Hi Andy, On Fri, 18 Apr 2025 at 01:16, Andy Yan <andys...@163.com> wrote: > I prefer the V1 version PATCH[0]. This is because we do not deal with > hardware-related > differences at this level. It involves a VOP-related restriction and we > always handle > limitiation like this within the VOP driver .
As said in the review for v1, this is not enough for generic userspace. If drmModeAddFB2WithModifiers() fails, userspace knows that the buffer can never work in that configuration, and it should fall back. If drmModeAtomicCommit(DRM_MODE_ATOMIC_TEST_ONLY) fails, userspace does not know that the buffer can _never_ work, because the failure may be transient or due to a combination of things. Returning the more localised error saves userspace a lot of work and enables a more optimal system. Cheers, Daniel
