Commits · e4de307e44837941a32c0749f421a22d42e6ad70 · Chen Yisong / angle

03 Sep, 2015 3 commits

Mark the pack and unpack buffer types as valid in ES3. · e4de307e

authored Sep 02, 2015

dEQP resetState generates a GL error and exits when trying to reset the
pixel buffers after a test.

BUG=angleproject:1149

Change-Id: I133788c6c2f17074bf5bbf5cd17b63338352765f
Reviewed-on: https://chromium-review.googlesource.com/297081Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

e4de307e

Only validate that color and alpha blend funcs are non-constant on D3D renderers. · a7e1d9aa

authored Sep 01, 2015

BUG=angleproject:1025

Change-Id: I9c6008cc2f268bda4afa2c67e0b3d3716e77ea31
Reviewed-on: https://chromium-review.googlesource.com/296681Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

a7e1d9aa

Force index/loop validation on the D3D9 backend · 814a54dd

authored Aug 27, 2015

D3D9 shader models have limited support for looping, so the Appendix A
index/loop limitations are necessary. This is consistent with applying
the limitations on D3D9_3 feature level and lower of the D3D11 backend
too.

Also, when the limitations are not in place, a workaround for dynamic
indexing of vectors and matrices will be applied. This workaround does not
work on the D3D9 backend, as demonstrated with some WebGL conformance
failures caused by it in Chrome on D3D9.

BUG=525188
TEST=angle_end2end_tests

Change-Id: I6411594f067f2740f55d7560427026b23b15dac7
Reviewed-on: https://chromium-review.googlesource.com/296940Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>

814a54dd

02 Sep, 2015 9 commits

translator: Add sh::OutputVariable type. · a0a9e12d

authored Sep 02, 2015

This replaces the dual-use of sh::Attribute, which can be a bit
confusing to people expecting a literal output variable.

Currently not used in Chromium, so should be safe to land.

BUG=angleproject:1146

Change-Id: I436f2bc9dc4ddc3709369cb2baa344c6b13a21a2
Reviewed-on: https://chromium-review.googlesource.com/296683Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

a0a9e12d

Make the end2end tests compile in the GN build. · beb81589

authored Sep 02, 2015

BUG=angleproject:929

Change-Id: Ifc5ad03579c3e187721842f7585f3f06b69cbf45
Reviewed-on: https://chromium-review.googlesource.com/296682Reviewed-by: Dirk Pranke <dpranke@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

beb81589

Revert "Move Uniform and UBO info to the gl::Program layer." · 48705cad

authored Sep 02, 2015

Seems to be failing dEQP-GLES2.functional.uniform_api.value.unused_uniforms.* on Linux (possibly Win as well)

BUG=angleproject:1123

This reverts commit 54f882c9.

Change-Id: I7dbbf40aae8dd9ebd35895df0dd338a3b6b9cc96
Reviewed-on: https://chromium-review.googlesource.com/297051Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

48705cad

Move Uniform and UBO info to the gl::Program layer. · 54f882c9

authored Sep 02, 2015

This data was previously stored entirely in the Impl level. Move
as much as possible to the GL level, using a read-only view in the
Impl level. Some information in D3D-specific, and should be stored
separately in the Impl.

This patch has a lot of refactoring that splits the D3D and GL info,
and moves as much validation as possible to the GL layer, where it
is shared between the back-ends.

BUG=angleproject:1123
TEST=end2end_tests,Canary WebGL (D3D11/GL), dEQP-GLES2+3.functional.uniform_api

Change-Id: I8b7e1380a61590a46ccee646a47ffe9bda287f6b
Reviewed-on: https://chromium-review.googlesource.com/295124Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tryjob-Request: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

54f882c9

Revert "Revert "Clean up interface to ProgramImpl::link."" · f5f4ad24

authored Sep 02, 2015

Should be fixed now with the fix the prior CL.

BUG=angleproject:1123

This reverts commit 63069dfc.

Change-Id: Ie1826673820e3a24f9bb3e4ff5c723ac47570041
Reviewed-on: https://chromium-review.googlesource.com/297050Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

f5f4ad24

Re-land "Compute packed varyings in ProgramD3D only." · ca03b35c

authored Sep 02, 2015

Instead of storing varying information in the shader, use a temporary
set when linking a D3D program. This also means we won't have to
modify information in the Shader object when linking a D3D program.

This completes the refactoring for PackedVaryings.

Re-land with fix for missing init of PackedVarying::vertexOnly.

BUG=angleproject:1123

Change-Id: If110809c3817d88b0370ac575d739d7385b067d9
Reviewed-on: https://chromium-review.googlesource.com/296731Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

ca03b35c

Only validate that stencil refs and masks are the same on D3D renderers. · 3a86ad38

authored Sep 01, 2015

BUG=angleproject:1025

Change-Id: If79046f9f81ad20c5f2f7296245b0cb74a4102aa
Reviewed-on: https://chromium-review.googlesource.com/296680Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

3a86ad38

Add back a suppression for a bot specific dEQP GLES2 failure on Windows · 93fc8f3e

authored Sep 02, 2015

BUG=angleproject:1127

Change-Id: I749c2c39dfaf60d5e39a00bd7954fcb1d7e931ac
Reviewed-on: https://chromium-review.googlesource.com/296960Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

93fc8f3e

ProgramBinaryTest: fix an unreachable code warning · efb6ac66

authored Sep 02, 2015

This warning was found by the GN build on Windows: the FAIL macro contains
a return which caused code after the FAIL to be unreachable.

BUG=angleproject:929

Change-Id: Id4bda39ea9f76d2b59c44391cb2de8a3bfb6947c
Reviewed-on: https://chromium-review.googlesource.com/296950Tested-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

efb6ac66

01 Sep, 2015 14 commits

Make the unittests compile in the GN build. · d4bcfa2c

authored Sep 01, 2015

This changes angle_unittests.gypi to use a variable per list of files so
that we can grab the list of files in conditions in BUILD.gn

BUG=angleproject:929

Change-Id: I7fc73fd84d8efc1ea545fe672b1dba1c429b317f
Reviewed-on: https://chromium-review.googlesource.com/296443Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

d4bcfa2c

BitSetIterator_unittest: fix an unreachable code warning · 94d099d0

authored Sep 01, 2015

Because gtest's FAIL macro does a return, MSVC complained that the
update part of a range-based for loop was unreachable. Fixed this by
having a boolean value set to true if we execute the loop body.

BUG=angleproject:929

Change-Id: I6c1fec7ed6cf1a3aba6f02b68c1015bee2feebfd
Reviewed-on: https://chromium-review.googlesource.com/296730Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

94d099d0

Revert "Compute packed varyings in ProgramD3D only." · c437046f

authored Sep 01, 2015

Likely the cause of Windows bot failures.

BUG=angleproject:1123

This reverts commit 532061bb.

Change-Id: Ia4f0161b97bfbf2adb4cafaa5fa5484f04fad245
Reviewed-on: https://chromium-review.googlesource.com/296710Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

c437046f

Revert "Clean up interface to ProgramImpl::link." · 63069dfc

authored Sep 01, 2015

Probably causing bot failures in a prior dependent patch.

BUG=angleproject:1123

This reverts commit b8359275.

Change-Id: I7642c4a66b190dda49cbb16611abc893f3eb6422
Reviewed-on: https://chromium-review.googlesource.com/296700Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

63069dfc

Change the FLATTEN heuristic to "ifs with a loop with a gradient" · 477b243b

authored Aug 31, 2015

This heuristic makes more sense than the previous "ifs with a
discontinuous loop" as the reason we need to flatten is that we need
gradients to be in branchless code.

Change the UnrollFlatten test accordingly.

Tested with:
 - the WebGL CTS
 - dev.miaumiau.cat/rayTracer "Skull Demo"
 - THe turbulenz engine GPU particle demo
 - Lots of ShaderToy Samples (inc. Volcanic, Metropolis and Hierarchical
   Voronoi)
 - Google Maps Earth mode
 - Lots of Chrome experiments
 - madebyevan.com/webgl-water

BUG=524297

Change-Id: Iaa727036fffcfde3952716a1ef33b6ee0546b69d
Reviewed-on: https://chromium-review.googlesource.com/296442Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

477b243b

Prefer glDepthRangef and glClearDepthf to the double version. · 12d59314

authored Aug 31, 2015

ES only has the float versions.

BUG=angleproject:1145

Change-Id: I6c96b553d20e8aa160f68fe1b3fa0200d64108a3
Reviewed-on: https://chromium-review.googlesource.com/296361Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

12d59314

Fix wrongly formatted dEQP GLES2 expectations · f2ee0752

authored Sep 01, 2015

The formatting problem was introduced in
770f8f86 and prevented the bots from
running the dEQP tests.

BUG=angleproject:1143

Change-Id: I427f6472625df19111ff56685b7e70b9e9aae866
Reviewed-on: https://chromium-review.googlesource.com/295199Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

f2ee0752

Make unittest #includes start from src · 2728f992

authored Aug 31, 2015

This is will help avoid adding many include directories for the GN
build.

BUG=angleproject:929

Change-Id: If8f7e98526df3560b8e410ec86271a2da0c6889d
Reviewed-on: https://chromium-review.googlesource.com/296480Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

2728f992

Mark TF Feedback array element capture as unsupported. · 89bb70e9

authored Aug 31, 2015

According to dEQP this is an undefined section of the spec. Mark it as
unsupported for now until we can investgate if it's possible on
certain platforms.

Noticed this when testing with the Program refactor.

BUG=angleproject:1123
TEST=dEQP-GLES3.functional.transform_feedback.array_element.*

Change-Id: I830dddf418e28fb4536f09500f4069def1d97b58
Reviewed-on: https://chromium-review.googlesource.com/295984Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

89bb70e9

Supress failures in RGB->L CopyTex[Sub]Image calls on Intel OpenGL. · aa40feea

authored Sep 01, 2015

BUG=angleproject:1113

Change-Id: Ie16a28046546fad2ff1c1af35ca67c434b9f5216
Reviewed-on: https://chromium-review.googlesource.com/296481Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

aa40feea

Implement drawRangeElements in RendererGL. · 47502236

authored Aug 25, 2015

BUG=angleproject:1136

Change-Id: I8b84b845da09f60bfd5e8f400da47d759f0d8f1a
Reviewed-on: https://chromium-review.googlesource.com/295234Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

47502236

Fix unsigned/signed comparison in VertexDataManager. · 846f107f

authored Sep 01, 2015

Affects Debug/Win32 only.

BUG=angleproject:1136

Change-Id: I29fbf06077014c313489be07ff983009afb652fd
Reviewed-on: https://chromium-review.googlesource.com/296362Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

846f107f

Clean up interface to ProgramImpl::link. · b8359275

authored Aug 26, 2015

We no longer need to pass mutable shaders with the changed to packed
varyings.

BUG=angleproject:1123

Change-Id: I0a4bbf5d770ff48e10e3de2146840d8987b1e826
Reviewed-on: https://chromium-review.googlesource.com/295194Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

b8359275

Compute packed varyings in ProgramD3D only. · 532061bb

authored Aug 31, 2015

Instead of storing varying information in the shader, use a temporary
set when linking a D3D program. This also means we won't have to
modify information in the Shader object when linking a D3D program.

This completes the refactoring for PackedVaryings.

BUG=angleproject:1123

Change-Id: I241610e87f7d14f3e18b0d8bd84f1a3509c05dfd
Reviewed-on: https://chromium-review.googlesource.com/295193Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

532061bb

31 Aug, 2015 9 commits

HLSLCompiler: fix an infinite loop when failing to compile · 0e06ea33

authored Aug 31, 2015

When compiling shaders that produce errors X3531 or X4014 even when macros
is nullptr, we would try the same permutation infinitely.

BUG=524297

Change-Id: Id2a26ce2a5344bff4566440342adad5031368994
Reviewed-on: https://chromium-review.googlesource.com/296441Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

0e06ea33

Implement the instanced draw calls in RendererGL. · 3cf12ce6

authored Aug 27, 2015

BUG=angleproject:1136

Change-Id: I1167365618bdc3ca37ac0f4c60809de32c7a9d78
Reviewed-on: https://chromium-review.googlesource.com/295733Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

3cf12ce6

Revert "Implement the instanced draw calls in RendererGL." · 5b49bb83

authored Aug 31, 2015

Causing crashes in some dEQP tests.

This reverts commit c1948415.

Change-Id: Idedb32484b4f2b09090346bc4228fe536190a621
Reviewed-on: https://chromium-review.googlesource.com/296440Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

5b49bb83

Fix dEQP renderer selection flag in documentation. · 23a22c87

authored Aug 31, 2015

BUG=None

Change-Id: Ibefeab05c9c9ac87263a23fd69e6725bf4998673
Reviewed-on: https://chromium-review.googlesource.com/295148Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

23a22c87

Implement the instanced draw calls in RendererGL. · c1948415

authored Aug 27, 2015

BUG=angleproject:1136

Change-Id: Id04e357b97c04b3843b25827f705fd802e6a68e3
Reviewed-on: https://chromium-review.googlesource.com/295233Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

c1948415

Re-land "Fix warnings with ANGLE and VS2015." · 929d7bca

authored Aug 31, 2015

With fix for VS2013.

BUG=None

Change-Id: I9f656efa6b540a03ce21f6b65fde0753c0d72538
Reviewed-on: https://chromium-review.googlesource.com/295157Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

929d7bca

Revert "Fix warnings with ANGLE and VS2015." · 344bd0af

authored Aug 31, 2015

Seems to break VS2013. Reverting.

This reverts commit 51e275b4.

Change-Id: I37843bf457122614007ba455f6ee751577b14604
Reviewed-on: https://chromium-review.googlesource.com/296290Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>

344bd0af

Fix warnings with ANGLE and VS2015. · 51e275b4

authored Aug 28, 2015

BUG=None

Change-Id: I47b08593a38d710b59051a3107933c08c2789ca2
Reviewed-on: https://chromium-review.googlesource.com/295722Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

51e275b4

Fix Chromium compilation error in VariablePack_test.cpp · 2a592150

authored Aug 31, 2015

BUG=

Change-Id: I59c9716774df4615da6dd412795639e055f45034
Reviewed-on: https://chromium-review.googlesource.com/295732Tested-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

2a592150

28 Aug, 2015 5 commits

Add row-sharing packing tests for VariablePacker_test.cpp · a9bee1b8

authored Aug 27, 2015

Recent test failures leads to suspicion that our variable packing algorithm is
buggy - turns out to be an underlying driver bug.

With the added test cases, such suspicion shouldn't even arise.

BUG=angleproject:1142
TEST=angle_unittests

Change-Id: I1fb3c5c7798d9ad17668a3d633286e031da79cab
Reviewed-on: https://chromium-review.googlesource.com/295901Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Tested-by: Zhenyao Mo <zmo@chromium.org>

a9bee1b8

Fix missing copy rules for DEQP EGL test cases and expectations. · 465e6f45

authored Aug 28, 2015

BUG=angleproject:1139

Change-Id: I9d58b2c0b1e6465b58b2f3d4aad4b1d3517ae3ec
Reviewed-on: https://chromium-review.googlesource.com/295147Tested-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

465e6f45

Only update the synced parts of mAppliedAttributes. · 3caa6528

authored Aug 27, 2015

Members such as divisor are not synced in
VertexArrayGL::updateAttribPointer so assigning the whole attribute causes
the divisor to not be synced.

BUG=angleproject:1136

Change-Id: I947a144f81ae67953947e363debd1d1cff78207a
Reviewed-on: https://chromium-review.googlesource.com/295145Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>

3caa6528

Add back suppressions for bot-specific dEQP GLES2 failures on Linux · 770f8f86

authored Aug 28, 2015

BUG=angleproject:1143

Change-Id: I5bd1a7de7b147062f2e093a30efe4ed06eb80ec2
Reviewed-on: https://chromium-review.googlesource.com/295247Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

770f8f86

Update the Windows dEQP GLES2 test expectations · ad915e9b

authored Aug 27, 2015

BUG=angleproject:1127

Change-Id: I6f38e5441c4a288e9c1f55fefa31d5d2f10a700c
Reviewed-on: https://chromium-review.googlesource.com/295246Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>

ad915e9b