "Switch to 8.1 SDK and d3dcompiler_47.dll" · 3cbdc34d

authored Nov 14, 2014

This reverts commit e1425d95.

This is to keep the GPU fyi bots green until the chrome side is landed. Chrome will be rolling to e1425d95...

Change-Id: Iff1c60eeb7143fb72afa8525b31f8f0de9385897
Reviewed-on: https://chromium-review.googlesource.com/229497Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

3cbdc34d

Switch to 8.1 SDK and d3dcompiler_47.dll · e1425d95

authored Nov 13, 2014

Change-Id: If43c7861ab7f47448fd726fd0b5c35f3a6eeb362
Reviewed-on: https://chromium-review.googlesource.com/229616
Commit-Queue: Scott Graham <scottmg@chromium.org>
Tested-by: Scott Graham <scottmg@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

e1425d95

Added Trim support for Windows Store application certification requirement · e9bdeb25

authored Nov 04, 2014

Change-Id: Iff58863bf4af756d77521aadb1ccbaa977f0b78a
Reviewed-on: https://chromium-review.googlesource.com/227530Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

e9bdeb25

Fix precision tracking of constructor return values · 21203702

authored Nov 13, 2014

Precision should be set for constructor return values if they are of a
built-in type. Structs should not be precision qualified.

BUG=angle:787

Change-Id: Ie5efd5be25a788ff6f01c5b989254572c00231eb
Reviewed-on: https://chromium-review.googlesource.com/229560Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Nicolas Capens <capn@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>

21203702

13 Nov, 2014 3 commits

Remove redundant rx:: scoping prefixes. · 30d6c255

authored Nov 13, 2014

In many places we can get rid of this scoping operator.

BUG=angle:789

Change-Id: If6c40c435a97473d2fd9d6f1fb5a2c709393eb00
Reviewed-on: https://chromium-review.googlesource.com/225566Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Brandon Jones <bajones@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

30d6c255

Remove the unused Renderer client version methods. · b38f2e5d

authored Nov 13, 2014

Previous patches enable us to get rid of this workaround. We should
in subsequent patches also try to clean up code that uses the
internal Renderer-only version of the context caps.

BUG=angle:789

Change-Id: Ic054090303cdc3f7ee8e45e4577c157559955472
Reviewed-on: https://chromium-review.googlesource.com/227714Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Brandon Jones <bajones@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

b38f2e5d

Pass GL context data into shader creation. · 245d3602

authored Nov 13, 2014

This allows us to remove the hack where we store the current GL
client version in the Renderer.

BUG=angle:789

Change-Id: I6526f11eaa922325a9a41df0df48b2152e022e26
Reviewed-on: https://chromium-review.googlesource.com/227713Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

245d3602

12 Nov, 2014 1 commit

Use the [[flatten]] attribute only when a loop is present. · 80bacde5

authored Nov 10, 2014

Flattening branch-heavy shaders that contained no loops caused regressions.
As a temporary workaround we only flatten ifs when there exists a loop.

BUG=395048

Change-Id: I95c40f0249643b98c62304a0f2a4563561d1fbbc
Reviewed-on: https://chromium-review.googlesource.com/228722Reviewed-by: Shannon Woods <shannonwoods@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

80bacde5

11 Nov, 2014 5 commits

Fix ProgramBinary being disabled when it should be enabled. · 560eef16

authored Nov 11, 2014

Change-Id: I6a21bca32ae5bf1e0e638c0cf46102bd964cd066
Reviewed-on: https://chromium-review.googlesource.com/229064Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

560eef16

Refactor Framebuffer attachment application. · ab75a056

authored Oct 15, 2014

Instead of passing texture and renderbufer IDs which requires a call to
gl::getNonLostContext to resolve the objects, pass the Texture and Renderbuffer
objects directly.

BUG=angleproject:733

Change-Id: Ia500a781643e43a17c8e9cea9f95847a7ff7b25d
Reviewed-on: https://chromium-review.googlesource.com/228280Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

ab75a056

Pass GL caps and version to compiler. · de8892b4

authored Nov 11, 2014

This allows us to get rid of some of the reliance on storing the
current context client version in the Renderer. A subsequent
patch will allow us to remove the client version ugly hack.

BUG=angle:789

Change-Id: I139e0f66e1d39e5cd41a484c841a7101b1f29540
Reviewed-on: https://chromium-review.googlesource.com/227712Reviewed-by: Brandon Jones <bajones@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

de8892b4

Move the Context ReadPixels logic into RendererD3D. · 87f6da1f

authored Nov 11, 2014

BUG=angle:789

Change-Id: I0a678f8fe31a3e7ddd9a675a0e3cb0342299b0e3
Reviewed-on: https://chromium-review.googlesource.com/226064Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

87f6da1f

Move BlitFramebuffer logic into RendererD3D. · a93bef2e

authored Nov 11, 2014

BUG=angle:789

Change-Id: Ib3f830ba2f915a5fa4fa552dccc13c24575aa843
Reviewed-on: https://chromium-review.googlesource.com/226063Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

a93bef2e