compressedTexSubImage2D does return error when COMPRESSED_RGB_ETC1_WEBGL is used
Would have been easier to understand the behavior for the developer in bug 229180.
// Not supported for ETC1_RGB8_OES textures.
Changed summary: "does" -> "does not".
In fact, the unreported error in validateCompressedTexSubDimensions is !USE(ANGLE) block. ANGLE does report INVALID_OPERATION and this should have been observable in the original report should I have understood to add a check for it. (Since it's from ANGLE and not a simulated error, it's not printed in the console).