1136426 - WebGL 1.0.3 conformance error: conformance/extensions/webgl-depth-texture.html

Status: RESOLVED WORKSFORME

(Core :: Graphics: CanvasWebGL, defect)

Description

[Luke]

User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:39.0) Gecko/20100101 Firefox/39.0
Build ID: 20150224030228

Steps to reproduce:

https://www.khronos.org/registry/webgl/sdk/tests/conformance/extensions/webgl-depth-texture.html?webglVersion=1


Actual results:

conformance/extensions/webgl-depth-texture.html (69 of 72 passed)

    failed: gl.getParameter(gl.DEPTH_BITS) should be >= 16. Was 0 (of type number).
    failed: gl.getParameter(gl.DEPTH_BITS) should be >= 16. Was 0 (of type number).
    failed: gl.getParameter(gl.DEPTH_BITS) should be >= 24. Was 0 (of type number).


Expected results:

Chrome passes all the tests. Firefox fails the test with both the ANGLE and OpenGL backends.

Comment 1

[Jukka Jylänki]

This was observed to fail on the following devices:
 - GIADA, MACBOOK_AIR_OSX, MACBOOK_AIR_WIN, MACBOOK_PRO_OSX, MACBOOK_PRO_WIN, MACMINI, MACPRO, NEXUS-10, SAMSUNG-GT, SURFACE, WINDBOX, HASWELL, HPOMEN

On SAMSUNG-GT, the test even crashed, see https://bugzilla.mozilla.org/show_bug.cgi?id=1178610 .

The test was observed to pass on the following devices:
 - SPARK, NEXUS-4, NEXUS-5

See https://bugzilla.mozilla.org/show_bug.cgi?id=1178601 for hardware configuration details of these systems.

Comment 2

[Kyle Fung]

Attachment: depth-bits-test.html

Reduced test case (needs to be put inside the extensions/ directory of the 1.0.3 test suite).

Comment 3

[Kyle Fung]

It seems that the depth attachment is not actually attached to the Framebuffer of WebGLFramebuffer, so when we query for the depth, we get 0 since the currently bound Framebuffer (the Framebuffer of WebGLFramebuffer) has no depth attachment.

I was thinking that we should actually attach depth buffers to the Framebuffers, but it was probably a conscious decision not to do it in the first place, so I am not too sure about that idea.

Any ideas Jeff?

Comment 4

[Ethan Lin[:ethlin]]

I tried the test on MacOS and Win10 with lastest m-c. The test has passed.