[Intel-gfx] [PATCH 6/7] drm/i915/guc: Optimize CTB writes and reads
Matthew Brost
matthew.brost at intel.com
Wed Jul 7 18:56:53 UTC 2021
On Wed, Jul 07, 2021 at 11:19:01AM -0700, John Harrison wrote:
> On 7/7/2021 10:50, Matthew Brost wrote:
> > On Tue, Jul 06, 2021 at 03:51:00PM -0700, John Harrison wrote:
> > > On 7/6/2021 15:20, Matthew Brost wrote:
> > > > CTB writes are now in the path of command submission and should be
> > > > optimized for performance. Rather than reading CTB descriptor values
> > > > (e.g. head, tail) which could result in accesses across the PCIe bus,
> > > > store shadow local copies and only read/write the descriptor values when
> > > > absolutely necessary. Also store the current space in the each channel
> > > > locally.
> > > >
> > > > v2:
> > > > (Michal)
> > > > - Add additional sanity checks for head / tail pointers
> > > > - Use GUC_CTB_HDR_LEN rather than magic 1
> > > > v3:
> > > > (Michal / John H)
> > > > - Drop redundant check of head value
> > > >
> > > > Signed-off-by: John Harrison <John.C.Harrison at Intel.com>
> > > > Signed-off-by: Matthew Brost <matthew.brost at intel.com>
> > > > ---
> > > > drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 88 +++++++++++++++--------
> > > > drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h | 6 ++
> > > > 2 files changed, 65 insertions(+), 29 deletions(-)
> > > >
> > > > diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
> > > > index db3e85b89573..4a73a1f03a9b 100644
> > > > --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
> > > > +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
> > > > @@ -130,6 +130,10 @@ static void guc_ct_buffer_desc_init(struct guc_ct_buffer_desc *desc)
> > > > static void guc_ct_buffer_reset(struct intel_guc_ct_buffer *ctb)
> > > > {
> > > > ctb->broken = false;
> > > > + ctb->tail = 0;
> > > > + ctb->head = 0;
> > > > + ctb->space = CIRC_SPACE(ctb->tail, ctb->head, ctb->size);
> > > > +
> > > > guc_ct_buffer_desc_init(ctb->desc);
> > > > }
> > > > @@ -383,10 +387,8 @@ static int ct_write(struct intel_guc_ct *ct,
> > > > {
> > > > struct intel_guc_ct_buffer *ctb = &ct->ctbs.send;
> > > > struct guc_ct_buffer_desc *desc = ctb->desc;
> > > > - u32 head = desc->head;
> > > > - u32 tail = desc->tail;
> > > > + u32 tail = ctb->tail;
> > > > u32 size = ctb->size;
> > > > - u32 used;
> > > > u32 header;
> > > > u32 hxg;
> > > > u32 type;
> > > > @@ -396,25 +398,22 @@ static int ct_write(struct intel_guc_ct *ct,
> > > > if (unlikely(desc->status))
> > > > goto corrupted;
> > > > - if (unlikely((tail | head) >= size)) {
> > > > + GEM_BUG_ON(tail > size);
> > > > +
> > > > +#ifdef CONFIG_DRM_I915_DEBUG_GUC
> > > > + if (unlikely(tail != READ_ONCE(desc->tail))) {
> > > > + CT_ERROR(ct, "Tail was modified %u != %u\n",
> > > > + desc->tail, ctb->tail);
> > > > + desc->status |= GUC_CTB_STATUS_MISMATCH;
> > > > + goto corrupted;
> > > > + }
> > > > + if (unlikely((desc->tail | desc->head) >= size)) {
> > > Same arguments below about head apply to tail here. Also, there is no #else
> > Yes, desc->tail can be removed from this check. Same for head below. Can
> > you fix this when merging?
> >
> > > check on ctb->head?
> > ctb->head variable isn't used in this path nor is ctb->tail in the
> > other. In the other path desc->tail is checked as it is read while
> > desc->head isn't needed to be read here. The other path can also likely
> > be reworked to pull the tail check outside of the if / else define
> > block.
> >
> > > > CT_ERROR(ct, "Invalid offsets head=%u tail=%u (size=%u)\n",
> > > > - head, tail, size);
> > > > + desc->head, desc->tail, size);
> > > > desc->status |= GUC_CTB_STATUS_OVERFLOW;
> > > > goto corrupted;
> > > > }
> > > > -
> > > > - /*
> > > > - * tail == head condition indicates empty. GuC FW does not support
> > > > - * using up the entire buffer to get tail == head meaning full.
> > > > - */
> > > > - if (tail < head)
> > > > - used = (size - head) + tail;
> > > > - else
> > > > - used = tail - head;
> > > > -
> > > > - /* make sure there is a space including extra dw for the header */
> > > > - if (unlikely(used + len + GUC_CTB_HDR_LEN >= size))
> > > > - return -ENOSPC;
> > > > +#endif
> > > > /*
> > > > * dw0: CT header (including fence)
> > > > @@ -453,7 +452,9 @@ static int ct_write(struct intel_guc_ct *ct,
> > > > write_barrier(ct);
> > > > /* now update descriptor */
> > > > + ctb->tail = tail;
> > > > WRITE_ONCE(desc->tail, tail);
> > > > + ctb->space -= len + GUC_CTB_HDR_LEN;
> > > > return 0;
> > > > @@ -469,7 +470,7 @@ static int ct_write(struct intel_guc_ct *ct,
> > > > * @req: pointer to pending request
> > > > * @status: placeholder for status
> > > > *
> > > > - * For each sent request, Guc shall send bac CT response message.
> > > > + * For each sent request, GuC shall send back CT response message.
> > > > * Our message handler will update status of tracked request once
> > > > * response message with given fence is received. Wait here and
> > > > * check for valid response status value.
> > > > @@ -525,24 +526,35 @@ static inline bool ct_deadlocked(struct intel_guc_ct *ct)
> > > > return ret;
> > > > }
> > > > -static inline bool h2g_has_room(struct intel_guc_ct_buffer *ctb, u32 len_dw)
> > > > +static inline bool h2g_has_room(struct intel_guc_ct *ct, u32 len_dw)
> > > > {
> > > > - struct guc_ct_buffer_desc *desc = ctb->desc;
> > > > - u32 head = READ_ONCE(desc->head);
> > > > + struct intel_guc_ct_buffer *ctb = &ct->ctbs.send;
> > > > + u32 head;
> > > > u32 space;
> > > > - space = CIRC_SPACE(desc->tail, head, ctb->size);
> > > > + if (ctb->space >= len_dw)
> > > > + return true;
> > > > +
> > > > + head = READ_ONCE(ctb->desc->head);
> > > > + if (unlikely(head > ctb->size)) {
> > > > + CT_ERROR(ct, "Corrupted descriptor head=%u tail=%u size=%u\n",
> > > > + ctb->desc->head, ctb->desc->tail, ctb->size);
> > > > + ctb->desc->status |= GUC_CTB_STATUS_OVERFLOW;
> > > > + ctb->broken = true;
> > > > + return false;
> > > > + }
> > > > +
> > > > + space = CIRC_SPACE(ctb->tail, head, ctb->size);
> > > > + ctb->space = space;
> > > > return space >= len_dw;
> > > > }
> > > > static int has_room_nb(struct intel_guc_ct *ct, u32 len_dw)
> > > > {
> > > > - struct intel_guc_ct_buffer *ctb = &ct->ctbs.send;
> > > > -
> > > > lockdep_assert_held(&ct->ctbs.send.lock);
> > > > - if (unlikely(!h2g_has_room(ctb, len_dw))) {
> > > > + if (unlikely(!h2g_has_room(ct, len_dw))) {
> > > > if (ct->stall_time == KTIME_MAX)
> > > > ct->stall_time = ktime_get();
> > > > @@ -612,7 +624,7 @@ static int ct_send(struct intel_guc_ct *ct,
> > > > */
> > > > retry:
> > > > spin_lock_irqsave(&ctb->lock, flags);
> > > > - if (unlikely(!h2g_has_room(ctb, len + GUC_CTB_HDR_LEN))) {
> > > > + if (unlikely(!h2g_has_room(ct, len + GUC_CTB_HDR_LEN))) {
> > > > if (ct->stall_time == KTIME_MAX)
> > > > ct->stall_time = ktime_get();
> > > > spin_unlock_irqrestore(&ctb->lock, flags);
> > > > @@ -732,7 +744,7 @@ static int ct_read(struct intel_guc_ct *ct, struct ct_incoming_msg **msg)
> > > > {
> > > > struct intel_guc_ct_buffer *ctb = &ct->ctbs.recv;
> > > > struct guc_ct_buffer_desc *desc = ctb->desc;
> > > > - u32 head = desc->head;
> > > > + u32 head = ctb->head;
> > > > u32 tail = desc->tail;
> > > > u32 size = ctb->size;
> > > > u32 *cmds = ctb->cmds;
> > > > @@ -747,12 +759,29 @@ static int ct_read(struct intel_guc_ct *ct, struct ct_incoming_msg **msg)
> > > > if (unlikely(desc->status))
> > > > goto corrupted;
> > > > - if (unlikely((tail | head) >= size)) {
> > > > + GEM_BUG_ON(head > size);
> > > > +
> > > > +#ifdef CONFIG_DRM_I915_DEBUG_GUC
> > > > + if (unlikely(head != READ_ONCE(desc->head))) {
> > > > + CT_ERROR(ct, "Head was modified %u != %u\n",
> > > > + desc->head, ctb->head);
> > > > + desc->status |= GUC_CTB_STATUS_MISMATCH;
> > > > + goto corrupted;
> > > > + }
> > > > + if (unlikely((desc->tail | desc->head) >= size)) {
> > > As per comment in other thread, the check on head here is redundant because
> > > you have already hit a BUG_ON(ctb->head > size) followed by
> > > CT_ERROR(ctb->head != desc->head). Therefore, you can't get here if
> > > 'desc->head > size'.
> > >
> > Yep. See above we can likely just delete this.
> > > > CT_ERROR(ct, "Invalid offsets head=%u tail=%u (size=%u)\n",
> > > > head, tail, size);
> > > > desc->status |= GUC_CTB_STATUS_OVERFLOW;
> > > > goto corrupted;
> > > > }
> > > > +#else
> > > > + if (unlikely(tail >= size)) {
> > > > + CT_ERROR(ct, "Invalid offsets tail=%u (size=%u)\n",
> > > > + tail, size);
> > > > + desc->status |= GUC_CTB_STATUS_OVERFLOW;
> > > > + goto corrupted;
> > > > + }
> > Now we can move this outside if/else define block as it is same check as
> > above. Again can you do this when you merge this?
> >
> > Matt
> Given that a) there are multiple changes which are not trivial one liners
> and b) I personally prefer keeping the CT_ERRORs and dropping the BUG_ON, I
> would recommend that you repost an updated patch how you want it changed.
> Shouldn't need to repost the whole set, just this one patch. And maybe get
> it reviewed by Michal as he seems to be in agreement with your preferred
> direction.
>
Ok, I sent it but I looks like patchworks didn't like it. Anyways we
should be able to review that patch.
Matt
> John.
>
>
> >
> > > > +#endif
> > > > /* tail == head condition indicates empty */
> > > > available = tail - head;
> > > > @@ -802,6 +831,7 @@ static int ct_read(struct intel_guc_ct *ct, struct ct_incoming_msg **msg)
> > > > }
> > > > CT_DEBUG(ct, "received %*ph\n", 4 * len, (*msg)->msg);
> > > > + ctb->head = head;
> > > > /* now update descriptor */
> > > > WRITE_ONCE(desc->head, head);
> > > > diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
> > > > index bee03794c1eb..edd1bba0445d 100644
> > > > --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
> > > > +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
> > > > @@ -33,6 +33,9 @@ struct intel_guc;
> > > > * @desc: pointer to the buffer descriptor
> > > > * @cmds: pointer to the commands buffer
> > > > * @size: size of the commands buffer in dwords
> > > > + * @head: local shadow copy of head in dwords
> > > > + * @tail: local shadow copy of tail in dwords
> > > > + * @space: local shadow copy of space in dwords
> > > > * @broken: flag to indicate if descriptor data is broken
> > > > */
> > > > struct intel_guc_ct_buffer {
> > > > @@ -40,6 +43,9 @@ struct intel_guc_ct_buffer {
> > > > struct guc_ct_buffer_desc *desc;
> > > > u32 *cmds;
> > > > u32 size;
> > > > + u32 tail;
> > > > + u32 head;
> > > > + u32 space;
> > > > bool broken;
> > > > };
>
More information about the Intel-gfx
mailing list