[Webkit-unassigned] [Bug 208188] New: [WPE] Many WebGL 2.0.0 conformance tests fail

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=208188

            Bug ID: 208188
           Summary: [WPE] Many WebGL 2.0.0 conformance tests fail
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: WPE WebKit
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: clord at igalia.com
                CC: bugs-noreply at webkitgtk.org

Bug 208078 recently enable a bunch of WebGL 2.0.0 conformance tests that were previously skipped. Many of them fail on WPE. From a local run:

Unexpected flakiness: timeouts (2)
  webgl/2.0.0/conformance2/state/gl-object-get-calls.html [ Failure Timeout Pass ]
  webgl/2.0.0/conformance2/transform_feedback/transform_feedback.html [ Failure Timeout Pass ]


Regressions: Unexpected text-only failures (26)
  webgl/2.0.0/conformance2/attribs/gl-vertexattribipointer-offsets.html [ Failure ]
  webgl/2.0.0/conformance2/attribs/gl-vertexattribipointer.html [ Failure ]
  webgl/2.0.0/conformance2/buffers/bound-buffer-size-change-test.html [ Failure ]
  webgl/2.0.0/conformance2/buffers/buffer-copying-contents.html [ Failure ]
  webgl/2.0.0/conformance2/glsl3/no-attribute-vertex-shader.html [ Failure ]
  webgl/2.0.0/conformance2/glsl3/shader-with-invalid-characters.html [ Failure ]
  webgl/2.0.0/conformance2/glsl3/shader-with-mis-matching-uniform-block.html [ Failure ]
  webgl/2.0.0/conformance2/glsl3/texture-offset-out-of-range.html [ Failure ]
  webgl/2.0.0/conformance2/misc/expando-loss-2.html [ Failure ]
  webgl/2.0.0/conformance2/misc/object-deletion-behaviour-2.html [ Failure ]
  webgl/2.0.0/conformance2/query/occlusion-query.html [ Failure ]
  webgl/2.0.0/conformance2/query/query.html [ Failure ]
  webgl/2.0.0/conformance2/reading/read-pixels-from-fbo-test.html [ Failure ]
  webgl/2.0.0/conformance2/reading/read-pixels-pack-parameters.html [ Failure ]
  webgl/2.0.0/conformance2/rendering/canvas-resizing-with-pbo-bound.html [ Failure ]
  webgl/2.0.0/conformance2/rendering/clear-func-buffer-type-match.html [ Failure ]
  webgl/2.0.0/conformance2/rendering/clipping-wide-points.html [ Failure ]
  webgl/2.0.0/conformance2/rendering/framebuffer-unsupported.html [ Failure ]
  webgl/2.0.0/conformance2/rendering/fs-color-type-mismatch-color-buffer-type.html [ Failure ]
  webgl/2.0.0/conformance2/rendering/rendering-sampling-feedback-loop.html [ Failure ]
  webgl/2.0.0/conformance2/samplers/sampler-drawing-test.html [ Failure ]
  webgl/2.0.0/conformance2/samplers/samplers.html [ Failure ]
  webgl/2.0.0/conformance2/textures/misc/tex-3d-size-limit.html [ Failure ]
  webgl/2.0.0/conformance2/textures/misc/tex-storage-2d.html [ Failure ]
  webgl/2.0.0/conformance2/transform_feedback/two-unreferenced-varyings.html [ Failure ]
  webgl/2.0.0/conformance2/transform_feedback/unwritten-output-defaults-to-zero.html [ Failure ]

Regressions: Unexpected timeouts (1)
  webgl/2.0.0/conformance2/textures/misc/copy-texture-image-luma-format.html [ Timeout ]


These tests remain disabled on iOS and were already explicitly disabled in GTK (ref bug 166536). Some WebGL 2.0.0 tests have been explicitly disabled in the past on WPE (ref bug 188346). It's unclear what the intention wrt to WebGL 2.0.0 on WPE is - should it work? If that's the case, I would suppose these failures need more thorough investigation. Perhaps they worked previously and the disabling of the tests went unnoticed, so breakage could creep in? If it was never intended to be reliable, then perhaps these should be skipped, as they were previously and are on iOS and GTK.

Feedback appreciated!

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20200225/61bac808/attachment.htm>