Some dEQP tests in dEQP-GLES3.functional.shaders.derivate.* started failing

[GoogleCodeExporter]

I found 44 tests failing from dEQP-GLES3.functional.shaders.derivate.* when I 
did a complete run of GLES3 shaders tests today. Specifically, static_loop and 
uniform_loop tests for dfdx, dfdy and fwidth.

These tests were passing earlier and started failing with the change:
https://chromium-review.googlesource.com/#/c/264791/

Original issue reported on code.google.com by apat...@nvidia.com on 24 Apr 2015 at 8:53

[GoogleCodeExporter]

Hm, that's unfortunate. Thanks for filing the issue - we should investigate.

Original comment by jmad...@chromium.org on 24 Apr 2015 at 11:56

[GoogleCodeExporter]

Project: angle/angle
Branch : master
Author : Corentin Wallez <cwallez@chromium.org>
Commit : a1884f2b65cdece93327f399ff9c9a9f263535ce

Code-Review  0 : Corentin Wallez
Code-Review  +1: Jamie Madill
Code-Review  +2: Geoff Lang
Verified     0 : Geoff Lang, Jamie Madill
Verified     +1: Corentin Wallez
Commit Queue   : Chumped
Change-Id      : I1315eac1ad36f726be52d7fda5facf3104341b1f
Reviewed-at    : https://chromium-review.googlesource.com/267814

Fixes for the tagging of discontinuous loops

The first fix was for all loops being considered discontinuous in
OutputHLSL because of a typo that produced a tautology.

The error was not detected by the unit tests because as an optimization we
do not generate Lod0 calls when they are not needed for the callee
function (which was correctly detected by the analysis in this case).
Fixed the unit tests by adding a call to a builtin gradient operation.

The second fix was for discard not being taken into account in the
analyses of the AST, which caused a WebGL test regression after the first
fix for conformance/glsl/bugs/conditional-discard-in-loop

BUG=angleproject:982

src/compiler/translator/ASTMetadataHLSL.cpp
src/compiler/translator/OutputHLSL.cpp
src/tests/compiler_tests/UnrollFlatten_test.cpp

Original comment by bugdro...@chromium.org on 29 Apr 2015 at 8:07

[GoogleCodeExporter]

Original comment by cwal...@chromium.org on 30 Apr 2015 at 2:16

• Changed state: Fixed