Message ID | CAPM=9txMo5f9QvPqdzt8g3CmUpyDFf2Q_0XS4V1FyjHX8WQPRA@mail.gmail.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | [git,pull] drm for 5.13-rc1 | expand |
On Tue, Apr 27, 2021 at 8:32 PM Dave Airlie <airlied@gmail.com> wrote: > > There aren't a massive amount of conflicts, only with vmwgfx when I > did a test merge into your master yesterday, I think you should be > able to handle them yourself, but let me know if you want me to push a > merged tree somewhere (or if I missed something). The conflict was easy enough to resolve, but was unusual in that my tree had vmwgfx fixes that weren't in the development tree (ie that commit 2ef4fb92363c: "drm/vmwgfx: Make sure bo's are unpinned before putting them back"). Usually when I merge stuff, I can see that the fixes that were pushed to my tree are also in the development branch. Not so this time. Linus
The pull request you sent on Wed, 28 Apr 2021 13:31:59 +1000:
> git://anongit.freedesktop.org/drm/drm tags/drm-next-2021-04-28
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/68a32ba14177d4a21c4a9a941cf1d7aea86d436f
Thank you!
On Wed, Apr 28, 2021 at 7:07 PM Linus Torvalds <torvalds@linux-foundation.org> wrote: > On Tue, Apr 27, 2021 at 8:32 PM Dave Airlie <airlied@gmail.com> wrote: > > > > There aren't a massive amount of conflicts, only with vmwgfx when I > > did a test merge into your master yesterday, I think you should be > > able to handle them yourself, but let me know if you want me to push a > > merged tree somewhere (or if I missed something). > > The conflict was easy enough to resolve, but was unusual in that my > tree had vmwgfx fixes that weren't in the development tree (ie that > commit 2ef4fb92363c: "drm/vmwgfx: Make sure bo's are unpinned before > putting them back"). > > Usually when I merge stuff, I can see that the fixes that were pushed > to my tree are also in the development branch. Not so this time. Maybe we're overdoing it a bit, but we're trying to not backmerge all the time for no reason at all. Only when someone requests it due to more more patches for the dev tree that need both stuff from -fixes and -next. Also keeps you entertained during the merge window :-) Plus I think it keeps us more honest with really just pushing minimal bugfixes to -fixes to keep conflicts reasonable and all that. But if it's a bit overdone I guess we can backmerge a bit more often. -Daniel
On Wed, Apr 28, 2021 at 11:14 AM Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > Maybe we're overdoing it a bit, but we're trying to not backmerge all > the time for no reason at all. Oh, I'm not complaining. I think it's reasonable if some particular issue doesn't hold up further development. So my email was more a statement of surprise at a new pattern than anything else. Linus
On Tue, Apr 27, 2021 at 8:32 PM Dave Airlie <airlied@gmail.com> wrote: > > This is the main drm pull request for 5.13. The usual lots of work all > over the place. [...] > > Mikita Lipski: > drm/amd/display: Add MST capability to trigger_hotplug interface Hmm. I've already merged this, but my clang build shows that this looks buggy: drivers/gpu/drm/amd/amdgpu/amdgpu_dm/amdgpu_dm_debugfs.c:3015:53: warning: address of 'aconnector->mst_port->mst_mgr' will always evaluate to 'true' [-Wpointer-bool-conversion] if (!(aconnector->port && &aconnector->mst_port->mst_mgr)) ~~ ~~~~~~~~~~~~~~~~~~~~~~^~~~~~~ and yeah, checking for the address of a member of a structure benign NULL doesn't really work. I'm assuming the '&' is just a left-over cut-and-paste error or something. Please fix after reviewing (I'm not going to blindly just remove the '&' just to silence the warning, since I don't know the code). Linus
Hi Linus, The patch to fix the warning is here (https://www.spinics.net/lists/amd-gfx/msg61614.html) I guess it just didn't propagate all the way to the release. Can it still be pulled into 5.13-rc1 release? From: Mikita Lipski <mikita.lipski@xxxxxxx> [why] Previous statement would always evaluate to true making it meaningless [how] Just check if a connector is MST by checking if its port exists. Reported-by: kernel test robot <lkp@xxxxxxxxx> Signed-off-by: Mikita Lipski <mikita.lipski@xxxxxxx> Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@xxxxxxx> Acked-by: Wayne Lin <waynelin@xxxxxxx> --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 656bc8f00a42..8bf0b566612b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -3030,7 +3030,7 @@ static int trigger_hpd_mst_set(void *data, u64 val) if (!aconnector->dc_link) continue; - if (!(aconnector->port && &aconnector->mst_port->mst_mgr)) + if (!aconnector->mst_port) continue; link = aconnector->dc_link;
On Wed, Apr 28, 2021 at 8:14 PM Mikita Lipski <mlipski@amd.com> wrote: > > Hi Linus, > > The patch to fix the warning is here > (https://www.spinics.net/lists/amd-gfx/msg61614.html) > > I guess it just didn't propagate all the way to the release. > Can it still be pulled into 5.13-rc1 release? I'll include it in my -fixes pull next week. Alex > > > From: Mikita Lipski <mikita.lipski@xxxxxxx> > > [why] > Previous statement would always evaluate to true > making it meaningless > [how] > Just check if a connector is MST by checking if its port exists. > > Reported-by: kernel test robot <lkp@xxxxxxxxx> > Signed-off-by: Mikita Lipski <mikita.lipski@xxxxxxx> > Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@xxxxxxx> > Acked-by: Wayne Lin <waynelin@xxxxxxx> > --- > drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c > b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c > index 656bc8f00a42..8bf0b566612b 100644 > --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c > +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c > @@ -3030,7 +3030,7 @@ static int trigger_hpd_mst_set(void *data, u64 val) > if (!aconnector->dc_link) > continue; > > - if (!(aconnector->port && &aconnector->mst_port->mst_mgr)) > + if (!aconnector->mst_port) > continue; > > link = aconnector->dc_link; > -- > 2.17.1 > > > > Thanks, > Mikita > > > On 2021-04-28 6:21 p.m., Linus Torvalds wrote: > > On Tue, Apr 27, 2021 at 8:32 PM Dave Airlie <airlied@gmail.com> wrote: > >> > >> This is the main drm pull request for 5.13. The usual lots of work all > >> over the place. [...] > >> > >> Mikita Lipski: > >> drm/amd/display: Add MST capability to trigger_hotplug interface > > > > Hmm. I've already merged this, but my clang build shows that this looks buggy: > > > > drivers/gpu/drm/amd/amdgpu/amdgpu_dm/amdgpu_dm_debugfs.c:3015:53: > > warning: address of 'aconnector->mst_port->mst_mgr' will always > > evaluate to 'true' [-Wpointer-bool-conversion] > > if (!(aconnector->port && > > &aconnector->mst_port->mst_mgr)) > > ~~ ~~~~~~~~~~~~~~~~~~~~~~^~~~~~~ > > > > and yeah, checking for the address of a member of a structure benign > > NULL doesn't really work. > > > > I'm assuming the '&' is just a left-over cut-and-paste error or something. > > > > Please fix after reviewing (I'm not going to blindly just remove the > > '&' just to silence the warning, since I don't know the code). > > > > Linus > > > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/dri-devel