On Thu, Sep 2, 2021 at 6:20 PM Tvrtko Ursulin
<tvrtko.ursulin@xxxxxxxxxxxxxxx> wrote:
> On 02/09/2021 16:05, Daniel Vetter wrote:
> > On Thu, Sep 2, 2021 at 2:42 PM Tvrtko Ursulin
> > <tvrtko.ursulin@xxxxxxxxxxxxxxx> wrote:
> >>
> >>
> >> On 13/08/2021 21:30, Daniel Vetter wrote:
> >>> The only reason for this really is the i915_gem_engines->fence
> >>> callback engines_notify(), which exists purely as a fairly funky
> >>> reference counting scheme for that. Otherwise all other callers are
> >>> from process context, and generally fairly benign locking context.
> >>
> >> There is reset which definitely isn't process context.
> >
> > gpu reset runs in process context. The tasklet context is the
> > engines_notify I'm talking about above.
>
> I haven't looked very deeply but please double check the path from
> execlists_submission_tasklet -> execlists_reset -> intel_engine_reset ->
> __intel_engine_reset -> execlists_reset_rewind -> execlists_reset_csb ->
> execlists_reset_active -> __i915_request_reset -> mark_guilty ->
> i915_gem_context_put.
Thanks for pointing this out, I'll add it to the commit message.
More stuff to fix, yay.
> >> Otherwise I did not really get from the commit message is this patch
> >> fixing an existing problem or preparing something for the future. If the
> >> former then as I wrote above - I am pretty sure there are call sites
> >> from the tasklet already.
> >>
> >> Regards,
> >>
> >> Tvrtko
> >>
> >>> Unfortunately untangling that requires some major surgery, and we have
> >>> a few i915_gem_context reference counting bugs that need fixing, and
> >>> they blow in the current hardirq calling context, so we need a
> >>> stop-gap measure.
> >
> > I guess this para wasn't clear, but subsequent patches fix the
> > refcount bugs and need this prep patch here.
>
> So up to where in the series are those fixes and where other stuff
> follows? Worth spliting and having cover letters perhaps? Is the fixing
> part applicable to the existing code or only comes to play with the
> syncobj single timeline changes?
There's Fixes: lines. One is timeline syncobj, the other is 2 years old.
-Daniel
>
> Regards,
>
> Tvrtko
>
> > -Daniel
> >
> >>>
> >>> Put a FIXME comment in when this should be removable again.
> >>>
> >>> Signed-off-by: Daniel Vetter <daniel.vetter@xxxxxxxxx>
> >>> Cc: Jon Bloomfield <jon.bloomfield@xxxxxxxxx>
> >>> Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> >>> Cc: Maarten Lankhorst <maarten.lankhorst@xxxxxxxxxxxxxxx>
> >>> Cc: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx>
> >>> Cc: Daniel Vetter <daniel.vetter@xxxxxxxx>
> >>> Cc: "Thomas Hellström" <thomas.hellstrom@xxxxxxxxxxxxxxx>
> >>> Cc: Matthew Auld <matthew.auld@xxxxxxxxx>
> >>> Cc: Lionel Landwerlin <lionel.g.landwerlin@xxxxxxxxx>
> >>> Cc: Dave Airlie <airlied@xxxxxxxxxx>
> >>> Cc: Jason Ekstrand <jason@xxxxxxxxxxxxxx>
> >>> ---
> >>> drivers/gpu/drm/i915/gem/i915_gem_context.c | 13 +++++++++++--
> >>> drivers/gpu/drm/i915/gem/i915_gem_context_types.h | 12 ++++++++++++
> >>> 2 files changed, 23 insertions(+), 2 deletions(-)
> >>>
> >>> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c
> >>> index fd169cf2f75a..051bc357ff65 100644
> >>> --- a/drivers/gpu/drm/i915/gem/i915_gem_context.c
> >>> +++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c
> >>> @@ -986,9 +986,10 @@ static struct i915_gem_engines *user_engines(struct i915_gem_context *ctx,
> >>> return err;
> >>> }
> >>>
> >>> -void i915_gem_context_release(struct kref *ref)
> >>> +static void i915_gem_context_release_work(struct work_struct *work)
> >>> {
> >>> - struct i915_gem_context *ctx = container_of(ref, typeof(*ctx), ref);
> >>> + struct i915_gem_context *ctx = container_of(work, typeof(*ctx),
> >>> + release_work);
> >>>
> >>> trace_i915_context_free(ctx);
> >>> GEM_BUG_ON(!i915_gem_context_is_closed(ctx));
> >>> @@ -1002,6 +1003,13 @@ void i915_gem_context_release(struct kref *ref)
> >>> kfree_rcu(ctx, rcu);
> >>> }
> >>>
> >>> +void i915_gem_context_release(struct kref *ref)
> >>> +{
> >>> + struct i915_gem_context *ctx = container_of(ref, typeof(*ctx), ref);
> >>> +
> >>> + queue_work(ctx->i915->wq, &ctx->release_work);
> >>> +}
> >>> +
> >>> static inline struct i915_gem_engines *
> >>> __context_engines_static(const struct i915_gem_context *ctx)
> >>> {
> >>> @@ -1303,6 +1311,7 @@ i915_gem_create_context(struct drm_i915_private *i915,
> >>> ctx->sched = pc->sched;
> >>> mutex_init(&ctx->mutex);
> >>> INIT_LIST_HEAD(&ctx->link);
> >>> + INIT_WORK(&ctx->release_work, i915_gem_context_release_work);
> >>>
> >>> spin_lock_init(&ctx->stale.lock);
> >>> INIT_LIST_HEAD(&ctx->stale.engines);
> >>> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context_types.h b/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
> >>> index 94c03a97cb77..0c38789bd4a8 100644
> >>> --- a/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
> >>> +++ b/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
> >>> @@ -288,6 +288,18 @@ struct i915_gem_context {
> >>> */
> >>> struct kref ref;
> >>>
> >>> + /**
> >>> + * @release_work:
> >>> + *
> >>> + * Work item for deferred cleanup, since i915_gem_context_put() tends to
> >>> + * be called from hardirq context.
> >>> + *
> >>> + * FIXME: The only real reason for this is &i915_gem_engines.fence, all
> >>> + * other callers are from process context and need at most some mild
> >>> + * shuffling to pull the i915_gem_context_put() call out of a spinlock.
> >>> + */
> >>> + struct work_struct release_work;
> >>> +
> >>> /**
> >>> * @rcu: rcu_head for deferred freeing.
> >>> */
> >>>
> >
> >
> >
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
