Hi Melissa, Thank you very much for your review. > On Thu, Dec 23, 2021 at 07:35:48PM -0100, Melissa Wen wrote: > What test did you run? Indeed, not all kms tests are passing and fixes > are welcome :) > > Last time, I used these testcases for overlay: kms_plane_cursor, > kms_atomic; and these tests were fine too: kms_cursor_crc, kms_writeback, > kms_flip For the different patches I have been working on I have tested mainly with kms_atomic, kms_plane_cursor and kms_plane_alpha_blend. For some reason, kms_cursor_crc suspends my PC. I still need to investigate the cause. I'll include a table with success/skip/fail tests before and after the patch on v2 :) > However, I think we need some limits for this number > of planes. I would suggest to just expand the enable_overlay option to > expose a predefined number of planes > [...] > I don't have a strong opinion on an exact/practical number. I took a > quick look at other drivers and exposing 8 planes seems reasonable to > me. 8 planes sound reasonable to me, I'll change it and send a revision of [1] as well using the new constant. Thanks again for taking the time to review this, José Expósito [1] https://lore.kernel.org/dri-devel/20211223081030.16629-1-jose.exposito89@xxxxxxxxx/T/
