On Mon, Mar 30, 2026 at 02:38:35PM -0700, Rosen Penev wrote: > On Mon, Mar 30, 2026 at 7:21 AM Timur Kristóf <[email protected]> wrote: > > > > On Monday, March 30, 2026 3:55:55 PM Central European Summer Time Christian > > König wrote: > > > On 3/30/26 15:16, Timur Kristóf wrote: > > > > On Friday, March 27, 2026 12:47:10 AM Central European Summer Time Rosen > > > > Penev> > > > > wrote: > > > >> From: Timur Kristóf <[email protected]> > > > >> > > > >> [ Upstream commit 7495962cbceb967e095233a5673ea71f3bcdee7e ] > > > >> > > > >> It already didn't work on DCE 8, > > > >> so there is no reason to assume it would on DCE 6. > > > >> > > > >> Signed-off-by: Timur Kristóf <[email protected]> > > > >> Reviewed-by: Rodrigo Siqueira <[email protected]> > > > >> Reviewed-by: Alex Deucher <[email protected]> > > > >> Reviewed-by: Alex Hung <[email protected]> > > > >> Signed-off-by: Alex Deucher <[email protected]> > > > >> Signed-off-by: Rosen Penev <[email protected]> > > > > > > > > This patch is incorrect and should not be backported. > > > > > > > > (Note that the error is already fixed upstream. For stable kernels IMO > > > > it's > > > > best to drop this one.) > > > > > > Is there some alternative which needs to be backported or should the old > > > kernel just work out of the box because we never enabled some feature > > > there? > > > > > > Apart from that the patch set looks good to me. > > > > > > > This patch had a typo and does the opposite of what it should, ie. it > > disables > > eDP fastboot on DCE10 and newer instead of disabling it on DCE8 and older. > > > > The upstream fix is here: > > https://lists.freedesktop.org/archives/amd-gfx/2026-February/138577.html > > which disables eDP fastboot on DCE10 and older. > Not sure what the process is here. I make sure everything can be git > cherry-pick ed. In that case, both should be present.
I agree, I don't understand the problem here. Just take the commits that are upstream including "fixes for the fixes". Timur, what specifically do you want to see happen here? thanks, greg k-h
