On ma, 2015-11-09 at 20:37 +0200, Ville Syrjälä wrote:
> On Mon, Nov 09, 2015 at 08:20:11PM +0200, Imre Deak wrote:
> > The device should be on when updating the GGTT PTEs, so add an
> > assert to
> > all relevant places.
> >
> > Signed-off-by: Imre Deak <imre.d...@intel.com>
> > ---
> > drivers/gpu/drm/i915/i915_gem_gtt.c | 8 ++++++++
> > 1 file changed, 8 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
> > b/drivers/gpu/drm/i915/i915_gem_gtt.c
> > index 016739e..d4a84b8 100644
> > --- a/drivers/gpu/drm/i915/i915_gem_gtt.c
> > +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
> > @@ -2358,6 +2358,8 @@ static void gen8_ggtt_insert_entries(struct
> > i915_address_space *vm,
> > struct sg_page_iter sg_iter;
> > dma_addr_t addr = 0; /* shut up gcc */
> >
> > + assert_device_not_suspended(vm->dev->dev_private);
> > +
>
> We do have the GFX_FLSH_CNTL_GEN6 write in the insert_entries hooks,
> which should already do the check. But I see no harm in doing it
> explicitly as well. Serves as documentation if nothing more. The
> clear_range hooks don't do any register access, so there we should
> definitely have this.
Yes, I think it's better for symmetry with the other updaters. And it's
a WARN_ONCE, so shouldn't add noise.
>
> > for_each_sg_page(st->sgl, &sg_iter, st->nents, 0) {
> > addr = sg_dma_address(sg_iter.sg) +
> > (sg_iter.sg_pgoffset << PAGE_SHIFT);
> > @@ -2404,6 +2406,8 @@ static void gen6_ggtt_insert_entries(struct
> > i915_address_space *vm,
> > struct sg_page_iter sg_iter;
> > dma_addr_t addr = 0;
> >
> > + assert_device_not_suspended(vm->dev->dev_private);
>
> There's already a dev_priv around in all of these AFAICS.
Yep, just copy pasted these.. Will fix them.
>
> > +
> > for_each_sg_page(st->sgl, &sg_iter, st->nents, 0) {
> > addr = sg_page_iter_dma_address(&sg_iter);
> > iowrite32(vm->pte_encode(addr, level, true,
> > flags), >t_entries[i]);
> > @@ -2442,6 +2446,8 @@ static void gen8_ggtt_clear_range(struct
> > i915_address_space *vm,
> > const int max_entries = gtt_total_entries(dev_priv->gtt) -
> > first_entry;
> > int i;
> >
> > + assert_device_not_suspended(vm->dev->dev_private);
> > +
> > if (WARN(num_entries > max_entries,
> > "First entry = %d; Num entries = %d (max=%d)\n",
> > first_entry, num_entries, max_entries))
> > @@ -2468,6 +2474,8 @@ static void gen6_ggtt_clear_range(struct
> > i915_address_space *vm,
> > const int max_entries = gtt_total_entries(dev_priv->gtt) -
> > first_entry;
> > int i;
> >
> > + assert_device_not_suspended(vm->dev->dev_private);
> > +
> > if (WARN(num_entries > max_entries,
> > "First entry = %d; Num entries = %d (max=%d)\n",
> > first_entry, num_entries, max_entries))
> > --
> > 2.5.0
> >
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx@lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/intel-gfx
>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
