On Monday, July 27, 2020 9:26 AM, Kazlauskas, Nicholas <nicholas.kazlauskas@xxxxxxx> wrote:
> On 2020-07-27 1:40 a.m., Mazin Rezk wrote:
> > This patch fixes a race condition that causes a use-after-free during
> > amdgpu_dm_atomic_commit_tail. This can occur when 2 non-blocking commits
> > are requested and the second one finishes before the first. Essentially,
> > this bug occurs when the following sequence of events happens:
> >
> > 1. Non-blocking commit #1 is requested w/ a new dm_state #1 and is
> > deferred to the workqueue.
> >
> > 2. Non-blocking commit #2 is requested w/ a new dm_state #2 and is
> > deferred to the workqueue.
> >
> > 3. Commit #2 starts before commit #1, dm_state #1 is used in the
> > commit_tail and commit #2 completes, freeing dm_state #1.
> >
> > 4. Commit #1 starts after commit #2 completes, uses the freed dm_state
> > 1 and dereferences a freelist pointer while setting the context.
> >
> > Since this bug has only been spotted with fast commits, this patch fixes
> > the bug by clearing the dm_state instead of using the old dc_state for
> > fast updates. In addition, since dm_state is only used for its dc_state
> > and amdgpu_dm_atomic_commit_tail will retain the dc_state if none is found,
> > removing the dm_state should not have any consequences in fast updates.
> >
> > This use-after-free bug has existed for a while now, but only caused a
> > noticeable issue starting from 5.7-rc1 due to 3202fa62f ("slub: relocate
> > freelist pointer to middle of object") moving the freelist pointer from
> > dm_state->base (which was unused) to dm_state->context (which is
> > dereferenced).
> >
> > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=207383
> > Fixes: bd200d190f45 ("drm/amd/display: Don't replace the dc_state for fast updates")
> > Reported-by: Duncan <1i5t5.duncan@xxxxxxx>
> > Signed-off-by: Mazin Rezk <mnrzk@xxxxxxxxxxxxxx>
> > ---
> > .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 36 ++++++++++++++-----
> > 1 file changed, 27 insertions(+), 9 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> > index 86ffa0c2880f..710edc70e37e 100644
> > --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> > +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> > @@ -8717,20 +8717,38 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
> > * the same resource. If we have a new DC context as part of
> > * the DM atomic state from validation we need to free it and
> > * retain the existing one instead.
> > + *
> > + * Furthermore, since the DM atomic state only contains the DC
> > + * context and can safely be annulled, we can free the state
> > + * and clear the associated private object now to free
> > + * some memory and avoid a possible use-after-free later.
> > */
> > - struct dm_atomic_state *new_dm_state, *old_dm_state;
> >
> > - new_dm_state = dm_atomic_get_new_state(state);
> > - old_dm_state = dm_atomic_get_old_state(state);
> > + for (i = 0; i < state->num_private_objs; i++) {
> > + struct drm_private_obj *obj = state->private_objs[i].ptr;
> >
> > - if (new_dm_state && old_dm_state) {
> > - if (new_dm_state->context)
> > - dc_release_state(new_dm_state->context);
> > + if (obj->funcs == adev->dm.atomic_obj.funcs) {
> > + int j = state->num_private_objs-1;
> >
> > - new_dm_state->context = old_dm_state->context;
> > + dm_atomic_destroy_state(obj,
> > + state->private_objs[i].state);
> > +
> > + /* If i is not at the end of the array then the
> > + * last element needs to be moved to where i was
> > + * before the array can safely be truncated.
> > + */
> > + if (i != j)
> > + state->private_objs[i] =
> > + state->private_objs[j];
> >
> > - if (old_dm_state->context)
> > - dc_retain_state(old_dm_state->context);
> > + state->private_objs[j].ptr = NULL;
> > + state->private_objs[j].state = NULL;
> > + state->private_objs[j].old_state = NULL;
> > + state->private_objs[j].new_state = NULL;
> > +
> > + state->num_private_objs = j;
> > + break;
> > + }
>
> In the bug report itself I mentioned that I don't really like hacking
> around the DRM core for resolving this patch but to go into more
> specifics, it's really two issues of code maintenance:
>
> 1. It's iterating over internal structures and layout of private objects
> in the state and modifying the state. The core doesn't really guarantee
> how these things are going to be laid out and it may change in the future.
>
> 2. It's freeing an allocation we don't own from DM. DRM doesn't track
> this state elsewhere for purposes of freeing, but nothing is really
> stopping the core from doing this later down the line.
>
> The implementation itself is correct from a technical perspective, but
> I'd rather it reside in DRM as a helper for code maintenance purposes.
So would something like this in drm_atomic_helper.c work?
void drm_atomic_helper_delete_private_obj(struct drm_atomic_state *state,
int i)
{
struct drm_private_obj *obj = state->private_objs[i].ptr;
int end = state->num_private_objs-1;
obj->funcs->atomic_destroy_state(obj, state->private_objs[i].state);
/* If i is not at the end of the array then the last element
* needs to be moved to where i was before the array can safely
* be truncated.
*/
if (i != end)
state->private_objs[i] = state->private_objs[end];
state->private_objs[end].ptr = NULL;
state->private_objs[end].state = NULL;
state->private_objs[end].old_state = NULL;
state->private_objs[end].new_state = NULL;
state->num_private_objs = end;
}
I was considering doing something like that, but I wanted to avoid
modifying DRM core to fix a bug in amdgpu. I guess this makes more sense
though since it does seem rather unorthodox to make changes to
drm_atomic_state outside of DRM core. Perhaps there will be a use for this
function outside of this patch in the future.
Thanks,
Mazin Rezk
>
> Regards,
> Nicholas Kazlauskas
>
> > }
> > }
> >
> > --
> > 2.27.0
> >
> > _______________________________________________
> > dri-devel mailing list
> > dri-devel@xxxxxxxxxxxxxxxxxxxxx
> > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> >
_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
