On Fri, 2025-10-17 at 19:14 +0300, Ville Syrjala wrote: > From: Ville Syrjälä <[email protected]> > > Feels more sensible to check the original unmangled watermarks > against the allocated DDB size first, and only then do > skl_wm_check_vblank() on whatever is left. > > Signed-off-by: Ville Syrjälä <[email protected]> > ---
Sorry for the delay... Somehow I missed these 3 last patches. Reviewed-by: Luca Coelho <[email protected]> -- Cheers, Luca. > drivers/gpu/drm/i915/display/skl_watermark.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c > b/drivers/gpu/drm/i915/display/skl_watermark.c > index 256162da9afc..764770fab7e7 100644 > --- a/drivers/gpu/drm/i915/display/skl_watermark.c > +++ b/drivers/gpu/drm/i915/display/skl_watermark.c > @@ -2346,7 +2346,7 @@ static int skl_build_pipe_wm(struct intel_atomic_state > *state, > > crtc_state->wm.skl.optimal = crtc_state->wm.skl.raw; > > - return skl_wm_check_vblank(crtc_state); > + return 0; > } > > static bool skl_wm_level_equals(const struct skl_wm_level *l1, > @@ -2923,6 +2923,10 @@ skl_compute_wm(struct intel_atomic_state *state) > for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i) { > struct skl_pipe_wm *pipe_wm = &new_crtc_state->wm.skl.optimal; > > + ret = skl_wm_check_vblank(new_crtc_state); > + if (ret) > + return ret; > + > /* > * We store use_sagv_wm in the crtc state rather than relying on > * that bw state since we have no convenient way to get at the
