Re: [git pull] drm for 5.8-rc1

[Daniel Vetter]

On Wed, Jun 3, 2020 at 9:18 AM Thomas Zimmermann <tzimmermann@xxxxxxx> wrote:
>
> Hi
>
> Am 02.06.20 um 23:56 schrieb Linus Torvalds:
> > On Tue, Jun 2, 2020 at 2:21 PM Linus Torvalds
> > <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
> >>
> >> I'm still working through the rest of the merge, so far that was the
> >> only one that made me go "Whaa?".
> >
> > Hmm. I'm also ending up effectively reverting the drm commit
> > b28ad7deb2f2 ("drm/tidss: Use simple encoder") because commit
> > 9da67433f64e ("drm/tidss: fix crash related to accessing freed
> > memory") made the premise of that simply encoder commit no longer be
> > true.
>
> That's OK. The simple encoder is just for consolidating these
> almost-empty encoders at a single place.
>
> > If there is a better way to sort that out (ie something like "use
> > simple encoder but make it free things at destroy time"), I don't know
> > of it.
>
> There's now drmm_kmalloc() to auto-free the memory when DRM releases a
> device.

Yeah I think we discussed that tidss patch on dri-devel when it showed
up, right fix is to essentially undo it, replace with a
s/devm_kzalloc/drmm_kmalloc/ and then re-apply the simple encoder
conversion. We had (and I think still have) some details to sort out
in all this, so some back&forth is entirely expected here. Also it's
just driver unload, which at least for integrated gpu no user ever
cares about, only developers.
-Daniel

>
> Best regards
> Thomas
>
> >
> > I'll let you guys fight it out (added people involved with those
> > commits to the participants,
> >
> >                     Linus
> > _______________________________________________
> > dri-devel mailing list
> > dri-devel@xxxxxxxxxxxxxxxxxxxxx
> > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> >
>
> --
> Thomas Zimmermann
> Graphics Driver Developer
> SUSE Software Solutions Germany GmbH
> Maxfeldstr. 5, 90409 NÃrnberg, Germany
> (HRB 36809, AG NÃrnberg)
> GeschÃftsfÃhrer: Felix ImendÃrffer
>


-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch