[PATCH v2] drm/tegra: Do not use ->load() and ->unload() callbacks
thierry.reding at gmail.com
Fri Oct 25 10:59:06 UTC 2019
On Fri, Oct 25, 2019 at 12:10:49AM +0300, Dmitry Osipenko wrote:
> 24.10.2019 21:15, Michał Mirosław пишет:
> > On Thu, Oct 24, 2019 at 07:31:37PM +0200, Thierry Reding wrote:
> >> From: Thierry Reding <treding at nvidia.com>
> >> The ->load() and ->unload() drivers are midlayers and should be avoided
> >> in modern drivers. Fix this by moving the code into the driver ->probe()
> >> and ->remove() implementations, respectively.
> >> v2: kick out conflicting framebuffers before initializing fbdev
> >> Signed-off-by: Thierry Reding <treding at nvidia.com>
> >> ---
> >> Michał, Dmitry,
> >> do you guys have a way of testing that the removal of the conflicting
> >> framebuffer actually works?
> > [...]
> > I might be able to check during the weekend. Is this patch alone enough
> > for v5.3?
> I don't think it will apply cleanly on top of 5.3, but should work with
> linux-next or by cherry-picking 9d5a54987265.
I just noticed that I based this version on top of a local branch that
will cause conflicts if you apply this to either 5.3 or linux-next. I'll
resend this later rebased onto drm/tegra/for-next, so it should apply
cleanly on at least linux-next. drm/tegra/for-next also contains this:
Author: Thierry Reding <treding at nvidia.com>
Date: Wed Sep 25 13:26:59 2019 +0200
drm/tegra: Fix ordering of cleanup code
Commit Fixes: b9f8b09ce256 ("drm/tegra: Setup shared IOMMU domain after
initialization") changed the initialization order of the IOMMU related
bits but didn't update the cleanup path accordingly. This asymmetry can
cause failures during error recovery.
Fixes: b9f8b09ce256 ("drm/tegra: Setup shared IOMMU domain after initialization")
Signed-off-by: Thierry Reding <treding at nvidia.com>
Reviewed-by: Dmitry Osipenko <digetx at gmail.com>
Tested-by: Dmitry Osipenko <digetx at gmail.com>
which will conflict with the ->load() and ->unload() removal patch.
If for some reason you can't use linux-next, it shouldn't be terribly
complicated to backport this to v5.3, though.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 833 bytes
Desc: not available
More information about the dri-devel