<html><head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> </head> <body> On 8/20/2021 15:44, Matthew Brost wrote: <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap="">Update context and full GPU reset to work with multi-lrc. The idea is parent context tracks all the active requests inflight for itself and its' children. The parent context owns the reset replaying / canceling</pre> </blockquote> its' -> its <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap=""> requests as needed. Signed-off-by: Matthew Brost <a class="moz-txt-link-rfc2396E" href="mailto:matthew.brost@intel.com"><matthew.brost@intel.com></a> --- drivers/gpu/drm/i915/gt/intel_context.c | 11 ++-- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 63 +++++++++++++------ 2 files changed, 51 insertions(+), 23 deletions(-) diff --git a/drivers/gpu/drm/i915/gt/intel_context.c b/drivers/gpu/drm/i915/gt/intel_context.c index 00d1aee6d199..5615be32879c 100644 --- a/drivers/gpu/drm/i915/gt/intel_context.c +++ b/drivers/gpu/drm/i915/gt/intel_context.c @@ -528,20 +528,21 @@ struct i915_request *intel_context_create_request(struct intel_context *ce) struct i915_request *intel_context_find_active_request(struct intel_context *ce) { + struct intel_context *parent = intel_context_to_parent(ce); struct i915_request *rq, *active = NULL; unsigned long flags; GEM_BUG_ON(!intel_engine_uses_guc(ce->engine));</pre> </blockquote> Should this not check the parent as well/instead? And to be clear, this can be called on regular contexts (where ce == parent) and on both the parent or child contexts of multi-LRC contexts (where ce may or may not match parent)? <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap=""> - spin_lock_irqsave(&ce->guc_state.lock, flags); - list_for_each_entry_reverse(rq, &ce->guc_state.requests, + spin_lock_irqsave(&parent->guc_state.lock, flags); + list_for_each_entry_reverse(rq, &parent->guc_state.requests, sched.link) { - if (i915_request_completed(rq)) + if (i915_request_completed(rq) && rq->context == ce)</pre> </blockquote> 'rq->context == ce' means: <ol> <li>single-LRC context, rq is owned by ce </li> <li>multi-LRC context, ce is child, rq really belongs to ce but is being tracked by parent</li> <li>multi-LRC context, ce is parent, rq really is owned by ce </li> </ol> So when 'rq->ce != ce', it means that the request is owned by a different child to 'ce' but within the same multi-LRC group. So we want to ignore that request and keep searching until we find one that is really owned by the target ce? <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap=""> break; - active = rq; + active = (rq->context == ce) ? rq : active;</pre> </blockquote> Would be clearer to say 'if(rq->ce != ce) continue;' and leave 'active = rq;' alone? And again, the intention is to ignore requests that are owned by other members of the same multi-LRC group? Would be good to add some documentation to this function to explain the above (assuming my description is correct?). <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap=""> } - spin_unlock_irqrestore(&ce->guc_state.lock, flags); + spin_unlock_irqrestore(&parent->guc_state.lock, flags); return active; } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index f0b60fecf253..e34e0ea9136a 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -670,6 +670,11 @@ static int rq_prio(const struct i915_request *rq) return rq->sched.attr.priority; } +static inline bool is_multi_lrc(struct intel_context *ce) +{ + return intel_context_is_parallel(ce); +} + static bool is_multi_lrc_rq(struct i915_request *rq) { return intel_context_is_parallel(rq->context); @@ -1179,10 +1184,13 @@ __unwind_incomplete_requests(struct intel_context *ce) static void __guc_reset_context(struct intel_context *ce, bool stalled) { + bool local_stalled; struct i915_request *rq; unsigned long flags; u32 head; + int i, number_children = ce->guc_number_children;</pre> </blockquote> If this is a child context, does it not need to pull the child count from the parent? Likewise the list/link pointers below? Or does each child context have a full list of its siblings + parent? <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap=""> bool skip = false; + struct intel_context *parent = ce; intel_context_get(ce); @@ -1209,25 +1217,34 @@ static void __guc_reset_context(struct intel_context *ce, bool stalled) if (unlikely(skip)) goto out_put; - rq = intel_context_find_active_request(ce); - if (!rq) { - head = ce->ring->tail; - stalled = false; - goto out_replay; - } + for (i = 0; i < number_children + 1; ++i) { + if (!intel_context_is_pinned(ce)) + goto next_context; + + local_stalled = false; + rq = intel_context_find_active_request(ce); + if (!rq) { + head = ce->ring->tail; + goto out_replay; + } - if (!i915_request_started(rq)) - stalled = false; + GEM_BUG_ON(i915_active_is_idle(&ce->active)); + head = intel_ring_wrap(ce->ring, rq->head); - GEM_BUG_ON(i915_active_is_idle(&ce->active)); - head = intel_ring_wrap(ce->ring, rq->head); - __i915_request_reset(rq, stalled); + if (i915_request_started(rq))</pre> </blockquote> Why change the ordering of the started test versus the wrap/reset call? Is it significant? Why is it now important to be reversed? <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap=""> + local_stalled = true; + __i915_request_reset(rq, local_stalled && stalled); out_replay: - guc_reset_state(ce, head, stalled); - __unwind_incomplete_requests(ce); + guc_reset_state(ce, head, local_stalled && stalled); +next_context: + if (i != number_children) + ce = list_next_entry(ce, guc_child_link);</pre> </blockquote> Can this not be put in to the step clause of the for statement? <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap=""> + } + + __unwind_incomplete_requests(parent); out_put: - intel_context_put(ce); + intel_context_put(parent);</pre> </blockquote> As above, I think this function would benefit from some comments to explain exactly what is being done and why. John. <blockquote type="cite" cite="mid:20210820224446.30620-18-matthew.brost@intel.com"> <pre class="moz-quote-pre" wrap=""> } void intel_guc_submission_reset(struct intel_guc *guc, bool stalled) @@ -1248,7 +1265,8 @@ void intel_guc_submission_reset(struct intel_guc *guc, bool stalled) xa_unlock(&guc->context_lookup); - if (intel_context_is_pinned(ce)) + if (intel_context_is_pinned(ce) && + !intel_context_is_child(ce)) __guc_reset_context(ce, stalled); intel_context_put(ce); @@ -1340,7 +1358,8 @@ void intel_guc_submission_cancel_requests(struct intel_guc *guc) xa_unlock(&guc->context_lookup); - if (intel_context_is_pinned(ce)) + if (intel_context_is_pinned(ce) && + !intel_context_is_child(ce)) guc_cancel_context_requests(ce); intel_context_put(ce); @@ -2031,6 +2050,8 @@ static struct i915_sw_fence *guc_context_block(struct intel_context *ce) u16 guc_id; bool enabled; + GEM_BUG_ON(intel_context_is_child(ce)); + spin_lock_irqsave(&ce->guc_state.lock, flags); incr_context_blocked(ce); @@ -2068,6 +2089,7 @@ static void guc_context_unblock(struct intel_context *ce) bool enable; GEM_BUG_ON(context_enabled(ce)); + GEM_BUG_ON(intel_context_is_child(ce)); spin_lock_irqsave(&ce->guc_state.lock, flags); @@ -2099,11 +2121,14 @@ static void guc_context_unblock(struct intel_context *ce) static void guc_context_cancel_request(struct intel_context *ce, struct i915_request *rq) { + struct intel_context *block_context = + request_to_scheduling_context(rq); + if (i915_sw_fence_signaled(&rq->submit)) { struct i915_sw_fence *fence; intel_context_get(ce); - fence = guc_context_block(ce); + fence = guc_context_block(block_context); i915_sw_fence_wait(fence); if (!i915_request_completed(rq)) { __i915_request_skip(rq); @@ -2117,7 +2142,7 @@ static void guc_context_cancel_request(struct intel_context *ce, */ flush_work(&ce_to_guc(ce)->ct.requests.worker); - guc_context_unblock(ce); + guc_context_unblock(block_context); intel_context_put(ce); } } @@ -2143,6 +2168,8 @@ static void guc_context_ban(struct intel_context *ce, struct i915_request *rq) intel_wakeref_t wakeref; unsigned long flags; + GEM_BUG_ON(intel_context_is_child(ce)); + guc_flush_submissions(guc); spin_lock_irqsave(&ce->guc_state.lock, flags); </pre> </blockquote> </body> </html>