Re: [PATCH v6 1/4] drm/xe/ggtt: Introduce full GGTT range size

["Piotr Piórkowski" <piotr.piorkowski@intel.com>]

Lis, Tomasz <tomasz.lis@intel.com> wrote on śro [2026-paź-21 03:00:38 +0200]:
> 
> On 2026-10-20 11:55, Piórkowski, Piotr wrote:
> > From: Piotr Piórkowski <piotr.piorkowski@intel.com>
> > 
> > Separate the full assigned GGTT range size from the size available for
> > regular allocations.
> > 
> > For now both sizes are identical. A subsequent change will reserve part of
> > the full range for VF provisioning.
> > 
> > Assisted-by: Claude:claude-5-sonnet
> > Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
> > Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > Cc: Maarten Lankhorst <dev@lankhorst.se>
> > ---
> >   drivers/gpu/drm/xe/xe_ggtt.c | 11 +++++++----
> >   1 file changed, 7 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
> > index 773777afc68a2..abcc929c08535 100644
> > --- a/drivers/gpu/drm/xe/xe_ggtt.c
> > +++ b/drivers/gpu/drm/xe/xe_ggtt.c
> > @@ -113,9 +113,11 @@ struct xe_ggtt_pt_ops {
> >   struct xe_ggtt {
> >   	/** @tile: Back pointer to tile where this GGTT belongs */
> >   	struct xe_tile *tile;
> > -	/** @start: Start offset of GGTT */
> > +	/** @start: Start offset of the assigned GGTT range */
> >   	u64 start;
> > -	/** @size: Total usable size of this GGTT */
> > +	/** @full_size: Full size of the assigned GGTT range */
> > +	u64 full_size;
> > +	/** @size: Size of the GGTT range for regular use */
> >   	u64 size;
> >   	/**
> >   	 * @flags: Flags for this GGTT.
> > @@ -240,7 +242,7 @@ u64 xe_ggtt_size(struct xe_ggtt *ggtt)
> >   static void xe_ggtt_set_pte(struct xe_ggtt *ggtt, u64 addr, u64 pte)
> >   {
> >   	xe_tile_assert(ggtt->tile, !(addr & XE_PTE_MASK));
> > -	xe_tile_assert(ggtt->tile, addr < ggtt->start + ggtt->size);
> > +	xe_tile_assert(ggtt->tile, addr < ggtt->start + ggtt->full_size);
> 
> Looks like the `start` was introduced in "Start using ggtt->start in
> preparation of balloon removal"
> 
> back in January, but asserts remained only for the top bound. Shouldn't we
> check the bottom
> 
> bound as well? Though that should probably be a separate patch.

Good point, thanks. I'll add the bottom-bound check as a separate patch at the start of the series.

Piotr

> -Tomasz
> 
> >   	writeq(pte, &ggtt->gsm[addr >> XE_PTE_SHIFT]);
> >   }
> > @@ -254,7 +256,7 @@ static void xe_ggtt_set_pte_and_flush(struct xe_ggtt *ggtt, u64 addr, u64 pte)
> >   static u64 xe_ggtt_get_pte(struct xe_ggtt *ggtt, u64 addr)
> >   {
> >   	xe_tile_assert(ggtt->tile, !(addr & XE_PTE_MASK));
> > -	xe_tile_assert(ggtt->tile, addr < ggtt->start + ggtt->size);
> > +	xe_tile_assert(ggtt->tile, addr < ggtt->start + ggtt->full_size);
> >   	return readq(&ggtt->gsm[addr >> XE_PTE_SHIFT]);
> >   }
> > @@ -359,6 +361,7 @@ static const struct xe_ggtt_pt_ops xelpg_pt_wa_ops = {
> >   static void __xe_ggtt_init_early(struct xe_ggtt *ggtt, u64 start, u64 size)
> >   {
> >   	ggtt->start = start;
> > +	ggtt->full_size = size;
> >   	ggtt->size = size;
> >   	drm_mm_init(&ggtt->mm, 0, size);
> >   }

--