GuC submission support
Subsystems (3)
Re: [PATCH 19/51] drm/i915/guc: GuC virtual engines
In reply to Daniele Ceraolo SpurioQuoted 109 lines · Daniele Ceraolo Spurio
Quoted 106 lines · Matthew Brost
Quoted 54 lines · Daniele Ceraolo Spurio
Quoted 52 lines · Matthew Brost
Implement GuC virtual engines. Rather simple implementation, basically just allocate an engine, setup context enter / exit function to virtual engine specific functions, set all other variables / functions to guc versions, and set the engine mask to that of all the siblings.
v2: Update to work with proto-ctx
Cc Daniele Ceraolo Spurio <[email protected]>Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gem/i915_gem_context.c | 8 +- drivers/gpu/drm/i915/gem/i915_gem_context.h | 1 + drivers/gpu/drm/i915/gt/intel_context_types.h | 6 + drivers/gpu/drm/i915/gt/intel_engine.h | 27 +- drivers/gpu/drm/i915/gt/intel_engine_cs.c | 14 + .../drm/i915/gt/intel_execlists_submission.c | 29 ++- .../drm/i915/gt/intel_execlists_submission.h | 4 - drivers/gpu/drm/i915/gt/selftest_execlists.c | 12 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 240 +++++++++++++++++- .../gpu/drm/i915/gt/uc/intel_guc_submission.h | 2 + 10 files changed, 308 insertions(+), 35 deletions(-)drivers/gpu/drm/i915/gem/i915_gem_context.c +2
-6@@ -74,7 +74,6 @@#include "gt/intel_context_param.h" #include "gt/intel_engine_heartbeat.h" #include "gt/intel_engine_user.h" -#include "gt/intel_execlists_submission.h" /* virtual_engine */ #include "gt/intel_gpu_commands.h" #include "gt/intel_ring.h" @@ -363,9 +362,6 @@ set_proto_ctx_engines_balance(struct i915_user_extension __user *base,if (!HAS_EXECLISTS (i915))return -ENODEV ; -if (intel_uc_uses_guc_submission (&i915->gt.uc)) -return -ENODEV ;/* not implement yet */ -if (get_user (idx, &ext->engine_index))return -EFAULT ;@@ -950,8 +946,8 @@ static struct i915_gem_engines *user_engines(struct i915_gem_context *ctx,break ;case I915_GEM_ENGINE_TYPE_BALANCED : - ce =intel_execlists_create_virtual (pe[n].siblings, - pe[n].num_siblings); + ce =intel_engine_create_virtual (pe[n].siblings, + pe[n].num_siblings);break ;case I915_GEM_ENGINE_TYPE_INVALID :drivers/gpu/drm/i915/gem/i915_gem_context.h +0
-0Quoted 9 lines
drivers/gpu/drm/i915/gem/i915_gem_context.h line 12
Quoted 6 lines
drivers/gpu/drm/i915/gem/i915_gem_context.h +1
-0@@ -10,6 +10,7 @@#include "i915_gem_context_types.h" #include "gt/intel_context.h" +#include "gt/intel_engine.h" Apologies for the late question, but why do you need this include in this header? nothing else is changing within the file.
It likely doesn't need to be included. Let me see what the build / CI says if I remove it.
Quoted 39 lines
Quoted 25 lines
drivers/gpu/drm/i915/gem/i915_gem_context.h +0
-0#include "i915_drv.h" #include "i915_gem.h" drivers/gpu/drm/i915/gt/intel_context_types.h +6
-0@@ -47,6 +47,12 @@ struct intel_context_ops {void (*reset)(struct intel_context *ce);void (*destroy)(struct kref *kref); + +/* virtual engine/context interface */ +struct intel_context *(*create_virtual)(struct intel_engine_cs **engine, +unsigned int count); +struct intel_engine_cs *(*get_sibling)(struct intel_engine_cs *engine, +unsigned int sibling); };struct intel_context {drivers/gpu/drm/i915/gt/intel_engine.h +1
-0@@ -273,13 +273,38 @@ intel_engine_has_preempt_reset(const struct intel_engine_cs *engine)return intel_engine_has_preemption (engine); } +struct intel_context *drivers/gpu/drm/i915/gt/intel_engine.h intel_engine_has_preempt_reset(const struct intel_engine_cs *engine) · line 285
Quoted 10 lines
drivers/gpu/drm/i915/gt/intel_engine.h +10
-0+intel_engine_create_virtual(struct intel_engine_cs **siblings, + unsigned int count); + +static inline bool +intel_virtual_engine_has_heartbeat (const struct intel_engine_cs *engine) +{ +if (intel_engine_uses_guc (engine)) +return intel_guc_virtual_engine_has_heartbeat (engine); +else +GEM_BUG_ON ("Only should be called in GuC submission" );I insist that this needs a better explanation, as I've commented on the previous rev. Given that this is a shared file, it is not immediately evident why this call shouldn't be called with the execlists backend.
Quoted 8 lines
Quoted 2 lines
Sure, can add a comment. Basically it comes down to it is always called an active request which in execlists mode is always a physical engine.
I had added a sample comment in the previous review in case you want to copy it. I remember you explaining this to me a while back, but it always takes me a moment to recall it when I look at this code and I expect someone not familiar with the backend implementation would be even more confused.
Let me see if I can dig up that comment.
Quoted 495 lines
Daniele
Quoted 492 lines
Matt
Quoted 489 lines
Daniele
Quoted 487 lines
drivers/gpu/drm/i915/gt/intel_engine.h +15
-1+ + return false; +} +static inline bool intel_engine_has_heartbeat (const struct intel_engine_cs *engine) {if (!IS_ACTIVE (CONFIG_DRM_I915_HEARTBEAT_INTERVAL ))return false; -return READ_ONCE (engine->props.heartbeat_interval_ms); +if (intel_engine_is_virtual (engine)) +return intel_virtual_engine_has_heartbeat (engine); +else +return READ_ONCE (engine->props.heartbeat_interval_ms); +} + +static inline struct intel_engine_cs * +intel_engine_get_sibling (struct intel_engine_cs *engine,unsigned int sibling) +{ +GEM_BUG_ON (!intel_engine_is_virtual (engine)); +return engine->cops->get_sibling (engine, sibling); }#endif /* _INTEL_RINGBUFFER_H_ */ drivers/gpu/drm/i915/gt/intel_engine_cs.c +14
-0@@ -1737,6 +1737,20 @@ ktime_t intel_engine_get_busy_time(struct intel_engine_cs *engine, ktime_t *now)return total; } +struct intel_context * +intel_engine_create_virtual (struct intel_engine_cs **siblings, +unsigned int count) +{ +if (count ==0 ) +return ERR_PTR (-EINVAL ); + +if (count ==1 ) +return intel_context_create (siblings[0 ]); + +GEM_BUG_ON (!siblings[0 ]->cops->create_virtual); +return siblings[0 ]->cops->create_virtual (siblings, count); +} +static bool match_ring (struct i915_request *rq) {u32 ring =ENGINE_READ (rq->engine,RING_START );drivers/gpu/drm/i915/gt/intel_execlists_submission.c +20
-9@@ -193,6 +193,9 @@ static struct virtual_engine *to_virtual_engine(struct intel_engine_cs *engine) return container_of(engine, struct virtual_engine, base); } +static struct intel_context * +execlists_create_virtual (struct intel_engine_cs **siblings,unsigned int count); +static struct i915_request * __active_request(const struct intel_timeline * const tl, struct i915_request *rq,@@ -2548,6 +2551,8 @@ static const struct intel_context_ops execlists_context_ops = { .reset = lrc_reset, .destroy = lrc_destroy, + + .create_virtual = execlists_create_virtual, };static int emit_pdps (struct i915_request *rq)@@ -3493,6 +3498,17 @@ static void virtual_context_exit(struct intel_context *ce)intel_engine_pm_put (ve->siblings[n]); } +static struct intel_engine_cs * +virtual_get_sibling (struct intel_engine_cs *engine,unsigned int sibling) +{ +struct virtual_engine *ve =to_virtual_engine (engine); + +if (sibling >= ve->num_siblings) +return NULL ; + +return ve->siblings[sibling]; +} +static const struct intel_context_ops virtual_context_ops = { .flags =COPS_HAS_INFLIGHT ,@@ -3507,6 +3523,8 @@ static const struct intel_context_ops virtual_context_ops = { .exit = virtual_context_exit, .destroy = virtual_context_destroy, + + .get_sibling = virtual_get_sibling, };static intel_engine_mask_t virtual_submission_mask (struct virtual_engine *ve)@@ -3655,20 +3673,13 @@ static void virtual_submit_request(struct i915_request *rq)spin_unlock_irqrestore (&ve->base.sched_engine->lock, flags); } -struct intel_context * -intel_execlists_create_virtual (struct intel_engine_cs **siblings, -unsigned int count) +static struct intel_context * +execlists_create_virtual (struct intel_engine_cs **siblings,unsigned int count) {struct virtual_engine *ve;unsigned int n;int err; -if (count ==0 ) -return ERR_PTR (-EINVAL ); - -if (count ==1 ) -return intel_context_create (siblings[0 ]); - ve =kzalloc (struct_size (ve, siblings, count),GFP_KERNEL );if (!ve)return ERR_PTR (-ENOMEM );drivers/gpu/drm/i915/gt/intel_execlists_submission.h +0
-4@@ -32,10 +32,6 @@ void intel_execlists_show_requests(struct intel_engine_cs *engine, int indent), unsigned int max); -struct intel_context * -intel_execlists_create_virtual (struct intel_engine_cs **siblings, -unsigned int count); -bool intel_engine_in_execlists_submission_mode (const struct intel_engine_cs *engine);drivers/gpu/drm/i915/gt/selftest_execlists.c +6
-6@@ -3727,7 +3727,7 @@ static int nop_virtual_engine(struct intel_gt *gt,GEM_BUG_ON (!nctx || nctx >ARRAY_SIZE (ve));for (n =0 ; n < nctx; n++) { - ve[n] =intel_execlists_create_virtual (siblings, nsibling); + ve[n] =intel_engine_create_virtual (siblings, nsibling);if (IS_ERR (ve[n])) { err =PTR_ERR (ve[n]); nctx = n;@@ -3923,7 +3923,7 @@ static int mask_virtual_engine(struct intel_gt *gt,* restrict it to our desired engine within the virtual engine. */ - ve =intel_execlists_create_virtual (siblings, nsibling); + ve =intel_engine_create_virtual (siblings, nsibling);if (IS_ERR (ve)) { err =PTR_ERR (ve); goto out_close;@@ -4054,7 +4054,7 @@ static int slicein_virtual_engine(struct intel_gt *gt,i915_request_add (rq); } - ce =intel_execlists_create_virtual (siblings, nsibling); + ce =intel_engine_create_virtual (siblings, nsibling);if (IS_ERR (ce)) { err =PTR_ERR (ce); goto out;@@ -4106,7 +4106,7 @@ static int sliceout_virtual_engine(struct intel_gt *gt,/* XXX We do not handle oversubscription and fairness with normal rq */ for (n =0 ; n < nsibling; n++) { - ce =intel_execlists_create_virtual (siblings, nsibling); + ce =intel_engine_create_virtual (siblings, nsibling);if (IS_ERR (ce)) { err =PTR_ERR (ce); goto out;@@ -4208,7 +4208,7 @@ static int preserved_virtual_engine(struct intel_gt *gt,if (err) goto out_scratch; - ve =intel_execlists_create_virtual (siblings, nsibling); + ve =intel_engine_create_virtual (siblings, nsibling);if (IS_ERR (ve)) { err =PTR_ERR (ve); goto out_scratch;@@ -4348,7 +4348,7 @@ static int reset_virtual_engine(struct intel_gt *gt,if (igt_spinner_init (&spin, gt))return -ENOMEM ; - ve =intel_execlists_create_virtual (siblings, nsibling); + ve =intel_engine_create_virtual (siblings, nsibling);if (IS_ERR (ve)) { err =PTR_ERR (ve); goto out_spin;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +231
-9@@ -60,6 +60,15 @@* */ +/* GuC Virtual Engine */ +struct guc_virtual_engine { +struct intel_engine_cs base; +struct intel_context context; +}; + +static struct intel_context * +guc_create_virtual (struct intel_engine_cs **siblings,unsigned int count); +#define GUC_REQUEST_SIZE 64/* bytes */ /* @@ -925,20 +934,35 @@ static int guc_lrc_desc_pin(struct intel_context *ce)return ret; } -static int guc_context_pre_pin (struct intel_context *ce, -struct i915_gem_ww_ctx *ww, -void **vaddr) +static int __guc_context_pre_pin (struct intel_context *ce, +struct intel_engine_cs *engine, +struct i915_gem_ww_ctx *ww, +void **vaddr) { -return lrc_pre_pin (ce, ce->engine, ww, vaddr); +return lrc_pre_pin (ce, engine, ww, vaddr); } -static int guc_context_pin (struct intel_context *ce,void *vaddr) +static int __guc_context_pin (struct intel_context *ce, +struct intel_engine_cs *engine, +void *vaddr) {if (i915_ggtt_offset (ce->state) != (ce->lrc.lrca &CTX_GTT_ADDRESS_MASK ))set_bit (CONTEXT_LRCA_DIRTY , &ce->flags); -return lrc_pin (ce, ce->engine, vaddr); +return lrc_pin (ce, engine, vaddr); +} + +static int guc_context_pre_pin (struct intel_context *ce, +struct i915_gem_ww_ctx *ww, +void **vaddr) +{ +return __guc_context_pre_pin (ce, ce->engine, ww, vaddr); +} + +static int guc_context_pin (struct intel_context *ce,void *vaddr) +{ +return __guc_context_pin (ce, ce->engine, vaddr); }static void guc_context_unpin (struct intel_context *ce)@@ -1043,6 +1067,21 @@ static inline void guc_lrc_desc_unpin(struct intel_context *ce)deregister_context (ce, ce->guc_id); } +static void __guc_context_destroy (struct intel_context *ce) +{ +lrc_fini (ce); +intel_context_fini (ce); + +if (intel_engine_is_virtual (ce->engine)) { +struct guc_virtual_engine *ve = +container_of (ce,typeof (*ve), context); + +kfree (ve); + }else { +intel_context_free (ce); + } +} +static void guc_context_destroy (struct kref *kref) {struct intel_context *ce =container_of (kref,typeof (*ce), ref);@@ -1059,7 +1098,7 @@ static void guc_context_destroy(struct kref *kref)if (context_guc_id_invalid (ce) || !lrc_desc_registered (guc, ce->guc_id)) {release_guc_id (guc, ce); -lrc_destroy (kref); +__guc_context_destroy (ce);return ; }@@ -1075,7 +1114,7 @@ static void guc_context_destroy(struct kref *kref)if (context_guc_id_invalid (ce)) {__release_guc_id (guc, ce);spin_unlock_irqrestore (&guc->contexts_lock, flags); -lrc_destroy (kref); +__guc_context_destroy (ce);return ; }@@ -1120,6 +1159,8 @@ static const struct intel_context_ops guc_context_ops = { .reset = lrc_reset, .destroy = guc_context_destroy, + + .create_virtual = guc_create_virtual, };static void __guc_signal_context_fence (struct intel_context *ce)@@ -1248,6 +1289,83 @@ static int guc_request_alloc(struct i915_request *rq)return 0 ; } +static struct intel_engine_cs * +guc_virtual_get_sibling (struct intel_engine_cs *ve,unsigned int sibling) +{ +struct intel_engine_cs *engine; +intel_engine_mask_t tmp, mask = ve->mask; +unsigned int num_siblings =0 ; + +for_each_engine_masked (engine, ve->gt, mask, tmp) +if (num_siblings++ == sibling) +return engine; + +return NULL ; +} + +static int guc_virtual_context_pre_pin (struct intel_context *ce, +struct i915_gem_ww_ctx *ww, +void **vaddr) +{ +struct intel_engine_cs *engine =guc_virtual_get_sibling (ce->engine,0 ); + +return __guc_context_pre_pin (ce, engine, ww, vaddr); +} + +static int guc_virtual_context_pin (struct intel_context *ce,void *vaddr) +{ +struct intel_engine_cs *engine =guc_virtual_get_sibling (ce->engine,0 ); + +return __guc_context_pin (ce, engine, vaddr); +} + +static void guc_virtual_context_enter (struct intel_context *ce) +{ +intel_engine_mask_t tmp, mask = ce->engine->mask; +struct intel_engine_cs *engine; + +for_each_engine_masked (engine, ce->engine->gt, mask, tmp) +intel_engine_pm_get (engine); + +intel_timeline_enter (ce->timeline); +} + +static void guc_virtual_context_exit (struct intel_context *ce) +{ +intel_engine_mask_t tmp, mask = ce->engine->mask; +struct intel_engine_cs *engine; + +for_each_engine_masked (engine, ce->engine->gt, mask, tmp) +intel_engine_pm_put (engine); + +intel_timeline_exit (ce->timeline); +} + +static int guc_virtual_context_alloc (struct intel_context *ce) +{ +struct intel_engine_cs *engine =guc_virtual_get_sibling (ce->engine,0 ); + +return lrc_alloc (ce, engine); +} + +static const struct intel_context_ops virtual_guc_context_ops = { + .alloc = guc_virtual_context_alloc, + + .pre_pin = guc_virtual_context_pre_pin, + .pin = guc_virtual_context_pin, + .unpin = guc_context_unpin, + .post_unpin = guc_context_post_unpin, + + .enter = guc_virtual_context_enter, + .exit = guc_virtual_context_exit, + + .sched_disable = guc_context_sched_disable, + + .destroy = guc_context_destroy, + + .get_sibling = guc_virtual_get_sibling, +}; +static void sanitize_hwsp (struct intel_engine_cs *engine) {struct intel_timeline *tl;@@ -1559,7 +1677,7 @@ int intel_guc_deregister_done_process_msg(struct intel_guc *guc, }else if (context_destroyed (ce)) {/* Context has been destroyed */ release_guc_id (guc, ce); -lrc_destroy (&ce->ref); +__guc_context_destroy (ce); }decr_outstanding_submission_g2h (guc);@@ -1674,3 +1792,107 @@ void intel_guc_submission_print_context_info(struct intel_guc *guc, atomic_read(&ce->guc_sched_state_no_lock)); } } + +static struct intel_context * +guc_create_virtual (struct intel_engine_cs **siblings,unsigned int count) +{ +struct guc_virtual_engine *ve; +struct intel_guc *guc; +unsigned int n; +int err; + + ve =kzalloc (sizeof (*ve),GFP_KERNEL ); +if (!ve) +return ERR_PTR (-ENOMEM ); + + guc = &siblings[0 ]->gt->uc.guc; + + ve->base.i915 = siblings[0 ]->i915; + ve->base.gt = siblings[0 ]->gt; + ve->base.uncore = siblings[0 ]->uncore; + ve->base.id =-1 ; + + ve->base.uabi_class =I915_ENGINE_CLASS_INVALID ; + ve->base.instance =I915_ENGINE_CLASS_INVALID_VIRTUAL ; + ve->base.uabi_instance =I915_ENGINE_CLASS_INVALID_VIRTUAL ; + ve->base.saturated =ALL_ENGINES ; + ve->base.breadcrumbs =intel_breadcrumbs_create (&ve->base); +if (!ve->base.breadcrumbs) { +kfree (ve); +return ERR_PTR (-ENOMEM ); + } + +snprintf (ve->base.name,sizeof (ve->base.name),"virtual" ); + + ve->base.sched_engine =i915_sched_engine_get (guc->sched_engine); + + ve->base.cops = &virtual_guc_context_ops; + ve->base.request_alloc = guc_request_alloc; + + ve->base.submit_request = guc_submit_request; + + ve->base.flags =I915_ENGINE_IS_VIRTUAL ; + +intel_context_init (&ve->context, &ve->base); + +for (n =0 ; n < count; n++) { +struct intel_engine_cs *sibling = siblings[n]; + +GEM_BUG_ON (!is_power_of_2 (sibling->mask)); +if (sibling->mask & ve->base.mask) { +DRM_DEBUG ("duplicate %s entry in load balancer\n" , + sibling->name); + err = -EINVAL ; + goto err_put; + } + + ve->base.mask |= sibling->mask; + +if (n !=0 && ve->base.class != sibling->class) { +DRM_DEBUG ("invalid mixing of engine class, sibling %d, already %d\n" , + sibling->class, ve->base.class); + err = -EINVAL ; + goto err_put; + }else if (n ==0 ) { + ve->base.class = sibling->class; + ve->base.uabi_class = sibling->uabi_class; +snprintf (ve->base.name,sizeof (ve->base.name), +"v%dx%d" , ve->base.class, count); + ve->base.context_size = sibling->context_size; + + ve->base.emit_bb_start = sibling->emit_bb_start; + ve->base.emit_flush = sibling->emit_flush; + ve->base.emit_init_breadcrumb = + sibling->emit_init_breadcrumb; + ve->base.emit_fini_breadcrumb = + sibling->emit_fini_breadcrumb; + ve->base.emit_fini_breadcrumb_dw = + sibling->emit_fini_breadcrumb_dw; + + ve->base.flags |= sibling->flags; + + ve->base.props.timeslice_duration_ms = + sibling->props.timeslice_duration_ms; + } + } + +return &ve->context; + +err_put: +intel_context_put (&ve->context); +return ERR_PTR (err); +} + + + +bool intel_guc_virtual_engine_has_heartbeat (const struct intel_engine_cs *ve) +{ +struct intel_engine_cs *engine; +intel_engine_mask_t tmp, mask = ve->mask; + +for_each_engine_masked (engine, ve->gt, mask, tmp) +if (READ_ONCE (engine->props.heartbeat_interval_ms)) +return true; + +return false; +}drivers/gpu/drm/i915/gt/uc/intel_guc_submission.h +2
-0@@ -26,6 +26,8 @@ void intel_guc_submission_print_info(struct intel_guc *guc,void intel_guc_submission_print_context_info (struct intel_guc *guc,struct drm_printer *p); +bool intel_guc_virtual_engine_has_heartbeat (const struct intel_engine_cs *ve); +static inline bool intel_guc_submission_is_supported (struct intel_guc *guc) {/* XXX: GuC submission is unavailable for now */
PATCH 20/51 drm/i915: Track 'serial' counts for virtual engines
In reply to Matthew BrostFrom: John Harrison <[email protected]>
The serial number tracking of engines happens at the backend of request submission and was expecting to only be given physical engines. However, in GuC submission mode, the decomposition of virtual to physical engines does not happen in i915. Instead, requests are submitted to their virtual engine mask all the way through to the hardware (i.e. to GuC). This would mean that the heart beat code thinks the physical engines are idle due to the serial number not incrementing.
This patch updates the tracking to decompose virtual engines into their physical constituents and tracks the request against each. This is not entirely accurate as the GuC will only be issuing the request to one physical engine. However, it is the best that i915 can do given that it has no knowledge of the GuC's scheduling decisions.
--- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 ++ .../gpu/drm/i915/gt/intel_execlists_submission.c | 6 ++++++ drivers/gpu/drm/i915/gt/intel_ring_submission.c | 6 ++++++ drivers/gpu/drm/i915/gt/mock_engine.c | 6 ++++++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 16 ++++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 4 +++- 6 files changed, 39 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/gt/intel_engine_types.h +2 -0
@@ -388,6 +388,8 @@ struct intel_engine_cs {void (*park)(struct intel_engine_cs *engine);void (*unpark)(struct intel_engine_cs *engine); +void (*bump_serial)(struct intel_engine_cs *engine); +void (*set_default_submission)(struct intel_engine_cs *engine);const struct intel_context_ops *cops;
drivers/gpu/drm/i915/gt/intel_execlists_submission.c +6 -0
@@ -3191,6 +3191,11 @@ static void execlists_release(struct intel_engine_cs *engine)lrc_fini_wa_ctx (engine); } +static void execlist_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +static void logical_ring_default_vfuncs (struct intel_engine_cs *engine) {
@@ -3200,6 +3205,7 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &execlists_context_ops; engine->request_alloc = execlists_request_alloc; + engine->bump_serial = execlist_bump_serial; engine->reset.prepare = execlists_reset_prepare; engine->reset.rewind = execlists_reset_rewind;
drivers/gpu/drm/i915/gt/intel_ring_submission.c +6 -0
@@ -1047,6 +1047,11 @@ static void setup_irq(struct intel_engine_cs *engine) } } +static void ring_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +static void setup_common (struct intel_engine_cs *engine) {struct drm_i915_private *i915 = engine->i915;
@@ -1066,6 +1071,7 @@ static void setup_common(struct intel_engine_cs *engine) engine->cops = &ring_context_ops; engine->request_alloc = ring_request_alloc; + engine->bump_serial = ring_bump_serial;/* * Using a global execution timeline; the previous final breadcrumb is
drivers/gpu/drm/i915/gt/mock_engine.c +6 -0
@@ -292,6 +292,11 @@ static void mock_engine_release(struct intel_engine_cs *engine)intel_engine_fini_retire (engine); } +static void mock_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +struct intel_engine_cs *mock_engine (struct drm_i915_private *i915,const char *name,int id)
@@ -318,6 +323,7 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, engine->base.cops = &mock_context_ops; engine->base.request_alloc = mock_request_alloc; + engine->base.bump_serial = mock_bump_serial; engine->base.emit_flush = mock_emit_flush; engine->base.emit_fini_breadcrumb = mock_emit_breadcrumb; engine->base.submit_request = mock_submit_request;
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +16 -0
@@ -1485,6 +1485,20 @@ static void guc_release(struct intel_engine_cs *engine)lrc_fini_wa_ctx (engine); } +static void guc_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} + +static void virtual_guc_bump_serial (struct intel_engine_cs *engine) +{ +struct intel_engine_cs *e; +intel_engine_mask_t tmp, mask = engine->mask; + +for_each_engine_masked (e, engine->gt, mask, tmp) + e->serial++; +} +static void guc_default_vfuncs (struct intel_engine_cs *engine) {/* Default vfuncs which can be overridden by each engine. */
@@ -1493,6 +1507,7 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &guc_context_ops; engine->request_alloc = guc_request_alloc; + engine->bump_serial = guc_bump_serial; engine->sched_engine->schedule = i915_schedule;
@@ -1828,6 +1843,7 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) ve->base.cops = &virtual_guc_context_ops; ve->base.request_alloc = guc_request_alloc; + ve->base.bump_serial = virtual_guc_bump_serial; ve->base.submit_request = guc_submit_request;
drivers/gpu/drm/i915/i915_request.c +3 -1
@@ -669,7 +669,9 @@ bool __i915_request_submit(struct i915_request *request) request->ring->vaddr + request->postfix);trace_i915_request_execute (request); - engine->serial++; +if (engine->bump_serial) + engine->bump_serial (engine); + result = true;GEM_BUG_ON (test_bit (I915_FENCE_FLAG_ACTIVE , &request->fence.flags));
Signature · 1 line
2.28.0
Re: [PATCH 20/51] drm/i915: Track 'serial' counts for virtual engines
Quoted 11 lines · Matthew Brost
From: John Harrison <[email protected]>
The serial number tracking of engines happens at the backend of request submission and was expecting to only be given physical engines. However, in GuC submission mode, the decomposition of virtual to physical engines does not happen in i915. Instead, requests are submitted to their virtual engine mask all the way through to the hardware (i.e. to GuC). This would mean that the heart beat code thinks the physical engines are idle due to the serial number not incrementing.
Quoted 8 lines
This patch updates the tracking to decompose virtual engines into their physical constituents and tracks the request against each. This is not entirely accurate as the GuC will only be issuing the request to one physical engine. However, it is the best that i915 can do given that it has no knowledge of the GuC's scheduling decisions.
Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>
Still needs to pull in Tvrtko's updated subject and description.
John.
Quoted 150 lines
--- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 ++ .../gpu/drm/i915/gt/intel_execlists_submission.c | 6 ++++++ drivers/gpu/drm/i915/gt/intel_ring_submission.c | 6 ++++++ drivers/gpu/drm/i915/gt/mock_engine.c | 6 ++++++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 16 ++++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 4 +++- 6 files changed, 39 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/gt/intel_engine_types.h +2
-0@@ -388,6 +388,8 @@ struct intel_engine_cs {void (*park)(struct intel_engine_cs *engine);void (*unpark)(struct intel_engine_cs *engine); +void (*bump_serial)(struct intel_engine_cs *engine); +void (*set_default_submission)(struct intel_engine_cs *engine);const struct intel_context_ops *cops;drivers/gpu/drm/i915/gt/intel_execlists_submission.c +6
-0@@ -3191,6 +3191,11 @@ static void execlists_release(struct intel_engine_cs *engine)lrc_fini_wa_ctx (engine); } +static void execlist_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +static void logical_ring_default_vfuncs (struct intel_engine_cs *engine) {@@ -3200,6 +3205,7 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &execlists_context_ops; engine->request_alloc = execlists_request_alloc; + engine->bump_serial = execlist_bump_serial; engine->reset.prepare = execlists_reset_prepare; engine->reset.rewind = execlists_reset_rewind;drivers/gpu/drm/i915/gt/intel_ring_submission.c +6
-0@@ -1047,6 +1047,11 @@ static void setup_irq(struct intel_engine_cs *engine) } } +static void ring_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +static void setup_common (struct intel_engine_cs *engine) {struct drm_i915_private *i915 = engine->i915;@@ -1066,6 +1071,7 @@ static void setup_common(struct intel_engine_cs *engine) engine->cops = &ring_context_ops; engine->request_alloc = ring_request_alloc; + engine->bump_serial = ring_bump_serial;/* * Using a global execution timeline; the previous final breadcrumb is drivers/gpu/drm/i915/gt/mock_engine.c +6
-0@@ -292,6 +292,11 @@ static void mock_engine_release(struct intel_engine_cs *engine)intel_engine_fini_retire (engine); } +static void mock_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +struct intel_engine_cs *mock_engine (struct drm_i915_private *i915,const char *name,int id)@@ -318,6 +323,7 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, engine->base.cops = &mock_context_ops; engine->base.request_alloc = mock_request_alloc; + engine->base.bump_serial = mock_bump_serial; engine->base.emit_flush = mock_emit_flush; engine->base.emit_fini_breadcrumb = mock_emit_breadcrumb; engine->base.submit_request = mock_submit_request;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +16
-0@@ -1485,6 +1485,20 @@ static void guc_release(struct intel_engine_cs *engine)lrc_fini_wa_ctx (engine); } +static void guc_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} + +static void virtual_guc_bump_serial (struct intel_engine_cs *engine) +{ +struct intel_engine_cs *e; +intel_engine_mask_t tmp, mask = engine->mask; + +for_each_engine_masked (e, engine->gt, mask, tmp) + e->serial++; +} +static void guc_default_vfuncs (struct intel_engine_cs *engine) {/* Default vfuncs which can be overridden by each engine. */ @@ -1493,6 +1507,7 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &guc_context_ops; engine->request_alloc = guc_request_alloc; + engine->bump_serial = guc_bump_serial; engine->sched_engine->schedule = i915_schedule;@@ -1828,6 +1843,7 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) ve->base.cops = &virtual_guc_context_ops; ve->base.request_alloc = guc_request_alloc; + ve->base.bump_serial = virtual_guc_bump_serial; ve->base.submit_request = guc_submit_request;drivers/gpu/drm/i915/i915_request.c +3
-1@@ -669,7 +669,9 @@ bool __i915_request_submit(struct i915_request *request) request->ring->vaddr + request->postfix);trace_i915_request_execute (request); - engine->serial++; +if (engine->bump_serial) + engine->bump_serial (engine); + result = true;GEM_BUG_ON (test_bit (I915_FENCE_FLAG_ACTIVE , &request->fence.flags));
Re: [PATCH 20/51] drm/i915: Track 'serial' counts for virtual engines
Quoted 18 lines · John Harrison
Quoted 17 lines · Matthew Brost
From: John Harrison <[email protected]>
The serial number tracking of engines happens at the backend of request submission and was expecting to only be given physical engines. However, in GuC submission mode, the decomposition of virtual to physical engines does not happen in i915. Instead, requests are submitted to their virtual engine mask all the way through to the hardware (i.e. to GuC). This would mean that the heart beat code thinks the physical engines are idle due to the serial number not incrementing.
This patch updates the tracking to decompose virtual engines into their physical constituents and tracks the request against each. This is not entirely accurate as the GuC will only be issuing the request to one physical engine. However, it is the best that i915 can do given that it has no knowledge of the GuC's scheduling decisions.
Quoted 4 lines
Quoted 2 lines
Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>Still needs to pull in Tvrtko's updated subject and description.
Forgot this in the post, already have it fixed locally.
Matt
Quoted 134 lines
John.
Quoted 131 lines
--- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 ++ .../gpu/drm/i915/gt/intel_execlists_submission.c | 6 ++++++ drivers/gpu/drm/i915/gt/intel_ring_submission.c | 6 ++++++ drivers/gpu/drm/i915/gt/mock_engine.c | 6 ++++++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 16 ++++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 4 +++- 6 files changed, 39 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/gt/intel_engine_types.h +2
-0@@ -388,6 +388,8 @@ struct intel_engine_cs {void (*park)(struct intel_engine_cs *engine);void (*unpark)(struct intel_engine_cs *engine); +void (*bump_serial)(struct intel_engine_cs *engine); +void (*set_default_submission)(struct intel_engine_cs *engine);const struct intel_context_ops *cops;drivers/gpu/drm/i915/gt/intel_execlists_submission.c +6
-0@@ -3191,6 +3191,11 @@ static void execlists_release(struct intel_engine_cs *engine)lrc_fini_wa_ctx (engine); } +static void execlist_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +static void logical_ring_default_vfuncs (struct intel_engine_cs *engine) {@@ -3200,6 +3205,7 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &execlists_context_ops; engine->request_alloc = execlists_request_alloc; + engine->bump_serial = execlist_bump_serial; engine->reset.prepare = execlists_reset_prepare; engine->reset.rewind = execlists_reset_rewind;drivers/gpu/drm/i915/gt/intel_ring_submission.c +6
-0@@ -1047,6 +1047,11 @@ static void setup_irq(struct intel_engine_cs *engine) } } +static void ring_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +static void setup_common (struct intel_engine_cs *engine) {struct drm_i915_private *i915 = engine->i915;@@ -1066,6 +1071,7 @@ static void setup_common(struct intel_engine_cs *engine) engine->cops = &ring_context_ops; engine->request_alloc = ring_request_alloc; + engine->bump_serial = ring_bump_serial;/* * Using a global execution timeline; the previous final breadcrumb is drivers/gpu/drm/i915/gt/mock_engine.c +6
-0@@ -292,6 +292,11 @@ static void mock_engine_release(struct intel_engine_cs *engine)intel_engine_fini_retire (engine); } +static void mock_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +struct intel_engine_cs *mock_engine (struct drm_i915_private *i915,const char *name,int id)@@ -318,6 +323,7 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, engine->base.cops = &mock_context_ops; engine->base.request_alloc = mock_request_alloc; + engine->base.bump_serial = mock_bump_serial; engine->base.emit_flush = mock_emit_flush; engine->base.emit_fini_breadcrumb = mock_emit_breadcrumb; engine->base.submit_request = mock_submit_request;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +16
-0@@ -1485,6 +1485,20 @@ static void guc_release(struct intel_engine_cs *engine)lrc_fini_wa_ctx (engine); } +static void guc_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} + +static void virtual_guc_bump_serial (struct intel_engine_cs *engine) +{ +struct intel_engine_cs *e; +intel_engine_mask_t tmp, mask = engine->mask; + +for_each_engine_masked (e, engine->gt, mask, tmp) + e->serial++; +} +static void guc_default_vfuncs (struct intel_engine_cs *engine) {/* Default vfuncs which can be overridden by each engine. */ @@ -1493,6 +1507,7 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &guc_context_ops; engine->request_alloc = guc_request_alloc; + engine->bump_serial = guc_bump_serial; engine->sched_engine->schedule = i915_schedule;@@ -1828,6 +1843,7 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) ve->base.cops = &virtual_guc_context_ops; ve->base.request_alloc = guc_request_alloc; + ve->base.bump_serial = virtual_guc_bump_serial; ve->base.submit_request = guc_submit_request;drivers/gpu/drm/i915/i915_request.c +3
-1@@ -669,7 +669,9 @@ bool __i915_request_submit(struct i915_request *request) request->ring->vaddr + request->postfix);trace_i915_request_execute (request); - engine->serial++; +if (engine->bump_serial) + engine->bump_serial (engine); + result = true;GEM_BUG_ON (test_bit (I915_FENCE_FLAG_ACTIVE , &request->fence.flags));
Matthew Brost
Reviewed-byRe: [PATCH 20/51] drm/i915: Track 'serial' counts for virtual engines
Quoted 19 lines · Matthew Brost
Quoted 18 lines · John Harrison
Quoted 17 lines · Matthew Brost
From: John Harrison <[email protected]>
The serial number tracking of engines happens at the backend of request submission and was expecting to only be given physical engines. However, in GuC submission mode, the decomposition of virtual to physical engines does not happen in i915. Instead, requests are submitted to their virtual engine mask all the way through to the hardware (i.e. to GuC). This would mean that the heart beat code thinks the physical engines are idle due to the serial number not incrementing.
This patch updates the tracking to decompose virtual engines into their physical constituents and tracks the request against each. This is not entirely accurate as the GuC will only be issuing the request to one physical engine. However, it is the best that i915 can do given that it has no knowledge of the GuC's scheduling decisions.
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>Still needs to pull in Tvrtko's updated subject and description.
Forgot this in the post, already have it fixed locally.
In addition to updated comment, I took his advise to have a default behavior without the vfunc to just increment engine->serial.
With those changes:
Quoted 136 lines
Matt
Quoted 134 lines
John.
Quoted 131 lines
--- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 ++ .../gpu/drm/i915/gt/intel_execlists_submission.c | 6 ++++++ drivers/gpu/drm/i915/gt/intel_ring_submission.c | 6 ++++++ drivers/gpu/drm/i915/gt/mock_engine.c | 6 ++++++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 16 ++++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 4 +++- 6 files changed, 39 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/gt/intel_engine_types.h +2
-0@@ -388,6 +388,8 @@ struct intel_engine_cs {void (*park)(struct intel_engine_cs *engine);void (*unpark)(struct intel_engine_cs *engine); +void (*bump_serial)(struct intel_engine_cs *engine); +void (*set_default_submission)(struct intel_engine_cs *engine);const struct intel_context_ops *cops;drivers/gpu/drm/i915/gt/intel_execlists_submission.c +6
-0@@ -3191,6 +3191,11 @@ static void execlists_release(struct intel_engine_cs *engine)lrc_fini_wa_ctx (engine); } +static void execlist_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +static void logical_ring_default_vfuncs (struct intel_engine_cs *engine) {@@ -3200,6 +3205,7 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &execlists_context_ops; engine->request_alloc = execlists_request_alloc; + engine->bump_serial = execlist_bump_serial; engine->reset.prepare = execlists_reset_prepare; engine->reset.rewind = execlists_reset_rewind;drivers/gpu/drm/i915/gt/intel_ring_submission.c +6
-0@@ -1047,6 +1047,11 @@ static void setup_irq(struct intel_engine_cs *engine) } } +static void ring_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +static void setup_common (struct intel_engine_cs *engine) {struct drm_i915_private *i915 = engine->i915;@@ -1066,6 +1071,7 @@ static void setup_common(struct intel_engine_cs *engine) engine->cops = &ring_context_ops; engine->request_alloc = ring_request_alloc; + engine->bump_serial = ring_bump_serial;/* * Using a global execution timeline; the previous final breadcrumb is drivers/gpu/drm/i915/gt/mock_engine.c +6
-0@@ -292,6 +292,11 @@ static void mock_engine_release(struct intel_engine_cs *engine)intel_engine_fini_retire (engine); } +static void mock_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} +struct intel_engine_cs *mock_engine (struct drm_i915_private *i915,const char *name,int id)@@ -318,6 +323,7 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, engine->base.cops = &mock_context_ops; engine->base.request_alloc = mock_request_alloc; + engine->base.bump_serial = mock_bump_serial; engine->base.emit_flush = mock_emit_flush; engine->base.emit_fini_breadcrumb = mock_emit_breadcrumb; engine->base.submit_request = mock_submit_request;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +16
-0@@ -1485,6 +1485,20 @@ static void guc_release(struct intel_engine_cs *engine)lrc_fini_wa_ctx (engine); } +static void guc_bump_serial (struct intel_engine_cs *engine) +{ + engine->serial++; +} + +static void virtual_guc_bump_serial (struct intel_engine_cs *engine) +{ +struct intel_engine_cs *e; +intel_engine_mask_t tmp, mask = engine->mask; + +for_each_engine_masked (e, engine->gt, mask, tmp) + e->serial++; +} +static void guc_default_vfuncs (struct intel_engine_cs *engine) {/* Default vfuncs which can be overridden by each engine. */ @@ -1493,6 +1507,7 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &guc_context_ops; engine->request_alloc = guc_request_alloc; + engine->bump_serial = guc_bump_serial; engine->sched_engine->schedule = i915_schedule;@@ -1828,6 +1843,7 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) ve->base.cops = &virtual_guc_context_ops; ve->base.request_alloc = guc_request_alloc; + ve->base.bump_serial = virtual_guc_bump_serial; ve->base.submit_request = guc_submit_request;drivers/gpu/drm/i915/i915_request.c +3
-1@@ -669,7 +669,9 @@ bool __i915_request_submit(struct i915_request *request) request->ring->vaddr + request->postfix);trace_i915_request_execute (request); - engine->serial++; +if (engine->bump_serial) + engine->bump_serial (engine); + result = true;GEM_BUG_ON (test_bit (I915_FENCE_FLAG_ACTIVE , &request->fence.flags));
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 21/51 drm/i915: Hold reference to intel_context over life of i915_request
In reply to Matthew BrostHold a reference to the intel_context over life of an i915_request. Without this an i915_request can exist after the context has been destroyed (e.g. request retired, context closed, but user space holds a reference to the request from an out fence). In the case of GuC submission + virtual engine, the engine that the request references is also destroyed which can trigger bad pointer dref in fence ops (e.g. i915_fence_get_driver_name). We could likely change i915_fence_get_driver_name to avoid touching the engine but let's just be safe and hold the intel_context reference.
v2:
(John Harrison)
- Update comment explaining how GuC mode and execlists mode deal with virtual engines differently
--- drivers/gpu/drm/i915/i915_request.c | 55 ++++++++++++----------------- 1 file changed, 23 insertions(+), 32 deletions(-)
drivers/gpu/drm/i915/i915_request.c +23 -32
@@ -125,39 +125,17 @@ static void i915_fence_release(struct dma_fence *fence)i915_sw_fence_fini (&rq->semaphore);/* -* Keep one request on each engine for reserved use under mempressure -* -* We do not hold a reference to the engine here and so have to be -* very careful in what rq->engine we poke. The virtual engine is -* referenced via the rq->context and we released that ref during -* i915_request_retire(), ergo we must not dereference a virtual -* engine here. Not that we would want to, as the only consumer of -* the reserved engine->request_pool is the power management parking, -* which must-not-fail, and that is only run on the physical engines. -* -* Since the request must have been executed to be have completed, -* we know that it will have been processed by the HW and will -* not be unsubmitted again, so rq->engine and rq->execution_mask -* at this point is stable. rq->execution_mask will be a single -* bit if the last and _only_ engine it could execution on was a -* physical engine, if it's multiple bits then it started on and -* could still be on a virtual engine. Thus if the mask is not a -* power-of-two we assume that rq->engine may still be a virtual -* engine and so a dangling invalid pointer that we cannot dereference -* -* For example, consider the flow of a bonded request through a virtual -* engine. The request is created with a wide engine mask (all engines -* that we might execute on). On processing the bond, the request mask -* is reduced to one or more engines. If the request is subsequently -* bound to a single engine, it will then be constrained to only -* execute on that engine and never returned to the virtual engine -* after timeslicing away, see __unwind_incomplete_requests(). Thus we -* know that if the rq->execution_mask is a single bit, rq->engine -* can be a physical engine with the exact corresponding mask. +* Keep one request on each engine for reserved use under mempressure, +* do not use with virtual engines as this really is only needed for +* kernel contexts. */ -if (is_power_of_2 (rq->execution_mask) && - !cmpxchg (&rq->engine->request_pool,NULL , rq)) +if (!intel_engine_is_virtual (rq->engine) && + !cmpxchg (&rq->engine->request_pool,NULL , rq)) { +intel_context_put (rq->context);return ; + } + +intel_context_put (rq->context);kmem_cache_free (global.slab_requests, rq); }
@@ -954,7 +932,19 @@ __i915_request_create(struct intel_context *ce, gfp_t gfp) } } - rq->context = ce; +/* +* Hold a reference to the intel_context over life of an i915_request. +* Without this an i915_request can exist after the context has been +* destroyed (e.g. request retired, context closed, but user space holds +* a reference to the request from an out fence). In the case of GuC +* submission + virtual engine, the engine that the request references +* is also destroyed which can trigger bad pointer dref in fence ops +* (e.g. i915_fence_get_driver_name). We could likely change these +* functions to avoid touching the engine but let's just be safe and +* hold the intel_context reference. In execlist mode the request always +* eventually points to a physical engine so this isn't an issue. +*/ + rq->context =intel_context_get (ce); rq->engine = ce->engine; rq->ring = ce->ring; rq->execution_mask = ce->engine->mask;
@@ -1031,6 +1021,7 @@ __i915_request_create(struct intel_context *ce, gfp_t gfp)GEM_BUG_ON (!list_empty (&rq->sched.waiters_list)); err_free: +intel_context_put (ce);kmem_cache_free (global.slab_requests, rq); err_unreserve:intel_context_unpin (ce);
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 22/51 drm/i915/guc: Disable bonding extension with GuC submission
In reply to Matthew BrostUpdate the bonding extension to return -ENODEV when using GuC submission as this extension fundamentally will not work with the GuC submission interface.
--- drivers/gpu/drm/i915/gem/i915_gem_context.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/gpu/drm/i915/gem/i915_gem_context.c +5 -0
@@ -491,6 +491,11 @@ set_proto_ctx_engines_bond(struct i915_user_extension __user *base, void *data)return -EINVAL ; } +if (intel_engine_uses_guc (master)) { +DRM_DEBUG ("bonding extension not supported with GuC submission" ); +return -ENODEV ; + } +if (get_user (num_bonds, &ext->num_bonds))return -EFAULT ;
Signature · 1 line
2.28.0
PATCH 23/51 drm/i915/guc: Direct all breadcrumbs for a class to single breadcrumbs
In reply to Matthew BrostWith GuC virtual engines the physical engine which a request executes and completes on isn't known to the i915. Therefore we can't attach a request to a physical engines breadcrumbs. To work around this we create a single breadcrumbs per engine class when using GuC submission and direct all physical engine interrupts to this breadcrumbs.
v2:
(John H)
- Rework header file structure so intel_engine_mask_t can be in intel_engine_types.h
--- drivers/gpu/drm/i915/gt/intel_breadcrumbs.c | 41 +++++------- drivers/gpu/drm/i915/gt/intel_breadcrumbs.h | 16 ++++- .../gpu/drm/i915/gt/intel_breadcrumbs_types.h | 7 ++ drivers/gpu/drm/i915/gt/intel_engine.h | 3 + drivers/gpu/drm/i915/gt/intel_engine_cs.c | 28 +++++++- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 +- .../drm/i915/gt/intel_execlists_submission.c | 2 +- drivers/gpu/drm/i915/gt/mock_engine.c | 4 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 67 +++++++++++++++++-- 9 files changed, 133 insertions(+), 37 deletions(-)
drivers/gpu/drm/i915/gt/intel_breadcrumbs.c +17 -24
@@ -15,28 +15,14 @@#include "intel_gt_pm.h" #include "intel_gt_requests.h" -static bool irq_enable (struct intel_engine_cs *engine) +static bool irq_enable (struct intel_breadcrumbs *b) { -if (!engine->irq_enable) -return false; - -/* Caller disables interrupts */ -spin_lock (&engine->gt->irq_lock); - engine->irq_enable (engine); -spin_unlock (&engine->gt->irq_lock); - -return true; +return intel_engine_irq_enable (b->irq_engine); } -static void irq_disable (struct intel_engine_cs *engine) +static void irq_disable (struct intel_breadcrumbs *b) { -if (!engine->irq_disable) -return ; - -/* Caller disables interrupts */ -spin_lock (&engine->gt->irq_lock); - engine->irq_disable (engine); -spin_unlock (&engine->gt->irq_lock); +intel_engine_irq_disable (b->irq_engine); }static void __intel_breadcrumbs_arm_irq (struct intel_breadcrumbs *b)
@@ -57,7 +43,7 @@ static void __intel_breadcrumbs_arm_irq(struct intel_breadcrumbs *b)WRITE_ONCE (b->irq_armed, true);/* Requests may have completed before we could enable the interrupt. */ -if (!b->irq_enabled++ &&irq_enable (b->irq_engine)) +if (!b->irq_enabled++ && b->irq_enable (b))irq_work_queue (&b->irq_work); }
@@ -76,7 +62,7 @@ static void __intel_breadcrumbs_disarm_irq(struct intel_breadcrumbs *b) {GEM_BUG_ON (!b->irq_enabled);if (!--b->irq_enabled) -irq_disable (b->irq_engine); + b->irq_disable (b);WRITE_ONCE (b->irq_armed, false);intel_gt_pm_put_async (b->irq_engine->gt);
@@ -281,7 +267,7 @@ intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)if (!b)return NULL ; - b->irq_engine = irq_engine; +kref_init (&b->ref);spin_lock_init (&b->signalers_lock);INIT_LIST_HEAD (&b->signalers);
@@ -290,6 +276,10 @@ intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)spin_lock_init (&b->irq_lock);init_irq_work (&b->irq_work, signal_irq_work); + b->irq_engine = irq_engine; + b->irq_enable = irq_enable; + b->irq_disable = irq_disable; +return b; }
@@ -303,9 +293,9 @@ void intel_breadcrumbs_reset(struct intel_breadcrumbs *b)spin_lock_irqsave (&b->irq_lock, flags);if (b->irq_enabled) -irq_enable (b->irq_engine); + b->irq_enable (b);else -irq_disable (b->irq_engine); + b->irq_disable (b);spin_unlock_irqrestore (&b->irq_lock, flags); }
@@ -325,11 +315,14 @@ void __intel_breadcrumbs_park(struct intel_breadcrumbs *b) } } -void intel_breadcrumbs_free (struct intel_breadcrumbs *b) +void intel_breadcrumbs_free (struct kref *kref) { +struct intel_breadcrumbs *b =container_of (kref,typeof (*b), ref); +irq_work_sync (&b->irq_work);GEM_BUG_ON (!list_empty (&b->signalers));GEM_BUG_ON (b->irq_armed); +kfree (b); }
drivers/gpu/drm/i915/gt/intel_breadcrumbs.h +14 -2
@@ -9,7 +9,7 @@#include <linux/atomic.h> #include <linux/irq_work.h> -#include "intel_engine_types.h" +#include "intel_breadcrumbs_types.h" struct drm_printer ;struct i915_request ;
@@ -17,7 +17,7 @@ struct intel_breadcrumbs;struct intel_breadcrumbs *intel_breadcrumbs_create (struct intel_engine_cs *irq_engine); -void intel_breadcrumbs_free (struct intel_breadcrumbs *b); +void intel_breadcrumbs_free (struct kref *kref);void intel_breadcrumbs_reset (struct intel_breadcrumbs *b);void __intel_breadcrumbs_park (struct intel_breadcrumbs *b);
@@ -48,4 +48,16 @@ void i915_request_cancel_breadcrumb(struct i915_request *request);void intel_context_remove_breadcrumbs (struct intel_context *ce,struct intel_breadcrumbs *b); +static inline struct intel_breadcrumbs * +intel_breadcrumbs_get (struct intel_breadcrumbs *b) +{ +kref_get (&b->ref); +return b; +} + +static inline void intel_breadcrumbs_put (struct intel_breadcrumbs *b) +{ +kref_put (&b->ref, intel_breadcrumbs_free); +} +#endif /* __INTEL_BREADCRUMBS__ */
drivers/gpu/drm/i915/gt/intel_breadcrumbs_types.h +7 -0
@@ -7,10 +7,13 @@#define __INTEL_BREADCRUMBS_TYPES__#include <linux/irq_work.h> +#include <linux/kref.h> #include <linux/list.h> #include <linux/spinlock.h> #include <linux/types.h> +#include "intel_engine_types.h" +/* * Rather than have every client wait upon all user interrupts, * with the herd waking after every interrupt and each doing the
@@ -29,6 +32,7 @@* the overhead of waking that client is much preferred. */ struct intel_breadcrumbs { +struct kref ref;atomic_t active;spinlock_t signalers_lock;/* protects the list of signalers */
@@ -42,7 +46,10 @@ struct intel_breadcrumbs {bool irq_armed;/* Not all breadcrumbs are attached to physical HW */ +intel_engine_mask_t engine_mask;struct intel_engine_cs *irq_engine; +bool (*irq_enable)(struct intel_breadcrumbs *b); + void (*irq_disable)(struct intel_breadcrumbs *b); };#endif /* __INTEL_BREADCRUMBS_TYPES__ */
drivers/gpu/drm/i915/gt/intel_engine.h +3 -0
@@ -212,6 +212,9 @@ void intel_engine_get_instdone(const struct intel_engine_cs *engine,void intel_engine_init_execlists (struct intel_engine_cs *engine); +bool intel_engine_irq_enable (struct intel_engine_cs *engine); +void intel_engine_irq_disable (struct intel_engine_cs *engine); +static inline void __intel_engine_reset (struct intel_engine_cs *engine,bool stalled) {
drivers/gpu/drm/i915/gt/intel_engine_cs.c +26 -2
@@ -739,7 +739,7 @@ static int engine_setup_common(struct intel_engine_cs *engine) err_cmd_parser:i915_sched_engine_put (engine->sched_engine); err_sched_engine: -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs); err_status:cleanup_status_page (engine);return err;
@@ -948,7 +948,7 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine)GEM_BUG_ON (!list_empty (&engine->sched_engine->requests));i915_sched_engine_put (engine->sched_engine); -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs);intel_engine_fini_retire (engine);intel_engine_cleanup_cmd_parser (engine);
@@ -1265,6 +1265,30 @@ bool intel_engines_are_idle(struct intel_gt *gt)return true; } +bool intel_engine_irq_enable (struct intel_engine_cs *engine) +{ +if (!engine->irq_enable) +return false; + +/* Caller disables interrupts */ +spin_lock (&engine->gt->irq_lock); + engine->irq_enable (engine); +spin_unlock (&engine->gt->irq_lock); + +return true; +} + +void intel_engine_irq_disable (struct intel_engine_cs *engine) +{ +if (!engine->irq_disable) +return ; + +/* Caller disables interrupts */ +spin_lock (&engine->gt->irq_lock); + engine->irq_disable (engine); +spin_unlock (&engine->gt->irq_lock); +} +void intel_engines_reset_default_submission (struct intel_gt *gt) {struct intel_engine_cs *engine;
drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -1
@@ -21,7 +21,6 @@#include "i915_pmu.h" #include "i915_priolist_types.h" #include "i915_selftest.h" -#include "intel_breadcrumbs_types.h" #include "intel_sseu.h" #include "intel_timeline_types.h" #include "intel_uncore.h"
@@ -63,6 +62,7 @@ struct i915_sched_engine;struct intel_gt ;struct intel_ring ;struct intel_uncore ; +struct intel_breadcrumbs ;typedef u8 intel_engine_mask_t ;#define ALL_ENGINES ((intel_engine_mask_t)~0ul)
drivers/gpu/drm/i915/gt/intel_execlists_submission.c +1 -1
@@ -3407,7 +3407,7 @@ static void rcu_virtual_context_destroy(struct work_struct *wrk)intel_context_fini (&ve->context);if (ve->base.breadcrumbs) -intel_breadcrumbs_free (ve->base.breadcrumbs); +intel_breadcrumbs_put (ve->base.breadcrumbs);if (ve->base.sched_engine)i915_sched_engine_put (ve->base.sched_engine);intel_engine_free_request_pool (&ve->base);
drivers/gpu/drm/i915/gt/mock_engine.c +2 -2
@@ -284,7 +284,7 @@ static void mock_engine_release(struct intel_engine_cs *engine)GEM_BUG_ON (timer_pending (&mock->hw_delay));i915_sched_engine_put (engine->sched_engine); -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs);intel_context_unpin (engine->kernel_context);intel_context_put (engine->kernel_context);
@@ -376,7 +376,7 @@ int mock_engine_init(struct intel_engine_cs *engine)return 0 ; err_breadcrumbs: -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs); err_schedule:i915_sched_engine_put (engine->sched_engine);return -ENOMEM ;
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +62 -5
97 lines not on this page · raw message
Signature · 1 line
2.28.0
Re: [PATCH 23/51] drm/i915/guc: Direct all breadcrumbs for a class to single breadcrumbs
Quoted 6 lines · Matthew Brost
With GuC virtual engines the physical engine which a request executes and completes on isn't known to the i915. Therefore we can't attach a request to a physical engines breadcrumbs. To work around this we create a single breadcrumbs per engine class when using GuC submission and direct all physical engine interrupts to this breadcrumbs.
Quoted 7 lines
v2:
(John H)
- Rework header file structure sointel_engine_mask_tcan be inintel_engine_types.hSigned-off-by Matthew Brost <[email protected]>CC John Harrison <[email protected]>
Quoted 419 lines
--- drivers/gpu/drm/i915/gt/intel_breadcrumbs.c | 41 +++++------- drivers/gpu/drm/i915/gt/intel_breadcrumbs.h | 16 ++++- .../gpu/drm/i915/gt/intel_breadcrumbs_types.h | 7 ++ drivers/gpu/drm/i915/gt/intel_engine.h | 3 + drivers/gpu/drm/i915/gt/intel_engine_cs.c | 28 +++++++- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 +- .../drm/i915/gt/intel_execlists_submission.c | 2 +- drivers/gpu/drm/i915/gt/mock_engine.c | 4 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 67 +++++++++++++++++-- 9 files changed, 133 insertions(+), 37 deletions(-)drivers/gpu/drm/i915/gt/intel_breadcrumbs.c +17
-24@@ -15,28 +15,14 @@#include "intel_gt_pm.h" #include "intel_gt_requests.h" -static bool irq_enable (struct intel_engine_cs *engine) +static bool irq_enable (struct intel_breadcrumbs *b) { -if (!engine->irq_enable) -return false; - -/* Caller disables interrupts */ -spin_lock (&engine->gt->irq_lock); - engine->irq_enable (engine); -spin_unlock (&engine->gt->irq_lock); - -return true; +return intel_engine_irq_enable (b->irq_engine); } -static void irq_disable (struct intel_engine_cs *engine) +static void irq_disable (struct intel_breadcrumbs *b) { -if (!engine->irq_disable) -return ; - -/* Caller disables interrupts */ -spin_lock (&engine->gt->irq_lock); - engine->irq_disable (engine); -spin_unlock (&engine->gt->irq_lock); +intel_engine_irq_disable (b->irq_engine); }static void __intel_breadcrumbs_arm_irq (struct intel_breadcrumbs *b)@@ -57,7 +43,7 @@ static void __intel_breadcrumbs_arm_irq(struct intel_breadcrumbs *b)WRITE_ONCE (b->irq_armed, true);/* Requests may have completed before we could enable the interrupt. */ -if (!b->irq_enabled++ &&irq_enable (b->irq_engine)) +if (!b->irq_enabled++ && b->irq_enable (b))irq_work_queue (&b->irq_work); }@@ -76,7 +62,7 @@ static void __intel_breadcrumbs_disarm_irq(struct intel_breadcrumbs *b) {GEM_BUG_ON (!b->irq_enabled);if (!--b->irq_enabled) -irq_disable (b->irq_engine); + b->irq_disable (b);WRITE_ONCE (b->irq_armed, false);intel_gt_pm_put_async (b->irq_engine->gt);@@ -281,7 +267,7 @@ intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)if (!b)return NULL ; - b->irq_engine = irq_engine; +kref_init (&b->ref);spin_lock_init (&b->signalers_lock);INIT_LIST_HEAD (&b->signalers);@@ -290,6 +276,10 @@ intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)spin_lock_init (&b->irq_lock);init_irq_work (&b->irq_work, signal_irq_work); + b->irq_engine = irq_engine; + b->irq_enable = irq_enable; + b->irq_disable = irq_disable; +return b; }@@ -303,9 +293,9 @@ void intel_breadcrumbs_reset(struct intel_breadcrumbs *b)spin_lock_irqsave (&b->irq_lock, flags);if (b->irq_enabled) -irq_enable (b->irq_engine); + b->irq_enable (b);else -irq_disable (b->irq_engine); + b->irq_disable (b);spin_unlock_irqrestore (&b->irq_lock, flags); }@@ -325,11 +315,14 @@ void __intel_breadcrumbs_park(struct intel_breadcrumbs *b) } } -void intel_breadcrumbs_free (struct intel_breadcrumbs *b) +void intel_breadcrumbs_free (struct kref *kref) { +struct intel_breadcrumbs *b =container_of (kref,typeof (*b), ref); +irq_work_sync (&b->irq_work);GEM_BUG_ON (!list_empty (&b->signalers));GEM_BUG_ON (b->irq_armed); +kfree (b); }drivers/gpu/drm/i915/gt/intel_breadcrumbs.h +14
-2@@ -9,7 +9,7 @@#include <linux/atomic.h> #include <linux/irq_work.h> -#include "intel_engine_types.h" +#include "intel_breadcrumbs_types.h" struct drm_printer ;struct i915_request ;@@ -17,7 +17,7 @@ struct intel_breadcrumbs;struct intel_breadcrumbs *intel_breadcrumbs_create (struct intel_engine_cs *irq_engine); -void intel_breadcrumbs_free (struct intel_breadcrumbs *b); +void intel_breadcrumbs_free (struct kref *kref);void intel_breadcrumbs_reset (struct intel_breadcrumbs *b);void __intel_breadcrumbs_park (struct intel_breadcrumbs *b);@@ -48,4 +48,16 @@ void i915_request_cancel_breadcrumb(struct i915_request *request);void intel_context_remove_breadcrumbs (struct intel_context *ce,struct intel_breadcrumbs *b); +static inline struct intel_breadcrumbs * +intel_breadcrumbs_get (struct intel_breadcrumbs *b) +{ +kref_get (&b->ref); +return b; +} + +static inline void intel_breadcrumbs_put (struct intel_breadcrumbs *b) +{ +kref_put (&b->ref, intel_breadcrumbs_free); +} +#endif /* __INTEL_BREADCRUMBS__ */ drivers/gpu/drm/i915/gt/intel_breadcrumbs_types.h +7
-0@@ -7,10 +7,13 @@#define __INTEL_BREADCRUMBS_TYPES__#include <linux/irq_work.h> +#include <linux/kref.h> #include <linux/list.h> #include <linux/spinlock.h> #include <linux/types.h> +#include "intel_engine_types.h" +/* * Rather than have every client wait upon all user interrupts, * with the herd waking after every interrupt and each doing the @@ -29,6 +32,7 @@* the overhead of waking that client is much preferred. */ struct intel_breadcrumbs { +struct kref ref;atomic_t active;spinlock_t signalers_lock;/* protects the list of signalers */ @@ -42,7 +46,10 @@ struct intel_breadcrumbs {bool irq_armed;/* Not all breadcrumbs are attached to physical HW */ +intel_engine_mask_t engine_mask;struct intel_engine_cs *irq_engine; +bool (*irq_enable)(struct intel_breadcrumbs *b); + void (*irq_disable)(struct intel_breadcrumbs *b); };#endif /* __INTEL_BREADCRUMBS_TYPES__ */ drivers/gpu/drm/i915/gt/intel_engine.h +3
-0@@ -212,6 +212,9 @@ void intel_engine_get_instdone(const struct intel_engine_cs *engine,void intel_engine_init_execlists (struct intel_engine_cs *engine); +bool intel_engine_irq_enable (struct intel_engine_cs *engine); +void intel_engine_irq_disable (struct intel_engine_cs *engine); +static inline void __intel_engine_reset (struct intel_engine_cs *engine,bool stalled) {drivers/gpu/drm/i915/gt/intel_engine_cs.c +26
-2@@ -739,7 +739,7 @@ static int engine_setup_common(struct intel_engine_cs *engine) err_cmd_parser:i915_sched_engine_put (engine->sched_engine); err_sched_engine: -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs); err_status:cleanup_status_page (engine);return err;@@ -948,7 +948,7 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine)GEM_BUG_ON (!list_empty (&engine->sched_engine->requests));i915_sched_engine_put (engine->sched_engine); -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs);intel_engine_fini_retire (engine);intel_engine_cleanup_cmd_parser (engine);@@ -1265,6 +1265,30 @@ bool intel_engines_are_idle(struct intel_gt *gt)return true; } +bool intel_engine_irq_enable (struct intel_engine_cs *engine) +{ +if (!engine->irq_enable) +return false; + +/* Caller disables interrupts */ +spin_lock (&engine->gt->irq_lock); + engine->irq_enable (engine); +spin_unlock (&engine->gt->irq_lock); + +return true; +} + +void intel_engine_irq_disable (struct intel_engine_cs *engine) +{ +if (!engine->irq_disable) +return ; + +/* Caller disables interrupts */ +spin_lock (&engine->gt->irq_lock); + engine->irq_disable (engine); +spin_unlock (&engine->gt->irq_lock); +} +void intel_engines_reset_default_submission (struct intel_gt *gt) {struct intel_engine_cs *engine;drivers/gpu/drm/i915/gt/intel_engine_types.h +1
-1@@ -21,7 +21,6 @@#include "i915_pmu.h" #include "i915_priolist_types.h" #include "i915_selftest.h" -#include "intel_breadcrumbs_types.h" #include "intel_sseu.h" #include "intel_timeline_types.h" #include "intel_uncore.h" @@ -63,6 +62,7 @@ struct i915_sched_engine;struct intel_gt ;struct intel_ring ;struct intel_uncore ; +struct intel_breadcrumbs ;typedef u8 intel_engine_mask_t ;#define ALL_ENGINES ((intel_engine_mask_t)~0ul)drivers/gpu/drm/i915/gt/intel_execlists_submission.c +1
-1@@ -3407,7 +3407,7 @@ static void rcu_virtual_context_destroy(struct work_struct *wrk)intel_context_fini (&ve->context);if (ve->base.breadcrumbs) -intel_breadcrumbs_free (ve->base.breadcrumbs); +intel_breadcrumbs_put (ve->base.breadcrumbs);if (ve->base.sched_engine)i915_sched_engine_put (ve->base.sched_engine);intel_engine_free_request_pool (&ve->base);drivers/gpu/drm/i915/gt/mock_engine.c +2
-2@@ -284,7 +284,7 @@ static void mock_engine_release(struct intel_engine_cs *engine)GEM_BUG_ON (timer_pending (&mock->hw_delay));i915_sched_engine_put (engine->sched_engine); -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs);intel_context_unpin (engine->kernel_context);intel_context_put (engine->kernel_context);@@ -376,7 +376,7 @@ int mock_engine_init(struct intel_engine_cs *engine)return 0 ; err_breadcrumbs: -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs); err_schedule:i915_sched_engine_put (engine->sched_engine);return -ENOMEM ;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +62
-5@@ -1076,6 +1076,9 @@ static void __guc_context_destroy(struct intel_context *ce)struct guc_virtual_engine *ve =container_of (ce,typeof (*ve), context); +if (ve->base.breadcrumbs) +intel_breadcrumbs_put (ve->base.breadcrumbs); +kfree (ve); }else {intel_context_free (ce);@@ -1366,6 +1369,62 @@ static const struct intel_context_ops virtual_guc_context_ops = { .get_sibling = guc_virtual_get_sibling, }; +static bool +guc_irq_enable_breadcrumbs (struct intel_breadcrumbs *b) +{ +struct intel_engine_cs *sibling; +intel_engine_mask_t tmp, mask = b->engine_mask; +bool result = false; + +for_each_engine_masked (sibling, b->irq_engine->gt, mask, tmp) + result |=intel_engine_irq_enable (sibling); + +return result; +} + +static void +guc_irq_disable_breadcrumbs (struct intel_breadcrumbs *b) +{ +struct intel_engine_cs *sibling; +intel_engine_mask_t tmp, mask = b->engine_mask; + +for_each_engine_masked (sibling, b->irq_engine->gt, mask, tmp) +intel_engine_irq_disable (sibling); +} + +static void guc_init_breadcrumbs (struct intel_engine_cs *engine) +{ +int i; + +/* +* In GuC submission mode we do not know which physical engine a request +* will be scheduled on, this creates a problem because the breadcrumb +* interrupt is per physical engine. To work around this we attach +* requests and direct all breadcrumb interrupts to the first instance +* of an engine per class. In addition all breadcrumb interrupts are +* enabled / disabled across an engine class in unison. +*/ +for (i =0 ; i <MAX_ENGINE_INSTANCE ; ++i) { +struct intel_engine_cs *sibling = + engine->gt->engine_class[engine->class][i]; + +if (sibling) { +if (engine->breadcrumbs != sibling->breadcrumbs) { +intel_breadcrumbs_put (engine->breadcrumbs); + engine->breadcrumbs = +intel_breadcrumbs_get (sibling->breadcrumbs); + } +break ; + } + } + +if (engine->breadcrumbs) { + engine->breadcrumbs->engine_mask |= engine->mask; + engine->breadcrumbs->irq_enable = guc_irq_enable_breadcrumbs; + engine->breadcrumbs->irq_disable = guc_irq_disable_breadcrumbs; + } +} +static void sanitize_hwsp (struct intel_engine_cs *engine) {struct intel_timeline *tl;@@ -1589,6 +1648,7 @@ int intel_guc_submission_setup(struct intel_engine_cs *engine)guc_default_vfuncs (engine);guc_default_irqs (engine); +guc_init_breadcrumbs (engine);if (engine->class ==RENDER_CLASS )rcs_submission_override (engine);@@ -1831,11 +1891,6 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) ve->base.instance =I915_ENGINE_CLASS_INVALID_VIRTUAL ; ve->base.uabi_instance =I915_ENGINE_CLASS_INVALID_VIRTUAL ; ve->base.saturated =ALL_ENGINES ; - ve->base.breadcrumbs =intel_breadcrumbs_create (&ve->base); -if (!ve->base.breadcrumbs) { -kfree (ve); -return ERR_PTR (-ENOMEM ); - }snprintf (ve->base.name,sizeof (ve->base.name),"virtual" );@@ -1884,6 +1939,8 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) sibling->emit_fini_breadcrumb; ve->base.emit_fini_breadcrumb_dw = sibling->emit_fini_breadcrumb_dw; + ve->base.breadcrumbs = +intel_breadcrumbs_get (sibling->breadcrumbs); ve->base.flags |= sibling->flags;
Re: [Intel-gfx] [PATCH 23/51] drm/i915/guc: Direct all breadcrumbs for a class to single breadcrumbs
In reply to Matthew BrostQuoted 376 lines · Matthew Brost
With GuC virtual engines the physical engine which a request executes and completes on isn't known to the i915. Therefore we can't attach a request to a physical engines breadcrumbs. To work around this we create a single breadcrumbs per engine class when using GuC submission and direct all physical engine interrupts to this breadcrumbs.
v2:
(John H)
- Rework header file structure sointel_engine_mask_tcan be inintel_engine_types.hSigned-off-by Matthew Brost <[email protected]>CC John Harrison <[email protected]>--- drivers/gpu/drm/i915/gt/intel_breadcrumbs.c | 41 +++++------- drivers/gpu/drm/i915/gt/intel_breadcrumbs.h | 16 ++++- .../gpu/drm/i915/gt/intel_breadcrumbs_types.h | 7 ++ drivers/gpu/drm/i915/gt/intel_engine.h | 3 + drivers/gpu/drm/i915/gt/intel_engine_cs.c | 28 +++++++- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 +- .../drm/i915/gt/intel_execlists_submission.c | 2 +- drivers/gpu/drm/i915/gt/mock_engine.c | 4 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 67 +++++++++++++++++-- 9 files changed, 133 insertions(+), 37 deletions(-)drivers/gpu/drm/i915/gt/intel_breadcrumbs.c +17
-24@@ -15,28 +15,14 @@#include "intel_gt_pm.h" #include "intel_gt_requests.h" -static bool irq_enable (struct intel_engine_cs *engine) +static bool irq_enable (struct intel_breadcrumbs *b) { -if (!engine->irq_enable) -return false; - -/* Caller disables interrupts */ -spin_lock (&engine->gt->irq_lock); - engine->irq_enable (engine); -spin_unlock (&engine->gt->irq_lock); - -return true; +return intel_engine_irq_enable (b->irq_engine); } -static void irq_disable (struct intel_engine_cs *engine) +static void irq_disable (struct intel_breadcrumbs *b) { -if (!engine->irq_disable) -return ; - -/* Caller disables interrupts */ -spin_lock (&engine->gt->irq_lock); - engine->irq_disable (engine); -spin_unlock (&engine->gt->irq_lock); +intel_engine_irq_disable (b->irq_engine); }static void __intel_breadcrumbs_arm_irq (struct intel_breadcrumbs *b)@@ -57,7 +43,7 @@ static void __intel_breadcrumbs_arm_irq(struct intel_breadcrumbs *b)WRITE_ONCE (b->irq_armed, true);/* Requests may have completed before we could enable the interrupt. */ -if (!b->irq_enabled++ &&irq_enable (b->irq_engine)) +if (!b->irq_enabled++ && b->irq_enable (b))irq_work_queue (&b->irq_work); }@@ -76,7 +62,7 @@ static void __intel_breadcrumbs_disarm_irq(struct intel_breadcrumbs *b) {GEM_BUG_ON (!b->irq_enabled);if (!--b->irq_enabled) -irq_disable (b->irq_engine); + b->irq_disable (b);WRITE_ONCE (b->irq_armed, false);intel_gt_pm_put_async (b->irq_engine->gt);@@ -281,7 +267,7 @@ intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)if (!b)return NULL ; - b->irq_engine = irq_engine; +kref_init (&b->ref);spin_lock_init (&b->signalers_lock);INIT_LIST_HEAD (&b->signalers);@@ -290,6 +276,10 @@ intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)spin_lock_init (&b->irq_lock);init_irq_work (&b->irq_work, signal_irq_work); + b->irq_engine = irq_engine; + b->irq_enable = irq_enable; + b->irq_disable = irq_disable; +return b; }@@ -303,9 +293,9 @@ void intel_breadcrumbs_reset(struct intel_breadcrumbs *b)spin_lock_irqsave (&b->irq_lock, flags);if (b->irq_enabled) -irq_enable (b->irq_engine); + b->irq_enable (b);else -irq_disable (b->irq_engine); + b->irq_disable (b);spin_unlock_irqrestore (&b->irq_lock, flags); }@@ -325,11 +315,14 @@ void __intel_breadcrumbs_park(struct intel_breadcrumbs *b) } } -void intel_breadcrumbs_free (struct intel_breadcrumbs *b) +void intel_breadcrumbs_free (struct kref *kref) { +struct intel_breadcrumbs *b =container_of (kref,typeof (*b), ref); +irq_work_sync (&b->irq_work);GEM_BUG_ON (!list_empty (&b->signalers));GEM_BUG_ON (b->irq_armed); +kfree (b); }drivers/gpu/drm/i915/gt/intel_breadcrumbs.h +14
-2@@ -9,7 +9,7 @@#include <linux/atomic.h> #include <linux/irq_work.h> -#include "intel_engine_types.h" +#include "intel_breadcrumbs_types.h" struct drm_printer ;struct i915_request ;@@ -17,7 +17,7 @@ struct intel_breadcrumbs;struct intel_breadcrumbs *intel_breadcrumbs_create (struct intel_engine_cs *irq_engine); -void intel_breadcrumbs_free (struct intel_breadcrumbs *b); +void intel_breadcrumbs_free (struct kref *kref);void intel_breadcrumbs_reset (struct intel_breadcrumbs *b);void __intel_breadcrumbs_park (struct intel_breadcrumbs *b);@@ -48,4 +48,16 @@ void i915_request_cancel_breadcrumb(struct i915_request *request);void intel_context_remove_breadcrumbs (struct intel_context *ce,struct intel_breadcrumbs *b); +static inline struct intel_breadcrumbs * +intel_breadcrumbs_get (struct intel_breadcrumbs *b) +{ +kref_get (&b->ref); +return b; +} + +static inline void intel_breadcrumbs_put (struct intel_breadcrumbs *b) +{ +kref_put (&b->ref, intel_breadcrumbs_free); +} +#endif /* __INTEL_BREADCRUMBS__ */ drivers/gpu/drm/i915/gt/intel_breadcrumbs_types.h +7
-0@@ -7,10 +7,13 @@#define __INTEL_BREADCRUMBS_TYPES__#include <linux/irq_work.h> +#include <linux/kref.h> #include <linux/list.h> #include <linux/spinlock.h> #include <linux/types.h> +#include "intel_engine_types.h" +/* * Rather than have every client wait upon all user interrupts, * with the herd waking after every interrupt and each doing the @@ -29,6 +32,7 @@* the overhead of waking that client is much preferred. */ struct intel_breadcrumbs { +struct kref ref;atomic_t active;spinlock_t signalers_lock;/* protects the list of signalers */ @@ -42,7 +46,10 @@ struct intel_breadcrumbs {bool irq_armed;/* Not all breadcrumbs are attached to physical HW */ +intel_engine_mask_t engine_mask;struct intel_engine_cs *irq_engine; +bool (*irq_enable)(struct intel_breadcrumbs *b); + void (*irq_disable)(struct intel_breadcrumbs *b); };#endif /* __INTEL_BREADCRUMBS_TYPES__ */ drivers/gpu/drm/i915/gt/intel_engine.h +3
-0@@ -212,6 +212,9 @@ void intel_engine_get_instdone(const struct intel_engine_cs *engine,void intel_engine_init_execlists (struct intel_engine_cs *engine); +bool intel_engine_irq_enable (struct intel_engine_cs *engine); +void intel_engine_irq_disable (struct intel_engine_cs *engine); +static inline void __intel_engine_reset (struct intel_engine_cs *engine,bool stalled) {drivers/gpu/drm/i915/gt/intel_engine_cs.c +26
-2@@ -739,7 +739,7 @@ static int engine_setup_common(struct intel_engine_cs *engine) err_cmd_parser:i915_sched_engine_put (engine->sched_engine); err_sched_engine: -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs); err_status:cleanup_status_page (engine);return err;@@ -948,7 +948,7 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine)GEM_BUG_ON (!list_empty (&engine->sched_engine->requests));i915_sched_engine_put (engine->sched_engine); -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs);intel_engine_fini_retire (engine);intel_engine_cleanup_cmd_parser (engine);@@ -1265,6 +1265,30 @@ bool intel_engines_are_idle(struct intel_gt *gt)return true; } +bool intel_engine_irq_enable (struct intel_engine_cs *engine) +{ +if (!engine->irq_enable) +return false; + +/* Caller disables interrupts */ +spin_lock (&engine->gt->irq_lock); + engine->irq_enable (engine); +spin_unlock (&engine->gt->irq_lock); + +return true; +} + +void intel_engine_irq_disable (struct intel_engine_cs *engine) +{ +if (!engine->irq_disable) +return ; + +/* Caller disables interrupts */ +spin_lock (&engine->gt->irq_lock); + engine->irq_disable (engine); +spin_unlock (&engine->gt->irq_lock); +} +void intel_engines_reset_default_submission (struct intel_gt *gt) {struct intel_engine_cs *engine;drivers/gpu/drm/i915/gt/intel_engine_types.h +1
-1@@ -21,7 +21,6 @@#include "i915_pmu.h" #include "i915_priolist_types.h" #include "i915_selftest.h" -#include "intel_breadcrumbs_types.h" #include "intel_sseu.h" #include "intel_timeline_types.h" #include "intel_uncore.h" @@ -63,6 +62,7 @@ struct i915_sched_engine;struct intel_gt ;struct intel_ring ;struct intel_uncore ; +struct intel_breadcrumbs ;typedef u8 intel_engine_mask_t ;#define ALL_ENGINES ((intel_engine_mask_t)~0ul)drivers/gpu/drm/i915/gt/intel_execlists_submission.c +1
-1@@ -3407,7 +3407,7 @@ static void rcu_virtual_context_destroy(struct work_struct *wrk)intel_context_fini (&ve->context);if (ve->base.breadcrumbs) -intel_breadcrumbs_free (ve->base.breadcrumbs); +intel_breadcrumbs_put (ve->base.breadcrumbs);if (ve->base.sched_engine)i915_sched_engine_put (ve->base.sched_engine);intel_engine_free_request_pool (&ve->base);drivers/gpu/drm/i915/gt/mock_engine.c +2
-2@@ -284,7 +284,7 @@ static void mock_engine_release(struct intel_engine_cs *engine)GEM_BUG_ON (timer_pending (&mock->hw_delay));i915_sched_engine_put (engine->sched_engine); -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs);intel_context_unpin (engine->kernel_context);intel_context_put (engine->kernel_context);@@ -376,7 +376,7 @@ int mock_engine_init(struct intel_engine_cs *engine)return 0 ; err_breadcrumbs: -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs); err_schedule:i915_sched_engine_put (engine->sched_engine);return -ENOMEM ;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +35
-0@@ -1076,6 +1076,9 @@ static void __guc_context_destroy(struct intel_context *ce)struct guc_virtual_engine *ve =container_of (ce,typeof (*ve), context); +if (ve->base.breadcrumbs) +intel_breadcrumbs_put (ve->base.breadcrumbs); +kfree (ve); }else {intel_context_free (ce);@@ -1366,6 +1369,62 @@ static const struct intel_context_ops virtual_guc_context_ops = { .get_sibling = guc_virtual_get_sibling, }; +static bool +guc_irq_enable_breadcrumbs (struct intel_breadcrumbs *b) +{ +struct intel_engine_cs *sibling; +intel_engine_mask_t tmp, mask = b->engine_mask; +bool result = false; + +for_each_engine_masked (sibling, b->irq_engine->gt, mask, tmp) + result |=intel_engine_irq_enable (sibling); + +return result; +} + +static void +guc_irq_disable_breadcrumbs (struct intel_breadcrumbs *b) +{ +struct intel_engine_cs *sibling; +intel_engine_mask_t tmp, mask = b->engine_mask; + +for_each_engine_masked (sibling, b->irq_engine->gt, mask, tmp) +intel_engine_irq_disable (sibling); +} + +static void guc_init_breadcrumbs (struct intel_engine_cs *engine) +{ +int i; + +/* +* In GuC submission mode we do not know which physical engine a request +* will be scheduled on, this creates a problem because the breadcrumb +* interrupt is per physical engine. To work around this we attach +* requests and direct all breadcrumb interrupts to the first instance
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static const struct intel_context_ops virtual_guc_context_ops = { · line 1413
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +10
-0+* of an engine per class. In addition all breadcrumb interrupts are +* enabled / disabled across an engine class in unison. +*/ +for (i =0 ; i <MAX_ENGINE_INSTANCE ; ++i) { +struct intel_engine_cs *sibling = + engine->gt->engine_class[engine->class][i]; + +if (sibling) { +if (engine->breadcrumbs != sibling->breadcrumbs) { +intel_breadcrumbs_put (engine->breadcrumbs);
So it's still only me that is bothered by the lazyness of the proposed init path?
intel_engines_init
for_each_engine
engine_setup_common
engine->breadcrumbs = intel_breadcrumbs_create(engine);
intel_guc_submission_setup
intel_breadcrumbs_put(engine->breadcrumbs);
...In other words leaves the code to allocate N objects in common and then it frees all but one of them, overriding vfuncs in the remaining instance.
Why not pull out the setup out of common, when it is obviously not common any more?
Regards,
Tvrtko
Quoted 47 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +17
-5+ engine->breadcrumbs = +intel_breadcrumbs_get (sibling->breadcrumbs); + } +break ; + } + } + +if (engine->breadcrumbs) { + engine->breadcrumbs->engine_mask |= engine->mask; + engine->breadcrumbs->irq_enable = guc_irq_enable_breadcrumbs; + engine->breadcrumbs->irq_disable = guc_irq_disable_breadcrumbs; + } +} +static void sanitize_hwsp (struct intel_engine_cs *engine) {struct intel_timeline *tl;@@ -1589,6 +1648,7 @@ int intel_guc_submission_setup(struct intel_engine_cs *engine)guc_default_vfuncs (engine);guc_default_irqs (engine); +guc_init_breadcrumbs (engine);if (engine->class ==RENDER_CLASS )rcs_submission_override (engine);@@ -1831,11 +1891,6 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) ve->base.instance =I915_ENGINE_CLASS_INVALID_VIRTUAL ; ve->base.uabi_instance =I915_ENGINE_CLASS_INVALID_VIRTUAL ; ve->base.saturated =ALL_ENGINES ; - ve->base.breadcrumbs =intel_breadcrumbs_create (&ve->base); -if (!ve->base.breadcrumbs) { -kfree (ve); -return ERR_PTR (-ENOMEM ); - }snprintf (ve->base.name,sizeof (ve->base.name),"virtual" );@@ -1884,6 +1939,8 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) sibling->emit_fini_breadcrumb; ve->base.emit_fini_breadcrumb_dw = sibling->emit_fini_breadcrumb_dw; + ve->base.breadcrumbs = +intel_breadcrumbs_get (sibling->breadcrumbs); ve->base.flags |= sibling->flags;
Re: [Intel-gfx] [PATCH 23/51] drm/i915/guc: Direct all breadcrumbs for a class to single breadcrumbs
Quoted 347 lines · Tvrtko Ursulin
Quoted 345 lines · Matthew Brost
With GuC virtual engines the physical engine which a request executes and completes on isn't known to the i915. Therefore we can't attach a request to a physical engines breadcrumbs. To work around this we create a single breadcrumbs per engine class when using GuC submission and direct all physical engine interrupts to this breadcrumbs.
v2:
(John H)
- Rework header file structure sointel_engine_mask_tcan be inintel_engine_types.hSigned-off-by Matthew Brost <[email protected]>CC John Harrison <[email protected]>--- drivers/gpu/drm/i915/gt/intel_breadcrumbs.c | 41 +++++------- drivers/gpu/drm/i915/gt/intel_breadcrumbs.h | 16 ++++- .../gpu/drm/i915/gt/intel_breadcrumbs_types.h | 7 ++ drivers/gpu/drm/i915/gt/intel_engine.h | 3 + drivers/gpu/drm/i915/gt/intel_engine_cs.c | 28 +++++++- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 +- .../drm/i915/gt/intel_execlists_submission.c | 2 +- drivers/gpu/drm/i915/gt/mock_engine.c | 4 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 67 +++++++++++++++++-- 9 files changed, 133 insertions(+), 37 deletions(-)drivers/gpu/drm/i915/gt/intel_breadcrumbs.c +17
-24@@ -15,28 +15,14 @@#include "intel_gt_pm.h" #include "intel_gt_requests.h" -static bool irq_enable (struct intel_engine_cs *engine) +static bool irq_enable (struct intel_breadcrumbs *b) { -if (!engine->irq_enable) -return false; - -/* Caller disables interrupts */ -spin_lock (&engine->gt->irq_lock); - engine->irq_enable (engine); -spin_unlock (&engine->gt->irq_lock); - -return true; +return intel_engine_irq_enable (b->irq_engine); } -static void irq_disable (struct intel_engine_cs *engine) +static void irq_disable (struct intel_breadcrumbs *b) { -if (!engine->irq_disable) -return ; - -/* Caller disables interrupts */ -spin_lock (&engine->gt->irq_lock); - engine->irq_disable (engine); -spin_unlock (&engine->gt->irq_lock); +intel_engine_irq_disable (b->irq_engine); }static void __intel_breadcrumbs_arm_irq (struct intel_breadcrumbs *b)@@ -57,7 +43,7 @@ static void __intel_breadcrumbs_arm_irq(struct intel_breadcrumbs *b)WRITE_ONCE (b->irq_armed, true);/* Requests may have completed before we could enable the interrupt. */ -if (!b->irq_enabled++ &&irq_enable (b->irq_engine)) +if (!b->irq_enabled++ && b->irq_enable (b))irq_work_queue (&b->irq_work); }@@ -76,7 +62,7 @@ static void __intel_breadcrumbs_disarm_irq(struct intel_breadcrumbs *b) {GEM_BUG_ON (!b->irq_enabled);if (!--b->irq_enabled) -irq_disable (b->irq_engine); + b->irq_disable (b);WRITE_ONCE (b->irq_armed, false);intel_gt_pm_put_async (b->irq_engine->gt);@@ -281,7 +267,7 @@ intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)if (!b)return NULL ; - b->irq_engine = irq_engine; +kref_init (&b->ref);spin_lock_init (&b->signalers_lock);INIT_LIST_HEAD (&b->signalers);@@ -290,6 +276,10 @@ intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)spin_lock_init (&b->irq_lock);init_irq_work (&b->irq_work, signal_irq_work); + b->irq_engine = irq_engine; + b->irq_enable = irq_enable; + b->irq_disable = irq_disable; +return b; }@@ -303,9 +293,9 @@ void intel_breadcrumbs_reset(struct intel_breadcrumbs *b)spin_lock_irqsave (&b->irq_lock, flags);if (b->irq_enabled) -irq_enable (b->irq_engine); + b->irq_enable (b);else -irq_disable (b->irq_engine); + b->irq_disable (b);spin_unlock_irqrestore (&b->irq_lock, flags); }@@ -325,11 +315,14 @@ void __intel_breadcrumbs_park(struct intel_breadcrumbs *b) } } -void intel_breadcrumbs_free (struct intel_breadcrumbs *b) +void intel_breadcrumbs_free (struct kref *kref) { +struct intel_breadcrumbs *b =container_of (kref,typeof (*b), ref); +irq_work_sync (&b->irq_work);GEM_BUG_ON (!list_empty (&b->signalers));GEM_BUG_ON (b->irq_armed); +kfree (b); }drivers/gpu/drm/i915/gt/intel_breadcrumbs.h +14
-2@@ -9,7 +9,7 @@#include <linux/atomic.h> #include <linux/irq_work.h> -#include "intel_engine_types.h" +#include "intel_breadcrumbs_types.h" struct drm_printer ;struct i915_request ;@@ -17,7 +17,7 @@ struct intel_breadcrumbs;struct intel_breadcrumbs *intel_breadcrumbs_create (struct intel_engine_cs *irq_engine); -void intel_breadcrumbs_free (struct intel_breadcrumbs *b); +void intel_breadcrumbs_free (struct kref *kref);void intel_breadcrumbs_reset (struct intel_breadcrumbs *b);void __intel_breadcrumbs_park (struct intel_breadcrumbs *b);@@ -48,4 +48,16 @@ void i915_request_cancel_breadcrumb(struct i915_request *request);void intel_context_remove_breadcrumbs (struct intel_context *ce,struct intel_breadcrumbs *b); +static inline struct intel_breadcrumbs * +intel_breadcrumbs_get (struct intel_breadcrumbs *b) +{ +kref_get (&b->ref); +return b; +} + +static inline void intel_breadcrumbs_put (struct intel_breadcrumbs *b) +{ +kref_put (&b->ref, intel_breadcrumbs_free); +} +#endif /* __INTEL_BREADCRUMBS__ */ drivers/gpu/drm/i915/gt/intel_breadcrumbs_types.h +7
-0@@ -7,10 +7,13 @@#define __INTEL_BREADCRUMBS_TYPES__#include <linux/irq_work.h> +#include <linux/kref.h> #include <linux/list.h> #include <linux/spinlock.h> #include <linux/types.h> +#include "intel_engine_types.h" +/* * Rather than have every client wait upon all user interrupts, * with the herd waking after every interrupt and each doing the @@ -29,6 +32,7 @@* the overhead of waking that client is much preferred. */ struct intel_breadcrumbs { +struct kref ref;atomic_t active;spinlock_t signalers_lock;/* protects the list of signalers */ @@ -42,7 +46,10 @@ struct intel_breadcrumbs {bool irq_armed;/* Not all breadcrumbs are attached to physical HW */ +intel_engine_mask_t engine_mask;struct intel_engine_cs *irq_engine; +bool (*irq_enable)(struct intel_breadcrumbs *b); + void (*irq_disable)(struct intel_breadcrumbs *b); };#endif /* __INTEL_BREADCRUMBS_TYPES__ */ drivers/gpu/drm/i915/gt/intel_engine.h +3
-0@@ -212,6 +212,9 @@ void intel_engine_get_instdone(const struct intel_engine_cs *engine,void intel_engine_init_execlists (struct intel_engine_cs *engine); +bool intel_engine_irq_enable (struct intel_engine_cs *engine); +void intel_engine_irq_disable (struct intel_engine_cs *engine); +static inline void __intel_engine_reset (struct intel_engine_cs *engine,bool stalled) {drivers/gpu/drm/i915/gt/intel_engine_cs.c +26
-2@@ -739,7 +739,7 @@ static int engine_setup_common(struct intel_engine_cs *engine) err_cmd_parser:i915_sched_engine_put (engine->sched_engine); err_sched_engine: -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs); err_status:cleanup_status_page (engine);return err;@@ -948,7 +948,7 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine)GEM_BUG_ON (!list_empty (&engine->sched_engine->requests));i915_sched_engine_put (engine->sched_engine); -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs);intel_engine_fini_retire (engine);intel_engine_cleanup_cmd_parser (engine);@@ -1265,6 +1265,30 @@ bool intel_engines_are_idle(struct intel_gt *gt)return true; } +bool intel_engine_irq_enable (struct intel_engine_cs *engine) +{ +if (!engine->irq_enable) +return false; + +/* Caller disables interrupts */ +spin_lock (&engine->gt->irq_lock); + engine->irq_enable (engine); +spin_unlock (&engine->gt->irq_lock); + +return true; +} + +void intel_engine_irq_disable (struct intel_engine_cs *engine) +{ +if (!engine->irq_disable) +return ; + +/* Caller disables interrupts */ +spin_lock (&engine->gt->irq_lock); + engine->irq_disable (engine); +spin_unlock (&engine->gt->irq_lock); +} +void intel_engines_reset_default_submission (struct intel_gt *gt) {struct intel_engine_cs *engine;drivers/gpu/drm/i915/gt/intel_engine_types.h +1
-1@@ -21,7 +21,6 @@#include "i915_pmu.h" #include "i915_priolist_types.h" #include "i915_selftest.h" -#include "intel_breadcrumbs_types.h" #include "intel_sseu.h" #include "intel_timeline_types.h" #include "intel_uncore.h" @@ -63,6 +62,7 @@ struct i915_sched_engine;struct intel_gt ;struct intel_ring ;struct intel_uncore ; +struct intel_breadcrumbs ;typedef u8 intel_engine_mask_t ;#define ALL_ENGINES ((intel_engine_mask_t)~0ul)drivers/gpu/drm/i915/gt/intel_execlists_submission.c +1
-1@@ -3407,7 +3407,7 @@ static void rcu_virtual_context_destroy(struct work_struct *wrk)intel_context_fini (&ve->context);if (ve->base.breadcrumbs) -intel_breadcrumbs_free (ve->base.breadcrumbs); +intel_breadcrumbs_put (ve->base.breadcrumbs);if (ve->base.sched_engine)i915_sched_engine_put (ve->base.sched_engine);intel_engine_free_request_pool (&ve->base);drivers/gpu/drm/i915/gt/mock_engine.c +2
-2@@ -284,7 +284,7 @@ static void mock_engine_release(struct intel_engine_cs *engine)GEM_BUG_ON (timer_pending (&mock->hw_delay));i915_sched_engine_put (engine->sched_engine); -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs);intel_context_unpin (engine->kernel_context);intel_context_put (engine->kernel_context);@@ -376,7 +376,7 @@ int mock_engine_init(struct intel_engine_cs *engine)return 0 ; err_breadcrumbs: -intel_breadcrumbs_free (engine->breadcrumbs); +intel_breadcrumbs_put (engine->breadcrumbs); err_schedule:i915_sched_engine_put (engine->sched_engine);return -ENOMEM ;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +40
-0@@ -1076,6 +1076,9 @@ static void __guc_context_destroy(struct intel_context *ce)struct guc_virtual_engine *ve =container_of (ce,typeof (*ve), context); +if (ve->base.breadcrumbs) +intel_breadcrumbs_put (ve->base.breadcrumbs); +kfree (ve); }else {intel_context_free (ce);@@ -1366,6 +1369,62 @@ static const struct intel_context_ops virtual_guc_context_ops = { .get_sibling = guc_virtual_get_sibling, }; +static bool +guc_irq_enable_breadcrumbs (struct intel_breadcrumbs *b) +{ +struct intel_engine_cs *sibling; +intel_engine_mask_t tmp, mask = b->engine_mask; +bool result = false; + +for_each_engine_masked (sibling, b->irq_engine->gt, mask, tmp) + result |=intel_engine_irq_enable (sibling); + +return result; +} + +static void +guc_irq_disable_breadcrumbs (struct intel_breadcrumbs *b) +{ +struct intel_engine_cs *sibling; +intel_engine_mask_t tmp, mask = b->engine_mask; + +for_each_engine_masked (sibling, b->irq_engine->gt, mask, tmp) +intel_engine_irq_disable (sibling); +} + +static void guc_init_breadcrumbs (struct intel_engine_cs *engine) +{ +int i; + +/* +* In GuC submission mode we do not know which physical engine a request +* will be scheduled on, this creates a problem because the breadcrumb +* interrupt is per physical engine. To work around this we attach +* requests and direct all breadcrumb interrupts to the first instance +* of an engine per class. In addition all breadcrumb interrupts are +* enabled / disabled across an engine class in unison. +*/ +for (i =0 ; i <MAX_ENGINE_INSTANCE ; ++i) { + struct intel_engine_cs *sibling =
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static const struct intel_context_ops virtual_guc_context_ops = { · line 1413
Quoted 5 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +5
-0+ engine->gt->engine_class[engine->class][i]; + +if (sibling) { +if (engine->breadcrumbs != sibling->breadcrumbs) { +intel_breadcrumbs_put (engine->breadcrumbs);So it's still only me that is bothered by the lazyness of the proposed init path?
IMO this really isn't all that bad nor the first place I've seen something like this done. In a quick 2 minute search of the i915 I found that we do basically the exact same thing in intel_engine_create_pinned_context for the ce->vm. I doubt this is the only place in the driver that does something similar to this too.
Quoted 8 lines
intel_engines_init
for_each_engineengine_setup_common engine->breadcrumbs = intel_breadcrumbs_create(engine); intel_guc_submission_setup intel_breadcrumbs_put(engine->breadcrumbs); ...
Quoted 6 lines
In other words leaves the code to allocate N objects in common and then it frees all but one of them, overriding vfuncs in the remaining instance.
Why not pull out the setup out of common, when it is obviously not common any more?
We can write down your concerns and revisit after this series merges. That will give a clear side by side of what we have and what you purpose. It should be easy at that point to make a decision and move on.
Matt
Quoted 37 lines
Regards,
Tvrtko
Quoted 33 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +15
-5+ engine->breadcrumbs = +intel_breadcrumbs_get (sibling->breadcrumbs); + } +break ; + } + } + +if (engine->breadcrumbs) { + engine->breadcrumbs->engine_mask |= engine->mask; + engine->breadcrumbs->irq_enable = guc_irq_enable_breadcrumbs; + engine->breadcrumbs->irq_disable = guc_irq_disable_breadcrumbs; + } +} +static void sanitize_hwsp (struct intel_engine_cs *engine) {struct intel_timeline *tl;@@ -1589,6 +1648,7 @@ int intel_guc_submission_setup(struct intel_engine_cs *engine)guc_default_vfuncs (engine);guc_default_irqs (engine); +guc_init_breadcrumbs (engine);if (engine->class ==RENDER_CLASS )rcs_submission_override (engine);@@ -1831,11 +1891,6 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) ve->base.instance =I915_ENGINE_CLASS_INVALID_VIRTUAL ; ve->base.uabi_instance =I915_ENGINE_CLASS_INVALID_VIRTUAL ; ve->base.saturated =ALL_ENGINES ; - ve->base.breadcrumbs =intel_breadcrumbs_create (&ve->base); -if (!ve->base.breadcrumbs) { -kfree (ve); -return ERR_PTR (-ENOMEM ); - }snprintf (ve->base.name,sizeof (ve->base.name),"virtual" );
Quoted 8 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) · line 1945
Quoted 8 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +2
-0@@ -1884,6 +1939,8 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) sibling->emit_fini_breadcrumb; ve->base.emit_fini_breadcrumb_dw = sibling->emit_fini_breadcrumb_dw; + ve->base.breadcrumbs = +intel_breadcrumbs_get (sibling->breadcrumbs); ve->base.flags |= sibling->flags;
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 24/51 drm/i915: Add i915_sched_engine destroy vfunc
In reply to Matthew BrostThis help the backends clean up when the schedule engine object gets destroyed.
--- drivers/gpu/drm/i915/i915_scheduler.c | 3 ++- drivers/gpu/drm/i915/i915_scheduler.h | 4 +--- drivers/gpu/drm/i915/i915_scheduler_types.h | 5 +++++ 3 files changed, 8 insertions(+), 4 deletions(-)
drivers/gpu/drm/i915/i915_scheduler.c +2 -1
@@ -431,7 +431,7 @@ void i915_request_show_with_schedule(struct drm_printer *m,rcu_read_unlock (); } -void i915_sched_engine_free (struct kref *kref) +static void default_destroy (struct kref *kref) {struct i915_sched_engine *sched_engine =container_of (kref,typeof (*sched_engine), ref);
@@ -453,6 +453,7 @@ i915_sched_engine_create(unsigned int subclass) sched_engine->queue =RB_ROOT_CACHED ; sched_engine->queue_priority_hint =INT_MIN ; + sched_engine->destroy = default_destroy;INIT_LIST_HEAD (&sched_engine->requests);INIT_LIST_HEAD (&sched_engine->hold);
drivers/gpu/drm/i915/i915_scheduler.h +1 -3
@@ -51,8 +51,6 @@ static inline void i915_priolist_free(struct i915_priolist *p)struct i915_sched_engine *i915_sched_engine_create (unsigned int subclass); -void i915_sched_engine_free (struct kref *kref); -static inline struct i915_sched_engine *i915_sched_engine_get (struct i915_sched_engine *sched_engine) {
@@ -63,7 +61,7 @@ i915_sched_engine_get(struct i915_sched_engine *sched_engine)static inline void i915_sched_engine_put (struct i915_sched_engine *sched_engine) { -kref_put (&sched_engine->ref, i915_sched_engine_free); +kref_put (&sched_engine->ref, sched_engine->destroy); } static inline bool
drivers/gpu/drm/i915/i915_scheduler_types.h +5 -0
@@ -163,6 +163,11 @@ struct i915_sched_engine {*/ void *private_data; +/** +* @destroy: destroy schedule engine / cleanup in backend +*/ +void (*destroy)(struct kref *kref); +/** * @kick_backend: kick backend after a request's priority has changed */
Signature · 1 line
2.28.0
Re: [PATCH 24/51] drm/i915: Add i915_sched_engine destroy vfunc
Quoted 1 line · Matthew Brost
This help the backends clean up when the schedule engine object gets
help -> helps. Although, I would say it's more like 'this is required to allow backend specific cleanup'. It doesn't just make life a bit easier, it allows us to not leak stuff and/or dereference null pointers!
Either way...
Quoted 68 lines
destroyed.
Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/i915_scheduler.c | 3 ++- drivers/gpu/drm/i915/i915_scheduler.h | 4 +--- drivers/gpu/drm/i915/i915_scheduler_types.h | 5 +++++ 3 files changed, 8 insertions(+), 4 deletions(-)drivers/gpu/drm/i915/i915_scheduler.c +2
-1@@ -431,7 +431,7 @@ void i915_request_show_with_schedule(struct drm_printer *m,rcu_read_unlock (); } -void i915_sched_engine_free (struct kref *kref) +static void default_destroy (struct kref *kref) {struct i915_sched_engine *sched_engine =container_of (kref,typeof (*sched_engine), ref);@@ -453,6 +453,7 @@ i915_sched_engine_create(unsigned int subclass) sched_engine->queue =RB_ROOT_CACHED ; sched_engine->queue_priority_hint =INT_MIN ; + sched_engine->destroy = default_destroy;INIT_LIST_HEAD (&sched_engine->requests);INIT_LIST_HEAD (&sched_engine->hold);drivers/gpu/drm/i915/i915_scheduler.h +1
-3@@ -51,8 +51,6 @@ static inline void i915_priolist_free(struct i915_priolist *p)struct i915_sched_engine *i915_sched_engine_create (unsigned int subclass); -void i915_sched_engine_free (struct kref *kref); -static inline struct i915_sched_engine *i915_sched_engine_get (struct i915_sched_engine *sched_engine) {@@ -63,7 +61,7 @@ i915_sched_engine_get(struct i915_sched_engine *sched_engine)static inline void i915_sched_engine_put (struct i915_sched_engine *sched_engine) { -kref_put (&sched_engine->ref, i915_sched_engine_free); +kref_put (&sched_engine->ref, sched_engine->destroy); } static inline booldrivers/gpu/drm/i915/i915_scheduler_types.h +5
-0@@ -163,6 +163,11 @@ struct i915_sched_engine {*/ void *private_data; +/** +* @destroy: destroy schedule engine / cleanup in backend +*/ +void (*destroy)(struct kref *kref); +/** * @kick_backend: kick backend after a request's priority has changed */
Re: [PATCH 24/51] drm/i915: Add i915_sched_engine destroy vfunc
Quoted 8 lines · John Harrison
Quoted 1 line · Matthew Brost
This help the backends clean up when the schedule engine object gets
help -> helps. Although, I would say it's more like 'this is required to allow backend specific cleanup'. It doesn't just make life a bit easier, it allows us to not leak stuff and/or dereference null pointers!
Either way...
Reviewed-by John Harrison <[email protected]>
I'll reword it.
Matt
Quoted 55 lines
Quoted 55 lines
destroyed.
Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/i915_scheduler.c | 3 ++- drivers/gpu/drm/i915/i915_scheduler.h | 4 +--- drivers/gpu/drm/i915/i915_scheduler_types.h | 5 +++++ 3 files changed, 8 insertions(+), 4 deletions(-)drivers/gpu/drm/i915/i915_scheduler.c +2
-1@@ -431,7 +431,7 @@ void i915_request_show_with_schedule(struct drm_printer *m,rcu_read_unlock (); } -void i915_sched_engine_free (struct kref *kref) +static void default_destroy (struct kref *kref) {struct i915_sched_engine *sched_engine =container_of (kref,typeof (*sched_engine), ref);@@ -453,6 +453,7 @@ i915_sched_engine_create(unsigned int subclass) sched_engine->queue =RB_ROOT_CACHED ; sched_engine->queue_priority_hint =INT_MIN ; + sched_engine->destroy = default_destroy;INIT_LIST_HEAD (&sched_engine->requests);INIT_LIST_HEAD (&sched_engine->hold);drivers/gpu/drm/i915/i915_scheduler.h +1
-3@@ -51,8 +51,6 @@ static inline void i915_priolist_free(struct i915_priolist *p)struct i915_sched_engine *i915_sched_engine_create (unsigned int subclass); -void i915_sched_engine_free (struct kref *kref); -static inline struct i915_sched_engine *i915_sched_engine_get (struct i915_sched_engine *sched_engine) {@@ -63,7 +61,7 @@ i915_sched_engine_get(struct i915_sched_engine *sched_engine)static inline void i915_sched_engine_put (struct i915_sched_engine *sched_engine) { -kref_put (&sched_engine->ref, i915_sched_engine_free); +kref_put (&sched_engine->ref, sched_engine->destroy); } static inline booldrivers/gpu/drm/i915/i915_scheduler_types.h +0
-0@@ -163,6 +163,11 @@ struct i915_sched_engine {*/ void *private_data;
Quoted 9 lines
drivers/gpu/drm/i915/i915_scheduler_types.h struct i915_sched_engine { · line 173
Quoted 8 lines
drivers/gpu/drm/i915/i915_scheduler_types.h +5
-0+/** +* @destroy: destroy schedule engine / cleanup in backend +*/ +void (*destroy)(struct kref *kref); +/** * @kick_backend: kick backend after a request's priority has changed */
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 25/51 drm/i915: Move active request tracking to a vfunc
In reply to Matthew BrostMove active request tracking to a backend vfunc rather than assuming all backends want to do this in the maner. In the case execlists / ring submission the tracking is on the physical engine while with GuC submission it is on the context.
--- drivers/gpu/drm/i915/gt/intel_context.c | 3 ++ drivers/gpu/drm/i915/gt/intel_context_types.h | 7 ++++ drivers/gpu/drm/i915/gt/intel_engine_types.h | 6 +++ .../drm/i915/gt/intel_execlists_submission.c | 40 ++++++++++++++++++ .../gpu/drm/i915/gt/intel_ring_submission.c | 22 ++++++++++ drivers/gpu/drm/i915/gt/mock_engine.c | 30 ++++++++++++++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 33 +++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 41 ++----------------- drivers/gpu/drm/i915/i915_request.h | 2 + 9 files changed, 147 insertions(+), 37 deletions(-)
drivers/gpu/drm/i915/gt/intel_context.c +3 -0
@@ -393,6 +393,9 @@ intel_context_init(struct intel_context *ce, struct intel_engine_cs *engine)spin_lock_init (&ce->guc_state.lock);INIT_LIST_HEAD (&ce->guc_state.fences); +spin_lock_init (&ce->guc_active.lock); +INIT_LIST_HEAD (&ce->guc_active.requests); + ce->guc_id =GUC_INVALID_LRC_ID ;INIT_LIST_HEAD (&ce->guc_id_link);
drivers/gpu/drm/i915/gt/intel_context_types.h +7 -0
@@ -162,6 +162,13 @@ struct intel_context {struct list_head fences; } guc_state; +struct { +/** lock: protects everything in guc_active */ +spinlock_t lock; +/** requests: active requests on this context */ +struct list_head requests; + } guc_active; +/* GuC scheduling state flags that do not require a lock. */ atomic_t guc_sched_state_no_lock;
drivers/gpu/drm/i915/gt/intel_engine_types.h +6 -0
@@ -420,6 +420,12 @@ struct intel_engine_cs {void (*release)(struct intel_engine_cs *engine); +/* +* Add / remove request from engine active tracking +*/ +void (*add_active_request)(struct i915_request *rq); +void (*remove_active_request)(struct i915_request *rq); +struct intel_engine_execlists execlists;/*
drivers/gpu/drm/i915/gt/intel_execlists_submission.c +40 -0
@@ -3106,6 +3106,42 @@ static void execlists_park(struct intel_engine_cs *engine)cancel_timer (&engine->execlists.preempt); } +static void add_to_engine (struct i915_request *rq) +{ +lockdep_assert_held (&rq->engine->sched_engine->lock); +list_move_tail (&rq->sched.link, &rq->engine->sched_engine->requests); +} + +static void remove_from_engine (struct i915_request *rq) +{ +struct intel_engine_cs *engine, *locked; + +/* +* Virtual engines complicate acquiring the engine timeline lock, +* as their rq->engine pointer is not stable until under that +* engine lock. The simple ploy we use is to take the lock then +* check that the rq still belongs to the newly locked engine. +*/ + locked =READ_ONCE (rq->engine); +spin_lock_irq (&locked->sched_engine->lock); +while (unlikely (locked != (engine =READ_ONCE (rq->engine)))) { +spin_unlock (&locked->sched_engine->lock); +spin_lock (&engine->sched_engine->lock); + locked = engine; + } +list_del_init (&rq->sched.link); + +clear_bit (I915_FENCE_FLAG_PQUEUE , &rq->fence.flags); +clear_bit (I915_FENCE_FLAG_HOLD , &rq->fence.flags); + +/* Prevent further __await_execution() registering a cb, then flush */ +set_bit (I915_FENCE_FLAG_ACTIVE , &rq->fence.flags); + +spin_unlock_irq (&locked->sched_engine->lock); + +i915_request_notify_execute_cb_imm (rq); +} +static bool can_preempt (struct intel_engine_cs *engine) {if (GRAPHICS_VER (engine->i915) >8 )
@@ -3206,6 +3242,8 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &execlists_context_ops; engine->request_alloc = execlists_request_alloc; engine->bump_serial = execlist_bump_serial; + engine->add_active_request = add_to_engine; + engine->remove_active_request = remove_from_engine; engine->reset.prepare = execlists_reset_prepare; engine->reset.rewind = execlists_reset_rewind;
@@ -3797,6 +3835,8 @@ execlists_create_virtual(struct intel_engine_cs **siblings, unsigned int count) "v%dx%d", ve->base.class, count); ve->base.context_size = sibling->context_size; + ve->base.add_active_request = sibling->add_active_request; + ve->base.remove_active_request = sibling->remove_active_request; ve->base.emit_bb_start = sibling->emit_bb_start; ve->base.emit_flush = sibling->emit_flush; ve->base.emit_init_breadcrumb = sibling->emit_init_breadcrumb;
drivers/gpu/drm/i915/gt/intel_ring_submission.c +22 -0
@@ -1052,6 +1052,25 @@ static void ring_bump_serial(struct intel_engine_cs *engine) engine->serial++; } +static void add_to_engine (struct i915_request *rq) +{ +lockdep_assert_held (&rq->engine->sched_engine->lock); +list_move_tail (&rq->sched.link, &rq->engine->sched_engine->requests); +} + +static void remove_from_engine (struct i915_request *rq) +{ +spin_lock_irq (&rq->engine->sched_engine->lock); +list_del_init (&rq->sched.link); + +/* Prevent further __await_execution() registering a cb, then flush */ +set_bit (I915_FENCE_FLAG_ACTIVE , &rq->fence.flags); + +spin_unlock_irq (&rq->engine->sched_engine->lock); + +i915_request_notify_execute_cb_imm (rq); +} +static void setup_common (struct intel_engine_cs *engine) {struct drm_i915_private *i915 = engine->i915;
@@ -1069,6 +1088,9 @@ static void setup_common(struct intel_engine_cs *engine) engine->reset.cancel = reset_cancel; engine->reset.finish = reset_finish; + engine->add_active_request = add_to_engine; + engine->remove_active_request = remove_from_engine; + engine->cops = &ring_context_ops; engine->request_alloc = ring_request_alloc; engine->bump_serial = ring_bump_serial;
drivers/gpu/drm/i915/gt/mock_engine.c +30 -0
@@ -235,6 +235,34 @@ static void mock_submit_request(struct i915_request *request)spin_unlock_irqrestore (&engine->hw_lock, flags); } +static void mock_add_to_engine (struct i915_request *rq) +{ +lockdep_assert_held (&rq->engine->sched_engine->lock); +list_move_tail (&rq->sched.link, &rq->engine->sched_engine->requests); +} + +static void mock_remove_from_engine (struct i915_request *rq) +{ +struct intel_engine_cs *engine, *locked; + +/* +* Virtual engines complicate acquiring the engine timeline lock, +* as their rq->engine pointer is not stable until under that +* engine lock. The simple ploy we use is to take the lock then +* check that the rq still belongs to the newly locked engine. +*/ + + locked =READ_ONCE (rq->engine); +spin_lock_irq (&locked->sched_engine->lock); +while (unlikely (locked != (engine =READ_ONCE (rq->engine)))) { +spin_unlock (&locked->sched_engine->lock); +spin_lock (&engine->sched_engine->lock); + locked = engine; + } +list_del_init (&rq->sched.link); +spin_unlock_irq (&locked->sched_engine->lock); +} +static void mock_reset_prepare (struct intel_engine_cs *engine) { }
@@ -327,6 +355,8 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, engine->base.emit_flush = mock_emit_flush; engine->base.emit_fini_breadcrumb = mock_emit_breadcrumb; engine->base.submit_request = mock_submit_request; + engine->base.add_active_request = mock_add_to_engine; + engine->base.remove_active_request = mock_remove_from_engine; engine->base.reset.prepare = mock_reset_prepare; engine->base.reset.rewind = mock_reset_rewind;
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +33 -0
@@ -1147,6 +1147,33 @@ static int guc_context_alloc(struct intel_context *ce)return lrc_alloc (ce, ce->engine); } +static void add_to_context (struct i915_request *rq) +{ +struct intel_context *ce = rq->context; + +spin_lock (&ce->guc_active.lock); +list_move_tail (&rq->sched.link, &ce->guc_active.requests); +spin_unlock (&ce->guc_active.lock); +} + +static void remove_from_context (struct i915_request *rq) +{ +struct intel_context *ce = rq->context; + +spin_lock_irq (&ce->guc_active.lock); + +list_del_init (&rq->sched.link); +clear_bit (I915_FENCE_FLAG_PQUEUE , &rq->fence.flags); + +/* Prevent further __await_execution() registering a cb, then flush */ +set_bit (I915_FENCE_FLAG_ACTIVE , &rq->fence.flags); + +spin_unlock_irq (&ce->guc_active.lock); + +atomic_dec (&ce->guc_id_ref); +i915_request_notify_execute_cb_imm (rq); +} +static const struct intel_context_ops guc_context_ops = { .alloc = guc_context_alloc,
@@ -1567,6 +1594,8 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &guc_context_ops; engine->request_alloc = guc_request_alloc; engine->bump_serial = guc_bump_serial; + engine->add_active_request = add_to_context; + engine->remove_active_request = remove_from_context; engine->sched_engine->schedule = i915_schedule;
@@ -1931,6 +1960,10 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) "v%dx%d", ve->base.class, count); ve->base.context_size = sibling->context_size; + ve->base.add_active_request = + sibling->add_active_request; + ve->base.remove_active_request = + sibling->remove_active_request; ve->base.emit_bb_start = sibling->emit_bb_start; ve->base.emit_flush = sibling->emit_flush; ve->base.emit_init_breadcrumb =
drivers/gpu/drm/i915/i915_request.c +4 -37
71 lines not on this page · raw message
drivers/gpu/drm/i915/i915_request.h +2 -0
@@ -647,4 +647,6 @@ bool i915_request_active_engine(struct i915_request *rq, struct intel_engine_cs **active); +void i915_request_notify_execute_cb_imm (struct i915_request *rq); +#endif /* I915_REQUEST_H */
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 25/51] drm/i915: Move active request tracking to a vfunc
Quoted 2 lines · Matthew Brost
Move active request tracking to a backend vfunc rather than assuming all backends want to do this in the maner. In the case execlists /
maner -> manner.
In the case *of* execlists
With those fixed...
Quoted 367 lines
ring submission the tracking is on the physical engine while with GuC submission it is on the context.
Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gt/intel_context.c | 3 ++ drivers/gpu/drm/i915/gt/intel_context_types.h | 7 ++++ drivers/gpu/drm/i915/gt/intel_engine_types.h | 6 +++ .../drm/i915/gt/intel_execlists_submission.c | 40 ++++++++++++++++++ .../gpu/drm/i915/gt/intel_ring_submission.c | 22 ++++++++++ drivers/gpu/drm/i915/gt/mock_engine.c | 30 ++++++++++++++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 33 +++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 41 ++----------------- drivers/gpu/drm/i915/i915_request.h | 2 + 9 files changed, 147 insertions(+), 37 deletions(-)drivers/gpu/drm/i915/gt/intel_context.c +3
-0@@ -393,6 +393,9 @@ intel_context_init(struct intel_context *ce, struct intel_engine_cs *engine)spin_lock_init (&ce->guc_state.lock);INIT_LIST_HEAD (&ce->guc_state.fences); +spin_lock_init (&ce->guc_active.lock); +INIT_LIST_HEAD (&ce->guc_active.requests); + ce->guc_id =GUC_INVALID_LRC_ID ;INIT_LIST_HEAD (&ce->guc_id_link);drivers/gpu/drm/i915/gt/intel_context_types.h +7
-0@@ -162,6 +162,13 @@ struct intel_context {struct list_head fences; } guc_state; +struct { +/** lock: protects everything in guc_active */ +spinlock_t lock; +/** requests: active requests on this context */ +struct list_head requests; + } guc_active; +/* GuC scheduling state flags that do not require a lock. */ atomic_t guc_sched_state_no_lock;drivers/gpu/drm/i915/gt/intel_engine_types.h +6
-0@@ -420,6 +420,12 @@ struct intel_engine_cs {void (*release)(struct intel_engine_cs *engine); +/* +* Add / remove request from engine active tracking +*/ +void (*add_active_request)(struct i915_request *rq); +void (*remove_active_request)(struct i915_request *rq); +struct intel_engine_execlists execlists;/* drivers/gpu/drm/i915/gt/intel_execlists_submission.c +40
-0@@ -3106,6 +3106,42 @@ static void execlists_park(struct intel_engine_cs *engine)cancel_timer (&engine->execlists.preempt); } +static void add_to_engine (struct i915_request *rq) +{ +lockdep_assert_held (&rq->engine->sched_engine->lock); +list_move_tail (&rq->sched.link, &rq->engine->sched_engine->requests); +} + +static void remove_from_engine (struct i915_request *rq) +{ +struct intel_engine_cs *engine, *locked; + +/* +* Virtual engines complicate acquiring the engine timeline lock, +* as their rq->engine pointer is not stable until under that +* engine lock. The simple ploy we use is to take the lock then +* check that the rq still belongs to the newly locked engine. +*/ + locked =READ_ONCE (rq->engine); +spin_lock_irq (&locked->sched_engine->lock); +while (unlikely (locked != (engine =READ_ONCE (rq->engine)))) { +spin_unlock (&locked->sched_engine->lock); +spin_lock (&engine->sched_engine->lock); + locked = engine; + } +list_del_init (&rq->sched.link); + +clear_bit (I915_FENCE_FLAG_PQUEUE , &rq->fence.flags); +clear_bit (I915_FENCE_FLAG_HOLD , &rq->fence.flags); + +/* Prevent further __await_execution() registering a cb, then flush */ +set_bit (I915_FENCE_FLAG_ACTIVE , &rq->fence.flags); + +spin_unlock_irq (&locked->sched_engine->lock); + +i915_request_notify_execute_cb_imm (rq); +} +static bool can_preempt (struct intel_engine_cs *engine) {if (GRAPHICS_VER (engine->i915) >8 )@@ -3206,6 +3242,8 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &execlists_context_ops; engine->request_alloc = execlists_request_alloc; engine->bump_serial = execlist_bump_serial; + engine->add_active_request = add_to_engine; + engine->remove_active_request = remove_from_engine; engine->reset.prepare = execlists_reset_prepare; engine->reset.rewind = execlists_reset_rewind;@@ -3797,6 +3835,8 @@ execlists_create_virtual(struct intel_engine_cs **siblings, unsigned int count) "v%dx%d", ve->base.class, count); ve->base.context_size = sibling->context_size; + ve->base.add_active_request = sibling->add_active_request; + ve->base.remove_active_request = sibling->remove_active_request; ve->base.emit_bb_start = sibling->emit_bb_start; ve->base.emit_flush = sibling->emit_flush; ve->base.emit_init_breadcrumb = sibling->emit_init_breadcrumb;drivers/gpu/drm/i915/gt/intel_ring_submission.c +22
-0@@ -1052,6 +1052,25 @@ static void ring_bump_serial(struct intel_engine_cs *engine) engine->serial++; } +static void add_to_engine (struct i915_request *rq) +{ +lockdep_assert_held (&rq->engine->sched_engine->lock); +list_move_tail (&rq->sched.link, &rq->engine->sched_engine->requests); +} + +static void remove_from_engine (struct i915_request *rq) +{ +spin_lock_irq (&rq->engine->sched_engine->lock); +list_del_init (&rq->sched.link); + +/* Prevent further __await_execution() registering a cb, then flush */ +set_bit (I915_FENCE_FLAG_ACTIVE , &rq->fence.flags); + +spin_unlock_irq (&rq->engine->sched_engine->lock); + +i915_request_notify_execute_cb_imm (rq); +} +static void setup_common (struct intel_engine_cs *engine) {struct drm_i915_private *i915 = engine->i915;@@ -1069,6 +1088,9 @@ static void setup_common(struct intel_engine_cs *engine) engine->reset.cancel = reset_cancel; engine->reset.finish = reset_finish; + engine->add_active_request = add_to_engine; + engine->remove_active_request = remove_from_engine; + engine->cops = &ring_context_ops; engine->request_alloc = ring_request_alloc; engine->bump_serial = ring_bump_serial;drivers/gpu/drm/i915/gt/mock_engine.c +30
-0@@ -235,6 +235,34 @@ static void mock_submit_request(struct i915_request *request)spin_unlock_irqrestore (&engine->hw_lock, flags); } +static void mock_add_to_engine (struct i915_request *rq) +{ +lockdep_assert_held (&rq->engine->sched_engine->lock); +list_move_tail (&rq->sched.link, &rq->engine->sched_engine->requests); +} + +static void mock_remove_from_engine (struct i915_request *rq) +{ +struct intel_engine_cs *engine, *locked; + +/* +* Virtual engines complicate acquiring the engine timeline lock, +* as their rq->engine pointer is not stable until under that +* engine lock. The simple ploy we use is to take the lock then +* check that the rq still belongs to the newly locked engine. +*/ + + locked =READ_ONCE (rq->engine); +spin_lock_irq (&locked->sched_engine->lock); +while (unlikely (locked != (engine =READ_ONCE (rq->engine)))) { +spin_unlock (&locked->sched_engine->lock); +spin_lock (&engine->sched_engine->lock); + locked = engine; + } +list_del_init (&rq->sched.link); +spin_unlock_irq (&locked->sched_engine->lock); +} +static void mock_reset_prepare (struct intel_engine_cs *engine) { }@@ -327,6 +355,8 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, engine->base.emit_flush = mock_emit_flush; engine->base.emit_fini_breadcrumb = mock_emit_breadcrumb; engine->base.submit_request = mock_submit_request; + engine->base.add_active_request = mock_add_to_engine; + engine->base.remove_active_request = mock_remove_from_engine; engine->base.reset.prepare = mock_reset_prepare; engine->base.reset.rewind = mock_reset_rewind;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +33
-0@@ -1147,6 +1147,33 @@ static int guc_context_alloc(struct intel_context *ce)return lrc_alloc (ce, ce->engine); } +static void add_to_context (struct i915_request *rq) +{ +struct intel_context *ce = rq->context; + +spin_lock (&ce->guc_active.lock); +list_move_tail (&rq->sched.link, &ce->guc_active.requests); +spin_unlock (&ce->guc_active.lock); +} + +static void remove_from_context (struct i915_request *rq) +{ +struct intel_context *ce = rq->context; + +spin_lock_irq (&ce->guc_active.lock); + +list_del_init (&rq->sched.link); +clear_bit (I915_FENCE_FLAG_PQUEUE , &rq->fence.flags); + +/* Prevent further __await_execution() registering a cb, then flush */ +set_bit (I915_FENCE_FLAG_ACTIVE , &rq->fence.flags); + +spin_unlock_irq (&ce->guc_active.lock); + +atomic_dec (&ce->guc_id_ref); +i915_request_notify_execute_cb_imm (rq); +} +static const struct intel_context_ops guc_context_ops = { .alloc = guc_context_alloc,@@ -1567,6 +1594,8 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->cops = &guc_context_ops; engine->request_alloc = guc_request_alloc; engine->bump_serial = guc_bump_serial; + engine->add_active_request = add_to_context; + engine->remove_active_request = remove_from_context; engine->sched_engine->schedule = i915_schedule;@@ -1931,6 +1960,10 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count) "v%dx%d", ve->base.class, count); ve->base.context_size = sibling->context_size; + ve->base.add_active_request = + sibling->add_active_request; + ve->base.remove_active_request = + sibling->remove_active_request; ve->base.emit_bb_start = sibling->emit_bb_start; ve->base.emit_flush = sibling->emit_flush; ve->base.emit_init_breadcrumb =drivers/gpu/drm/i915/i915_request.c +4
-37@@ -182,7 +182,7 @@ static bool irq_work_imm(struct irq_work *wrk)return false; } -static void __notify_execute_cb_imm (struct i915_request *rq) +void i915_request_notify_execute_cb_imm (struct i915_request *rq) {__notify_execute_cb (rq, irq_work_imm); }@@ -256,37 +256,6 @@ i915_request_active_engine(struct i915_request *rq,return ret; } - -static void remove_from_engine (struct i915_request *rq) -{ -struct intel_engine_cs *engine, *locked; - -/* -* Virtual engines complicate acquiring the engine timeline lock, -* as their rq->engine pointer is not stable until under that -* engine lock. The simple ploy we use is to take the lock then -* check that the rq still belongs to the newly locked engine. -*/ - locked =READ_ONCE (rq->engine); -spin_lock_irq (&locked->sched_engine->lock); -while (unlikely (locked != (engine =READ_ONCE (rq->engine)))) { -spin_unlock (&locked->sched_engine->lock); -spin_lock (&engine->sched_engine->lock); - locked = engine; - } -list_del_init (&rq->sched.link); - -clear_bit (I915_FENCE_FLAG_PQUEUE , &rq->fence.flags); -clear_bit (I915_FENCE_FLAG_HOLD , &rq->fence.flags); - -/* Prevent further __await_execution() registering a cb, then flush */ -set_bit (I915_FENCE_FLAG_ACTIVE , &rq->fence.flags); - -spin_unlock_irq (&locked->sched_engine->lock); - -__notify_execute_cb_imm (rq); -} -static void __rq_init_watchdog (struct i915_request *rq) { rq->watchdog.timer.function =NULL ;@@ -383,9 +352,7 @@ bool i915_request_retire(struct i915_request *rq)* after removing the breadcrumb and signaling it, so that we do not * inadvertently attach the breadcrumb to a completed request. */ -if (!list_empty (&rq->sched.link)) -remove_from_engine (rq); -atomic_dec (&rq->context->guc_id_ref); + rq->engine->remove_active_request (rq);GEM_BUG_ON (!llist_empty (&rq->execute_cb));__list_del_entry (&rq->link);/* poison neither prev/next (RCU walks) */ @@ -516,7 +483,7 @@ __await_execution(struct i915_request *rq,if (llist_add (&cb->work.node.llist, &signal->execute_cb)) {if (i915_request_is_active (signal) ||__request_in_flight (signal)) -__notify_execute_cb_imm (signal); +i915_request_notify_execute_cb_imm (signal); }return 0 ;@@ -653,7 +620,7 @@ bool __i915_request_submit(struct i915_request *request) result = true;GEM_BUG_ON (test_bit (I915_FENCE_FLAG_ACTIVE , &request->fence.flags)); -list_move_tail (&request->sched.link, &engine->sched_engine->requests); + engine->add_active_request (request); active:clear_bit (I915_FENCE_FLAG_PQUEUE , &request->fence.flags);set_bit (I915_FENCE_FLAG_ACTIVE , &request->fence.flags);drivers/gpu/drm/i915/i915_request.h +2
-0@@ -647,4 +647,6 @@ bool i915_request_active_engine(struct i915_request *rq, struct intel_engine_cs **active); +void i915_request_notify_execute_cb_imm (struct i915_request *rq); +#endif /* I915_REQUEST_H */
PATCH 26/51 drm/i915/guc: Reset implementation for new GuC interface
In reply to Matthew BrostReset implementation for new GuC interface. This is the legacy reset implementation which is called when the i915 owns the engine hang check. Future patches will offload the engine hang check to GuC but we will continue to maintain this legacy path as a fallback and this code path is also required if the GuC dies.
With the new GuC interface it is not possible to reset individual engines - it is only possible to reset the GPU entirely. This patch forces an entire chip reset if any engine hangs.
v2:
(Michal)
- Check for -EPIPE rather than -EIO (CT deadlock/corrupt check)
v3:
(John H)
- Split into a series of smaller patches
--- drivers/gpu/drm/i915/gt/intel_gt_pm.c | 6 +- drivers/gpu/drm/i915/gt/intel_reset.c | 18 +- drivers/gpu/drm/i915/gt/uc/intel_guc.c | 13 - drivers/gpu/drm/i915/gt/uc/intel_guc.h | 8 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 562 ++++++++++++++---- drivers/gpu/drm/i915/gt/uc/intel_uc.c | 39 +- drivers/gpu/drm/i915/gt/uc/intel_uc.h | 3 + 7 files changed, 515 insertions(+), 134 deletions(-)
drivers/gpu/drm/i915/gt/intel_gt_pm.c +4 -2
@@ -174,8 +174,6 @@ static void gt_sanitize(struct intel_gt *gt, bool force)if (intel_gt_is_wedged (gt))intel_gt_unset_wedged (gt); -intel_uc_sanitize (>->uc); -for_each_engine (engine, gt, id)if (engine->reset.prepare) engine->reset.prepare (engine);
@@ -191,6 +189,8 @@ static void gt_sanitize(struct intel_gt *gt, bool force)__intel_engine_reset (engine, false); } +intel_uc_reset (>->uc, false); +for_each_engine (engine, gt, id)if (engine->reset.finish) engine->reset.finish (engine);
@@ -243,6 +243,8 @@ int intel_gt_resume(struct intel_gt *gt) goto err_wedged; } +intel_uc_reset_finish (>->uc); +intel_rps_enable (>->rps);intel_llc_enable (>->llc);
drivers/gpu/drm/i915/gt/intel_reset.c +12 -6
@@ -826,6 +826,8 @@ static int gt_reset(struct intel_gt *gt, intel_engine_mask_t stalled_mask)__intel_engine_reset (engine, stalled_mask & engine->mask);local_bh_enable (); +intel_uc_reset (>->uc, true); +intel_ggtt_restore_fences (gt->ggtt);return err;
@@ -850,6 +852,8 @@ static void reset_finish(struct intel_gt *gt, intel_engine_mask_t awake)if (awake & engine->mask)intel_engine_pm_put (engine); } + +intel_uc_reset_finish (>->uc); }static void nop_submit_request (struct i915_request *request)
@@ -903,6 +907,7 @@ static void __intel_gt_set_wedged(struct intel_gt *gt)for_each_engine (engine, gt, id)if (engine->reset.cancel) engine->reset.cancel (engine); +intel_uc_cancel_requests (>->uc);local_bh_enable ();reset_finish (gt, awake);
@@ -1191,6 +1196,9 @@ int __intel_engine_reset_bh(struct intel_engine_cs *engine, const char *msg)ENGINE_TRACE (engine,"flags=%lx\n" , gt->reset.flags);GEM_BUG_ON (!test_bit (I915_RESET_ENGINE + engine->id, >->reset.flags)); +if (intel_engine_uses_guc (engine)) +return -ENODEV ; +if (!intel_engine_pm_get_if_awake (engine))return 0 ;
@@ -1201,13 +1209,10 @@ int __intel_engine_reset_bh(struct intel_engine_cs *engine, const char *msg) "Resetting %s for %s\n", engine->name, msg);atomic_inc (&engine->i915->gpu_error.reset_engine_count[engine->uabi_class]); -if (intel_engine_uses_guc (engine)) - ret =intel_guc_reset_engine (&engine->gt->uc.guc, engine); -else - ret =intel_gt_reset_engine (engine); + ret =intel_gt_reset_engine (engine);if (ret) {/* If we fail here, we expect to fallback to a global reset */ -ENGINE_TRACE (engine,"Failed to reset, err: %d\n" , ret); +ENGINE_TRACE (engine,"Failed to reset %s, err: %d\n" , engine->name, ret); goto out; }
@@ -1341,7 +1346,8 @@ void intel_gt_handle_error(struct intel_gt *gt,* Try engine reset when available. We fall back to full reset if * single reset fails. */ -if (intel_has_reset_engine (gt) && !intel_gt_is_wedged (gt)) { +if (!intel_uc_uses_guc_submission (>->uc) && +intel_has_reset_engine (gt) && !intel_gt_is_wedged (gt)) {local_bh_disable ();for_each_engine_masked (engine, gt, engine_mask, tmp) {BUILD_BUG_ON (I915_RESET_MODESET >=I915_RESET_ENGINE );
drivers/gpu/drm/i915/gt/uc/intel_guc.c +0 -13
@@ -572,19 +572,6 @@ int intel_guc_suspend(struct intel_guc *guc)return 0 ; } -/** -* intel_guc_reset_engine() - ask GuC to reset an engine -* @guc: intel_guc structure -* @engine: engine to be reset -*/ -int intel_guc_reset_engine (struct intel_guc *guc, -struct intel_engine_cs *engine) -{ -/* XXX: to be implemented with submission interface rework */ - -return -ENODEV ; -} -/** * intel_guc_resume() - notify GuC resuming from suspend state * @guc: the guc
drivers/gpu/drm/i915/gt/uc/intel_guc.h +5 -3
@@ -242,14 +242,16 @@ static inline void intel_guc_disable_msg(struct intel_guc *guc, u32 mask)int intel_guc_wait_for_idle (struct intel_guc *guc,long timeout); -int intel_guc_reset_engine (struct intel_guc *guc, -struct intel_engine_cs *engine); -int intel_guc_deregister_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len);int intel_guc_sched_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len); +void intel_guc_submission_reset_prepare (struct intel_guc *guc); +void intel_guc_submission_reset (struct intel_guc *guc,bool stalled); +void intel_guc_submission_reset_finish (struct intel_guc *guc); +void intel_guc_submission_cancel_requests (struct intel_guc *guc); +void intel_guc_load_status (struct intel_guc *guc,struct drm_printer *p);#endif
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +453 -109
856 lines not on this page · raw message
drivers/gpu/drm/i915/gt/uc/intel_uc.c +38 -1
@@ -565,12 +565,49 @@ void intel_uc_reset_prepare(struct intel_uc *uc) {struct intel_guc *guc = &uc->guc; -if (!intel_guc_is_ready (guc)) + +/* Nothing to do if GuC isn't supported */ +if (!intel_uc_supports_guc (uc))return ; +/* Firmware expected to be running when this function is called */ +if (!intel_guc_is_ready (guc)) + goto sanitize; + +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset_prepare (guc); + +sanitize:__uc_sanitize (uc); } +void intel_uc_reset (struct intel_uc *uc,bool stalled) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware can not be running when this function is called */ +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset (guc, stalled); +} + +void intel_uc_reset_finish (struct intel_uc *uc) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware expected to be running when this function is called */ +if (intel_guc_is_fw_running (guc) &&intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset_finish (guc); +} + +void intel_uc_cancel_requests (struct intel_uc *uc) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware can not be running when this function is called */ +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_cancel_requests (guc); +} +void intel_uc_runtime_suspend (struct intel_uc *uc) {struct intel_guc *guc = &uc->guc;
drivers/gpu/drm/i915/gt/uc/intel_uc.h +3 -0
@@ -37,6 +37,9 @@ void intel_uc_driver_late_release(struct intel_uc *uc);void intel_uc_driver_remove (struct intel_uc *uc);void intel_uc_init_mmio (struct intel_uc *uc);void intel_uc_reset_prepare (struct intel_uc *uc); +void intel_uc_reset (struct intel_uc *uc,bool stalled); +void intel_uc_reset_finish (struct intel_uc *uc); +void intel_uc_cancel_requests (struct intel_uc *uc);void intel_uc_suspend (struct intel_uc *uc);void intel_uc_runtime_suspend (struct intel_uc *uc);int intel_uc_resume (struct intel_uc *uc);
Signature · 1 line
2.28.0
Re: [PATCH 26/51] drm/i915/guc: Reset implementation for new GuC interface
Quoted 10 lines · Matthew Brost
Reset implementation for new GuC interface. This is the legacy reset implementation which is called when the i915 owns the engine hang check. Future patches will offload the engine hang check to GuC but we will continue to maintain this legacy path as a fallback and this code path is also required if the GuC dies.
With the new GuC interface it is not possible to reset individual engines - it is only possible to reset the GPU entirely. This patch forces an entire chip reset if any engine hangs.
Quoted 6 lines
v2:
(Michal)
- Check for-EPIPErather than-EIO(CT deadlock/corrupt check)
v3:
(John H)
- Split into a series of smaller patches
While the split happened, it doesn't look like any of the other comments were address. Repeated below for clarity. Also, Tvrtko has a bunch of outstanding comments too.
Quoted 308 lines
Cc John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gt/intel_gt_pm.c | 6 +- drivers/gpu/drm/i915/gt/intel_reset.c | 18 +- drivers/gpu/drm/i915/gt/uc/intel_guc.c | 13 - drivers/gpu/drm/i915/gt/uc/intel_guc.h | 8 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 562 ++++++++++++++---- drivers/gpu/drm/i915/gt/uc/intel_uc.c | 39 +- drivers/gpu/drm/i915/gt/uc/intel_uc.h | 3 + 7 files changed, 515 insertions(+), 134 deletions(-)drivers/gpu/drm/i915/gt/intel_gt_pm.c +4
-2@@ -174,8 +174,6 @@ static void gt_sanitize(struct intel_gt *gt, bool force)if (intel_gt_is_wedged (gt))intel_gt_unset_wedged (gt); -intel_uc_sanitize (>->uc); -for_each_engine (engine, gt, id)if (engine->reset.prepare) engine->reset.prepare (engine);@@ -191,6 +189,8 @@ static void gt_sanitize(struct intel_gt *gt, bool force)__intel_engine_reset (engine, false); } +intel_uc_reset (>->uc, false); +for_each_engine (engine, gt, id)if (engine->reset.finish) engine->reset.finish (engine);@@ -243,6 +243,8 @@ int intel_gt_resume(struct intel_gt *gt) goto err_wedged; } +intel_uc_reset_finish (>->uc); +intel_rps_enable (>->rps);intel_llc_enable (>->llc);drivers/gpu/drm/i915/gt/intel_reset.c +12
-6@@ -826,6 +826,8 @@ static int gt_reset(struct intel_gt *gt, intel_engine_mask_t stalled_mask)__intel_engine_reset (engine, stalled_mask & engine->mask);local_bh_enable (); +intel_uc_reset (>->uc, true); +intel_ggtt_restore_fences (gt->ggtt);return err;@@ -850,6 +852,8 @@ static void reset_finish(struct intel_gt *gt, intel_engine_mask_t awake)if (awake & engine->mask)intel_engine_pm_put (engine); } + +intel_uc_reset_finish (>->uc); }static void nop_submit_request (struct i915_request *request)@@ -903,6 +907,7 @@ static void __intel_gt_set_wedged(struct intel_gt *gt)for_each_engine (engine, gt, id)if (engine->reset.cancel) engine->reset.cancel (engine); +intel_uc_cancel_requests (>->uc);local_bh_enable ();reset_finish (gt, awake);@@ -1191,6 +1196,9 @@ int __intel_engine_reset_bh(struct intel_engine_cs *engine, const char *msg)ENGINE_TRACE (engine,"flags=%lx\n" , gt->reset.flags);GEM_BUG_ON (!test_bit (I915_RESET_ENGINE + engine->id, >->reset.flags)); +if (intel_engine_uses_guc (engine)) +return -ENODEV ; +if (!intel_engine_pm_get_if_awake (engine))return 0 ;@@ -1201,13 +1209,10 @@ int __intel_engine_reset_bh(struct intel_engine_cs *engine, const char *msg) "Resetting %s for %s\n", engine->name, msg);atomic_inc (&engine->i915->gpu_error.reset_engine_count[engine->uabi_class]); -if (intel_engine_uses_guc (engine)) - ret =intel_guc_reset_engine (&engine->gt->uc.guc, engine); -else - ret =intel_gt_reset_engine (engine); + ret =intel_gt_reset_engine (engine);if (ret) {/* If we fail here, we expect to fallback to a global reset */ -ENGINE_TRACE (engine,"Failed to reset, err: %d\n" , ret); +ENGINE_TRACE (engine,"Failed to reset %s, err: %d\n" , engine->name, ret); goto out; }@@ -1341,7 +1346,8 @@ void intel_gt_handle_error(struct intel_gt *gt,* Try engine reset when available. We fall back to full reset if * single reset fails. */ -if (intel_has_reset_engine (gt) && !intel_gt_is_wedged (gt)) { +if (!intel_uc_uses_guc_submission (>->uc) && +intel_has_reset_engine (gt) && !intel_gt_is_wedged (gt)) {local_bh_disable ();for_each_engine_masked (engine, gt, engine_mask, tmp) {BUILD_BUG_ON (I915_RESET_MODESET >=I915_RESET_ENGINE );drivers/gpu/drm/i915/gt/uc/intel_guc.c +0
-13@@ -572,19 +572,6 @@ int intel_guc_suspend(struct intel_guc *guc)return 0 ; } -/** -* intel_guc_reset_engine() - ask GuC to reset an engine -* @guc: intel_guc structure -* @engine: engine to be reset -*/ -int intel_guc_reset_engine (struct intel_guc *guc, -struct intel_engine_cs *engine) -{ -/* XXX: to be implemented with submission interface rework */ - -return -ENODEV ; -} -/** * intel_guc_resume() - notify GuC resuming from suspend state * @guc: the guc drivers/gpu/drm/i915/gt/uc/intel_guc.h +5
-3@@ -242,14 +242,16 @@ static inline void intel_guc_disable_msg(struct intel_guc *guc, u32 mask)int intel_guc_wait_for_idle (struct intel_guc *guc,long timeout); -int intel_guc_reset_engine (struct intel_guc *guc, -struct intel_engine_cs *engine); -int intel_guc_deregister_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len);int intel_guc_sched_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len); +void intel_guc_submission_reset_prepare (struct intel_guc *guc); +void intel_guc_submission_reset (struct intel_guc *guc,bool stalled); +void intel_guc_submission_reset_finish (struct intel_guc *guc); +void intel_guc_submission_cancel_requests (struct intel_guc *guc); +void intel_guc_load_status (struct intel_guc *guc,struct drm_printer *p);#endif drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +60
-15@@ -141,7 +141,7 @@ context_wait_for_deregister_to_register(struct intel_context *ce)static inline void set_context_wait_for_deregister_to_register (struct intel_context *ce) { -/* Only should be called from guc_lrc_desc_pin() */ +/* Only should be called from guc_lrc_desc_pin() without lock */ ce->guc_state.sched_state |=SCHED_STATE_WAIT_FOR_DEREGISTER_TO_REGISTER ; }@@ -241,15 +241,31 @@ static int guc_lrc_desc_pool_create(struct intel_guc *guc)static void guc_lrc_desc_pool_destroy (struct intel_guc *guc) { + guc->lrc_desc_pool_vaddr =NULL ;i915_vma_unpin_and_release (&guc->lrc_desc_pool,I915_VMA_RELEASE_MAP ); } +static inline bool guc_submission_initialized (struct intel_guc *guc) +{ +return guc->lrc_desc_pool_vaddr !=NULL ; +} +static inline void reset_lrc_desc (struct intel_guc *guc,u32 id) { -struct guc_lrc_desc *desc =__get_lrc_desc (guc, id); +if (likely (guc_submission_initialized (guc))) { +struct guc_lrc_desc *desc =__get_lrc_desc (guc, id); +unsigned long flags; -memset (desc,0 ,sizeof (*desc)); -xa_erase_irq (&guc->context_lookup, id); +memset (desc,0 ,sizeof (*desc)); + +/* +* xarray API doesn't have xa_erase_irqsave wrapper, so calling +* the lower level functions directly. +*/ +xa_lock_irqsave (&guc->context_lookup, flags); +__xa_erase (&guc->context_lookup, id); +xa_unlock_irqrestore (&guc->context_lookup, flags); + } }static inline bool lrc_desc_registered (struct intel_guc *guc,u32 id)@@ -260,7 +276,15 @@ static inline bool lrc_desc_registered(struct intel_guc *guc, u32 id)static inline void set_lrc_desc_registered (struct intel_guc *guc,u32 id,struct intel_context *ce) { -xa_store_irq (&guc->context_lookup, id, ce,GFP_ATOMIC ); +unsigned long flags; + +/* +* xarray API doesn't have xa_save_irqsave wrapper, so calling the +* lower level functions directly. +*/ +xa_lock_irqsave (&guc->context_lookup, flags); +__xa_store (&guc->context_lookup, id, ce,GFP_ATOMIC ); +xa_unlock_irqrestore (&guc->context_lookup, flags); } static int guc_submission_send_busy_loop(struct intel_guc* guc,@@ -326,6 +350,8 @@ int intel_guc_wait_for_idle(struct intel_guc *guc, long timeout) true, timeout); } +static int guc_lrc_desc_pin (struct intel_context *ce,bool loop); +static int guc_add_request (struct intel_guc *guc,struct i915_request *rq) {int err;@@ -333,11 +359,22 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)u32 action[3 ];int len =0 ;u32 g2h_len_dw =0 ; -bool enabled =context_enabled (ce); +bool enabled;GEM_BUG_ON (!atomic_read (&ce->guc_id_ref));GEM_BUG_ON (context_guc_id_invalid (ce)); +/* +* Corner case where the GuC firmware was blown away and reloaded while +* this context was pinned. +*/ +if (unlikely (!lrc_desc_registered (guc, ce->guc_id))) { + err =guc_lrc_desc_pin (ce, false); +if (unlikely (err)) + goto out; + } + enabled =context_enabled (ce); +if (!enabled) { action[len++] =INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_SET ; action[len++] = ce->guc_id;@@ -360,6 +397,7 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)intel_context_put (ce); } +out:return err; }@@ -414,15 +452,10 @@ static int guc_dequeue_one_context(struct intel_guc *guc)if (submit) {guc_set_lrc_tail (last); resubmit: -/* -* We only check for -EBUSY here even though it is possible for -* -EDEADLK to be returned. If -EDEADLK is returned, the GuC has -* died and a full GT reset needs to be done. The hangcheck will -* eventually detect that the GuC has died and trigger this -* reset so no need to handle -EDEADLK here. -*/ ret =guc_add_request (guc, last); -if (ret == -EBUSY ) { +if (unlikely (ret == -EPIPE )) + goto deadlk; +else if (ret == -EBUSY ) {tasklet_schedule (&sched_engine->tasklet); guc->stalled_request = last;return false;@@ -432,6 +465,11 @@ static int guc_dequeue_one_context(struct intel_guc *guc) guc->stalled_request =NULL ;return submit; + +deadlk: + sched_engine->tasklet.callback =NULL ; +tasklet_disable_nosync (&sched_engine->tasklet); +return false; }static void guc_submission_tasklet (struct tasklet_struct *t)@@ -458,27 +496,166 @@ static void cs_irq_handler(struct intel_engine_cs *engine, u16 iir)intel_engine_signal_breadcrumbs (engine); } -static void guc_reset_prepare (struct intel_engine_cs *engine) +static void __guc_context_destroy (struct intel_context *ce); +static void release_guc_id (struct intel_guc *guc,struct intel_context *ce); +static void guc_signal_context_fence (struct intel_context *ce); + +static void scrub_guc_desc_for_outstanding_g2h (struct intel_guc *guc) +{ +struct intel_context *ce; +unsigned long index, flags;
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void cs_irq_handler(struct intel_engine_cs *engine, u16 iir) · line 516
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +10
-0+bool pending_disable, pending_enable, deregister, destroyed; + +xa_for_each (&guc->context_lookup, index, ce) { +/* Flush context */ +spin_lock_irqsave (&ce->guc_state.lock, flags); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + +/* +* Once we are at this point submission_disabled() is guaranteed +* to visible to all callers who set the below flags (see above
Still needs to be 'to be visible'
Quoted 77 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +77
-0+* flush and flushes in reset_prepare). If submission_disabled() +* is set, the caller shouldn't set these flags. +*/ + + destroyed =context_destroyed (ce); + pending_enable =context_pending_enable (ce); + pending_disable =context_pending_disable (ce); + deregister =context_wait_for_deregister_to_register (ce); +init_sched_state (ce); + +if (pending_enable || destroyed || deregister) { +atomic_dec (&guc->outstanding_submission_g2h); +if (deregister) +guc_signal_context_fence (ce); +if (destroyed) { +release_guc_id (guc, ce); +__guc_context_destroy (ce); + } +if (pending_enable|| deregister) +intel_context_put (ce); + } + +/* Not mutualy exclusive with above if statement. */ +if (pending_disable) { +guc_signal_context_fence (ce); +intel_context_sched_disable_unpin (ce); +atomic_dec (&guc->outstanding_submission_g2h); +intel_context_put (ce); + } + } +} + +static inline bool +submission_disabled (struct intel_guc *guc) +{ +struct i915_sched_engine *const sched_engine = guc->sched_engine; + +return unlikely (!sched_engine || + !__tasklet_is_enabled (&sched_engine->tasklet)); +} + +static void disable_submission (struct intel_guc *guc) +{ +struct i915_sched_engine *const sched_engine = guc->sched_engine; + +if (__tasklet_is_enabled (&sched_engine->tasklet)) { +GEM_BUG_ON (!guc->ct.enabled); +__tasklet_disable_sync_once (&sched_engine->tasklet); + sched_engine->tasklet.callback =NULL ; + } +} + +static void enable_submission (struct intel_guc *guc) +{ +struct i915_sched_engine *const sched_engine = guc->sched_engine; +unsigned long flags; + +spin_lock_irqsave (&guc->sched_engine->lock, flags); + sched_engine->tasklet.callback = guc_submission_tasklet; +wmb (); +if (!__tasklet_is_enabled (&sched_engine->tasklet) && +__tasklet_enable (&sched_engine->tasklet)) { +GEM_BUG_ON (!guc->ct.enabled); + +/* And kick in case we missed a new request submission. */ +tasklet_hi_schedule (&sched_engine->tasklet); + } +spin_unlock_irqrestore (&guc->sched_engine->lock, flags); +} + +static void guc_flush_submissions (struct intel_guc *guc) +{ +struct i915_sched_engine *const sched_engine = guc->sched_engine; +unsigned long flags; + +spin_lock_irqsave (&sched_engine->lock, flags); +spin_unlock_irqrestore (&sched_engine->lock, flags);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void cs_irq_handler(struct intel_engine_cs *engine, u16 iir) · line 602
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +8
-1+} + +void intel_guc_submission_reset_prepare (struct intel_guc *guc) { -ENGINE_TRACE (engine,"\n" ); +int i; + +if (unlikely (!guc_submission_initialized (guc))) +/* Reset called during driver load? GuC not yet initialised! */ +return ;
Still think this should have braces. Maybe not syntactically required but just looks wrong.
Quoted 183 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +137
-24+ +disable_submission (guc); + guc->interrupts.disable (guc); + +/* Flush IRQ handler */ +spin_lock_irq (&guc_to_gt (guc)->irq_lock); +spin_unlock_irq (&guc_to_gt (guc)->irq_lock); + +guc_flush_submissions (guc);/* -* Prevent request submission to the hardware until we have -* completed the reset in i915_gem_reset_finish(). If a request -* is completed by one engine, it may then queue a request -* to a second via its execlists->tasklet *just* as we are -* calling engine->init_hw() and also writing the ELSP. -* Turning off the execlists->tasklet until the reset is over -* prevents the race. -*/ -__tasklet_disable_sync_once (&engine->sched_engine->tasklet); +* Handle any outstanding G2Hs before reset. Call IRQ handler directly +* each pass as interrupt have been disabled. We always scrub for +* outstanding G2H as it is possible for outstanding_submission_g2h to +* be incremented after the context state update. +*/ +for (i =0 ; i <4 &&atomic_read (&guc->outstanding_submission_g2h); ++i) { +intel_guc_to_host_event_handler (guc); +#define wait_for_reset (guc, wait_var) \ + guc_wait_for_pending_msg(guc, wait_var, false, (HZ / 20)) +do { +wait_for_reset (guc, &guc->outstanding_submission_g2h); + }while (!list_empty (&guc->ct.requests.incoming)); + } +scrub_guc_desc_for_outstanding_g2h (guc); +} + +static struct intel_engine_cs * +guc_virtual_get_sibling (struct intel_engine_cs *ve,unsigned int sibling) +{ +struct intel_engine_cs *engine; +intel_engine_mask_t tmp, mask = ve->mask; +unsigned int num_siblings =0 ; + +for_each_engine_masked (engine, ve->gt, mask, tmp) +if (num_siblings++ == sibling) +return engine; + +return NULL ; +} + +static inline struct intel_engine_cs * +__context_to_physical_engine (struct intel_context *ce) +{ +struct intel_engine_cs *engine = ce->engine; + +if (intel_engine_is_virtual (engine)) + engine =guc_virtual_get_sibling (engine,0 ); + +return engine; } -static void guc_reset_state (struct intel_context *ce, -struct intel_engine_cs *engine, -u32 head, -bool scrub) +static void guc_reset_state (struct intel_context *ce,u32 head,bool scrub) { +struct intel_engine_cs *engine =__context_to_physical_engine (ce); +GEM_BUG_ON (!intel_context_is_pinned (ce));/* @@ -496,42 +673,147 @@ static void guc_reset_state(struct intel_context *ce,lrc_update_regs (ce, engine, head); } -static void guc_reset_rewind (struct intel_engine_cs *engine,bool stalled) +static void guc_reset_nop (struct intel_engine_cs *engine) { -struct intel_engine_execlists *const execlists = &engine->execlists; -struct i915_request *rq; +} + +static void guc_rewind_nop (struct intel_engine_cs *engine,bool stalled) +{ +} + +static void +__unwind_incomplete_requests (struct intel_context *ce) +{ +struct i915_request *rq, *rn; +struct list_head *pl; +int prio =I915_PRIORITY_INVALID ; +struct i915_sched_engine *const sched_engine = + ce->engine->sched_engine;unsigned long flags; -spin_lock_irqsave (&engine->sched_engine->lock, flags); +spin_lock_irqsave (&sched_engine->lock, flags); +spin_lock (&ce->guc_active.lock); +list_for_each_entry_safe (rq, rn, + &ce->guc_active.requests, + sched.link) { +if (i915_request_completed (rq)) +continue ; + +list_del_init (&rq->sched.link); +spin_unlock (&ce->guc_active.lock); + +__i915_request_unsubmit (rq); + +/* Push the request back into the queue for later resubmission. */ +GEM_BUG_ON (rq_prio (rq) ==I915_PRIORITY_INVALID ); +if (rq_prio (rq) != prio) { + prio =rq_prio (rq); + pl =i915_sched_lookup_priolist (sched_engine, prio); + } +GEM_BUG_ON (i915_sched_engine_is_empty (sched_engine)); -/* Push back any incomplete requests for replay after the reset. */ - rq =execlists_unwind_incomplete_requests (execlists); -if (!rq) - goto out_unlock; +list_add_tail (&rq->sched.link, pl); +set_bit (I915_FENCE_FLAG_PQUEUE , &rq->fence.flags); + +spin_lock (&ce->guc_active.lock); + } +spin_unlock (&ce->guc_active.lock); +spin_unlock_irqrestore (&sched_engine->lock, flags); +} + +static struct i915_request *context_find_active_request (struct intel_context *ce) +{ +struct i915_request *rq, *active =NULL ; +unsigned long flags; + +spin_lock_irqsave (&ce->guc_active.lock, flags); +list_for_each_entry_reverse (rq, &ce->guc_active.requests, + sched.link) { +if (i915_request_completed (rq)) +break ; + + active = rq; + } +spin_unlock_irqrestore (&ce->guc_active.lock, flags); + +return active; +} + +static void __guc_reset_context (struct intel_context *ce,bool stalled) +{ +struct i915_request *rq; +u32 head; + +/* +* GuC will implicitly mark the context as non-schedulable +* when it sends the reset notification. Make sure our state +* reflects this change. The context will be marked enabled +* on resubmission. +*/ +clr_context_enabled (ce); + + rq =context_find_active_request (ce); +if (!rq) { + head = ce->ring->tail; + stalled = false; + 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);__i915_request_reset (rq, stalled); -guc_reset_state (rq->context, engine, rq->head, stalled); -out_unlock: -spin_unlock_irqrestore (&engine->sched_engine->lock, flags); +out_replay: +guc_reset_state (ce, head, stalled); +__unwind_incomplete_requests (ce);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void guc_reset_state(struct intel_context *ce, · line 781
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +10
-0+} + +void intel_guc_submission_reset (struct intel_guc *guc,bool stalled) +{ +struct intel_context *ce; +unsigned long index; + +if (unlikely (!guc_submission_initialized (guc))) +/* Reset called during driver load? GuC not yet initialised! */ +return ;
And again.
Quoted 81 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +31
-17+ +xa_for_each (&guc->context_lookup, index, ce) +if (intel_context_is_pinned (ce)) +__guc_reset_context (ce, stalled); + +/* GuC is blown away, drop all references to contexts */ +xa_destroy (&guc->context_lookup); +} + +static void guc_cancel_context_requests (struct intel_context *ce) +{ +struct i915_sched_engine *sched_engine =ce_to_guc (ce)->sched_engine; +struct i915_request *rq; +unsigned long flags; + +/* Mark all executing requests as skipped. */ +spin_lock_irqsave (&sched_engine->lock, flags); +spin_lock (&ce->guc_active.lock); +list_for_each_entry (rq, &ce->guc_active.requests, sched.link) +i915_request_put (i915_request_mark_eio (rq)); +spin_unlock (&ce->guc_active.lock); +spin_unlock_irqrestore (&sched_engine->lock, flags); } -static void guc_reset_cancel (struct intel_engine_cs *engine) +static void +guc_cancel_sched_engine_requests (struct i915_sched_engine *sched_engine) { -struct i915_sched_engine *const sched_engine = engine->sched_engine;struct i915_request *rq, *rn;struct rb_node *rb;unsigned long flags;/* Can be called during boot if GuC fails to load */ -if (!engine->gt) +if (!sched_engine)return ; -ENGINE_TRACE (engine,"\n" ); -/* * Before we call engine->cancel_requests(), we should have exclusive * access to the submission state. This is arranged for us by the @@ -548,21 +830,16 @@ static void guc_reset_cancel(struct intel_engine_cs *engine)*/ spin_lock_irqsave (&sched_engine->lock, flags); -/* Mark all executing requests as skipped. */ -list_for_each_entry (rq, &sched_engine->requests, sched.link) { -i915_request_set_error_once (rq, -EIO ); -i915_request_mark_complete (rq); - } -/* Flush the queued requests to the timeline list (for retiring). */ while ((rb =rb_first_cached (&sched_engine->queue))) {struct i915_priolist *p =to_priolist (rb);priolist_for_each_request_consume (rq, rn, p) {list_del_init (&rq->sched.link); +__i915_request_submit (rq); -dma_fence_set_error (&rq->fence, -EIO ); -i915_request_mark_complete (rq); + +i915_request_put (i915_request_mark_eio (rq)); }rb_erase_cached (&p->node, &sched_engine->queue);@@ -577,14 +854,38 @@ static void guc_reset_cancel(struct intel_engine_cs *engine)spin_unlock_irqrestore (&sched_engine->lock, flags); } -static void guc_reset_finish (struct intel_engine_cs *engine) +void intel_guc_submission_cancel_requests (struct intel_guc *guc) { -if (__tasklet_enable (&engine->sched_engine->tasklet)) -/* And kick in case we missed a new request submission. */ -tasklet_hi_schedule (&engine->sched_engine->tasklet); +struct intel_context *ce; +unsigned long index;
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void guc_reset_cancel(struct intel_engine_cs *engine) · line 869
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +8
-2-ENGINE_TRACE (engine,"depth->%d\n" , -atomic_read (&engine->sched_engine->tasklet.count)); +xa_for_each (&guc->context_lookup, index, ce) +if (intel_context_is_pinned (ce)) +guc_cancel_context_requests (ce); + +guc_cancel_sched_engine_requests (guc->sched_engine); + +/* GuC is blown away, drop all references to contexts */ +xa_destroy (&guc->context_lookup);
And again, most of this code is common to 'intel_guc_submission_reset' so could be moved to a shared helper function.
John.
Quoted 434 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +112
-50+} + +void intel_guc_submission_reset_finish (struct intel_guc *guc) +{ +/* Reset called during driver load or during wedge? */ +if (unlikely (!guc_submission_initialized (guc) || +test_bit (I915_WEDGED , &guc_to_gt (guc)->reset.flags))) +return ; + +/* +* Technically possible for either of these values to be non-zero here, +* but very unlikely + harmless. Regardless let's add a warn so we can +* see in CI if this happens frequently / a precursor to taking down the +* machine. +*/ +GEM_WARN_ON (atomic_read (&guc->outstanding_submission_g2h)); +atomic_set (&guc->outstanding_submission_g2h,0 ); + +enable_submission (guc); }/* @@ -651,6 +952,9 @@ static int guc_bypass_tasklet_submit(struct intel_guc *guc,else trace_i915_request_guc_submit (rq); +if (unlikely (ret == -EPIPE )) +disable_submission (guc); +return ret; }@@ -663,7 +967,8 @@ static void guc_submit_request(struct i915_request *rq)/* Will be called from irq-context when using foreign fences. */ spin_lock_irqsave (&sched_engine->lock, flags); -if (guc->stalled_request || !i915_sched_engine_is_empty (sched_engine)) +if (submission_disabled (guc) || guc->stalled_request || + !i915_sched_engine_is_empty (sched_engine))queue_request (sched_engine, rq,rq_prio (rq));else if (guc_bypass_tasklet_submit (guc, rq) == -EBUSY )tasklet_hi_schedule (&sched_engine->tasklet);@@ -800,7 +1105,8 @@ static void unpin_guc_id(struct intel_guc *guc, struct intel_context *ce)static int __guc_action_register_context (struct intel_guc *guc,u32 guc_id, -u32 offset) +u32 offset, +bool loop) {u32 action[] = {INTEL_GUC_ACTION_REGISTER_CONTEXT ,@@ -809,10 +1115,10 @@ static int __guc_action_register_context(struct intel_guc *guc, };return guc_submission_send_busy_loop (guc, action,ARRAY_SIZE (action), -0 , true); +0 , loop); } -static int register_context (struct intel_context *ce) +static int register_context (struct intel_context *ce,bool loop) {struct intel_guc *guc =ce_to_guc (ce); u32 offset = intel_guc_ggtt_offset(guc, guc->lrc_desc_pool) +@@ -820,11 +1126,12 @@ static int register_context(struct intel_context *ce)trace_intel_context_register (ce); -return __guc_action_register_context (guc, ce->guc_id, offset); +return __guc_action_register_context (guc, ce->guc_id, offset, loop); }static int __guc_action_deregister_context (struct intel_guc *guc, -u32 guc_id) +u32 guc_id, +bool loop) {u32 action[] = {INTEL_GUC_ACTION_DEREGISTER_CONTEXT ,@@ -833,16 +1140,16 @@ static int __guc_action_deregister_context(struct intel_guc *guc,return guc_submission_send_busy_loop (guc, action,ARRAY_SIZE (action),G2H_LEN_DW_DEREGISTER_CONTEXT , - true); + loop); } -static int deregister_context (struct intel_context *ce,u32 guc_id) +static int deregister_context (struct intel_context *ce,u32 guc_id,bool loop) {struct intel_guc *guc =ce_to_guc (ce);trace_intel_context_deregister (ce); -return __guc_action_deregister_context (guc, guc_id); +return __guc_action_deregister_context (guc, guc_id, loop); }static intel_engine_mask_t adjust_engine_mask (u8 class,intel_engine_mask_t mask)@@ -871,7 +1178,7 @@ static void guc_context_policy_init(struct intel_engine_cs *engine, desc->preemption_timeout =CONTEXT_POLICY_DEFAULT_PREEMPTION_TIME_US ; } -static int guc_lrc_desc_pin (struct intel_context *ce) +static int guc_lrc_desc_pin (struct intel_context *ce,bool loop) {struct intel_runtime_pm *runtime_pm = &ce->engine->gt->i915->runtime_pm;@@ -917,18 +1224,45 @@ static int guc_lrc_desc_pin(struct intel_context *ce)*/ if (context_registered) {trace_intel_context_steal_guc_id (ce); -set_context_wait_for_deregister_to_register (ce); -intel_context_get (ce); +if (!loop) { +set_context_wait_for_deregister_to_register (ce); +intel_context_get (ce); + }else { +bool disabled; +unsigned long flags; + +/* Seal race with Reset */ +spin_lock_irqsave (&ce->guc_state.lock, flags); + disabled =submission_disabled (guc); +if (likely (!disabled)) { +set_context_wait_for_deregister_to_register (ce); +intel_context_get (ce); + } +spin_unlock_irqrestore (&ce->guc_state.lock, flags); +if (unlikely (disabled)) { +reset_lrc_desc (guc, desc_idx); +return 0 ;/* Will get registered later */ + } + }/* * If stealing the guc_id, this ce has the same guc_id as the * context whos guc_id was stole. */ with_intel_runtime_pm (runtime_pm, wakeref) - ret = deregister_context(ce, ce->guc_id); + ret = deregister_context(ce, ce->guc_id, loop); +if (unlikely (ret == -EBUSY )) { +clr_context_wait_for_deregister_to_register (ce); +intel_context_put (ce); + }else if (unlikely (ret == -ENODEV )) + ret =0 ;/* Will get registered later */ }else {with_intel_runtime_pm (runtime_pm, wakeref) - ret = register_context(ce); + ret = register_context(ce, loop); +if (unlikely (ret == -EBUSY )) +reset_lrc_desc (guc, desc_idx); +else if (unlikely (ret == -ENODEV )) + ret =0 ;/* Will get registered later */ }return ret;@@ -991,7 +1325,6 @@ static void __guc_context_sched_disable(struct intel_guc *guc,GEM_BUG_ON (guc_id ==GUC_INVALID_LRC_ID );trace_intel_context_sched_disable (ce); -intel_context_get (ce);guc_submission_send_busy_loop (guc, action,ARRAY_SIZE (action),G2H_LEN_DW_SCHED_CONTEXT_MODE_SET , true);@@ -1003,6 +1336,7 @@ static u16 prep_context_pending_disable(struct intel_context *ce)set_context_pending_disable (ce);clr_context_enabled (ce); +intel_context_get (ce);return ce->guc_id; }@@ -1015,7 +1349,7 @@ static void guc_context_sched_disable(struct intel_context *ce)u16 guc_id;intel_wakeref_t wakeref; -if (context_guc_id_invalid (ce) || +if (submission_disabled (guc) ||context_guc_id_invalid (ce) || !lrc_desc_registered (guc, ce->guc_id)) {clr_context_enabled (ce); goto unpin;@@ -1036,6 +1370,7 @@ static void guc_context_sched_disable(struct intel_context *ce)* a request before we set the 'context_pending_disable' flag here. */ if (unlikely (atomic_add_unless (&ce->pin_count,-2 ,2 ))) { +spin_unlock_irqrestore (&ce->guc_state.lock, flags);return ; } guc_id =prep_context_pending_disable (ce);@@ -1052,19 +1387,13 @@ static void guc_context_sched_disable(struct intel_context *ce)static inline void guc_lrc_desc_unpin (struct intel_context *ce) { -struct intel_engine_cs *engine = ce->engine; -struct intel_guc *guc = &engine->gt->uc.guc; -unsigned long flags; +struct intel_guc *guc =ce_to_guc (ce);GEM_BUG_ON (!lrc_desc_registered (guc, ce->guc_id));GEM_BUG_ON (ce !=__get_context (guc, ce->guc_id));GEM_BUG_ON (context_enabled (ce)); -spin_lock_irqsave (&ce->guc_state.lock, flags); -set_context_destroyed (ce); -spin_unlock_irqrestore (&ce->guc_state.lock, flags); - -deregister_context (ce, ce->guc_id); +deregister_context (ce, ce->guc_id, true); }static void __guc_context_destroy (struct intel_context *ce)@@ -1092,13 +1421,15 @@ static void guc_context_destroy(struct kref *kref)struct intel_guc *guc = &ce->engine->gt->uc.guc;intel_wakeref_t wakeref;unsigned long flags; +bool disabled;/* * If the guc_id is invalid this context has been stolen and we can free * it immediately. Also can be freed immediately if the context is not * registered with the GuC. */ -if (context_guc_id_invalid (ce) || +if (submission_disabled (guc) || +context_guc_id_invalid (ce) || !lrc_desc_registered (guc, ce->guc_id)) {release_guc_id (guc, ce);__guc_context_destroy (ce);@@ -1125,6 +1456,18 @@ static void guc_context_destroy(struct kref *kref)list_del_init (&ce->guc_id_link);spin_unlock_irqrestore (&guc->contexts_lock, flags); +/* Seal race with Reset */ +spin_lock_irqsave (&ce->guc_state.lock, flags); + disabled =submission_disabled (guc); +if (likely (!disabled)) +set_context_destroyed (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); +if (unlikely (disabled)) { +release_guc_id (guc, ce); +__guc_context_destroy (ce); +return ; + } +/* * We defer GuC context deregistration until the context is destroyed * in order to save on CTBs. With this optimization ideally we only need @@ -1212,8 +1555,6 @@ static void guc_signal_context_fence(struct intel_context *ce) {unsigned long flags; -GEM_BUG_ON (!context_wait_for_deregister_to_register (ce)); -spin_lock_irqsave (&ce->guc_state.lock, flags);clr_context_wait_for_deregister_to_register (ce);__guc_signal_context_fence (ce);@@ -1222,8 +1563,9 @@ static void guc_signal_context_fence(struct intel_context *ce)static bool context_needs_register (struct intel_context *ce,bool new_guc_id) { -return new_guc_id ||test_bit (CONTEXT_LRCA_DIRTY , &ce->flags) || - !lrc_desc_registered (ce_to_guc (ce), ce->guc_id); +return (new_guc_id ||test_bit (CONTEXT_LRCA_DIRTY , &ce->flags) || + !lrc_desc_registered (ce_to_guc (ce), ce->guc_id)) && + !submission_disabled (ce_to_guc (ce)); }static int guc_request_alloc (struct i915_request *rq)@@ -1281,8 +1623,12 @@ static int guc_request_alloc(struct i915_request *rq)if (unlikely (ret <0 ))return ret;;if (context_needs_register (ce, !!ret)) { - ret =guc_lrc_desc_pin (ce); + ret =guc_lrc_desc_pin (ce, true);if (unlikely (ret)) {/* unwind */ +if (ret == -EPIPE ) { +disable_submission (guc); + goto out;/* GPU will be reset */ + }atomic_dec (&ce->guc_id_ref);unpin_guc_id (guc, ce);return ret;@@ -1319,20 +1665,6 @@ static int guc_request_alloc(struct i915_request *rq)return 0 ; } -static struct intel_engine_cs * -guc_virtual_get_sibling (struct intel_engine_cs *ve,unsigned int sibling) -{ -struct intel_engine_cs *engine; -intel_engine_mask_t tmp, mask = ve->mask; -unsigned int num_siblings =0 ; - -for_each_engine_masked (engine, ve->gt, mask, tmp) -if (num_siblings++ == sibling) -return engine; - -return NULL ; -} -static int guc_virtual_context_pre_pin (struct intel_context *ce,struct i915_gem_ww_ctx *ww,void **vaddr)@@ -1528,7 +1860,7 @@ static inline void guc_kernel_context_pin(struct intel_guc *guc, {if (context_guc_id_invalid (ce))pin_guc_id (guc, ce); -guc_lrc_desc_pin (ce); +guc_lrc_desc_pin (ce, true); }static inline void guc_init_lrc_mapping (struct intel_guc *guc)@@ -1599,10 +1931,10 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->sched_engine->schedule = i915_schedule; - engine->reset.prepare = guc_reset_prepare; - engine->reset.rewind = guc_reset_rewind; - engine->reset.cancel = guc_reset_cancel; - engine->reset.finish = guc_reset_finish; + engine->reset.prepare = guc_reset_nop; + engine->reset.rewind = guc_rewind_nop; + engine->reset.cancel = guc_reset_nop; + engine->reset.finish = guc_reset_nop; engine->emit_flush = gen8_emit_flush_xcs; engine->emit_init_breadcrumb = gen8_emit_init_breadcrumb;@@ -1651,6 +1983,17 @@ static inline void guc_default_irqs(struct intel_engine_cs *engine)intel_engine_set_irq_handler (engine, cs_irq_handler); } +static void guc_sched_engine_destroy (struct kref *kref) +{ +struct i915_sched_engine *sched_engine = +container_of (kref,typeof (*sched_engine), ref); +struct intel_guc *guc = sched_engine->private_data; + + guc->sched_engine =NULL ; +tasklet_kill (&sched_engine->tasklet);/* flush the callback */ +kfree (sched_engine); +} +int intel_guc_submission_setup (struct intel_engine_cs *engine) {struct drm_i915_private *i915 = engine->i915;@@ -1669,6 +2012,7 @@ int intel_guc_submission_setup(struct intel_engine_cs *engine) guc->sched_engine->schedule = i915_schedule; guc->sched_engine->private_data = guc; + guc->sched_engine->destroy = guc_sched_engine_destroy;tasklet_setup (&guc->sched_engine->tasklet, guc_submission_tasklet); }@@ -1775,7 +2119,7 @@ int intel_guc_deregister_done_process_msg(struct intel_guc *guc,* register this context. */ with_intel_runtime_pm(runtime_pm, wakeref) -register_context (ce); +register_context (ce, true);guc_signal_context_fence (ce);intel_context_put (ce); }else if (context_destroyed (ce)) {drivers/gpu/drm/i915/gt/uc/intel_uc.c +38
-1@@ -565,12 +565,49 @@ void intel_uc_reset_prepare(struct intel_uc *uc) {struct intel_guc *guc = &uc->guc; -if (!intel_guc_is_ready (guc)) + +/* Nothing to do if GuC isn't supported */ +if (!intel_uc_supports_guc (uc))return ; +/* Firmware expected to be running when this function is called */ +if (!intel_guc_is_ready (guc)) + goto sanitize; + +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset_prepare (guc); + +sanitize:__uc_sanitize (uc); } +void intel_uc_reset (struct intel_uc *uc,bool stalled) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware can not be running when this function is called */ +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset (guc, stalled); +} + +void intel_uc_reset_finish (struct intel_uc *uc) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware expected to be running when this function is called */ +if (intel_guc_is_fw_running (guc) &&intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset_finish (guc); +} + +void intel_uc_cancel_requests (struct intel_uc *uc) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware can not be running when this function is called */ +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_cancel_requests (guc); +} +void intel_uc_runtime_suspend (struct intel_uc *uc) {struct intel_guc *guc = &uc->guc;drivers/gpu/drm/i915/gt/uc/intel_uc.h +3
-0@@ -37,6 +37,9 @@ void intel_uc_driver_late_release(struct intel_uc *uc);void intel_uc_driver_remove (struct intel_uc *uc);void intel_uc_init_mmio (struct intel_uc *uc);void intel_uc_reset_prepare (struct intel_uc *uc); +void intel_uc_reset (struct intel_uc *uc,bool stalled); +void intel_uc_reset_finish (struct intel_uc *uc); +void intel_uc_cancel_requests (struct intel_uc *uc);void intel_uc_suspend (struct intel_uc *uc);void intel_uc_runtime_suspend (struct intel_uc *uc);int intel_uc_resume (struct intel_uc *uc);
Re: [PATCH 26/51] drm/i915/guc: Reset implementation for new GuC interface
Quoted 11 lines · John Harrison
Quoted 10 lines · Matthew Brost
Reset implementation for new GuC interface. This is the legacy reset implementation which is called when the i915 owns the engine hang check. Future patches will offload the engine hang check to GuC but we will continue to maintain this legacy path as a fallback and this code path is also required if the GuC dies.
With the new GuC interface it is not possible to reset individual engines - it is only possible to reset the GPU entirely. This patch forces an entire chip reset if any engine hangs.
Quoted 8 lines
Quoted 6 lines
v2:
(Michal)
- Check for-EPIPErather than-EIO(CT deadlock/corrupt check)
v3:
(John H)
- Split into a series of smaller patchesWhile the split happened, it doesn't look like any of the other comments were address. Repeated below for clarity. Also, Tvrtko has a bunch of
Opps, I thought you asked for the split and didn't realize you had comments.
I thought I addressed Tvrtko's comments but a I'll circle back to them.
Quoted 283 lines
outstanding comments too.
Quoted 281 lines
Cc John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gt/intel_gt_pm.c | 6 +- drivers/gpu/drm/i915/gt/intel_reset.c | 18 +- drivers/gpu/drm/i915/gt/uc/intel_guc.c | 13 - drivers/gpu/drm/i915/gt/uc/intel_guc.h | 8 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 562 ++++++++++++++---- drivers/gpu/drm/i915/gt/uc/intel_uc.c | 39 +- drivers/gpu/drm/i915/gt/uc/intel_uc.h | 3 + 7 files changed, 515 insertions(+), 134 deletions(-)drivers/gpu/drm/i915/gt/intel_gt_pm.c +4
-2@@ -174,8 +174,6 @@ static void gt_sanitize(struct intel_gt *gt, bool force)if (intel_gt_is_wedged (gt))intel_gt_unset_wedged (gt); -intel_uc_sanitize (>->uc); -for_each_engine (engine, gt, id)if (engine->reset.prepare) engine->reset.prepare (engine);@@ -191,6 +189,8 @@ static void gt_sanitize(struct intel_gt *gt, bool force)__intel_engine_reset (engine, false); } +intel_uc_reset (>->uc, false); +for_each_engine (engine, gt, id)if (engine->reset.finish) engine->reset.finish (engine);@@ -243,6 +243,8 @@ int intel_gt_resume(struct intel_gt *gt) goto err_wedged; } +intel_uc_reset_finish (>->uc); +intel_rps_enable (>->rps);intel_llc_enable (>->llc);drivers/gpu/drm/i915/gt/intel_reset.c +12
-6@@ -826,6 +826,8 @@ static int gt_reset(struct intel_gt *gt, intel_engine_mask_t stalled_mask)__intel_engine_reset (engine, stalled_mask & engine->mask);local_bh_enable (); +intel_uc_reset (>->uc, true); +intel_ggtt_restore_fences (gt->ggtt);return err;@@ -850,6 +852,8 @@ static void reset_finish(struct intel_gt *gt, intel_engine_mask_t awake)if (awake & engine->mask)intel_engine_pm_put (engine); } + +intel_uc_reset_finish (>->uc); }static void nop_submit_request (struct i915_request *request)@@ -903,6 +907,7 @@ static void __intel_gt_set_wedged(struct intel_gt *gt)for_each_engine (engine, gt, id)if (engine->reset.cancel) engine->reset.cancel (engine); +intel_uc_cancel_requests (>->uc);local_bh_enable ();reset_finish (gt, awake);@@ -1191,6 +1196,9 @@ int __intel_engine_reset_bh(struct intel_engine_cs *engine, const char *msg)ENGINE_TRACE (engine,"flags=%lx\n" , gt->reset.flags);GEM_BUG_ON (!test_bit (I915_RESET_ENGINE + engine->id, >->reset.flags)); +if (intel_engine_uses_guc (engine)) +return -ENODEV ; +if (!intel_engine_pm_get_if_awake (engine))return 0 ;@@ -1201,13 +1209,10 @@ int __intel_engine_reset_bh(struct intel_engine_cs *engine, const char *msg) "Resetting %s for %s\n", engine->name, msg);atomic_inc (&engine->i915->gpu_error.reset_engine_count[engine->uabi_class]); -if (intel_engine_uses_guc (engine)) - ret =intel_guc_reset_engine (&engine->gt->uc.guc, engine); -else - ret =intel_gt_reset_engine (engine); + ret =intel_gt_reset_engine (engine);if (ret) {/* If we fail here, we expect to fallback to a global reset */ -ENGINE_TRACE (engine,"Failed to reset, err: %d\n" , ret); +ENGINE_TRACE (engine,"Failed to reset %s, err: %d\n" , engine->name, ret); goto out; }@@ -1341,7 +1346,8 @@ void intel_gt_handle_error(struct intel_gt *gt,* Try engine reset when available. We fall back to full reset if * single reset fails. */ -if (intel_has_reset_engine (gt) && !intel_gt_is_wedged (gt)) { +if (!intel_uc_uses_guc_submission (>->uc) && +intel_has_reset_engine (gt) && !intel_gt_is_wedged (gt)) {local_bh_disable ();for_each_engine_masked (engine, gt, engine_mask, tmp) {BUILD_BUG_ON (I915_RESET_MODESET >=I915_RESET_ENGINE );drivers/gpu/drm/i915/gt/uc/intel_guc.c +0
-13@@ -572,19 +572,6 @@ int intel_guc_suspend(struct intel_guc *guc)return 0 ; } -/** -* intel_guc_reset_engine() - ask GuC to reset an engine -* @guc: intel_guc structure -* @engine: engine to be reset -*/ -int intel_guc_reset_engine (struct intel_guc *guc, -struct intel_engine_cs *engine) -{ -/* XXX: to be implemented with submission interface rework */ - -return -ENODEV ; -} -/** * intel_guc_resume() - notify GuC resuming from suspend state * @guc: the guc drivers/gpu/drm/i915/gt/uc/intel_guc.h +5
-3@@ -242,14 +242,16 @@ static inline void intel_guc_disable_msg(struct intel_guc *guc, u32 mask)int intel_guc_wait_for_idle (struct intel_guc *guc,long timeout); -int intel_guc_reset_engine (struct intel_guc *guc, -struct intel_engine_cs *engine); -int intel_guc_deregister_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len);int intel_guc_sched_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len); +void intel_guc_submission_reset_prepare (struct intel_guc *guc); +void intel_guc_submission_reset (struct intel_guc *guc,bool stalled); +void intel_guc_submission_reset_finish (struct intel_guc *guc); +void intel_guc_submission_cancel_requests (struct intel_guc *guc); +void intel_guc_load_status (struct intel_guc *guc,struct drm_printer *p);#endif drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +63
-15@@ -141,7 +141,7 @@ context_wait_for_deregister_to_register(struct intel_context *ce)static inline void set_context_wait_for_deregister_to_register (struct intel_context *ce) { -/* Only should be called from guc_lrc_desc_pin() */ +/* Only should be called from guc_lrc_desc_pin() without lock */ ce->guc_state.sched_state |=SCHED_STATE_WAIT_FOR_DEREGISTER_TO_REGISTER ; }@@ -241,15 +241,31 @@ static int guc_lrc_desc_pool_create(struct intel_guc *guc)static void guc_lrc_desc_pool_destroy (struct intel_guc *guc) { + guc->lrc_desc_pool_vaddr =NULL ;i915_vma_unpin_and_release (&guc->lrc_desc_pool,I915_VMA_RELEASE_MAP ); } +static inline bool guc_submission_initialized (struct intel_guc *guc) +{ +return guc->lrc_desc_pool_vaddr !=NULL ; +} +static inline void reset_lrc_desc (struct intel_guc *guc,u32 id) { -struct guc_lrc_desc *desc =__get_lrc_desc (guc, id); +if (likely (guc_submission_initialized (guc))) { +struct guc_lrc_desc *desc =__get_lrc_desc (guc, id); +unsigned long flags; -memset (desc,0 ,sizeof (*desc)); -xa_erase_irq (&guc->context_lookup, id); +memset (desc,0 ,sizeof (*desc)); + +/* +* xarray API doesn't have xa_erase_irqsave wrapper, so calling +* the lower level functions directly. +*/ +xa_lock_irqsave (&guc->context_lookup, flags); +__xa_erase (&guc->context_lookup, id); +xa_unlock_irqrestore (&guc->context_lookup, flags); + } }static inline bool lrc_desc_registered (struct intel_guc *guc,u32 id)@@ -260,7 +276,15 @@ static inline bool lrc_desc_registered(struct intel_guc *guc, u32 id)static inline void set_lrc_desc_registered (struct intel_guc *guc,u32 id,struct intel_context *ce) { -xa_store_irq (&guc->context_lookup, id, ce,GFP_ATOMIC ); +unsigned long flags; + +/* +* xarray API doesn't have xa_save_irqsave wrapper, so calling the +* lower level functions directly. +*/ +xa_lock_irqsave (&guc->context_lookup, flags); +__xa_store (&guc->context_lookup, id, ce,GFP_ATOMIC ); +xa_unlock_irqrestore (&guc->context_lookup, flags); } static int guc_submission_send_busy_loop(struct intel_guc* guc,@@ -326,6 +350,8 @@ int intel_guc_wait_for_idle(struct intel_guc *guc, long timeout) true, timeout); } +static int guc_lrc_desc_pin (struct intel_context *ce,bool loop); +static int guc_add_request (struct intel_guc *guc,struct i915_request *rq) {int err;@@ -333,11 +359,22 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)u32 action[3 ];int len =0 ;u32 g2h_len_dw =0 ; -bool enabled =context_enabled (ce); +bool enabled;GEM_BUG_ON (!atomic_read (&ce->guc_id_ref));GEM_BUG_ON (context_guc_id_invalid (ce)); +/* +* Corner case where the GuC firmware was blown away and reloaded while +* this context was pinned. +*/ +if (unlikely (!lrc_desc_registered (guc, ce->guc_id))) { + err =guc_lrc_desc_pin (ce, false); +if (unlikely (err)) + goto out; + } + enabled =context_enabled (ce); +if (!enabled) { action[len++] =INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_SET ; action[len++] = ce->guc_id;@@ -360,6 +397,7 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)intel_context_put (ce); } +out:return err; }@@ -414,15 +452,10 @@ static int guc_dequeue_one_context(struct intel_guc *guc)if (submit) {guc_set_lrc_tail (last); resubmit: -/* -* We only check for -EBUSY here even though it is possible for -* -EDEADLK to be returned. If -EDEADLK is returned, the GuC has -* died and a full GT reset needs to be done. The hangcheck will -* eventually detect that the GuC has died and trigger this -* reset so no need to handle -EDEADLK here. -*/ ret =guc_add_request (guc, last); -if (ret == -EBUSY ) { +if (unlikely (ret == -EPIPE )) + goto deadlk; +else if (ret == -EBUSY ) {tasklet_schedule (&sched_engine->tasklet); guc->stalled_request = last;return false;@@ -432,6 +465,11 @@ static int guc_dequeue_one_context(struct intel_guc *guc) guc->stalled_request =NULL ;return submit; + +deadlk: + sched_engine->tasklet.callback =NULL ; +tasklet_disable_nosync (&sched_engine->tasklet); +return false; }static void guc_submission_tasklet (struct tasklet_struct *t)@@ -458,27 +496,166 @@ static void cs_irq_handler(struct intel_engine_cs *engine, u16 iir)intel_engine_signal_breadcrumbs (engine); } -static void guc_reset_prepare (struct intel_engine_cs *engine) +static void __guc_context_destroy (struct intel_context *ce); +static void release_guc_id (struct intel_guc *guc,struct intel_context *ce); +static void guc_signal_context_fence (struct intel_context *ce); + +static void scrub_guc_desc_for_outstanding_g2h (struct intel_guc *guc) +{ +struct intel_context *ce; +unsigned long index, flags; +bool pending_disable, pending_enable, deregister, destroyed; + +xa_for_each (&guc->context_lookup, index, ce) {
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void cs_irq_handler(struct intel_engine_cs *engine, u16 iir) · line 516
Quoted 7 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +7
-0+/* Flush context */ +spin_lock_irqsave (&ce->guc_state.lock, flags); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + +/* +* Once we are at this point submission_disabled() is guaranteed +* to visible to all callers who set the below flags (see above Still needs to be 'to be visible'
Fixed.
Quoted 81 lines
Quoted 81 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +80
-0+* flush and flushes in reset_prepare). If submission_disabled() +* is set, the caller shouldn't set these flags. +*/ + + destroyed =context_destroyed (ce); + pending_enable =context_pending_enable (ce); + pending_disable =context_pending_disable (ce); + deregister =context_wait_for_deregister_to_register (ce); +init_sched_state (ce); + +if (pending_enable || destroyed || deregister) { +atomic_dec (&guc->outstanding_submission_g2h); +if (deregister) +guc_signal_context_fence (ce); +if (destroyed) { +release_guc_id (guc, ce); +__guc_context_destroy (ce); + } +if (pending_enable|| deregister) +intel_context_put (ce); + } + +/* Not mutualy exclusive with above if statement. */ +if (pending_disable) { +guc_signal_context_fence (ce); +intel_context_sched_disable_unpin (ce); +atomic_dec (&guc->outstanding_submission_g2h); +intel_context_put (ce); + } + } +} + +static inline bool +submission_disabled (struct intel_guc *guc) +{ +struct i915_sched_engine *const sched_engine = guc->sched_engine; + +return unlikely (!sched_engine || + !__tasklet_is_enabled (&sched_engine->tasklet)); +} + +static void disable_submission (struct intel_guc *guc) +{ +struct i915_sched_engine *const sched_engine = guc->sched_engine; + +if (__tasklet_is_enabled (&sched_engine->tasklet)) { +GEM_BUG_ON (!guc->ct.enabled); +__tasklet_disable_sync_once (&sched_engine->tasklet); + sched_engine->tasklet.callback =NULL ; + } +} + +static void enable_submission (struct intel_guc *guc) +{ +struct i915_sched_engine *const sched_engine = guc->sched_engine; +unsigned long flags; + +spin_lock_irqsave (&guc->sched_engine->lock, flags); + sched_engine->tasklet.callback = guc_submission_tasklet; +wmb (); +if (!__tasklet_is_enabled (&sched_engine->tasklet) && +__tasklet_enable (&sched_engine->tasklet)) { +GEM_BUG_ON (!guc->ct.enabled); + +/* And kick in case we missed a new request submission. */ +tasklet_hi_schedule (&sched_engine->tasklet); + } +spin_unlock_irqrestore (&guc->sched_engine->lock, flags); +} + +static void guc_flush_submissions (struct intel_guc *guc) +{ +struct i915_sched_engine *const sched_engine = guc->sched_engine; +unsigned long flags; + +spin_lock_irqsave (&sched_engine->lock, flags); +spin_unlock_irqrestore (&sched_engine->lock, flags); +} + +void intel_guc_submission_reset_prepare (struct intel_guc *guc) {
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void cs_irq_handler(struct intel_engine_cs *engine, u16 iir) · line 602
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +5
-1-ENGINE_TRACE (engine,"\n" ); +int i; + +if (unlikely (!guc_submission_initialized (guc))) +/* Reset called during driver load? GuC not yet initialised! */ +return ;Still think this should have braces. Maybe not syntactically required but just looks wrong.
Sure.
Quoted 177 lines
Quoted 177 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +140
-24+ +disable_submission (guc); + guc->interrupts.disable (guc); + +/* Flush IRQ handler */ +spin_lock_irq (&guc_to_gt (guc)->irq_lock); +spin_unlock_irq (&guc_to_gt (guc)->irq_lock); + +guc_flush_submissions (guc);/* -* Prevent request submission to the hardware until we have -* completed the reset in i915_gem_reset_finish(). If a request -* is completed by one engine, it may then queue a request -* to a second via its execlists->tasklet *just* as we are -* calling engine->init_hw() and also writing the ELSP. -* Turning off the execlists->tasklet until the reset is over -* prevents the race. -*/ -__tasklet_disable_sync_once (&engine->sched_engine->tasklet); +* Handle any outstanding G2Hs before reset. Call IRQ handler directly +* each pass as interrupt have been disabled. We always scrub for +* outstanding G2H as it is possible for outstanding_submission_g2h to +* be incremented after the context state update. +*/ +for (i =0 ; i <4 &&atomic_read (&guc->outstanding_submission_g2h); ++i) { +intel_guc_to_host_event_handler (guc); +#define wait_for_reset (guc, wait_var) \ + guc_wait_for_pending_msg(guc, wait_var, false, (HZ / 20)) +do { +wait_for_reset (guc, &guc->outstanding_submission_g2h); + }while (!list_empty (&guc->ct.requests.incoming)); + } +scrub_guc_desc_for_outstanding_g2h (guc); +} + +static struct intel_engine_cs * +guc_virtual_get_sibling (struct intel_engine_cs *ve,unsigned int sibling) +{ +struct intel_engine_cs *engine; +intel_engine_mask_t tmp, mask = ve->mask; +unsigned int num_siblings =0 ; + +for_each_engine_masked (engine, ve->gt, mask, tmp) +if (num_siblings++ == sibling) +return engine; + +return NULL ; +} + +static inline struct intel_engine_cs * +__context_to_physical_engine (struct intel_context *ce) +{ +struct intel_engine_cs *engine = ce->engine; + +if (intel_engine_is_virtual (engine)) + engine =guc_virtual_get_sibling (engine,0 ); + +return engine; } -static void guc_reset_state (struct intel_context *ce, -struct intel_engine_cs *engine, -u32 head, -bool scrub) +static void guc_reset_state (struct intel_context *ce,u32 head,bool scrub) { +struct intel_engine_cs *engine =__context_to_physical_engine (ce); +GEM_BUG_ON (!intel_context_is_pinned (ce));/* @@ -496,42 +673,147 @@ static void guc_reset_state(struct intel_context *ce,lrc_update_regs (ce, engine, head); } -static void guc_reset_rewind (struct intel_engine_cs *engine,bool stalled) +static void guc_reset_nop (struct intel_engine_cs *engine) { -struct intel_engine_execlists *const execlists = &engine->execlists; -struct i915_request *rq; +} + +static void guc_rewind_nop (struct intel_engine_cs *engine,bool stalled) +{ +} + +static void +__unwind_incomplete_requests (struct intel_context *ce) +{ +struct i915_request *rq, *rn; +struct list_head *pl; +int prio =I915_PRIORITY_INVALID ; +struct i915_sched_engine *const sched_engine = + ce->engine->sched_engine;unsigned long flags; -spin_lock_irqsave (&engine->sched_engine->lock, flags); +spin_lock_irqsave (&sched_engine->lock, flags); +spin_lock (&ce->guc_active.lock); +list_for_each_entry_safe (rq, rn, + &ce->guc_active.requests, + sched.link) { +if (i915_request_completed (rq)) +continue ; + +list_del_init (&rq->sched.link); +spin_unlock (&ce->guc_active.lock); + +__i915_request_unsubmit (rq); + +/* Push the request back into the queue for later resubmission. */ +GEM_BUG_ON (rq_prio (rq) ==I915_PRIORITY_INVALID ); +if (rq_prio (rq) != prio) { + prio =rq_prio (rq); + pl =i915_sched_lookup_priolist (sched_engine, prio); + } +GEM_BUG_ON (i915_sched_engine_is_empty (sched_engine)); -/* Push back any incomplete requests for replay after the reset. */ - rq =execlists_unwind_incomplete_requests (execlists); -if (!rq) - goto out_unlock; +list_add_tail (&rq->sched.link, pl); +set_bit (I915_FENCE_FLAG_PQUEUE , &rq->fence.flags); + +spin_lock (&ce->guc_active.lock); + } +spin_unlock (&ce->guc_active.lock); +spin_unlock_irqrestore (&sched_engine->lock, flags); +} + +static struct i915_request *context_find_active_request (struct intel_context *ce) +{ +struct i915_request *rq, *active =NULL ; +unsigned long flags; + +spin_lock_irqsave (&ce->guc_active.lock, flags); +list_for_each_entry_reverse (rq, &ce->guc_active.requests, + sched.link) { +if (i915_request_completed (rq)) +break ; + + active = rq; + } +spin_unlock_irqrestore (&ce->guc_active.lock, flags); + +return active; +} + +static void __guc_reset_context (struct intel_context *ce,bool stalled) +{ +struct i915_request *rq; +u32 head; + +/* +* GuC will implicitly mark the context as non-schedulable +* when it sends the reset notification. Make sure our state +* reflects this change. The context will be marked enabled +* on resubmission. +*/ +clr_context_enabled (ce); + + rq =context_find_active_request (ce); +if (!rq) { + head = ce->ring->tail; + stalled = false; + 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);__i915_request_reset (rq, stalled); -guc_reset_state (rq->context, engine, rq->head, stalled); -out_unlock: -spin_unlock_irqrestore (&engine->sched_engine->lock, flags); +out_replay: +guc_reset_state (ce, head, stalled); +__unwind_incomplete_requests (ce); +} + +void intel_guc_submission_reset (struct intel_guc *guc,bool stalled)
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void guc_reset_state(struct intel_context *ce, · line 781
Quoted 7 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +7
-0+{ +struct intel_context *ce; +unsigned long index; + +if (unlikely (!guc_submission_initialized (guc))) +/* Reset called during driver load? GuC not yet initialised! */ +return ;And again.
Yep.
Quoted 77 lines
Quoted 77 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +33
-19+ +xa_for_each (&guc->context_lookup, index, ce) +if (intel_context_is_pinned (ce)) +__guc_reset_context (ce, stalled); + +/* GuC is blown away, drop all references to contexts */ +xa_destroy (&guc->context_lookup); +} + +static void guc_cancel_context_requests (struct intel_context *ce) +{ +struct i915_sched_engine *sched_engine =ce_to_guc (ce)->sched_engine; +struct i915_request *rq; +unsigned long flags; + +/* Mark all executing requests as skipped. */ +spin_lock_irqsave (&sched_engine->lock, flags); +spin_lock (&ce->guc_active.lock); +list_for_each_entry (rq, &ce->guc_active.requests, sched.link) +i915_request_put (i915_request_mark_eio (rq)); +spin_unlock (&ce->guc_active.lock); +spin_unlock_irqrestore (&sched_engine->lock, flags); } -static void guc_reset_cancel (struct intel_engine_cs *engine) +static void +guc_cancel_sched_engine_requests (struct i915_sched_engine *sched_engine) { -struct i915_sched_engine *const sched_engine = engine->sched_engine;struct i915_request *rq, *rn;struct rb_node *rb;unsigned long flags;/* Can be called during boot if GuC fails to load */ -if (!engine->gt) +if (!sched_engine)return ; -ENGINE_TRACE (engine,"\n" ); -/* * Before we call engine->cancel_requests(), we should have exclusive * access to the submission state. This is arranged for us by the @@ -548,21 +830,16 @@ static void guc_reset_cancel(struct intel_engine_cs *engine)*/ spin_lock_irqsave (&sched_engine->lock, flags); -/* Mark all executing requests as skipped. */ -list_for_each_entry (rq, &sched_engine->requests, sched.link) { -i915_request_set_error_once (rq, -EIO ); -i915_request_mark_complete (rq); - } -/* Flush the queued requests to the timeline list (for retiring). */ while ((rb =rb_first_cached (&sched_engine->queue))) {struct i915_priolist *p =to_priolist (rb);priolist_for_each_request_consume (rq, rn, p) {list_del_init (&rq->sched.link); +__i915_request_submit (rq); -dma_fence_set_error (&rq->fence, -EIO ); -i915_request_mark_complete (rq); + +i915_request_put (i915_request_mark_eio (rq)); }rb_erase_cached (&p->node, &sched_engine->queue);@@ -577,14 +854,38 @@ static void guc_reset_cancel(struct intel_engine_cs *engine)spin_unlock_irqrestore (&sched_engine->lock, flags); } -static void guc_reset_finish (struct intel_engine_cs *engine) +void intel_guc_submission_cancel_requests (struct intel_guc *guc) { -if (__tasklet_enable (&engine->sched_engine->tasklet)) -/* And kick in case we missed a new request submission. */ -tasklet_hi_schedule (&engine->sched_engine->tasklet); +struct intel_context *ce; +unsigned long index; -ENGINE_TRACE (engine,"depth->%d\n" , -atomic_read (&engine->sched_engine->tasklet.count)); +xa_for_each (&guc->context_lookup, index, ce) +if (intel_context_is_pinned (ce))
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void guc_reset_cancel(struct intel_engine_cs *engine) · line 869
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0+guc_cancel_context_requests (ce); + +guc_cancel_sched_engine_requests (guc->sched_engine); + +/* GuC is blown away, drop all references to contexts */ +xa_destroy (&guc->context_lookup);And again, most of this code is common to '
intel_guc_submission_reset' so could be moved to a shared helper function.
Similar code structure but different.
We use the same xa_for_each loop + pinned check but call __guc_reset_context vs guc_cancel_context_requests. Also here I don't think we can clear &guc->context_lookup until guc_cancel_sched_engine_requests completes. Not sure how to structure a helper that makes sense.
Matt
Quoted 386 lines
John.
Quoted 384 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +112
-50+} + +void intel_guc_submission_reset_finish (struct intel_guc *guc) +{ +/* Reset called during driver load or during wedge? */ +if (unlikely (!guc_submission_initialized (guc) || +test_bit (I915_WEDGED , &guc_to_gt (guc)->reset.flags))) +return ; + +/* +* Technically possible for either of these values to be non-zero here, +* but very unlikely + harmless. Regardless let's add a warn so we can +* see in CI if this happens frequently / a precursor to taking down the +* machine. +*/ +GEM_WARN_ON (atomic_read (&guc->outstanding_submission_g2h)); +atomic_set (&guc->outstanding_submission_g2h,0 ); + +enable_submission (guc); }/* @@ -651,6 +952,9 @@ static int guc_bypass_tasklet_submit(struct intel_guc *guc,else trace_i915_request_guc_submit (rq); +if (unlikely (ret == -EPIPE )) +disable_submission (guc); +return ret; }@@ -663,7 +967,8 @@ static void guc_submit_request(struct i915_request *rq)/* Will be called from irq-context when using foreign fences. */ spin_lock_irqsave (&sched_engine->lock, flags); -if (guc->stalled_request || !i915_sched_engine_is_empty (sched_engine)) +if (submission_disabled (guc) || guc->stalled_request || + !i915_sched_engine_is_empty (sched_engine))queue_request (sched_engine, rq,rq_prio (rq));else if (guc_bypass_tasklet_submit (guc, rq) == -EBUSY )tasklet_hi_schedule (&sched_engine->tasklet);@@ -800,7 +1105,8 @@ static void unpin_guc_id(struct intel_guc *guc, struct intel_context *ce)static int __guc_action_register_context (struct intel_guc *guc,u32 guc_id, -u32 offset) +u32 offset, +bool loop) {u32 action[] = {INTEL_GUC_ACTION_REGISTER_CONTEXT ,@@ -809,10 +1115,10 @@ static int __guc_action_register_context(struct intel_guc *guc, };return guc_submission_send_busy_loop (guc, action,ARRAY_SIZE (action), -0 , true); +0 , loop); } -static int register_context (struct intel_context *ce) +static int register_context (struct intel_context *ce,bool loop) {struct intel_guc *guc =ce_to_guc (ce); u32 offset = intel_guc_ggtt_offset(guc, guc->lrc_desc_pool) +@@ -820,11 +1126,12 @@ static int register_context(struct intel_context *ce)trace_intel_context_register (ce); -return __guc_action_register_context (guc, ce->guc_id, offset); +return __guc_action_register_context (guc, ce->guc_id, offset, loop); }static int __guc_action_deregister_context (struct intel_guc *guc, -u32 guc_id) +u32 guc_id, +bool loop) {u32 action[] = {INTEL_GUC_ACTION_DEREGISTER_CONTEXT ,@@ -833,16 +1140,16 @@ static int __guc_action_deregister_context(struct intel_guc *guc,return guc_submission_send_busy_loop (guc, action,ARRAY_SIZE (action),G2H_LEN_DW_DEREGISTER_CONTEXT , - true); + loop); } -static int deregister_context (struct intel_context *ce,u32 guc_id) +static int deregister_context (struct intel_context *ce,u32 guc_id,bool loop) {struct intel_guc *guc =ce_to_guc (ce);trace_intel_context_deregister (ce); -return __guc_action_deregister_context (guc, guc_id); +return __guc_action_deregister_context (guc, guc_id, loop); }static intel_engine_mask_t adjust_engine_mask (u8 class,intel_engine_mask_t mask)@@ -871,7 +1178,7 @@ static void guc_context_policy_init(struct intel_engine_cs *engine, desc->preemption_timeout =CONTEXT_POLICY_DEFAULT_PREEMPTION_TIME_US ; } -static int guc_lrc_desc_pin (struct intel_context *ce) +static int guc_lrc_desc_pin (struct intel_context *ce,bool loop) {struct intel_runtime_pm *runtime_pm = &ce->engine->gt->i915->runtime_pm;@@ -917,18 +1224,45 @@ static int guc_lrc_desc_pin(struct intel_context *ce)*/ if (context_registered) {trace_intel_context_steal_guc_id (ce); -set_context_wait_for_deregister_to_register (ce); -intel_context_get (ce); +if (!loop) { +set_context_wait_for_deregister_to_register (ce); +intel_context_get (ce); + }else { +bool disabled; +unsigned long flags; + +/* Seal race with Reset */ +spin_lock_irqsave (&ce->guc_state.lock, flags); + disabled =submission_disabled (guc); +if (likely (!disabled)) { +set_context_wait_for_deregister_to_register (ce); +intel_context_get (ce); + } +spin_unlock_irqrestore (&ce->guc_state.lock, flags); +if (unlikely (disabled)) { +reset_lrc_desc (guc, desc_idx); +return 0 ;/* Will get registered later */ + } + }/* * If stealing the guc_id, this ce has the same guc_id as the * context whos guc_id was stole. */ with_intel_runtime_pm (runtime_pm, wakeref) - ret = deregister_context(ce, ce->guc_id); + ret = deregister_context(ce, ce->guc_id, loop); +if (unlikely (ret == -EBUSY )) { +clr_context_wait_for_deregister_to_register (ce); +intel_context_put (ce); + }else if (unlikely (ret == -ENODEV )) + ret =0 ;/* Will get registered later */ }else {with_intel_runtime_pm (runtime_pm, wakeref) - ret = register_context(ce); + ret = register_context(ce, loop); +if (unlikely (ret == -EBUSY )) +reset_lrc_desc (guc, desc_idx); +else if (unlikely (ret == -ENODEV )) + ret =0 ;/* Will get registered later */ }return ret;@@ -991,7 +1325,6 @@ static void __guc_context_sched_disable(struct intel_guc *guc,GEM_BUG_ON (guc_id ==GUC_INVALID_LRC_ID );trace_intel_context_sched_disable (ce); -intel_context_get (ce);guc_submission_send_busy_loop (guc, action,ARRAY_SIZE (action),G2H_LEN_DW_SCHED_CONTEXT_MODE_SET , true);@@ -1003,6 +1336,7 @@ static u16 prep_context_pending_disable(struct intel_context *ce)set_context_pending_disable (ce);clr_context_enabled (ce); +intel_context_get (ce);return ce->guc_id; }@@ -1015,7 +1349,7 @@ static void guc_context_sched_disable(struct intel_context *ce)u16 guc_id;intel_wakeref_t wakeref; -if (context_guc_id_invalid (ce) || +if (submission_disabled (guc) ||context_guc_id_invalid (ce) || !lrc_desc_registered (guc, ce->guc_id)) {clr_context_enabled (ce); goto unpin;@@ -1036,6 +1370,7 @@ static void guc_context_sched_disable(struct intel_context *ce)* a request before we set the 'context_pending_disable' flag here. */ if (unlikely (atomic_add_unless (&ce->pin_count,-2 ,2 ))) { +spin_unlock_irqrestore (&ce->guc_state.lock, flags);return ; } guc_id =prep_context_pending_disable (ce);@@ -1052,19 +1387,13 @@ static void guc_context_sched_disable(struct intel_context *ce)static inline void guc_lrc_desc_unpin (struct intel_context *ce) { -struct intel_engine_cs *engine = ce->engine; -struct intel_guc *guc = &engine->gt->uc.guc; -unsigned long flags; +struct intel_guc *guc =ce_to_guc (ce);GEM_BUG_ON (!lrc_desc_registered (guc, ce->guc_id));GEM_BUG_ON (ce !=__get_context (guc, ce->guc_id));GEM_BUG_ON (context_enabled (ce)); -spin_lock_irqsave (&ce->guc_state.lock, flags); -set_context_destroyed (ce); -spin_unlock_irqrestore (&ce->guc_state.lock, flags); - -deregister_context (ce, ce->guc_id); +deregister_context (ce, ce->guc_id, true); }static void __guc_context_destroy (struct intel_context *ce)@@ -1092,13 +1421,15 @@ static void guc_context_destroy(struct kref *kref)struct intel_guc *guc = &ce->engine->gt->uc.guc;intel_wakeref_t wakeref;unsigned long flags; +bool disabled;/* * If the guc_id is invalid this context has been stolen and we can free * it immediately. Also can be freed immediately if the context is not * registered with the GuC. */ -if (context_guc_id_invalid (ce) || +if (submission_disabled (guc) || +context_guc_id_invalid (ce) || !lrc_desc_registered (guc, ce->guc_id)) {release_guc_id (guc, ce);__guc_context_destroy (ce);@@ -1125,6 +1456,18 @@ static void guc_context_destroy(struct kref *kref)list_del_init (&ce->guc_id_link);spin_unlock_irqrestore (&guc->contexts_lock, flags); +/* Seal race with Reset */ +spin_lock_irqsave (&ce->guc_state.lock, flags); + disabled =submission_disabled (guc); +if (likely (!disabled)) +set_context_destroyed (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); +if (unlikely (disabled)) { +release_guc_id (guc, ce); +__guc_context_destroy (ce); +return ; + } +/* * We defer GuC context deregistration until the context is destroyed * in order to save on CTBs. With this optimization ideally we only need @@ -1212,8 +1555,6 @@ static void guc_signal_context_fence(struct intel_context *ce) {unsigned long flags; -GEM_BUG_ON (!context_wait_for_deregister_to_register (ce)); -spin_lock_irqsave (&ce->guc_state.lock, flags);clr_context_wait_for_deregister_to_register (ce);__guc_signal_context_fence (ce);@@ -1222,8 +1563,9 @@ static void guc_signal_context_fence(struct intel_context *ce)static bool context_needs_register (struct intel_context *ce,bool new_guc_id) { -return new_guc_id ||test_bit (CONTEXT_LRCA_DIRTY , &ce->flags) || - !lrc_desc_registered (ce_to_guc (ce), ce->guc_id); +return (new_guc_id ||test_bit (CONTEXT_LRCA_DIRTY , &ce->flags) || + !lrc_desc_registered (ce_to_guc (ce), ce->guc_id)) && + !submission_disabled (ce_to_guc (ce)); }static int guc_request_alloc (struct i915_request *rq)@@ -1281,8 +1623,12 @@ static int guc_request_alloc(struct i915_request *rq)if (unlikely (ret <0 ))return ret;;if (context_needs_register (ce, !!ret)) { - ret =guc_lrc_desc_pin (ce); + ret =guc_lrc_desc_pin (ce, true);if (unlikely (ret)) {/* unwind */ +if (ret == -EPIPE ) { +disable_submission (guc); + goto out;/* GPU will be reset */ + }atomic_dec (&ce->guc_id_ref);unpin_guc_id (guc, ce);return ret;@@ -1319,20 +1665,6 @@ static int guc_request_alloc(struct i915_request *rq)return 0 ; } -static struct intel_engine_cs * -guc_virtual_get_sibling (struct intel_engine_cs *ve,unsigned int sibling) -{ -struct intel_engine_cs *engine; -intel_engine_mask_t tmp, mask = ve->mask; -unsigned int num_siblings =0 ; - -for_each_engine_masked (engine, ve->gt, mask, tmp) -if (num_siblings++ == sibling) -return engine; - -return NULL ; -} -static int guc_virtual_context_pre_pin (struct intel_context *ce,struct i915_gem_ww_ctx *ww,void **vaddr)@@ -1528,7 +1860,7 @@ static inline void guc_kernel_context_pin(struct intel_guc *guc, {if (context_guc_id_invalid (ce))pin_guc_id (guc, ce); -guc_lrc_desc_pin (ce); +guc_lrc_desc_pin (ce, true); }static inline void guc_init_lrc_mapping (struct intel_guc *guc)@@ -1599,10 +1931,10 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->sched_engine->schedule = i915_schedule; - engine->reset.prepare = guc_reset_prepare; - engine->reset.rewind = guc_reset_rewind; - engine->reset.cancel = guc_reset_cancel; - engine->reset.finish = guc_reset_finish; + engine->reset.prepare = guc_reset_nop; + engine->reset.rewind = guc_rewind_nop; + engine->reset.cancel = guc_reset_nop; + engine->reset.finish = guc_reset_nop; engine->emit_flush = gen8_emit_flush_xcs; engine->emit_init_breadcrumb = gen8_emit_init_breadcrumb;@@ -1651,6 +1983,17 @@ static inline void guc_default_irqs(struct intel_engine_cs *engine)intel_engine_set_irq_handler (engine, cs_irq_handler); } +static void guc_sched_engine_destroy (struct kref *kref) +{ +struct i915_sched_engine *sched_engine = +container_of (kref,typeof (*sched_engine), ref); +struct intel_guc *guc = sched_engine->private_data; + + guc->sched_engine =NULL ; +tasklet_kill (&sched_engine->tasklet);/* flush the callback */ +kfree (sched_engine); +} +int intel_guc_submission_setup (struct intel_engine_cs *engine) {struct drm_i915_private *i915 = engine->i915;@@ -1669,6 +2012,7 @@ int intel_guc_submission_setup(struct intel_engine_cs *engine) guc->sched_engine->schedule = i915_schedule; guc->sched_engine->private_data = guc; + guc->sched_engine->destroy = guc_sched_engine_destroy;tasklet_setup (&guc->sched_engine->tasklet, guc_submission_tasklet); }@@ -1775,7 +2119,7 @@ int intel_guc_deregister_done_process_msg(struct intel_guc *guc,* register this context. */ with_intel_runtime_pm(runtime_pm, wakeref) -register_context (ce); +register_context (ce, true);guc_signal_context_fence (ce);intel_context_put (ce); }else if (context_destroyed (ce)) {drivers/gpu/drm/i915/gt/uc/intel_uc.c +38
-1@@ -565,12 +565,49 @@ void intel_uc_reset_prepare(struct intel_uc *uc) {struct intel_guc *guc = &uc->guc; -if (!intel_guc_is_ready (guc)) + +/* Nothing to do if GuC isn't supported */ +if (!intel_uc_supports_guc (uc))return ; +/* Firmware expected to be running when this function is called */ +if (!intel_guc_is_ready (guc)) + goto sanitize; + +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset_prepare (guc); + +sanitize:__uc_sanitize (uc); } +void intel_uc_reset (struct intel_uc *uc,bool stalled) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware can not be running when this function is called */ +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset (guc, stalled); +} + +void intel_uc_reset_finish (struct intel_uc *uc) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware expected to be running when this function is called */ +if (intel_guc_is_fw_running (guc) &&intel_uc_uses_guc_submission (uc)) +intel_guc_submission_reset_finish (guc); +} + +void intel_uc_cancel_requests (struct intel_uc *uc) +{ +struct intel_guc *guc = &uc->guc; + +/* Firmware can not be running when this function is called */ +if (intel_uc_uses_guc_submission (uc)) +intel_guc_submission_cancel_requests (guc); +} +void intel_uc_runtime_suspend (struct intel_uc *uc) {struct intel_guc *guc = &uc->guc;drivers/gpu/drm/i915/gt/uc/intel_uc.h +0
-0@@ -37,6 +37,9 @@ void intel_uc_driver_late_release(struct intel_uc *uc);void intel_uc_driver_remove (struct intel_uc *uc);
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_uc.h void intel_uc_driver_late_release(struct intel_uc *uc); · line 45
Quoted 8 lines
drivers/gpu/drm/i915/gt/uc/intel_uc.h +3
-0void intel_uc_init_mmio (struct intel_uc *uc);void intel_uc_reset_prepare (struct intel_uc *uc); +void intel_uc_reset (struct intel_uc *uc,bool stalled); +void intel_uc_reset_finish (struct intel_uc *uc); +void intel_uc_cancel_requests (struct intel_uc *uc);void intel_uc_suspend (struct intel_uc *uc);void intel_uc_runtime_suspend (struct intel_uc *uc);int intel_uc_resume (struct intel_uc *uc);
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 27/51 drm/i915: Reset GPU immediately if submission is disabled
In reply to Matthew BrostIf submission is disabled by the backend for any reason, reset the GPU immediately in the heartbeat code as the backend can't be reenabled until the GPU is reset.
--- .../gpu/drm/i915/gt/intel_engine_heartbeat.c | 63 +++++++++++++++---- .../gpu/drm/i915/gt/intel_engine_heartbeat.h | 4 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 9 +++ drivers/gpu/drm/i915/i915_scheduler.c | 6 ++ drivers/gpu/drm/i915/i915_scheduler.h | 6 ++ drivers/gpu/drm/i915/i915_scheduler_types.h | 5 ++ 6 files changed, 80 insertions(+), 13 deletions(-)
drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c +50 -13
@@ -70,12 +70,30 @@ static void show_heartbeat(const struct i915_request *rq, {struct drm_printer p =drm_debug_printer ("heartbeat" ); -intel_engine_dump (engine, &p, -"%s heartbeat {seqno:%llx:%lld, prio:%d} not ticking\n" , - engine->name, - rq->fence.context, - rq->fence.seqno, - rq->sched.attr.priority); +if (!rq) { +intel_engine_dump (engine, &p, +"%s heartbeat not ticking\n" , + engine->name); + }else { +intel_engine_dump (engine, &p, +"%s heartbeat {seqno:%llx:%lld, prio:%d} not ticking\n" , + engine->name, + rq->fence.context, + rq->fence.seqno, + rq->sched.attr.priority); + } +} + +static void +reset_engine (struct intel_engine_cs *engine,struct i915_request *rq) +{ +if (IS_ENABLED (CONFIG_DRM_I915_DEBUG_GEM )) +show_heartbeat (rq, engine); + +intel_gt_handle_error (engine->gt, engine->mask, +I915_ERROR_CAPTURE , +"stopped heartbeat on %s" , + engine->name); }static void heartbeat (struct work_struct *wrk)
@@ -102,6 +120,11 @@ static void heartbeat(struct work_struct *wrk)if (intel_gt_is_wedged (engine->gt)) goto out; +if (i915_sched_engine_disabled (engine->sched_engine)) { +reset_engine (engine, engine->heartbeat.systole); + goto out; + } +if (engine->heartbeat.systole) {long delay =READ_ONCE (engine->props.heartbeat_interval_ms);
@@ -139,13 +162,7 @@ static void heartbeat(struct work_struct *wrk) engine->sched_engine->schedule (rq, &attr);local_bh_enable (); }else { -if (IS_ENABLED (CONFIG_DRM_I915_DEBUG_GEM )) -show_heartbeat (rq, engine); - -intel_gt_handle_error (engine->gt, engine->mask, -I915_ERROR_CAPTURE , -"stopped heartbeat on %s" , - engine->name); +reset_engine (engine, rq); } rq->emitted_jiffies = jiffies;
@@ -194,6 +211,26 @@ void intel_engine_park_heartbeat(struct intel_engine_cs *engine)i915_request_put (fetch_and_zero (&engine->heartbeat.systole)); } +void intel_gt_unpark_heartbeats (struct intel_gt *gt) +{ +struct intel_engine_cs *engine; +enum intel_engine_id id; + +for_each_engine (engine, gt, id) +if (intel_engine_pm_is_awake (engine)) +intel_engine_unpark_heartbeat (engine); + +} + +void intel_gt_park_heartbeats (struct intel_gt *gt) +{ +struct intel_engine_cs *engine; +enum intel_engine_id id; + +for_each_engine (engine, gt, id) +intel_engine_park_heartbeat (engine); +} +void intel_engine_init_heartbeat (struct intel_engine_cs *engine) {INIT_DELAYED_WORK (&engine->heartbeat.work, heartbeat);
drivers/gpu/drm/i915/gt/intel_engine_heartbeat.h +4 -0
@@ -7,6 +7,7 @@#define INTEL_ENGINE_HEARTBEAT_H struct intel_engine_cs ; +struct intel_gt ;void intel_engine_init_heartbeat (struct intel_engine_cs *engine);
@@ -16,6 +17,9 @@ int intel_engine_set_heartbeat(struct intel_engine_cs *engine,void intel_engine_park_heartbeat (struct intel_engine_cs *engine);void intel_engine_unpark_heartbeat (struct intel_engine_cs *engine); +void intel_gt_park_heartbeats (struct intel_gt *gt); +void intel_gt_unpark_heartbeats (struct intel_gt *gt); +int intel_engine_pulse (struct intel_engine_cs *engine);int intel_engine_flush_barriers (struct intel_engine_cs *engine);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +9 -0
@@ -10,6 +10,7 @@#include "gt/intel_breadcrumbs.h" #include "gt/intel_context.h" #include "gt/intel_engine_pm.h" +#include "gt/intel_engine_heartbeat.h" #include "gt/intel_gt.h" #include "gt/intel_gt_irq.h" #include "gt/intel_gt_pm.h"
@@ -601,6 +602,7 @@ void intel_guc_submission_reset_prepare(struct intel_guc *guc)/* Reset called during driver load? GuC not yet initialised! */ return ; +intel_gt_park_heartbeats (guc_to_gt (guc));disable_submission (guc); guc->interrupts.disable (guc);
@@ -886,6 +888,7 @@ void intel_guc_submission_reset_finish(struct intel_guc *guc)atomic_set (&guc->outstanding_submission_g2h,0 );enable_submission (guc); +intel_gt_unpark_heartbeats (guc_to_gt (guc)); }/*
@@ -1850,6 +1853,11 @@ static int guc_resume(struct intel_engine_cs *engine)return 0 ; } +static bool guc_sched_engine_disabled (struct i915_sched_engine *sched_engine) +{ +return !sched_engine->tasklet.callback; +} +static void guc_set_default_submission (struct intel_engine_cs *engine) { engine->submit_request = guc_submit_request;
@@ -2011,6 +2019,7 @@ int intel_guc_submission_setup(struct intel_engine_cs *engine)return -ENOMEM ; guc->sched_engine->schedule = i915_schedule; + guc->sched_engine->disabled = guc_sched_engine_disabled; guc->sched_engine->private_data = guc; guc->sched_engine->destroy = guc_sched_engine_destroy; tasklet_setup(&guc->sched_engine->tasklet,
drivers/gpu/drm/i915/i915_scheduler.c +6 -0
@@ -440,6 +440,11 @@ static void default_destroy(struct kref *kref)kfree (sched_engine); } +static bool default_disabled (struct i915_sched_engine *sched_engine) +{ +return false; +} +struct i915_sched_engine *i915_sched_engine_create (unsigned int subclass) {
@@ -454,6 +459,7 @@ i915_sched_engine_create(unsigned int subclass) sched_engine->queue =RB_ROOT_CACHED ; sched_engine->queue_priority_hint =INT_MIN ; sched_engine->destroy = default_destroy; + sched_engine->disabled = default_disabled;INIT_LIST_HEAD (&sched_engine->requests);INIT_LIST_HEAD (&sched_engine->hold);
drivers/gpu/drm/i915/i915_scheduler.h +6 -0
@@ -96,4 +96,10 @@ void i915_request_show_with_schedule(struct drm_printer *m,const char *prefix, int indent); +static inline bool +i915_sched_engine_disabled (struct i915_sched_engine *sched_engine) +{ +return sched_engine->disabled (sched_engine); +} +#endif /* _I915_SCHEDULER_H_ */
drivers/gpu/drm/i915/i915_scheduler_types.h +5 -0
@@ -168,6 +168,11 @@ struct i915_sched_engine {*/ void (*destroy)(struct kref *kref); +/** +* @disabled: check if backend has disabled submission +*/ +bool (*disabled)(struct i915_sched_engine *sched_engine); +/** * @kick_backend: kick backend after a request's priority has changed */
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 28/51 drm/i915/guc: Add disable interrupts to guc sanitize
In reply to Matthew BrostAdd disable GuC interrupts to intel_guc_sanitize(). Part of this requires moving the guc_*_interrupt wrapper function into header file intel_guc.h.
--- drivers/gpu/drm/i915/gt/uc/intel_guc.h | 16 ++++++++++++++++ drivers/gpu/drm/i915/gt/uc/intel_uc.c | 21 +++------------------ 2 files changed, 19 insertions(+), 18 deletions(-)
drivers/gpu/drm/i915/gt/uc/intel_guc.h +16 -0
@@ -217,9 +217,25 @@ static inline bool intel_guc_is_ready(struct intel_guc *guc)return intel_guc_is_fw_running (guc) &&intel_guc_ct_enabled (&guc->ct); } +static inline void intel_guc_reset_interrupts (struct intel_guc *guc) +{ + guc->interrupts.reset (guc); +} + +static inline void intel_guc_enable_interrupts (struct intel_guc *guc) +{ + guc->interrupts.enable (guc); +} + +static inline void intel_guc_disable_interrupts (struct intel_guc *guc) +{ + guc->interrupts.disable (guc); +} +static inline int intel_guc_sanitize (struct intel_guc *guc) {intel_uc_fw_sanitize (&guc->fw); +intel_guc_disable_interrupts (guc);intel_guc_ct_sanitize (&guc->ct); guc->mmio_msg =0 ;
drivers/gpu/drm/i915/gt/uc/intel_uc.c +3 -18
@@ -207,21 +207,6 @@ static void guc_handle_mmio_msg(struct intel_guc *guc)spin_unlock_irq (&guc->irq_lock); } -static void guc_reset_interrupts (struct intel_guc *guc) -{ - guc->interrupts.reset (guc); -} - -static void guc_enable_interrupts (struct intel_guc *guc) -{ - guc->interrupts.enable (guc); -} - -static void guc_disable_interrupts (struct intel_guc *guc) -{ - guc->interrupts.disable (guc); -} -static int guc_enable_communication (struct intel_guc *guc) {struct intel_gt *gt =guc_to_gt (guc);
@@ -242,7 +227,7 @@ static int guc_enable_communication(struct intel_guc *guc)guc_get_mmio_msg (guc);guc_handle_mmio_msg (guc); -guc_enable_interrupts (guc); +intel_guc_enable_interrupts (guc);/* check for CT messages received before we enabled interrupts */ spin_lock_irq (>->irq_lock);
@@ -265,7 +250,7 @@ static void guc_disable_communication(struct intel_guc *guc)*/ guc_clear_mmio_msg (guc); -guc_disable_interrupts (guc); +intel_guc_disable_interrupts (guc);intel_guc_ct_disable (&guc->ct);
@@ -463,7 +448,7 @@ static int __uc_init_hw(struct intel_uc *uc)if (ret) goto err_out; -guc_reset_interrupts (guc); +intel_guc_reset_interrupts (guc);/* WaEnableuKernelHeaderValidFix:skl */ /* WaEnableGuCBootHashCheckNotSet:skl,bxt,kbl */
Signature · 1 line
2.28.0
PATCH 29/51 drm/i915/guc: Suspend/resume implementation for new interface
In reply to Matthew BrostThe new GuC interface introduces an MMIO H2G command, INTEL_GUC_ACTION_RESET_CLIENT, which is used to implement suspend. This MMIO tears down any active contexts generating a context reset G2H CTB for each. Once that step completes the GuC tears down the CTB channels. It is safe to suspend once this MMIO H2G command completes and all G2H CTBs have been processed. In practice the i915 will likely never receive a G2H as suspend should only be called after the GPU is idle.
Resume is implemented in the same manner as before - simply reload the GuC firmware and reinitialize everything (e.g. CTB channels, contexts, etc..).
--- .../gpu/drm/i915/gt/uc/abi/guc_actions_abi.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_guc.c | 64 ++++++++----------- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 15 +++-- .../gpu/drm/i915/gt/uc/intel_guc_submission.h | 5 ++ drivers/gpu/drm/i915/gt/uc/intel_uc.c | 20 ++++-- 5 files changed, 54 insertions(+), 51 deletions(-)
drivers/gpu/drm/i915/gt/uc/abi/guc_actions_abi.h +1 -0
@@ -142,6 +142,7 @@ enum intel_guc_action {INTEL_GUC_ACTION_REGISTER_COMMAND_TRANSPORT_BUFFER =0x4505 ,INTEL_GUC_ACTION_DEREGISTER_COMMAND_TRANSPORT_BUFFER =0x4506 ,INTEL_GUC_ACTION_DEREGISTER_CONTEXT_DONE =0x4600 , +INTEL_GUC_ACTION_RESET_CLIENT =0x5B01 , INTEL_GUC_ACTION_LIMIT };
drivers/gpu/drm/i915/gt/uc/intel_guc.c +26 -38
84 lines not on this page · raw message
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +8 -7
@@ -304,10 +304,10 @@ static int guc_submission_send_busy_loop(struct intel_guc* guc,return err; } -static int guc_wait_for_pending_msg (struct intel_guc *guc, -atomic_t *wait_var, -bool interruptible, -long timeout) +int intel_guc_wait_for_pending_msg (struct intel_guc *guc, +atomic_t *wait_var, +bool interruptible, +long timeout) {const int state = interruptible ?TASK_INTERRUPTIBLE :TASK_UNINTERRUPTIBLE ;
@@ -347,8 +347,9 @@ static int guc_wait_for_pending_msg(struct intel_guc *guc,int intel_guc_wait_for_idle (struct intel_guc *guc,long timeout) { -return guc_wait_for_pending_msg (guc, &guc->outstanding_submission_g2h, - true, timeout); +return intel_guc_wait_for_pending_msg (guc, + &guc->outstanding_submission_g2h, + true, timeout); }static int guc_lrc_desc_pin (struct intel_context *ce,bool loop);
@@ -621,7 +622,7 @@ void intel_guc_submission_reset_prepare(struct intel_guc *guc)for (i =0 ; i <4 &&atomic_read (&guc->outstanding_submission_g2h); ++i) {intel_guc_to_host_event_handler (guc);#define wait_for_reset (guc, wait_var) \ - guc_wait_for_pending_msg(guc, wait_var, false, (HZ / 20)) + intel_guc_wait_for_pending_msg(guc, wait_var, false, (HZ / 20))do {wait_for_reset (guc, &guc->outstanding_submission_g2h); }while (!list_empty (&guc->ct.requests.incoming));
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.h +5 -0
@@ -28,6 +28,11 @@ void intel_guc_submission_print_context_info(struct intel_guc *guc,bool intel_guc_virtual_engine_has_heartbeat (const struct intel_engine_cs *ve); +int intel_guc_wait_for_pending_msg (struct intel_guc *guc, +atomic_t *wait_var, +bool interruptible, +long timeout); +static inline bool intel_guc_submission_is_supported (struct intel_guc *guc) {/* XXX: GuC submission is unavailable for now */
drivers/gpu/drm/i915/gt/uc/intel_uc.c +14 -6
@@ -596,14 +596,18 @@ void intel_uc_cancel_requests(struct intel_uc *uc)void intel_uc_runtime_suspend (struct intel_uc *uc) {struct intel_guc *guc = &uc->guc; -int err;if (!intel_guc_is_ready (guc))return ; - err =intel_guc_suspend (guc); -if (err) -DRM_DEBUG_DRIVER ("Failed to suspend GuC, err=%d" , err); +/* +* Wait for any outstanding CTB before tearing down communication /w the +* GuC. +*/ +#define OUTSTANDING_CTB_TIMEOUT_PERIOD (HZ / 5) +intel_guc_wait_for_pending_msg (guc, &guc->outstanding_submission_g2h, + false,OUTSTANDING_CTB_TIMEOUT_PERIOD ); +GEM_WARN_ON (atomic_read (&guc->outstanding_submission_g2h));guc_disable_communication (guc); }
@@ -612,12 +616,16 @@ void intel_uc_suspend(struct intel_uc *uc) {struct intel_guc *guc = &uc->guc;intel_wakeref_t wakeref; +int err;if (!intel_guc_is_ready (guc))return ; - with_intel_runtime_pm(uc_to_gt(uc)->uncore->rpm, wakeref) -intel_uc_runtime_suspend (uc); + with_intel_runtime_pm(&uc_to_gt(uc)->i915->runtime_pm, wakeref) { + err =intel_guc_suspend (guc); +if (err) +DRM_DEBUG_DRIVER ("Failed to suspend GuC, err=%d" , err); + } }static int __uc_resume (struct intel_uc *uc,bool enable_communication)
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 30/51 drm/i915/guc: Handle context reset notification
In reply to Matthew BrostGuC will issue a reset on detecting an engine hang and will notify the driver via a G2H message. The driver will service the notification by resetting the guilty context to a simple state or banning it completely.
v2:
(John Harrison)
- Move msg[0] lookup after length check
--- drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 ++ drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 3 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 36 +++++++++++++++++++ drivers/gpu/drm/i915/i915_trace.h | 10 ++++++ 4 files changed, 51 insertions(+)
drivers/gpu/drm/i915/gt/uc/intel_guc.h +2 -0
@@ -262,6 +262,8 @@ int intel_guc_deregister_done_process_msg(struct intel_guc *guc, const u32 *msg, u32 len);int intel_guc_sched_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len); +int intel_guc_context_reset_process_msg (struct intel_guc *guc, +const u32 *msg,u32 len);void intel_guc_submission_reset_prepare (struct intel_guc *guc);void intel_guc_submission_reset (struct intel_guc *guc,bool stalled);
drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +3 -0
@@ -981,6 +981,9 @@ static int ct_process_request(struct intel_guc_ct *ct, struct ct_incoming_msg *rcase INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_DONE : ret =intel_guc_sched_done_process_msg (guc, payload, len);break ; +case INTEL_GUC_ACTION_CONTEXT_RESET_NOTIFICATION : + ret =intel_guc_context_reset_process_msg (guc, payload, len); +break ;default : ret = -EOPNOTSUPP ;break ;
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +36 -0
@@ -2196,6 +2196,42 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc,return 0 ; } +static void guc_context_replay (struct intel_context *ce) +{ +struct i915_sched_engine *sched_engine = ce->engine->sched_engine; + +__guc_reset_context (ce, true); +tasklet_hi_schedule (&sched_engine->tasklet); +} + +static void guc_handle_context_reset (struct intel_guc *guc, +struct intel_context *ce) +{ +trace_intel_context_reset (ce); +guc_context_replay (ce); +} + +int intel_guc_context_reset_process_msg (struct intel_guc *guc, +const u32 *msg,u32 len) +{ +struct intel_context *ce; +int desc_idx; + +if (unlikely (len !=1 )) { +drm_dbg (&guc_to_gt (guc)->i915->drm,"Invalid length %u" , len); +return -EPROTO ; + } + + desc_idx = msg[0 ]; + ce =g2h_context_lookup (guc, desc_idx); +if (unlikely (!ce)) +return -EPROTO ; + +guc_handle_context_reset (guc, ce); + +return 0 ; +} +void intel_guc_submission_print_info (struct intel_guc *guc,struct drm_printer *p) {
drivers/gpu/drm/i915/i915_trace.h +10 -0
@@ -929,6 +929,11 @@ DECLARE_EVENT_CLASS(intel_context, __entry->guc_sched_state_no_lock) ); +DEFINE_EVENT (intel_context, intel_context_reset, + TP_PROTO(struct intel_context *ce), +TP_ARGS (ce) +); +DEFINE_EVENT (intel_context, intel_context_register, TP_PROTO(struct intel_context *ce), TP_ARGS(ce)
@@ -1026,6 +1031,11 @@ trace_i915_request_out(struct i915_request *rq) { } +static inline void +trace_intel_context_reset (struct intel_context *ce) +{ +} +static inline void trace_intel_context_register (struct intel_context *ce) {
Signature · 1 line
2.28.0
Re: [PATCH 30/51] drm/i915/guc: Handle context reset notification
Quoted 67 lines · Matthew Brost
GuC will issue a reset on detecting an engine hang and will notify the driver via a G2H message. The driver will service the notification by resetting the guilty context to a simple state or banning it completely.
v2:
(John Harrison)
- Move msg[0] lookup after length checkCc Matthew Brost <[email protected]>Cc John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 ++ drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 3 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 36 +++++++++++++++++++ drivers/gpu/drm/i915/i915_trace.h | 10 ++++++ 4 files changed, 51 insertions(+)drivers/gpu/drm/i915/gt/uc/intel_guc.h +2
-0@@ -262,6 +262,8 @@ int intel_guc_deregister_done_process_msg(struct intel_guc *guc, const u32 *msg, u32 len);int intel_guc_sched_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len); +int intel_guc_context_reset_process_msg (struct intel_guc *guc, +const u32 *msg,u32 len);void intel_guc_submission_reset_prepare (struct intel_guc *guc);void intel_guc_submission_reset (struct intel_guc *guc,bool stalled);drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +3
-0@@ -981,6 +981,9 @@ static int ct_process_request(struct intel_guc_ct *ct, struct ct_incoming_msg *rcase INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_DONE : ret =intel_guc_sched_done_process_msg (guc, payload, len);break ; +case INTEL_GUC_ACTION_CONTEXT_RESET_NOTIFICATION : + ret =intel_guc_context_reset_process_msg (guc, payload, len); +break ;default : ret = -EOPNOTSUPP ;break ;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +13
-0@@ -2196,6 +2196,42 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc,return 0 ; } +static void guc_context_replay (struct intel_context *ce) +{ +struct i915_sched_engine *sched_engine = ce->engine->sched_engine; + +__guc_reset_context (ce, true); +tasklet_hi_schedule (&sched_engine->tasklet); +} + +static void guc_handle_context_reset (struct intel_guc *guc, +struct intel_context *ce) +{ +trace_intel_context_reset (ce); +guc_context_replay (ce);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c int intel_guc_sched_done_process_msg(struct intel_guc *guc, · line 2221
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +10
-0+} + +int intel_guc_context_reset_process_msg (struct intel_guc *guc, +const u32 *msg,u32 len) +{ +struct intel_context *ce; +int desc_idx; + +if (unlikely (len !=1 )) { +drm_dbg (&guc_to_gt (guc)->i915->drm,"Invalid length %u" , len);
I think we decided that these should be drm_err rather than drm_dbg?
With that updated:
Quoted 44 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +13
-0+return -EPROTO ; + } + + desc_idx = msg[0 ]; + ce =g2h_context_lookup (guc, desc_idx); +if (unlikely (!ce)) +return -EPROTO ; + +guc_handle_context_reset (guc, ce); + +return 0 ; +} +void intel_guc_submission_print_info (struct intel_guc *guc,struct drm_printer *p) {drivers/gpu/drm/i915/i915_trace.h +10
-0@@ -929,6 +929,11 @@ DECLARE_EVENT_CLASS(intel_context, __entry->guc_sched_state_no_lock) ); +DEFINE_EVENT (intel_context, intel_context_reset, + TP_PROTO(struct intel_context *ce), +TP_ARGS (ce) +); +DEFINE_EVENT (intel_context, intel_context_register, TP_PROTO(struct intel_context *ce), TP_ARGS(ce)@@ -1026,6 +1031,11 @@ trace_i915_request_out(struct i915_request *rq) { } +static inline void +trace_intel_context_reset (struct intel_context *ce) +{ +} +static inline void trace_intel_context_register (struct intel_context *ce) {
Re: [PATCH 30/51] drm/i915/guc: Handle context reset notification
Quoted 69 lines · John Harrison
Quoted 68 lines · Matthew Brost
GuC will issue a reset on detecting an engine hang and will notify the driver via a G2H message. The driver will service the notification by resetting the guilty context to a simple state or banning it completely.
v2:
(John Harrison)
- Move msg[0] lookup after length checkCc Matthew Brost <[email protected]>Cc John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 ++ drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 3 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 36 +++++++++++++++++++ drivers/gpu/drm/i915/i915_trace.h | 10 ++++++ 4 files changed, 51 insertions(+)drivers/gpu/drm/i915/gt/uc/intel_guc.h +2
-0@@ -262,6 +262,8 @@ int intel_guc_deregister_done_process_msg(struct intel_guc *guc, const u32 *msg, u32 len);int intel_guc_sched_done_process_msg (struct intel_guc *guc,const u32 *msg,u32 len); +int intel_guc_context_reset_process_msg (struct intel_guc *guc, +const u32 *msg,u32 len);void intel_guc_submission_reset_prepare (struct intel_guc *guc);void intel_guc_submission_reset (struct intel_guc *guc,bool stalled);drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +3
-0@@ -981,6 +981,9 @@ static int ct_process_request(struct intel_guc_ct *ct, struct ct_incoming_msg *rcase INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_DONE : ret =intel_guc_sched_done_process_msg (guc, payload, len);break ; +case INTEL_GUC_ACTION_CONTEXT_RESET_NOTIFICATION : + ret =intel_guc_context_reset_process_msg (guc, payload, len); +break ;default : ret = -EOPNOTSUPP ;break ;drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +16
-0@@ -2196,6 +2196,42 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc,return 0 ; } +static void guc_context_replay (struct intel_context *ce) +{ +struct i915_sched_engine *sched_engine = ce->engine->sched_engine; + +__guc_reset_context (ce, true); +tasklet_hi_schedule (&sched_engine->tasklet); +} + +static void guc_handle_context_reset (struct intel_guc *guc, +struct intel_context *ce) +{ +trace_intel_context_reset (ce); +guc_context_replay (ce); +} + +int intel_guc_context_reset_process_msg(struct intel_guc *guc,
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c int intel_guc_sched_done_process_msg(struct intel_guc *guc, · line 2221
Quoted 7 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +7
-0+ const u32 *msg, u32 len) +{ +struct intel_context *ce; +int desc_idx; + +if (unlikely (len !=1 )) { +drm_dbg (&guc_to_gt (guc)->i915->drm,"Invalid length %u" , len);I think we decided that these should be
drm_errrather thandrm_dbg?
Yes, we did. Already fixed this message and all subsequent in my branch.
Matt
Quoted 37 lines
With that updated:
Reviewed-by John Harrison <[email protected]>Quoted 34 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +13
-0+return -EPROTO ; + } + + desc_idx = msg[0 ]; + ce =g2h_context_lookup (guc, desc_idx); +if (unlikely (!ce)) +return -EPROTO ; + +guc_handle_context_reset (guc, ce); + +return 0 ; +} +void intel_guc_submission_print_info (struct intel_guc *guc,struct drm_printer *p) {drivers/gpu/drm/i915/i915_trace.h +5
-0@@ -929,6 +929,11 @@ DECLARE_EVENT_CLASS(intel_context, __entry->guc_sched_state_no_lock) ); +DEFINE_EVENT (intel_context, intel_context_reset, + TP_PROTO(struct intel_context *ce), +TP_ARGS (ce) +); +DEFINE_EVENT (intel_context, intel_context_register, TP_PROTO(struct intel_context *ce), TP_ARGS(ce)@@ -1026,6 +1031,11 @@ trace_i915_request_out(struct i915_request *rq) { }
Quoted 9 lines
drivers/gpu/drm/i915/i915_trace.h trace_i915_request_out(struct i915_request *rq) · line 1041
Quoted 8 lines
drivers/gpu/drm/i915/i915_trace.h +5
-0+static inline void +trace_intel_context_reset (struct intel_context *ce) +{ +} +static inline void trace_intel_context_register (struct intel_context *ce) {
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 31/51 drm/i915/guc: Handle engine reset failure notification
In reply to Matthew BrostGuC will notify the driver, via G2H, if it fails to reset an engine. We recover by resorting to a full GPU reset.
--- drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 3 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 43 +++++++++++++++++++ 3 files changed, 48 insertions(+)
drivers/gpu/drm/i915/gt/uc/intel_guc.h +2 -0
@@ -264,6 +264,8 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc, const u32 *msg, u32 len);int intel_guc_context_reset_process_msg (struct intel_guc *guc,const u32 *msg,u32 len); +int intel_guc_engine_failure_process_msg (struct intel_guc *guc, +const u32 *msg,u32 len);void intel_guc_submission_reset_prepare (struct intel_guc *guc);void intel_guc_submission_reset (struct intel_guc *guc,bool stalled);
drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +3 -0
@@ -984,6 +984,9 @@ static int ct_process_request(struct intel_guc_ct *ct, struct ct_incoming_msg *rcase INTEL_GUC_ACTION_CONTEXT_RESET_NOTIFICATION : ret =intel_guc_context_reset_process_msg (guc, payload, len);break ; +case INTEL_GUC_ACTION_ENGINE_FAILURE_NOTIFICATION : + ret =intel_guc_engine_failure_process_msg (guc, payload, len); +break ;default : ret = -EOPNOTSUPP ;break ;
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +43 -0
49 lines not on this page · raw message
Signature · 1 line
2.28.0
PATCH 32/51 drm/i915/guc: Enable the timer expired interrupt for GuC
In reply to Matthew BrostThe GuC can implement execution qunatums, detect hung contexts and other such things but it requires the timer expired interrupt to do so.
--- drivers/gpu/drm/i915/gt/intel_rps.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/gpu/drm/i915/gt/intel_rps.c +4 -0
@@ -1877,6 +1877,10 @@ void intel_rps_init(struct intel_rps *rps)if (GRAPHICS_VER (i915) >=8 &&GRAPHICS_VER (i915) <11 ) rps->pm_intrmsk_mbz |=GEN8_PMINTR_DISABLE_REDIRECT_TO_GUC ; + +/* GuC needs ARAT expired interrupt unmasked */ +if (intel_uc_uses_guc_submission (&rps_to_gt (rps)->uc)) + rps->pm_intrmsk_mbz |=ARAT_EXPIRED_INTRMSK ; }void intel_rps_sanitize (struct intel_rps *rps)
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 33/51 drm/i915/guc: Provide mmio list to be saved/restored on engine reset
In reply to Matthew BrostFrom: John Harrison <[email protected]>
The driver must provide GuC with a list of mmio registers that should be saved/restored during a GuC-based engine reset. Unfortunately, the list must be dynamically allocated as its size is variable. That means the driver must generate the list twice - once to work out the size and a second time to actually save it.
v2:
(Alan / CI)
- GEN7_GT_MODE -> GEN6_GT_MODE to fix WA selftest failure
--- drivers/gpu/drm/i915/gt/intel_workarounds.c | 46 ++-- .../gpu/drm/i915/gt/intel_workarounds_types.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_guc.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 199 +++++++++++++++++- drivers/gpu/drm/i915/i915_reg.h | 1 + 5 files changed, 222 insertions(+), 26 deletions(-)
drivers/gpu/drm/i915/gt/intel_workarounds.c +24 -22
@@ -150,13 +150,14 @@ static void _wa_add(struct i915_wa_list *wal, const struct i915_wa *wa) }static void wa_add (struct i915_wa_list *wal,i915_reg_t reg, -u32 clear,u32 set,u32 read_mask) +u32 clear,u32 set,u32 read_mask,bool masked_reg) {struct i915_wa wa = { .reg = reg, .clr = clear, .set = set, .read = read_mask, + .masked_reg = masked_reg, };_wa_add (wal, &wa);
@@ -165,7 +166,7 @@ static void wa_add(struct i915_wa_list *wal, i915_reg_t reg,static void wa_write_clr_set (struct i915_wa_list *wal,i915_reg_t reg,u32 clear,u32 set) { -wa_add (wal, reg, clear, set, clear); +wa_add (wal, reg, clear, set, clear, false); } static void
@@ -200,20 +201,20 @@ wa_write_clr(struct i915_wa_list *wal, i915_reg_t reg, u32 clr)static void wa_masked_en (struct i915_wa_list *wal,i915_reg_t reg,u32 val) { -wa_add (wal, reg,0 ,_MASKED_BIT_ENABLE (val), val); +wa_add (wal, reg,0 ,_MASKED_BIT_ENABLE (val), val, true); }static void wa_masked_dis (struct i915_wa_list *wal,i915_reg_t reg,u32 val) { -wa_add (wal, reg,0 ,_MASKED_BIT_DISABLE (val), val); +wa_add (wal, reg,0 ,_MASKED_BIT_DISABLE (val), val, true); }static void wa_masked_field_set (struct i915_wa_list *wal,i915_reg_t reg,u32 mask,u32 val) { -wa_add (wal, reg,0 ,_MASKED_FIELD (mask, val), mask); +wa_add (wal, reg,0 ,_MASKED_FIELD (mask, val), mask, true); } static void gen6_ctx_workarounds_init(struct intel_engine_cs *engine,
@@ -583,10 +584,10 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine, GEN11_BLEND_EMB_FIX_DISABLE_IN_RCC);/* WaEnableFloatBlendOptimization:icl */ -wa_write_clr_set (wal, -GEN10_CACHE_MODE_SS , -0 ,/* write-only, so skip validation */ -_MASKED_BIT_ENABLE (FLOAT_BLEND_OPTIMIZATION_ENABLE )); +wa_add (wal,GEN10_CACHE_MODE_SS ,0 , +_MASKED_BIT_ENABLE (FLOAT_BLEND_OPTIMIZATION_ENABLE ), +0 /* write-only, so skip validation */ , + true);/* WaDisableGPGPUMidThreadPreemption:icl */ wa_masked_field_set(wal, GEN8_CS_CHICKEN1,
@@ -631,7 +632,7 @@ static void gen12_ctx_gt_tuning_init(struct intel_engine_cs *engine,FF_MODE2 ,FF_MODE2_TDS_TIMER_MASK ,FF_MODE2_TDS_TIMER_128 , - 0); + 0, false); } static void gen12_ctx_workarounds_init(struct intel_engine_cs *engine,
@@ -669,7 +670,7 @@ static void gen12_ctx_workarounds_init(struct intel_engine_cs *engine,FF_MODE2 ,FF_MODE2_GS_TIMER_MASK ,FF_MODE2_GS_TIMER_224 , - 0); + 0, false);/* * Wa_14012131227:dg1
@@ -847,7 +848,7 @@ hsw_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)wa_add (wal,HSW_ROW_CHICKEN3 ,0 ,_MASKED_BIT_ENABLE (HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE ), -0 /* XXX does this reg exist? */ ); +0 /* XXX does this reg exist? */ , true);/* WaVSRefCountFullforceMissDisable:hsw */ wa_write_clr (wal,GEN7_FF_THREAD_MODE ,GEN7_FF_VS_REF_CNT_FFME );
@@ -1937,10 +1938,10 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)* disable bit, which we don't touch here, but it's good * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM). */ -wa_add (wal,GEN7_GT_MODE ,0 , -_MASKED_FIELD (GEN6_WIZ_HASHING_MASK , - GEN6_WIZ_HASHING_16x4), - GEN6_WIZ_HASHING_16x4); +wa_masked_field_set (wal, +GEN7_GT_MODE , +GEN6_WIZ_HASHING_MASK , + GEN6_WIZ_HASHING_16x4); }if (IS_GRAPHICS_VER (i915,6 ,7 ))
@@ -1990,10 +1991,10 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)* disable bit, which we don't touch here, but it's good * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM). */ -wa_add (wal, -GEN6_GT_MODE ,0 , -_MASKED_FIELD (GEN6_WIZ_HASHING_MASK , GEN6_WIZ_HASHING_16x4), - GEN6_WIZ_HASHING_16x4); +wa_masked_field_set (wal, +GEN6_GT_MODE , +GEN6_WIZ_HASHING_MASK , + GEN6_WIZ_HASHING_16x4);/* WaDisable_RenderCache_OperationalFlush:snb */ wa_masked_dis (wal,CACHE_MODE_0 ,RC_OP_FLUSH_ENABLE );
@@ -2014,7 +2015,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)wa_add (wal,MI_MODE ,0 ,_MASKED_BIT_ENABLE (VS_TIMER_DISPATCH ),/* XXX bit doesn't stick on Broadwater */ -IS_I965G (i915) ?0 :VS_TIMER_DISPATCH ); +IS_I965G (i915) ?0 :VS_TIMER_DISPATCH , true);if (GRAPHICS_VER (i915) ==4 )/*
@@ -2029,7 +2030,8 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)*/ wa_add (wal,ECOSKPD ,0 ,_MASKED_BIT_ENABLE (ECO_CONSTANT_BUFFER_SR_DISABLE ), -0 /* XXX bit doesn't stick on Broadwater */ ); +0 /* XXX bit doesn't stick on Broadwater */ , + true); } static void
drivers/gpu/drm/i915/gt/intel_workarounds_types.h +1 -0
@@ -15,6 +15,7 @@ struct i915_wa {u32 clr;u32 set;u32 read; +bool masked_reg; };struct i915_wa_list {
drivers/gpu/drm/i915/gt/uc/intel_guc.h +1 -0
@@ -59,6 +59,7 @@ struct intel_guc {struct i915_vma *ads_vma;struct __guc_ads_blob *ads_blob; +u32 ads_regset_size;struct i915_vma *lrc_desc_pool;void *lrc_desc_pool_vaddr;
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +195 -4
250 lines not on this page · raw message
drivers/gpu/drm/i915/i915_reg.h +1 -0
@@ -12309,6 +12309,7 @@ enum skl_power_gate {/* MOCS (Memory Object Control State) registers */ #define GEN9_LNCFCMOCS (i) _MMIO(0xb020 + (i) * 4)/* L3 Cache Control */ +#define GEN9_LNCFCMOCS_REG_COUNT 32#define __GEN9_RCS0_MOCS0 0xc800#define GEN9_GFX_MOCS (i) _MMIO(__GEN9_RCS0_MOCS0 + (i) * 4)
Signature · 1 line
2.28.0
Re: [PATCH 33/51] drm/i915/guc: Provide mmio list to be saved/restored on engine reset
Quoted 8 lines · Matthew Brost
From: John Harrison <[email protected]>
The driver must provide GuC with a list of mmio registers that should be saved/restored during a GuC-based engine reset. Unfortunately, the list must be dynamically allocated as its size is variable. That means the driver must generate the list twice - once to work out the size and a second time to actually save it.
Quoted 7 lines
v2:
(Alan / CI)
-GEN7_GT_MODE->GEN6_GT_MODEto fix WA selftest failureSigned-off-by John Harrison <[email protected]>Signed-off-by Fernando Pacheco <[email protected]>Signed-off-by Matthew Brost <[email protected]>
Everything looks structurally correct. Feel confident on my below RB but W/A are not my area of expertise. If any one else wanted to give it a look, I wouldn't mind.
With that:
Quoted 454 lines
Cc Daniele Ceraolo Spurio <[email protected]>Cc Tvrtko Ursulin <[email protected]>--- drivers/gpu/drm/i915/gt/intel_workarounds.c | 46 ++-- .../gpu/drm/i915/gt/intel_workarounds_types.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_guc.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 199 +++++++++++++++++- drivers/gpu/drm/i915/i915_reg.h | 1 + 5 files changed, 222 insertions(+), 26 deletions(-)drivers/gpu/drm/i915/gt/intel_workarounds.c +24
-22@@ -150,13 +150,14 @@ static void _wa_add(struct i915_wa_list *wal, const struct i915_wa *wa) }static void wa_add (struct i915_wa_list *wal,i915_reg_t reg, -u32 clear,u32 set,u32 read_mask) +u32 clear,u32 set,u32 read_mask,bool masked_reg) {struct i915_wa wa = { .reg = reg, .clr = clear, .set = set, .read = read_mask, + .masked_reg = masked_reg, };_wa_add (wal, &wa);@@ -165,7 +166,7 @@ static void wa_add(struct i915_wa_list *wal, i915_reg_t reg,static void wa_write_clr_set (struct i915_wa_list *wal,i915_reg_t reg,u32 clear,u32 set) { -wa_add (wal, reg, clear, set, clear); +wa_add (wal, reg, clear, set, clear, false); } static void@@ -200,20 +201,20 @@ wa_write_clr(struct i915_wa_list *wal, i915_reg_t reg, u32 clr)static void wa_masked_en (struct i915_wa_list *wal,i915_reg_t reg,u32 val) { -wa_add (wal, reg,0 ,_MASKED_BIT_ENABLE (val), val); +wa_add (wal, reg,0 ,_MASKED_BIT_ENABLE (val), val, true); }static void wa_masked_dis (struct i915_wa_list *wal,i915_reg_t reg,u32 val) { -wa_add (wal, reg,0 ,_MASKED_BIT_DISABLE (val), val); +wa_add (wal, reg,0 ,_MASKED_BIT_DISABLE (val), val, true); }static void wa_masked_field_set (struct i915_wa_list *wal,i915_reg_t reg,u32 mask,u32 val) { -wa_add (wal, reg,0 ,_MASKED_FIELD (mask, val), mask); +wa_add (wal, reg,0 ,_MASKED_FIELD (mask, val), mask, true); } static void gen6_ctx_workarounds_init(struct intel_engine_cs *engine,@@ -583,10 +584,10 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine, GEN11_BLEND_EMB_FIX_DISABLE_IN_RCC);/* WaEnableFloatBlendOptimization:icl */ -wa_write_clr_set (wal, -GEN10_CACHE_MODE_SS , -0 ,/* write-only, so skip validation */ -_MASKED_BIT_ENABLE (FLOAT_BLEND_OPTIMIZATION_ENABLE )); +wa_add (wal,GEN10_CACHE_MODE_SS ,0 , +_MASKED_BIT_ENABLE (FLOAT_BLEND_OPTIMIZATION_ENABLE ), +0 /* write-only, so skip validation */ , + true);/* WaDisableGPGPUMidThreadPreemption:icl */ wa_masked_field_set(wal, GEN8_CS_CHICKEN1,@@ -631,7 +632,7 @@ static void gen12_ctx_gt_tuning_init(struct intel_engine_cs *engine,FF_MODE2 ,FF_MODE2_TDS_TIMER_MASK ,FF_MODE2_TDS_TIMER_128 , - 0); + 0, false); } static void gen12_ctx_workarounds_init(struct intel_engine_cs *engine,@@ -669,7 +670,7 @@ static void gen12_ctx_workarounds_init(struct intel_engine_cs *engine,FF_MODE2 ,FF_MODE2_GS_TIMER_MASK ,FF_MODE2_GS_TIMER_224 , - 0); + 0, false);/* * Wa_14012131227:dg1 @@ -847,7 +848,7 @@ hsw_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)wa_add (wal,HSW_ROW_CHICKEN3 ,0 ,_MASKED_BIT_ENABLE (HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE ), -0 /* XXX does this reg exist? */ ); +0 /* XXX does this reg exist? */ , true);/* WaVSRefCountFullforceMissDisable:hsw */ wa_write_clr (wal,GEN7_FF_THREAD_MODE ,GEN7_FF_VS_REF_CNT_FFME );@@ -1937,10 +1938,10 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)* disable bit, which we don't touch here, but it's good * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM). */ -wa_add (wal,GEN7_GT_MODE ,0 , -_MASKED_FIELD (GEN6_WIZ_HASHING_MASK , - GEN6_WIZ_HASHING_16x4), - GEN6_WIZ_HASHING_16x4); +wa_masked_field_set (wal, +GEN7_GT_MODE , +GEN6_WIZ_HASHING_MASK , + GEN6_WIZ_HASHING_16x4); }if (IS_GRAPHICS_VER (i915,6 ,7 ))@@ -1990,10 +1991,10 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)* disable bit, which we don't touch here, but it's good * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM). */ -wa_add (wal, -GEN6_GT_MODE ,0 , -_MASKED_FIELD (GEN6_WIZ_HASHING_MASK , GEN6_WIZ_HASHING_16x4), - GEN6_WIZ_HASHING_16x4); +wa_masked_field_set (wal, +GEN6_GT_MODE , +GEN6_WIZ_HASHING_MASK , + GEN6_WIZ_HASHING_16x4);/* WaDisable_RenderCache_OperationalFlush:snb */ wa_masked_dis (wal,CACHE_MODE_0 ,RC_OP_FLUSH_ENABLE );@@ -2014,7 +2015,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)wa_add (wal,MI_MODE ,0 ,_MASKED_BIT_ENABLE (VS_TIMER_DISPATCH ),/* XXX bit doesn't stick on Broadwater */ -IS_I965G (i915) ?0 :VS_TIMER_DISPATCH ); +IS_I965G (i915) ?0 :VS_TIMER_DISPATCH , true);if (GRAPHICS_VER (i915) ==4 )/* @@ -2029,7 +2030,8 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)*/ wa_add (wal,ECOSKPD ,0 ,_MASKED_BIT_ENABLE (ECO_CONSTANT_BUFFER_SR_DISABLE ), -0 /* XXX bit doesn't stick on Broadwater */ ); +0 /* XXX bit doesn't stick on Broadwater */ , + true); } static voiddrivers/gpu/drm/i915/gt/intel_workarounds_types.h +1
-0@@ -15,6 +15,7 @@ struct i915_wa {u32 clr;u32 set;u32 read; +bool masked_reg; };struct i915_wa_list {drivers/gpu/drm/i915/gt/uc/intel_guc.h +1
-0@@ -59,6 +59,7 @@ struct intel_guc {struct i915_vma *ads_vma;struct __guc_ads_blob *ads_blob; +u32 ads_regset_size;struct i915_vma *lrc_desc_pool;void *lrc_desc_pool_vaddr;drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +195
-4@@ -3,6 +3,8 @@* Copyright © 2014-2019 Intel Corporation */ +#include <linux/bsearch.h> +#include "gt/intel_gt.h" #include "gt/intel_lrc.h" #include "intel_guc_ads.h" @@ -23,7 +25,12 @@* | guc_policies | * +---------------------------------------+ * | guc_gt_system_info | -* +---------------------------------------+ +* +---------------------------------------+ <== static +* | guc_mmio_reg[countA] (engine 0.0) | +* | guc_mmio_reg[countB] (engine 0.1) | +* | guc_mmio_reg[countC] (engine 1.0) | +* | ... | +* +---------------------------------------+ <== dynamic * | padding | * +---------------------------------------+ <== 4K aligned * | private data | @@ -35,16 +42,33 @@ struct __guc_ads_blob {struct guc_ads ads;struct guc_policies policies;struct guc_gt_system_info system_info; +/* From here on, location is dynamic! Refer to above diagram. */ +struct guc_mmio_reg regset[0 ]; } __packed; +static u32 guc_ads_regset_size (struct intel_guc *guc) +{ +GEM_BUG_ON (!guc->ads_regset_size); +return guc->ads_regset_size; +} +static u32 guc_ads_private_data_size (struct intel_guc *guc) {return PAGE_ALIGN (guc->fw.private_data_size); } +static u32 guc_ads_regset_offset (struct intel_guc *guc) +{ +return offsetof(struct __guc_ads_blob , regset); +} +static u32 guc_ads_private_data_offset (struct intel_guc *guc) { -return PAGE_ALIGN (sizeof (struct __guc_ads_blob )); +u32 offset; + + offset =guc_ads_regset_offset (guc) + +guc_ads_regset_size (guc); +return PAGE_ALIGN (offset); }static u32 guc_ads_blob_size (struct intel_guc *guc)@@ -83,6 +107,165 @@ static void guc_mapping_table_init(struct intel_gt *gt, } } +/* +* The save/restore register list must be pre-calculated to a temporary +* buffer of driver defined size before it can be generated in place +* inside the ADS. +*/ +#define MAX_MMIO_REGS 128/* Arbitrary size, increase as needed */ +struct temp_regset { +struct guc_mmio_reg *registers; +u32 used; +u32 size; +}; + +static int guc_mmio_reg_cmp (const void *a,const void *b) +{ +const struct guc_mmio_reg *ra = a; +const struct guc_mmio_reg *rb = b; + +return (int )ra->offset - (int )rb->offset; +} + +static void guc_mmio_reg_add (struct temp_regset *regset, +u32 offset,u32 flags) +{ +u32 count = regset->used; +struct guc_mmio_reg reg = { + .offset = offset, + .flags = flags, + }; +struct guc_mmio_reg *slot; + +GEM_BUG_ON (count >= regset->size); + +/* +* The mmio list is built using separate lists within the driver. +* It's possible that at some point we may attempt to add the same +* register more than once. Do not consider this an error; silently +* move on if the register is already in the list. +*/ +if (bsearch (®, regset->registers, count, +sizeof (reg), guc_mmio_reg_cmp)) +return ; + + slot = ®set->registers[count]; + regset->used++; + *slot = reg; + +while (slot-- > regset->registers) { +GEM_BUG_ON (slot[0 ].offset == slot[1 ].offset); +if (slot[1 ].offset > slot[0 ].offset) +break ; + +swap (slot[1 ], slot[0 ]); + } +} + +#define GUC_MMIO_REG_ADD (regset, reg, masked) \ + guc_mmio_reg_add(regset, \ + i915_mmio_reg_offset((reg)), \ + (masked) ? GUC_REGSET_MASKED : 0) + +static void guc_mmio_regset_init (struct temp_regset *regset, +struct intel_engine_cs *engine) +{ +const u32 base = engine->mmio_base; +struct i915_wa_list *wal = &engine->wa_list; +struct i915_wa *wa; +unsigned int i; + + regset->used =0 ; + +GUC_MMIO_REG_ADD (regset,RING_MODE_GEN7 (base), true); +GUC_MMIO_REG_ADD (regset,RING_HWS_PGA (base), false); +GUC_MMIO_REG_ADD (regset,RING_IMR (base), false); + +for (i =0 , wa = wal->list; i < wal->count; i++, wa++) +GUC_MMIO_REG_ADD (regset, wa->reg, wa->masked_reg); + +/* Be extra paranoid and include all whitelist registers. */ +for (i =0 ; i <RING_MAX_NONPRIV_SLOTS ; i++) +GUC_MMIO_REG_ADD (regset, +RING_FORCE_TO_NONPRIV (base, i), + false); + +/* add in local MOCS registers */ +for (i =0 ; i <GEN9_LNCFCMOCS_REG_COUNT ; i++) +GUC_MMIO_REG_ADD (regset,GEN9_LNCFCMOCS (i), false); +} + +static int guc_mmio_reg_state_query (struct intel_guc *guc) +{ +struct intel_gt *gt =guc_to_gt (guc); +struct intel_engine_cs *engine; +enum intel_engine_id id; +struct temp_regset temp_set; +u32 total; + +/* +* Need to actually build the list in order to filter out +* duplicates and other such data dependent constructions. +*/ + temp_set.size =MAX_MMIO_REGS ; + temp_set.registers =kmalloc_array (temp_set.size, +sizeof (*temp_set.registers), +GFP_KERNEL ); +if (!temp_set.registers) +return -ENOMEM ; + + total =0 ; +for_each_engine (engine, gt, id) { +guc_mmio_regset_init (&temp_set, engine); + total += temp_set.used; + } + +kfree (temp_set.registers); + +return total *sizeof (struct guc_mmio_reg ); +} + +static void guc_mmio_reg_state_init (struct intel_guc *guc, +struct __guc_ads_blob *blob) +{ +struct intel_gt *gt =guc_to_gt (guc); +struct intel_engine_cs *engine; +enum intel_engine_id id; +struct temp_regset temp_set; +struct guc_mmio_reg_set *ads_reg_set; +u32 addr_ggtt, offset; +u8 guc_class; + + offset =guc_ads_regset_offset (guc); + addr_ggtt =intel_guc_ggtt_offset (guc, guc->ads_vma) + offset; + temp_set.registers = (struct guc_mmio_reg *) (((u8 *) blob) + offset); + temp_set.size = guc->ads_regset_size /sizeof (temp_set.registers[0 ]); + +for_each_engine (engine, gt, id) { +/* Class index is checked in class converter */ +GEM_BUG_ON (engine->instance >=GUC_MAX_INSTANCES_PER_CLASS ); + + guc_class =engine_class_to_guc_class (engine->class); + ads_reg_set = &blob->ads.reg_state_list[guc_class][engine->instance]; + +guc_mmio_regset_init (&temp_set, engine); +if (!temp_set.used) { + ads_reg_set->address =0 ; + ads_reg_set->count =0 ; +continue ; + } + + ads_reg_set->address = addr_ggtt; + ads_reg_set->count = temp_set.used; + + temp_set.size -= temp_set.used; + temp_set.registers += temp_set.used; + addr_ggtt += temp_set.used *sizeof (struct guc_mmio_reg ); + } + +GEM_BUG_ON (temp_set.size); +} +/* * The first 80 dwords of the register state context, containing the * execlists and ppgtt registers. @@ -121,8 +304,7 @@ static void __guc_ads_init(struct intel_guc *guc)*/ blob->ads.golden_context_lrca[guc_class] =0 ; blob->ads.eng_state_size[guc_class] = -intel_engine_context_size (guc_to_gt (guc), - engine_class) - +intel_engine_context_size (gt, engine_class) - skipped_size; }@@ -153,6 +335,9 @@ static void __guc_ads_init(struct intel_guc *guc) blob->ads.scheduler_policies = base +ptr_offset (blob, policies); blob->ads.gt_system_info = base +ptr_offset (blob, system_info); +/* MMIO save/restore list */ +guc_mmio_reg_state_init (guc, blob); +/* Private Data */ blob->ads.private_data = base +guc_ads_private_data_offset (guc);@@ -173,6 +358,12 @@ int intel_guc_ads_create(struct intel_guc *guc)GEM_BUG_ON (guc->ads_vma); +/* Need to calculate the reg state size dynamically: */ + ret =guc_mmio_reg_state_query (guc); +if (ret <0 ) +return ret; + guc->ads_regset_size = ret; + size =guc_ads_blob_size (guc); ret = intel_guc_allocate_and_map_vma(guc, size, &guc->ads_vma,drivers/gpu/drm/i915/i915_reg.h +1
-0@@ -12309,6 +12309,7 @@ enum skl_power_gate {/* MOCS (Memory Object Control State) registers */ #define GEN9_LNCFCMOCS (i) _MMIO(0xb020 + (i) * 4)/* L3 Cache Control */ +#define GEN9_LNCFCMOCS_REG_COUNT 32#define __GEN9_RCS0_MOCS0 0xc800#define GEN9_GFX_MOCS (i) _MMIO(__GEN9_RCS0_MOCS0 + (i) * 4)Signature · 1 line
2.28.0
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 34/51 drm/i915/guc: Don't complain about reset races
In reply to Matthew BrostFrom: John Harrison <[email protected]>
It is impossible to seal all race conditions of resets occurring concurrent to other operations. At least, not without introducing excesive mutex locking. Instead, don't complain if it occurs. In particular, don't complain if trying to send a H2G during a reset. Whatever the H2G was about should get redone once the reset is over.
--- drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 5 ++++- drivers/gpu/drm/i915/gt/uc/intel_uc.c | 3 +++ drivers/gpu/drm/i915/gt/uc/intel_uc.h | 2 ++ 3 files changed, 9 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +4 -1
@@ -757,7 +757,10 @@ int intel_guc_ct_send(struct intel_guc_ct *ct, const u32 *action, u32 len,int ret;if (unlikely (!ct->enabled)) { -WARN (1 ,"Unexpected send: action=%#x\n" , *action); +struct intel_guc *guc =ct_to_guc (ct); + struct intel_uc *uc = container_of(guc, struct intel_uc, guc); + +WARN (!uc->reset_in_progress,"Unexpected send: action=%#x\n" , *action);return -ENODEV ; }
drivers/gpu/drm/i915/gt/uc/intel_uc.c +3 -0
@@ -550,6 +550,7 @@ void intel_uc_reset_prepare(struct intel_uc *uc) {struct intel_guc *guc = &uc->guc; + uc->reset_in_progress = true;/* Nothing to do if GuC isn't supported */ if (!intel_uc_supports_guc (uc))
@@ -579,6 +580,8 @@ void intel_uc_reset_finish(struct intel_uc *uc) {struct intel_guc *guc = &uc->guc; + uc->reset_in_progress = false; +/* Firmware expected to be running when this function is called */ if (intel_guc_is_fw_running (guc) &&intel_uc_uses_guc_submission (uc))intel_guc_submission_reset_finish (guc);
drivers/gpu/drm/i915/gt/uc/intel_uc.h +2 -0
@@ -30,6 +30,8 @@ struct intel_uc {/* Snapshot of GuC log from last failed load */ struct drm_i915_gem_object *load_err_log; + +bool reset_in_progress; };void intel_uc_init_early (struct intel_uc *uc);
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 35/51 drm/i915/guc: Enable GuC engine reset
In reply to Matthew BrostFrom: John Harrison <[email protected]>
Clear the 'disable resets' flag to allow GuC to reset hung contexts (detected via pre-emption timeout).
--- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +1 -2
@@ -81,8 +81,7 @@ static void guc_policies_init(struct guc_policies *policies) { policies->dpc_promote_time =GLOBAL_POLICY_DEFAULT_DPC_PROMOTE_TIME_US ; policies->max_num_work_items =GLOBAL_POLICY_MAX_NUM_WI ; -/* Disable automatic resets as not yet supported. */ - policies->global_flags =GLOBAL_POLICY_DISABLE_ENGINE_RESET ; + policies->global_flags =0 ; policies->is_valid =1 ; }
Signature · 1 line
2.28.0
PATCH 36/51 drm/i915/guc: Capture error state on context reset
In reply to Matthew BrostWe receive notification of an engine reset from GuC at its completion. Meaning GuC has potentially cleared any HW state we may have been interested in capturing. GuC resumes scheduling on the engine post-reset, as the resets are meant to be transparent, further muddling our error state.
There is ongoing work to define an API for a GuC debug state dump. The suggestion for now is to manually disable FW initiated resets in cases where debug state is needed.
--- drivers/gpu/drm/i915/gt/intel_context.c | 20 +++++++++++ drivers/gpu/drm/i915/gt/intel_context.h | 3 ++ drivers/gpu/drm/i915/gt/intel_engine.h | 21 ++++++++++- drivers/gpu/drm/i915/gt/intel_engine_cs.c | 11 ++++-- drivers/gpu/drm/i915/gt/intel_engine_types.h | 2 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 35 +++++++++---------- drivers/gpu/drm/i915/i915_gpu_error.c | 25 ++++++++++--- 7 files changed, 91 insertions(+), 26 deletions(-)
drivers/gpu/drm/i915/gt/intel_context.c +20 -0
@@ -515,6 +515,26 @@ struct i915_request *intel_context_create_request(struct intel_context *ce)return rq; } +struct i915_request *intel_context_find_active_request (struct intel_context *ce) +{ +struct i915_request *rq, *active =NULL ; +unsigned long flags; + +GEM_BUG_ON (!intel_engine_uses_guc (ce->engine)); + +spin_lock_irqsave (&ce->guc_active.lock, flags); +list_for_each_entry_reverse (rq, &ce->guc_active.requests, + sched.link) { +if (i915_request_completed (rq)) +break ; + + active = rq; + } +spin_unlock_irqrestore (&ce->guc_active.lock, flags); + +return active; +} +#if IS_ENABLED (CONFIG_DRM_I915_SELFTEST )#include "selftest_context.c" #endif
drivers/gpu/drm/i915/gt/intel_context.h +3 -0
@@ -200,6 +200,9 @@ int intel_context_prepare_remote_request(struct intel_context *ce,struct i915_request *intel_context_create_request (struct intel_context *ce); +struct i915_request * +intel_context_find_active_request (struct intel_context *ce); +static inline bool intel_context_is_barrier (const struct intel_context *ce) {return test_bit (CONTEXT_BARRIER_BIT , &ce->flags);
drivers/gpu/drm/i915/gt/intel_engine.h +20 -1
@@ -245,7 +245,7 @@ ktime_t intel_engine_get_busy_time(struct intel_engine_cs *engine, ktime_t *now);struct i915_request * -intel_engine_find_active_request (struct intel_engine_cs *engine); +intel_engine_execlist_find_hung_request (struct intel_engine_cs *engine);u32 intel_engine_context_size (struct intel_gt *gt,u8 class); struct intel_context *
@@ -310,4 +310,23 @@ intel_engine_get_sibling(struct intel_engine_cs *engine, unsigned int sibling) return engine->cops->get_sibling(engine, sibling); } +static inline void +intel_engine_set_hung_context (struct intel_engine_cs *engine, +struct intel_context *ce) +{ + engine->hung_ce = ce; +} + +static inline void +intel_engine_clear_hung_context (struct intel_engine_cs *engine) +{ +intel_engine_set_hung_context (engine,NULL ); +} + +static inline struct intel_context * +intel_engine_get_hung_context (struct intel_engine_cs *engine) +{ +return engine->hung_ce; +} +#endif /* _INTEL_RINGBUFFER_H_ */
drivers/gpu/drm/i915/gt/intel_engine_cs.c +9 -2
@@ -1672,7 +1672,7 @@ void intel_engine_dump(struct intel_engine_cs *engine,drm_printf (m,"\tRequests:\n" );spin_lock_irqsave (&engine->sched_engine->lock, flags); - rq =intel_engine_find_active_request (engine); + rq =intel_engine_execlist_find_hung_request (engine);if (rq) {struct intel_timeline *tl =get_timeline (rq);
@@ -1783,10 +1783,17 @@ static bool match_ring(struct i915_request *rq) }struct i915_request * -intel_engine_find_active_request (struct intel_engine_cs *engine) +intel_engine_execlist_find_hung_request (struct intel_engine_cs *engine) {struct i915_request *request, *active =NULL ; +/* +* This search does not work in GuC submission mode. However, the GuC +* will report the hanging context directly to the driver itself. So +* the driver should never get here when in GuC mode. +*/ +GEM_BUG_ON (intel_uc_uses_guc_submission (&engine->gt->uc)); +/* * We are called by the error capture, reset and to dump engine * state at random points in time. In particular, note that neither is
drivers/gpu/drm/i915/gt/intel_engine_types.h +2 -0
@@ -304,6 +304,8 @@ struct intel_engine_cs {/* keep a request in reserve for a [pm] barrier under oom */ struct i915_request *request_pool; +struct intel_context *hung_ce; +struct llist_head barrier_tasks;struct intel_context *kernel_context;/* pinned */
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +16 -19
@@ -724,24 +724,6 @@ __unwind_incomplete_requests(struct intel_context *ce)spin_unlock_irqrestore (&sched_engine->lock, flags); } -static struct i915_request *context_find_active_request (struct intel_context *ce) -{ -struct i915_request *rq, *active =NULL ; -unsigned long flags; - -spin_lock_irqsave (&ce->guc_active.lock, flags); -list_for_each_entry_reverse (rq, &ce->guc_active.requests, - sched.link) { -if (i915_request_completed (rq)) -break ; - - active = rq; - } -spin_unlock_irqrestore (&ce->guc_active.lock, flags); - -return active; -} -static void __guc_reset_context (struct intel_context *ce,bool stalled) {struct i915_request *rq;
@@ -755,7 +737,7 @@ static void __guc_reset_context(struct intel_context *ce, bool stalled)*/ clr_context_enabled (ce); - rq =context_find_active_request (ce); + rq =intel_context_find_active_request (ce);if (!rq) { head = ce->ring->tail; stalled = false;
@@ -2196,6 +2178,20 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc,return 0 ; } +static void capture_error_state (struct intel_guc *guc, +struct intel_context *ce) +{ +struct intel_gt *gt =guc_to_gt (guc); +struct drm_i915_private *i915 = gt->i915; +struct intel_engine_cs *engine =__context_to_physical_engine (ce); +intel_wakeref_t wakeref; + +intel_engine_set_hung_context (engine, ce); + with_intel_runtime_pm(&i915->runtime_pm, wakeref) +i915_capture_error_state (gt, engine->mask); +atomic_inc (&i915->gpu_error.reset_engine_count[engine->uabi_class]); +} +static void guc_context_replay (struct intel_context *ce) {struct i915_sched_engine *sched_engine = ce->engine->sched_engine;
@@ -2208,6 +2204,7 @@ static void guc_handle_context_reset(struct intel_guc *guc, struct intel_context *ce) {trace_intel_context_reset (ce); +capture_error_state (guc, ce);guc_context_replay (ce); }
drivers/gpu/drm/i915/i915_gpu_error.c +21 -4
@@ -1429,20 +1429,37 @@ capture_engine(struct intel_engine_cs *engine, {struct intel_engine_capture_vma *capture =NULL ;struct intel_engine_coredump *ee; -struct i915_request *rq; +struct intel_context *ce; +struct i915_request *rq =NULL ;unsigned long flags; ee =intel_engine_coredump_alloc (engine,GFP_KERNEL );if (!ee)return NULL ; -spin_lock_irqsave (&engine->sched_engine->lock, flags); - rq =intel_engine_find_active_request (engine); + ce =intel_engine_get_hung_context (engine); +if (ce) { +intel_engine_clear_hung_context (engine); + rq =intel_context_find_active_request (ce); +if (!rq || !i915_request_started (rq)) + goto no_request_capture; + }else { +/* +* Getting here with GuC enabled means it is a forced error capture +* with no actual hang. So, no need to attempt the execlist search. +*/ +if (!intel_uc_uses_guc_submission (&engine->gt->uc)) { +spin_lock_irqsave (&engine->sched_engine->lock, flags); + rq =intel_engine_execlist_find_hung_request (engine); +spin_unlock_irqrestore (&engine->sched_engine->lock, + flags); + } + }if (rq) capture =intel_engine_coredump_add_request (ee, rq,ATOMIC_MAYFAIL ); -spin_unlock_irqrestore (&engine->sched_engine->lock, flags);if (!capture) { +no_request_capture:kfree (ee);return NULL ; }
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 37/51 drm/i915/guc: Fix for error capture after full GPU reset with GuC
In reply to Matthew BrostFrom: John Harrison <[email protected]>
In the case of a full GPU reset (e.g. because GuC has died or because GuC's hang detection has been disabled), the driver can't rely on GuC reporting the guilty context. Instead, the driver needs to scan all active contexts and find one that is currently executing, as per the execlist mode behaviour. In GuC mode, this scan is different to execlist mode as the active request list is handled very differently.
Similarly, the request state dump in debugfs needs to be handled differently when in GuC submission mode.
Also refactured some of the request scanning code to avoid duplication across the multiple code paths that are now replicating it.
--- drivers/gpu/drm/i915/gt/intel_engine.h | 3 + drivers/gpu/drm/i915/gt/intel_engine_cs.c | 139 ++++++++++++------ .../gpu/drm/i915/gt/intel_engine_heartbeat.c | 8 + drivers/gpu/drm/i915/gt/intel_reset.c | 2 +- drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 67 +++++++++ .../gpu/drm/i915/gt/uc/intel_guc_submission.h | 3 + drivers/gpu/drm/i915/i915_request.c | 41 ++++++ drivers/gpu/drm/i915/i915_request.h | 11 ++ 9 files changed, 229 insertions(+), 47 deletions(-)
drivers/gpu/drm/i915/gt/intel_engine.h +3 -0
@@ -240,6 +240,9 @@ __printf(3, 4)void intel_engine_dump (struct intel_engine_cs *engine,struct drm_printer *m,const char *header, ...); +void intel_engine_dump_active_requests (struct list_head *requests, +struct i915_request *hung_rq, +struct drm_printer *m);ktime_t intel_engine_get_busy_time (struct intel_engine_cs *engine,ktime_t *now);
drivers/gpu/drm/i915/gt/intel_engine_cs.c +93 -46
165 lines not on this page · raw message
drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c +8 -0
@@ -90,6 +90,14 @@ reset_engine(struct intel_engine_cs *engine, struct i915_request *rq)if (IS_ENABLED (CONFIG_DRM_I915_DEBUG_GEM ))show_heartbeat (rq, engine); +if (intel_engine_uses_guc (engine)) +/* +* GuC itself is toast or GuC's hang detection +* is disabled. Either way, need to find the +* hang culprit manually. +*/ +intel_guc_find_hung_context (engine); + intel_gt_handle_error(engine->gt, engine->mask, I915_ERROR_CAPTURE, "stopped heartbeat on %s",
drivers/gpu/drm/i915/gt/intel_reset.c +1 -1
@@ -156,7 +156,7 @@ void __i915_request_reset(struct i915_request *rq, bool guilty)if (guilty) {i915_request_set_error_once (rq, -EIO );__i915_request_skip (rq); -if (mark_guilty (rq)) +if (mark_guilty (rq) && !intel_engine_uses_guc (rq->engine))skip_context (rq); }else {i915_request_set_error_once (rq, -EAGAIN );
drivers/gpu/drm/i915/gt/uc/intel_guc.h +2 -0
@@ -268,6 +268,8 @@ int intel_guc_context_reset_process_msg(struct intel_guc *guc,int intel_guc_engine_failure_process_msg (struct intel_guc *guc,const u32 *msg,u32 len); +void intel_guc_find_hung_context (struct intel_engine_cs *engine); +void intel_guc_submission_reset_prepare (struct intel_guc *guc);void intel_guc_submission_reset (struct intel_guc *guc,bool stalled);void intel_guc_submission_reset_finish (struct intel_guc *guc);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +67 -0
73 lines not on this page · raw message
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.h +3 -0
@@ -25,6 +25,9 @@ void intel_guc_submission_print_info(struct intel_guc *guc, struct drm_printer *p);void intel_guc_submission_print_context_info (struct intel_guc *guc,struct drm_printer *p); +void intel_guc_dump_active_requests (struct intel_engine_cs *engine, +struct i915_request *hung_rq, +struct drm_printer *m);bool intel_guc_virtual_engine_has_heartbeat (const struct intel_engine_cs *ve);
drivers/gpu/drm/i915/i915_request.c +41 -0
47 lines not on this page · raw message
drivers/gpu/drm/i915/i915_request.h +11 -0
@@ -649,4 +649,15 @@ i915_request_active_engine(struct i915_request *rq,void i915_request_notify_execute_cb_imm (struct i915_request *rq); +enum i915_request_state +{ +I915_REQUEST_UNKNOWN =0 , +I915_REQUEST_COMPLETE , +I915_REQUEST_PENDING , +I915_REQUEST_QUEUED , +I915_REQUEST_ACTIVE , +}; + +enum i915_request_state i915_test_request_state (struct i915_request *rq); +#endif /* I915_REQUEST_H */
Signature · 1 line
2.28.0
PATCH 38/51 drm/i915/guc: Hook GuC scheduling policies up
In reply to Matthew BrostFrom: John Harrison <[email protected]>
Use the official driver default scheduling policies for configuring the GuC scheduler rather than a bunch of hardcoded values.
v2:
(Matthew Brost)
- Move I915_ENGINE_WANT_FORCED_PREEMPTION to later patch
--- drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 44 ++++++++++++++++++- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 8 ++-- 3 files changed, 49 insertions(+), 5 deletions(-)
drivers/gpu/drm/i915/gt/uc/intel_guc.h +2 -0
@@ -270,6 +270,8 @@ int intel_guc_engine_failure_process_msg(struct intel_guc *guc,void intel_guc_find_hung_context (struct intel_engine_cs *engine); +int intel_guc_global_policies_update (struct intel_guc *guc); +void intel_guc_submission_reset_prepare (struct intel_guc *guc);void intel_guc_submission_reset (struct intel_guc *guc,bool stalled);void intel_guc_submission_reset_finish (struct intel_guc *guc);
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +42 -2
63 lines not on this page · raw message
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +5 -3
@@ -870,6 +870,7 @@ void intel_guc_submission_reset_finish(struct intel_guc *guc)GEM_WARN_ON (atomic_read (&guc->outstanding_submission_g2h));atomic_set (&guc->outstanding_submission_g2h,0 ); +intel_guc_global_policies_update (guc);enable_submission (guc);intel_gt_unpark_heartbeats (guc_to_gt (guc)); }
@@ -1160,8 +1161,9 @@ static void guc_context_policy_init(struct intel_engine_cs *engine, { desc->policy_flags =0 ; - desc->execution_quantum =CONTEXT_POLICY_DEFAULT_EXECUTION_QUANTUM_US ; - desc->preemption_timeout =CONTEXT_POLICY_DEFAULT_PREEMPTION_TIME_US ; +/* NB: For both of these, zero means disabled. */ + desc->execution_quantum = engine->props.timeslice_duration_ms *1000 ; + desc->preemption_timeout = engine->props.preempt_timeout_ms *1000 ; }static int guc_lrc_desc_pin (struct intel_context *ce,bool loop)
@@ -1937,13 +1939,13 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine) engine->set_default_submission = guc_set_default_submission; engine->flags |=I915_ENGINE_HAS_PREEMPTION ; + engine->flags |=I915_ENGINE_HAS_TIMESLICES ;/* * TODO: GuC supports timeslicing and semaphores as well, but they're * handled by the firmware so some minor tweaks are required before * enabling. * -* engine->flags |= I915_ENGINE_HAS_TIMESLICES; * engine->flags |= I915_ENGINE_HAS_SEMAPHORES; */
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 39/51 drm/i915/guc: Connect reset modparam updates to GuC policy flags
In reply to Matthew BrostFrom: John Harrison <[email protected]>
Changing the reset module parameter has no effect on a running GuC. The corresponding entry in the ADS must be updated and then the GuC informed via a Host2GuC message.
The new debugfs interface to module parameters allows this to happen. However, connecting the parameter data address back to anything useful is messy. One option would be to pass a new private data structure address through instead of just the parameter pointer. However, that means having a new (and different) data structure for each parameter and a new (and different) write function for each parameter. This method keeps everything generic by instead using a string lookup on the directory entry name.
--- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 2 +- drivers/gpu/drm/i915/i915_debugfs_params.c | 31 ++++++++++++++++++++++ 2 files changed, 32 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +1 -1
@@ -99,7 +99,7 @@ static int guc_action_policies_update(struct intel_guc *guc, u32 policy_offset) policy_offset }; -return intel_guc_send (guc, action,ARRAY_SIZE (action)); +return intel_guc_send_busy_loop (guc, action,ARRAY_SIZE (action),0 , true); }int intel_guc_global_policies_update (struct intel_guc *guc)
drivers/gpu/drm/i915/i915_debugfs_params.c +31 -0
@@ -6,9 +6,20 @@#include <linux/kernel.h> #include "i915_debugfs_params.h" +#include "gt/intel_gt.h" +#include "gt/uc/intel_guc.h" #include "i915_drv.h" #include "i915_params.h" +#define MATCH_DEBUGFS_NODE_NAME (_file, _name) (strcmp((_file)->f_path.dentry->d_name.name, (_name)) == 0) + +#define GET_I915 (i915, name, ptr) \ + do { \ + struct i915_params *params; \ + params = container_of(((void *) (ptr)), typeof(*params), name); \ + (i915) = container_of(params, typeof(*(i915)), params); \ + } while(0) +/* int param */ static int i915_param_int_show (struct seq_file *m,void *data) {
@@ -24,6 +35,16 @@ static int i915_param_int_open(struct inode *inode, struct file *file)return single_open (file, i915_param_int_show, inode->i_private); } +static int notify_guc (struct drm_i915_private *i915) +{ +int ret =0 ; + +if (intel_uc_uses_guc_submission (&i915->gt.uc)) + ret =intel_guc_global_policies_update (&i915->gt.uc.guc); + +return ret; +} +static ssize_t i915_param_int_write (struct file *file,const char __user *ubuf,size_t len,loff_t *offp)
@@ -81,8 +102,10 @@ static ssize_t i915_param_uint_write(struct file *file, const char __user *ubuf, size_t len, loff_t *offp) { + struct drm_i915_private *i915;struct seq_file *m = file->private_data;unsigned int *value = m->private; +unsigned int old = *value;int ret; ret =kstrtouint_from_user (ubuf, len,0 , value);
@@ -95,6 +118,14 @@ static ssize_t i915_param_uint_write(struct file *file, *value = b; } +if (!ret &&MATCH_DEBUGFS_NODE_NAME (file,"reset" )) { +GET_I915 (i915, reset, value); + + ret =notify_guc (i915); +if (ret) + *value = old; + } +return ret ?: len; }
Signature · 1 line
2.28.0
Re: [PATCH 39/51] drm/i915/guc: Connect reset modparam updates to GuC policy flags
Quoted 15 lines · Matthew Brost
From: John Harrison <[email protected]>
Changing the reset module parameter has no effect on a running GuC. The corresponding entry in the ADS must be updated and then the GuC informed via a Host2GuC message.
The new debugfs interface to module parameters allows this to happen. However, connecting the parameter data address back to anything useful is messy. One option would be to pass a new private data structure address through instead of just the parameter pointer. However, that means having a new (and different) data structure for each parameter and a new (and different) write function for each parameter. This method keeps everything generic by instead using a string lookup on the directory entry name.
Quoted 2 lines
Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>
Quoted 89 lines
--- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 2 +- drivers/gpu/drm/i915/i915_debugfs_params.c | 31 ++++++++++++++++++++++ 2 files changed, 32 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +1
-1@@ -99,7 +99,7 @@ static int guc_action_policies_update(struct intel_guc *guc, u32 policy_offset) policy_offset }; -return intel_guc_send (guc, action,ARRAY_SIZE (action)); +return intel_guc_send_busy_loop (guc, action,ARRAY_SIZE (action),0 , true); }int intel_guc_global_policies_update (struct intel_guc *guc)drivers/gpu/drm/i915/i915_debugfs_params.c +31
-0@@ -6,9 +6,20 @@#include <linux/kernel.h> #include "i915_debugfs_params.h" +#include "gt/intel_gt.h" +#include "gt/uc/intel_guc.h" #include "i915_drv.h" #include "i915_params.h" +#define MATCH_DEBUGFS_NODE_NAME (_file, _name) (strcmp((_file)->f_path.dentry->d_name.name, (_name)) == 0) + +#define GET_I915 (i915, name, ptr) \ + do { \ + struct i915_params *params; \ + params = container_of(((void *) (ptr)), typeof(*params), name); \ + (i915) = container_of(params, typeof(*(i915)), params); \ + } while(0) +/* int param */ static int i915_param_int_show (struct seq_file *m,void *data) {@@ -24,6 +35,16 @@ static int i915_param_int_open(struct inode *inode, struct file *file)return single_open (file, i915_param_int_show, inode->i_private); } +static int notify_guc (struct drm_i915_private *i915) +{ +int ret =0 ; + +if (intel_uc_uses_guc_submission (&i915->gt.uc)) + ret =intel_guc_global_policies_update (&i915->gt.uc.guc); + +return ret; +} +static ssize_t i915_param_int_write (struct file *file,const char __user *ubuf,size_t len,loff_t *offp)@@ -81,8 +102,10 @@ static ssize_t i915_param_uint_write(struct file *file, const char __user *ubuf, size_t len, loff_t *offp) { + struct drm_i915_private *i915;struct seq_file *m = file->private_data;unsigned int *value = m->private; +unsigned int old = *value;int ret; ret =kstrtouint_from_user (ubuf, len,0 , value);@@ -95,6 +118,14 @@ static ssize_t i915_param_uint_write(struct file *file, *value = b; } +if (!ret &&MATCH_DEBUGFS_NODE_NAME (file,"reset" )) { +GET_I915 (i915, reset, value); + + ret =notify_guc (i915); +if (ret) + *value = old; + } +return ret ?: len; }Signature · 1 line
2.28.0
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 40/51 drm/i915/guc: Include scheduling policies in the debugfs state dump
In reply to Matthew BrostFrom: John Harrison <[email protected]>
Added the scheduling policy parameters to the 'guc_info' debugfs state dump.
--- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 14 ++++++++++++++ drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h | 3 +++ drivers/gpu/drm/i915/gt/uc/intel_guc_debugfs.c | 2 ++ 3 files changed, 19 insertions(+)
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +14 -0
@@ -92,6 +92,20 @@ static void guc_policies_init(struct intel_guc *guc, struct guc_policies *polici policies->is_valid =1 ; } +void intel_guc_ads_print_policy_info (struct intel_guc *guc, +struct drm_printer *dp) +{ +struct __guc_ads_blob *blob = guc->ads_blob; + +if (unlikely (!blob)) +return ; + +drm_printf (dp,"Global scheduling policies:\n" ); +drm_printf (dp," DPC promote time = %u\n" , blob->policies.dpc_promote_time); +drm_printf (dp," Max num work items = %u\n" , blob->policies.max_num_work_items); +drm_printf (dp," Flags = %u\n" , blob->policies.global_flags); +} +static int guc_action_policies_update (struct intel_guc *guc,u32 policy_offset) {u32 action[] = {
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h +3 -0
@@ -7,9 +7,12 @@#define _INTEL_GUC_ADS_H_struct intel_guc ; +struct drm_printer ;int intel_guc_ads_create (struct intel_guc *guc);void intel_guc_ads_destroy (struct intel_guc *guc);void intel_guc_ads_reset (struct intel_guc *guc); +void intel_guc_ads_print_policy_info (struct intel_guc *guc, +struct drm_printer *p);#endif
drivers/gpu/drm/i915/gt/uc/intel_guc_debugfs.c +2 -0
@@ -10,6 +10,7 @@#include "intel_guc_debugfs.h" #include "intel_guc_log_debugfs.h" #include "gt/uc/intel_guc_ct.h" +#include "gt/uc/intel_guc_ads.h" #include "gt/uc/intel_guc_submission.h" static int guc_info_show (struct seq_file *m,void *data)
@@ -29,6 +30,7 @@ static int guc_info_show(struct seq_file *m, void *data)intel_guc_ct_print_info (&guc->ct, &p);intel_guc_submission_print_info (guc, &p); +intel_guc_ads_print_policy_info (guc, &p);return 0 ; }
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 41/51 drm/i915/guc: Add golden context to GuC ADS
In reply to Matthew BrostFrom: John Harrison <[email protected]>
The media watchdog mechanism involves GuC doing a silent reset and continue of the hung context. This requires the i915 driver provide a golden context to GuC in the ADS.
--- drivers/gpu/drm/i915/gt/intel_gt.c | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc.c | 5 + drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 213 ++++++++++++++++++--- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_uc.c | 5 + drivers/gpu/drm/i915/gt/uc/intel_uc.h | 1 + 7 files changed, 199 insertions(+), 30 deletions(-)
drivers/gpu/drm/i915/gt/intel_gt.c +2 -0
@@ -654,6 +654,8 @@ int intel_gt_init(struct intel_gt *gt)if (err) goto err_gt; +intel_uc_init_late (>->uc); + err =i915_inject_probe_error (gt->i915, -EIO );if (err) goto err_gt;
drivers/gpu/drm/i915/gt/uc/intel_guc.c +5 -0
@@ -180,6 +180,11 @@ void intel_guc_init_early(struct intel_guc *guc) } } +void intel_guc_init_late (struct intel_guc *guc) +{ +intel_guc_ads_init_late (guc); +} +static u32 guc_ctl_debug_flags (struct intel_guc *guc) {u32 level =intel_guc_log_get_level (&guc->log);
drivers/gpu/drm/i915/gt/uc/intel_guc.h +2 -0
@@ -60,6 +60,7 @@ struct intel_guc {struct i915_vma *ads_vma;struct __guc_ads_blob *ads_blob;u32 ads_regset_size; +u32 ads_golden_ctxt_size;struct i915_vma *lrc_desc_pool;void *lrc_desc_pool_vaddr;
@@ -176,6 +177,7 @@ static inline u32 intel_guc_ggtt_offset(struct intel_guc *guc, }void intel_guc_init_early (struct intel_guc *guc); +void intel_guc_init_late (struct intel_guc *guc);void intel_guc_init_send_regs (struct intel_guc *guc);void intel_guc_write_params (struct intel_guc *guc);int intel_guc_init (struct intel_guc *guc);
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +183 -30
284 lines not on this page · raw message
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h +1 -0
@@ -11,6 +11,7 @@ struct drm_printer;int intel_guc_ads_create (struct intel_guc *guc);void intel_guc_ads_destroy (struct intel_guc *guc); +void intel_guc_ads_init_late (struct intel_guc *guc);void intel_guc_ads_reset (struct intel_guc *guc);void intel_guc_ads_print_policy_info (struct intel_guc *guc,struct drm_printer *p);
drivers/gpu/drm/i915/gt/uc/intel_uc.c +5 -0
@@ -120,6 +120,11 @@ void intel_uc_init_early(struct intel_uc *uc) uc->ops = &uc_ops_off; } +void intel_uc_init_late (struct intel_uc *uc) +{ +intel_guc_init_late (&uc->guc); +} +void intel_uc_driver_late_release (struct intel_uc *uc) { }
drivers/gpu/drm/i915/gt/uc/intel_uc.h +1 -0
@@ -35,6 +35,7 @@ struct intel_uc { };void intel_uc_init_early (struct intel_uc *uc); +void intel_uc_init_late (struct intel_uc *uc);void intel_uc_driver_late_release (struct intel_uc *uc);void intel_uc_driver_remove (struct intel_uc *uc);void intel_uc_init_mmio (struct intel_uc *uc);
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 41/51] drm/i915/guc: Add golden context to GuC ADS
Quoted 285 lines · Matthew Brost
From: John Harrison <[email protected]>
The media watchdog mechanism involves GuC doing a silent reset and continue of the hung context. This requires the i915 driver provide a golden context to GuC in the ADS.
Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gt/intel_gt.c | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc.c | 5 + drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 213 ++++++++++++++++++--- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_uc.c | 5 + drivers/gpu/drm/i915/gt/uc/intel_uc.h | 1 + 7 files changed, 199 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/gt/intel_gt.c +2
-0@@ -654,6 +654,8 @@ int intel_gt_init(struct intel_gt *gt)if (err) goto err_gt; +intel_uc_init_late (>->uc); + err =i915_inject_probe_error (gt->i915, -EIO );if (err) goto err_gt;drivers/gpu/drm/i915/gt/uc/intel_guc.c +5
-0@@ -180,6 +180,11 @@ void intel_guc_init_early(struct intel_guc *guc) } } +void intel_guc_init_late (struct intel_guc *guc) +{ +intel_guc_ads_init_late (guc); +} +static u32 guc_ctl_debug_flags (struct intel_guc *guc) {u32 level =intel_guc_log_get_level (&guc->log);drivers/gpu/drm/i915/gt/uc/intel_guc.h +2
-0@@ -60,6 +60,7 @@ struct intel_guc {struct i915_vma *ads_vma;struct __guc_ads_blob *ads_blob;u32 ads_regset_size; +u32 ads_golden_ctxt_size;struct i915_vma *lrc_desc_pool;void *lrc_desc_pool_vaddr;@@ -176,6 +177,7 @@ static inline u32 intel_guc_ggtt_offset(struct intel_guc *guc, }void intel_guc_init_early (struct intel_guc *guc); +void intel_guc_init_late (struct intel_guc *guc);void intel_guc_init_send_regs (struct intel_guc *guc);void intel_guc_write_params (struct intel_guc *guc);int intel_guc_init (struct intel_guc *guc);drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +136
-26@@ -7,6 +7,7 @@#include "gt/intel_gt.h" #include "gt/intel_lrc.h" +#include "gt/shmem_utils.h" #include "intel_guc_ads.h" #include "intel_guc_fwif.h" #include "intel_uc.h" @@ -33,6 +34,10 @@* +---------------------------------------+ <== dynamic * | padding | * +---------------------------------------+ <== 4K aligned +* | golden contexts | +* +---------------------------------------+ +* | padding | +* +---------------------------------------+ <== 4K aligned * | private data | * +---------------------------------------+ * | padding | @@ -52,6 +57,11 @@ static u32 guc_ads_regset_size(struct intel_guc *guc)return guc->ads_regset_size; } +static u32 guc_ads_golden_ctxt_size (struct intel_guc *guc) +{ +return PAGE_ALIGN (guc->ads_golden_ctxt_size); +} +static u32 guc_ads_private_data_size (struct intel_guc *guc) {return PAGE_ALIGN (guc->fw.private_data_size);@@ -62,12 +72,23 @@ static u32 guc_ads_regset_offset(struct intel_guc *guc)return offsetof(struct __guc_ads_blob , regset); } -static u32 guc_ads_private_data_offset (struct intel_guc *guc) +static u32 guc_ads_golden_ctxt_offset (struct intel_guc *guc) {u32 offset; offset =guc_ads_regset_offset (guc) +guc_ads_regset_size (guc); + +return PAGE_ALIGN (offset); +} + +static u32 guc_ads_private_data_offset (struct intel_guc *guc) +{ +u32 offset; + + offset =guc_ads_golden_ctxt_offset (guc) + +guc_ads_golden_ctxt_size (guc); +return PAGE_ALIGN (offset); }@@ -319,53 +340,163 @@ static void guc_mmio_reg_state_init(struct intel_guc *guc,GEM_BUG_ON (temp_set.size); } -/* -* The first 80 dwords of the register state context, containing the -* execlists and ppgtt registers. -*/ -#define LR_HW_CONTEXT_SIZE (80 * sizeof(u32)) +static void fill_engine_enable_masks (struct intel_gt *gt, +struct guc_gt_system_info *info) +{ + info->engine_enabled_masks[GUC_RENDER_CLASS ] =1 ; + info->engine_enabled_masks[GUC_BLITTER_CLASS ] =1 ; + info->engine_enabled_masks[GUC_VIDEO_CLASS ] =VDBOX_MASK (gt); + info->engine_enabled_masks[GUC_VIDEOENHANCE_CLASS ] =VEBOX_MASK (gt); +} -static void __guc_ads_init (struct intel_guc *guc) +/* Skip execlist and PPGTT registers */ +#define LR_HW_CONTEXT_SIZE (80 * sizeof(u32)) +#define SKIP_SIZE (LRC_PPHWSP_SZ * PAGE_SIZE + LR_HW_CONTEXT_SIZE) + +static int guc_prep_golden_context (struct intel_guc *guc, +struct __guc_ads_blob *blob) {struct intel_gt *gt =guc_to_gt (guc); -struct drm_i915_private *i915 = gt->i915; +u32 addr_ggtt, offset; +u32 total_size =0 , alloc_size, real_size; +u8 engine_class, guc_class; +struct guc_gt_system_info *info, local_info; + +/* +* Reserve the memory for the golden contexts and point GuC at it but +* leave it empty for now. The context data will be filled in later +* once there is something available to put there. +* +* Note that the HWSP and ring context are not included. +* +* Note also that the storage must be pinned in the GGTT, so that the +* address won't change after GuC has been told where to find it. The +* GuC will also validate that the LRC base + size fall within the +* allowed GGTT range. +*/ +if (blob) { + offset =guc_ads_golden_ctxt_offset (guc); + addr_ggtt =intel_guc_ggtt_offset (guc, guc->ads_vma) + offset; + info = &blob->system_info; + }else { +memset (&local_info,0 ,sizeof (local_info)); + info = &local_info; +fill_engine_enable_masks (gt, info); + } + +for (engine_class =0 ; engine_class <=MAX_ENGINE_CLASS ; ++engine_class) { +if (engine_class ==OTHER_CLASS ) +continue ; + + guc_class =engine_class_to_guc_class (engine_class); + +if (!info->engine_enabled_masks[guc_class]) +continue ; + + real_size =intel_engine_context_size (gt, engine_class); + alloc_size =PAGE_ALIGN (real_size); + total_size += alloc_size; + +if (!blob) +continue ; + + blob->ads.eng_state_size[guc_class] = real_size; + blob->ads.golden_context_lrca[guc_class] = addr_ggtt; + addr_ggtt += alloc_size; + } + +if (!blob) +return total_size; + +GEM_BUG_ON (guc->ads_golden_ctxt_size != total_size); +return total_size; +} + +static struct intel_engine_cs *find_engine_state (struct intel_gt *gt,u8 engine_class) +{ +struct intel_engine_cs *engine; +enum intel_engine_id id; + +for_each_engine (engine, gt, id) { +if (engine->class != engine_class) +continue ; + +if (!engine->default_state) +continue ; + +return engine; + } + +return NULL ; +} + +static void guc_init_golden_context (struct intel_guc *guc) +{struct __guc_ads_blob *blob = guc->ads_blob; -const u32 skipped_size =LRC_PPHWSP_SZ *PAGE_SIZE +LR_HW_CONTEXT_SIZE ; -u32 base; +struct intel_engine_cs *engine; +struct intel_gt *gt =guc_to_gt (guc); +u32 addr_ggtt, offset; +u32 total_size =0 , alloc_size, real_size;u8 engine_class, guc_class; +u8 *ptr; -/* GuC scheduling policies */ -guc_policies_init (guc, &blob->policies); +if (!intel_uc_uses_guc_submission (>->uc)) +return ; + +GEM_BUG_ON (!blob);/* -* GuC expects a per-engine-class context image and size -* (minus hwsp and ring context). The context image will be -* used to reinitialize engines after a reset. It must exist -* and be pinned in the GGTT, so that the address won't change after -* we have told GuC where to find it. The context size will be used -* to validate that the LRC base + size fall within allowed GGTT. +* Go back and fill in the golden context data now that it is +* available. */ + offset =guc_ads_golden_ctxt_offset (guc); + addr_ggtt =intel_guc_ggtt_offset (guc, guc->ads_vma) + offset; + ptr = ((u8 *) blob) + offset; +for (engine_class =0 ; engine_class <=MAX_ENGINE_CLASS ; ++engine_class) {if (engine_class ==OTHER_CLASS )continue ; guc_class =engine_class_to_guc_class (engine_class); -/* -* TODO: Set context pointer to default state to allow -* GuC to re-init guilty contexts after internal reset. -*/ - blob->ads.golden_context_lrca[guc_class] =0 ; - blob->ads.eng_state_size[guc_class] = -intel_engine_context_size (gt, engine_class) - - skipped_size; +if (!blob->system_info.engine_enabled_masks[guc_class]) +continue ; + + real_size =intel_engine_context_size (gt, engine_class); + alloc_size =PAGE_ALIGN (real_size); + total_size += alloc_size; + + engine =find_engine_state (gt, engine_class); +if (!engine) { +drm_err (>->i915->drm,"No engine state recorded for class %d!\n" , engine_class);
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c static void guc_mmio_reg_state_init(struct intel_guc *guc, · line 481
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +10
-0+ blob->ads.eng_state_size[guc_class] =0 ; + blob->ads.golden_context_lrca[guc_class] =0 ; +continue ; + } + +GEM_BUG_ON (blob->ads.eng_state_size[guc_class] != real_size); +GEM_BUG_ON (blob->ads.golden_context_lrca[guc_class] != addr_ggtt); + addr_ggtt += alloc_size; + +shmem_read (engine->default_state,SKIP_SIZE , ptr +SKIP_SIZE , real_size);
There is a confirmed memory corruption here, should be:shmem_read(engine->default_state, SKIP_SIZE, ptr + SKIP_SIZE, real_size - SKIP_SIZE);
Also a bit of nit but I think s/SKIP_SIZE/GOLDEN_CONTEXT_SKIP_SIZE/g
As long you agree John, I'll fix both of these in the next rev.
Matt
Quoted 115 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +37
-4+ ptr += alloc_size; } +GEM_BUG_ON (guc->ads_golden_ctxt_size != total_size); +} + +static void __guc_ads_init (struct intel_guc *guc) +{ +struct intel_gt *gt =guc_to_gt (guc); +struct drm_i915_private *i915 = gt->i915; +struct __guc_ads_blob *blob = guc->ads_blob; +u32 base; + +/* GuC scheduling policies */ +guc_policies_init (guc, &blob->policies); +/* System info */ - blob->system_info.engine_enabled_masks[GUC_RENDER_CLASS ] =1 ; - blob->system_info.engine_enabled_masks[GUC_BLITTER_CLASS ] =1 ; - blob->system_info.engine_enabled_masks[GUC_VIDEO_CLASS ] =VDBOX_MASK (gt); - blob->system_info.engine_enabled_masks[GUC_VIDEOENHANCE_CLASS ] =VEBOX_MASK (gt); +fill_engine_enable_masks (gt, &blob->system_info); blob->system_info.generic_gt_sysinfo[GUC_GENERIC_GT_SYSINFO_SLICE_ENABLED ] =hweight8 (gt->info.sseu.slice_mask);@@ -380,6 +511,9 @@ static void __guc_ads_init(struct intel_guc *guc) GEN12_DOORBELLS_PER_SQIDI) + 1; } +/* Golden contexts for re-initialising after a watchdog reset */ +guc_prep_golden_context (guc, blob); +guc_mapping_table_init (guc_to_gt (guc), &blob->system_info); base =intel_guc_ggtt_offset (guc, guc->ads_vma);@@ -417,6 +551,13 @@ int intel_guc_ads_create(struct intel_guc *guc)return ret; guc->ads_regset_size = ret; +/* Likewise the golden contexts: */ + ret =guc_prep_golden_context (guc,NULL ); +if (ret <0 ) +return ret; + guc->ads_golden_ctxt_size = ret; + +/* Now the total size can be determined: */ size =guc_ads_blob_size (guc); ret = intel_guc_allocate_and_map_vma(guc, size, &guc->ads_vma,@@ -429,6 +570,18 @@ int intel_guc_ads_create(struct intel_guc *guc)return 0 ; } +void intel_guc_ads_init_late (struct intel_guc *guc) +{ +/* +* The golden context setup requires the saved engine state from +* __engines_record_defaults(). However, that requires engines to be +* operational which means the ADS must already have been configured. +* Fortunately, the golden context state is not needed until a hang +* occurs, so it can be filled in during this late init phase. +*/ +guc_init_golden_context (guc); +} +void intel_guc_ads_destroy (struct intel_guc *guc) {i915_vma_unpin_and_release (&guc->ads_vma,I915_VMA_RELEASE_MAP );drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h +1
-0@@ -11,6 +11,7 @@ struct drm_printer;int intel_guc_ads_create (struct intel_guc *guc);void intel_guc_ads_destroy (struct intel_guc *guc); +void intel_guc_ads_init_late (struct intel_guc *guc);void intel_guc_ads_reset (struct intel_guc *guc);void intel_guc_ads_print_policy_info (struct intel_guc *guc,struct drm_printer *p);drivers/gpu/drm/i915/gt/uc/intel_uc.c +5
-0@@ -120,6 +120,11 @@ void intel_uc_init_early(struct intel_uc *uc) uc->ops = &uc_ops_off; } +void intel_uc_init_late (struct intel_uc *uc) +{ +intel_guc_init_late (&uc->guc); +} +void intel_uc_driver_late_release (struct intel_uc *uc) { }drivers/gpu/drm/i915/gt/uc/intel_uc.h +1
-0@@ -35,6 +35,7 @@ struct intel_uc { };void intel_uc_init_early (struct intel_uc *uc); +void intel_uc_init_late (struct intel_uc *uc);void intel_uc_driver_late_release (struct intel_uc *uc);void intel_uc_driver_remove (struct intel_uc *uc);void intel_uc_init_mmio (struct intel_uc *uc);Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 41/51] drm/i915/guc: Add golden context to GuC ADS
Quoted 299 lines · Matthew Brost
Quoted 285 lines · Matthew Brost
From: John Harrison <[email protected]>
The media watchdog mechanism involves GuC doing a silent reset and continue of the hung context. This requires the i915 driver provide a golden context to GuC in the ADS.
Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gt/intel_gt.c | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc.c | 5 + drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 213 ++++++++++++++++++--- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_uc.c | 5 + drivers/gpu/drm/i915/gt/uc/intel_uc.h | 1 + 7 files changed, 199 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/gt/intel_gt.c +2
-0@@ -654,6 +654,8 @@ int intel_gt_init(struct intel_gt *gt)if (err) goto err_gt; +intel_uc_init_late (>->uc); + err =i915_inject_probe_error (gt->i915, -EIO );if (err) goto err_gt;drivers/gpu/drm/i915/gt/uc/intel_guc.c +5
-0@@ -180,6 +180,11 @@ void intel_guc_init_early(struct intel_guc *guc) } } +void intel_guc_init_late (struct intel_guc *guc) +{ +intel_guc_ads_init_late (guc); +} +static u32 guc_ctl_debug_flags (struct intel_guc *guc) {u32 level =intel_guc_log_get_level (&guc->log);drivers/gpu/drm/i915/gt/uc/intel_guc.h +2
-0@@ -60,6 +60,7 @@ struct intel_guc {struct i915_vma *ads_vma;struct __guc_ads_blob *ads_blob;u32 ads_regset_size; +u32 ads_golden_ctxt_size;struct i915_vma *lrc_desc_pool;void *lrc_desc_pool_vaddr;@@ -176,6 +177,7 @@ static inline u32 intel_guc_ggtt_offset(struct intel_guc *guc, }void intel_guc_init_early (struct intel_guc *guc); +void intel_guc_init_late (struct intel_guc *guc);void intel_guc_init_send_regs (struct intel_guc *guc);void intel_guc_write_params (struct intel_guc *guc);int intel_guc_init (struct intel_guc *guc);drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +136
-26@@ -7,6 +7,7 @@#include "gt/intel_gt.h" #include "gt/intel_lrc.h" +#include "gt/shmem_utils.h" #include "intel_guc_ads.h" #include "intel_guc_fwif.h" #include "intel_uc.h" @@ -33,6 +34,10 @@* +---------------------------------------+ <== dynamic * | padding | * +---------------------------------------+ <== 4K aligned +* | golden contexts | +* +---------------------------------------+ +* | padding | +* +---------------------------------------+ <== 4K aligned * | private data | * +---------------------------------------+ * | padding | @@ -52,6 +57,11 @@ static u32 guc_ads_regset_size(struct intel_guc *guc)return guc->ads_regset_size; } +static u32 guc_ads_golden_ctxt_size (struct intel_guc *guc) +{ +return PAGE_ALIGN (guc->ads_golden_ctxt_size); +} +static u32 guc_ads_private_data_size (struct intel_guc *guc) {return PAGE_ALIGN (guc->fw.private_data_size);@@ -62,12 +72,23 @@ static u32 guc_ads_regset_offset(struct intel_guc *guc)return offsetof(struct __guc_ads_blob , regset); } -static u32 guc_ads_private_data_offset (struct intel_guc *guc) +static u32 guc_ads_golden_ctxt_offset (struct intel_guc *guc) {u32 offset; offset =guc_ads_regset_offset (guc) +guc_ads_regset_size (guc); + +return PAGE_ALIGN (offset); +} + +static u32 guc_ads_private_data_offset (struct intel_guc *guc) +{ +u32 offset; + + offset =guc_ads_golden_ctxt_offset (guc) + +guc_ads_golden_ctxt_size (guc); +return PAGE_ALIGN (offset); }@@ -319,53 +340,163 @@ static void guc_mmio_reg_state_init(struct intel_guc *guc,GEM_BUG_ON (temp_set.size); } -/* -* The first 80 dwords of the register state context, containing the -* execlists and ppgtt registers. -*/ -#define LR_HW_CONTEXT_SIZE (80 * sizeof(u32)) +static void fill_engine_enable_masks (struct intel_gt *gt, +struct guc_gt_system_info *info) +{ + info->engine_enabled_masks[GUC_RENDER_CLASS ] =1 ; + info->engine_enabled_masks[GUC_BLITTER_CLASS ] =1 ; + info->engine_enabled_masks[GUC_VIDEO_CLASS ] =VDBOX_MASK (gt); + info->engine_enabled_masks[GUC_VIDEOENHANCE_CLASS ] =VEBOX_MASK (gt); +} -static void __guc_ads_init (struct intel_guc *guc) +/* Skip execlist and PPGTT registers */ +#define LR_HW_CONTEXT_SIZE (80 * sizeof(u32)) +#define SKIP_SIZE (LRC_PPHWSP_SZ * PAGE_SIZE + LR_HW_CONTEXT_SIZE) + +static int guc_prep_golden_context (struct intel_guc *guc, +struct __guc_ads_blob *blob) {struct intel_gt *gt =guc_to_gt (guc); -struct drm_i915_private *i915 = gt->i915; +u32 addr_ggtt, offset; +u32 total_size =0 , alloc_size, real_size; +u8 engine_class, guc_class; +struct guc_gt_system_info *info, local_info; + +/* +* Reserve the memory for the golden contexts and point GuC at it but +* leave it empty for now. The context data will be filled in later +* once there is something available to put there. +* +* Note that the HWSP and ring context are not included. +* +* Note also that the storage must be pinned in the GGTT, so that the +* address won't change after GuC has been told where to find it. The +* GuC will also validate that the LRC base + size fall within the +* allowed GGTT range. +*/ +if (blob) { + offset =guc_ads_golden_ctxt_offset (guc); + addr_ggtt =intel_guc_ggtt_offset (guc, guc->ads_vma) + offset; + info = &blob->system_info; + }else { +memset (&local_info,0 ,sizeof (local_info)); + info = &local_info; +fill_engine_enable_masks (gt, info); + } + +for (engine_class =0 ; engine_class <=MAX_ENGINE_CLASS ; ++engine_class) { +if (engine_class ==OTHER_CLASS ) +continue ; + + guc_class =engine_class_to_guc_class (engine_class); + +if (!info->engine_enabled_masks[guc_class]) +continue ; + + real_size =intel_engine_context_size (gt, engine_class); + alloc_size =PAGE_ALIGN (real_size); + total_size += alloc_size; + +if (!blob) +continue ; + + blob->ads.eng_state_size[guc_class] = real_size; + blob->ads.golden_context_lrca[guc_class] = addr_ggtt; + addr_ggtt += alloc_size; + } + +if (!blob) +return total_size; + +GEM_BUG_ON (guc->ads_golden_ctxt_size != total_size); +return total_size; +} + +static struct intel_engine_cs *find_engine_state (struct intel_gt *gt,u8 engine_class) +{ +struct intel_engine_cs *engine; +enum intel_engine_id id; + +for_each_engine (engine, gt, id) { +if (engine->class != engine_class) +continue ; + +if (!engine->default_state) +continue ; + +return engine; + } + +return NULL ; +} + +static void guc_init_golden_context (struct intel_guc *guc) +{struct __guc_ads_blob *blob = guc->ads_blob; -const u32 skipped_size =LRC_PPHWSP_SZ *PAGE_SIZE +LR_HW_CONTEXT_SIZE ; -u32 base; +struct intel_engine_cs *engine; +struct intel_gt *gt =guc_to_gt (guc); +u32 addr_ggtt, offset; +u32 total_size =0 , alloc_size, real_size;u8 engine_class, guc_class; +u8 *ptr; -/* GuC scheduling policies */ -guc_policies_init (guc, &blob->policies); +if (!intel_uc_uses_guc_submission (>->uc)) +return ; + +GEM_BUG_ON (!blob);/* -* GuC expects a per-engine-class context image and size -* (minus hwsp and ring context). The context image will be -* used to reinitialize engines after a reset. It must exist -* and be pinned in the GGTT, so that the address won't change after -* we have told GuC where to find it. The context size will be used -* to validate that the LRC base + size fall within allowed GGTT. +* Go back and fill in the golden context data now that it is +* available. */ + offset =guc_ads_golden_ctxt_offset (guc); + addr_ggtt =intel_guc_ggtt_offset (guc, guc->ads_vma) + offset; + ptr = ((u8 *) blob) + offset; +for (engine_class =0 ; engine_class <=MAX_ENGINE_CLASS ; ++engine_class) {if (engine_class ==OTHER_CLASS )continue ; guc_class =engine_class_to_guc_class (engine_class); -/* -* TODO: Set context pointer to default state to allow -* GuC to re-init guilty contexts after internal reset. -*/ - blob->ads.golden_context_lrca[guc_class] =0 ; - blob->ads.eng_state_size[guc_class] = -intel_engine_context_size (gt, engine_class) - - skipped_size; +if (!blob->system_info.engine_enabled_masks[guc_class]) +continue ; + + real_size =intel_engine_context_size (gt, engine_class); + alloc_size =PAGE_ALIGN (real_size); + total_size += alloc_size; + + engine =find_engine_state (gt, engine_class); +if (!engine) { +drm_err (>->i915->drm,"No engine state recorded for class %d!\n" , engine_class);drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c static void guc_mmio_reg_state_init(struct intel_guc *guc, · line 481
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +10
-0+ blob->ads.eng_state_size[guc_class] =0 ; + blob->ads.golden_context_lrca[guc_class] =0 ; +continue ; + } + +GEM_BUG_ON (blob->ads.eng_state_size[guc_class] != real_size); +GEM_BUG_ON (blob->ads.golden_context_lrca[guc_class] != addr_ggtt); + addr_ggtt += alloc_size; + +shmem_read (engine->default_state,SKIP_SIZE , ptr +SKIP_SIZE , real_size);There is a confirmed memory corruption here, should be:
shmem_read(engine->default_state, SKIP_SIZE, ptr + SKIP_SIZE, real_size - SKIP_SIZE);
Quoted 5 lines
Also a bit of nit but I think s/SKIP_SIZE/GOLDEN_CONTEXT_SKIP_SIZE/g
As long you agree John, I'll fix both of these in the next rev.
Matt
Yup. Good catch.
Rather than ballooning the name out, maybe make it and the LR_HW_ define local enums within this function? I don't think they are used anywhere else.
John.
Quoted 120 lines
Quoted 116 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +37
-4+ ptr += alloc_size; } +GEM_BUG_ON (guc->ads_golden_ctxt_size != total_size); +} + +static void __guc_ads_init (struct intel_guc *guc) +{ +struct intel_gt *gt =guc_to_gt (guc); +struct drm_i915_private *i915 = gt->i915; +struct __guc_ads_blob *blob = guc->ads_blob; +u32 base; + +/* GuC scheduling policies */ +guc_policies_init (guc, &blob->policies); +/* System info */ - blob->system_info.engine_enabled_masks[GUC_RENDER_CLASS ] =1 ; - blob->system_info.engine_enabled_masks[GUC_BLITTER_CLASS ] =1 ; - blob->system_info.engine_enabled_masks[GUC_VIDEO_CLASS ] =VDBOX_MASK (gt); - blob->system_info.engine_enabled_masks[GUC_VIDEOENHANCE_CLASS ] =VEBOX_MASK (gt); +fill_engine_enable_masks (gt, &blob->system_info); blob->system_info.generic_gt_sysinfo[GUC_GENERIC_GT_SYSINFO_SLICE_ENABLED ] =hweight8 (gt->info.sseu.slice_mask);@@ -380,6 +511,9 @@ static void __guc_ads_init(struct intel_guc *guc) GEN12_DOORBELLS_PER_SQIDI) + 1; } +/* Golden contexts for re-initialising after a watchdog reset */ +guc_prep_golden_context (guc, blob); +guc_mapping_table_init (guc_to_gt (guc), &blob->system_info); base =intel_guc_ggtt_offset (guc, guc->ads_vma);@@ -417,6 +551,13 @@ int intel_guc_ads_create(struct intel_guc *guc)return ret; guc->ads_regset_size = ret; +/* Likewise the golden contexts: */ + ret =guc_prep_golden_context (guc,NULL ); +if (ret <0 ) +return ret; + guc->ads_golden_ctxt_size = ret; + +/* Now the total size can be determined: */ size =guc_ads_blob_size (guc); ret = intel_guc_allocate_and_map_vma(guc, size, &guc->ads_vma,@@ -429,6 +570,18 @@ int intel_guc_ads_create(struct intel_guc *guc)return 0 ; } +void intel_guc_ads_init_late (struct intel_guc *guc) +{ +/* +* The golden context setup requires the saved engine state from +* __engines_record_defaults(). However, that requires engines to be +* operational which means the ADS must already have been configured. +* Fortunately, the golden context state is not needed until a hang +* occurs, so it can be filled in during this late init phase. +*/ +guc_init_golden_context (guc); +} +void intel_guc_ads_destroy (struct intel_guc *guc) {i915_vma_unpin_and_release (&guc->ads_vma,I915_VMA_RELEASE_MAP );drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h +1
-0@@ -11,6 +11,7 @@ struct drm_printer;int intel_guc_ads_create (struct intel_guc *guc);void intel_guc_ads_destroy (struct intel_guc *guc); +void intel_guc_ads_init_late (struct intel_guc *guc);void intel_guc_ads_reset (struct intel_guc *guc);void intel_guc_ads_print_policy_info (struct intel_guc *guc,struct drm_printer *p);drivers/gpu/drm/i915/gt/uc/intel_uc.c +5
-0@@ -120,6 +120,11 @@ void intel_uc_init_early(struct intel_uc *uc) uc->ops = &uc_ops_off; } +void intel_uc_init_late (struct intel_uc *uc) +{ +intel_guc_init_late (&uc->guc); +} +void intel_uc_driver_late_release (struct intel_uc *uc) { }drivers/gpu/drm/i915/gt/uc/intel_uc.h +1
-0@@ -35,6 +35,7 @@ struct intel_uc { };void intel_uc_init_early (struct intel_uc *uc); +void intel_uc_init_late (struct intel_uc *uc);void intel_uc_driver_late_release (struct intel_uc *uc);void intel_uc_driver_remove (struct intel_uc *uc);void intel_uc_init_mmio (struct intel_uc *uc);Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Matthew Brost
Reviewed-byRe: [Intel-gfx] [PATCH 41/51] drm/i915/guc: Add golden context to GuC ADS
Quoted 287 lines · John Harrison
Quoted 286 lines · Matthew Brost
Quoted 270 lines · Matthew Brost
From: John Harrison <[email protected]>
The media watchdog mechanism involves GuC doing a silent reset and continue of the hung context. This requires the i915 driver provide a golden context to GuC in the ADS.
Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/gt/intel_gt.c | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc.c | 5 + drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 213 ++++++++++++++++++--- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_uc.c | 5 + drivers/gpu/drm/i915/gt/uc/intel_uc.h | 1 + 7 files changed, 199 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/gt/intel_gt.c +2
-0@@ -654,6 +654,8 @@ int intel_gt_init(struct intel_gt *gt)if (err) goto err_gt; +intel_uc_init_late (>->uc); + err =i915_inject_probe_error (gt->i915, -EIO );if (err) goto err_gt;drivers/gpu/drm/i915/gt/uc/intel_guc.c +5
-0@@ -180,6 +180,11 @@ void intel_guc_init_early(struct intel_guc *guc) } } +void intel_guc_init_late (struct intel_guc *guc) +{ +intel_guc_ads_init_late (guc); +} +static u32 guc_ctl_debug_flags (struct intel_guc *guc) {u32 level =intel_guc_log_get_level (&guc->log);drivers/gpu/drm/i915/gt/uc/intel_guc.h +2
-0@@ -60,6 +60,7 @@ struct intel_guc {struct i915_vma *ads_vma;struct __guc_ads_blob *ads_blob;u32 ads_regset_size; +u32 ads_golden_ctxt_size;struct i915_vma *lrc_desc_pool;void *lrc_desc_pool_vaddr;@@ -176,6 +177,7 @@ static inline u32 intel_guc_ggtt_offset(struct intel_guc *guc, }void intel_guc_init_early (struct intel_guc *guc); +void intel_guc_init_late (struct intel_guc *guc);void intel_guc_init_send_regs (struct intel_guc *guc);void intel_guc_write_params (struct intel_guc *guc);int intel_guc_init (struct intel_guc *guc);drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +136
-26@@ -7,6 +7,7 @@#include "gt/intel_gt.h" #include "gt/intel_lrc.h" +#include "gt/shmem_utils.h" #include "intel_guc_ads.h" #include "intel_guc_fwif.h" #include "intel_uc.h" @@ -33,6 +34,10 @@* +---------------------------------------+ <== dynamic * | padding | * +---------------------------------------+ <== 4K aligned +* | golden contexts | +* +---------------------------------------+ +* | padding | +* +---------------------------------------+ <== 4K aligned * | private data | * +---------------------------------------+ * | padding | @@ -52,6 +57,11 @@ static u32 guc_ads_regset_size(struct intel_guc *guc)return guc->ads_regset_size; } +static u32 guc_ads_golden_ctxt_size (struct intel_guc *guc) +{ +return PAGE_ALIGN (guc->ads_golden_ctxt_size); +} +static u32 guc_ads_private_data_size (struct intel_guc *guc) {return PAGE_ALIGN (guc->fw.private_data_size);@@ -62,12 +72,23 @@ static u32 guc_ads_regset_offset(struct intel_guc *guc)return offsetof(struct __guc_ads_blob , regset); } -static u32 guc_ads_private_data_offset (struct intel_guc *guc) +static u32 guc_ads_golden_ctxt_offset (struct intel_guc *guc) {u32 offset; offset =guc_ads_regset_offset (guc) +guc_ads_regset_size (guc); + +return PAGE_ALIGN (offset); +} + +static u32 guc_ads_private_data_offset (struct intel_guc *guc) +{ +u32 offset; + + offset =guc_ads_golden_ctxt_offset (guc) + +guc_ads_golden_ctxt_size (guc); +return PAGE_ALIGN (offset); }@@ -319,53 +340,163 @@ static void guc_mmio_reg_state_init(struct intel_guc *guc,GEM_BUG_ON (temp_set.size); } -/* -* The first 80 dwords of the register state context, containing the -* execlists and ppgtt registers. -*/ -#define LR_HW_CONTEXT_SIZE (80 * sizeof(u32)) +static void fill_engine_enable_masks (struct intel_gt *gt, +struct guc_gt_system_info *info) +{ + info->engine_enabled_masks[GUC_RENDER_CLASS ] =1 ; + info->engine_enabled_masks[GUC_BLITTER_CLASS ] =1 ; + info->engine_enabled_masks[GUC_VIDEO_CLASS ] =VDBOX_MASK (gt); + info->engine_enabled_masks[GUC_VIDEOENHANCE_CLASS ] =VEBOX_MASK (gt); +} -static void __guc_ads_init (struct intel_guc *guc) +/* Skip execlist and PPGTT registers */ +#define LR_HW_CONTEXT_SIZE (80 * sizeof(u32)) +#define SKIP_SIZE (LRC_PPHWSP_SZ * PAGE_SIZE + LR_HW_CONTEXT_SIZE) + +static int guc_prep_golden_context (struct intel_guc *guc, +struct __guc_ads_blob *blob) {struct intel_gt *gt =guc_to_gt (guc); -struct drm_i915_private *i915 = gt->i915; +u32 addr_ggtt, offset; +u32 total_size =0 , alloc_size, real_size; +u8 engine_class, guc_class; +struct guc_gt_system_info *info, local_info; + +/* +* Reserve the memory for the golden contexts and point GuC at it but +* leave it empty for now. The context data will be filled in later +* once there is something available to put there. +* +* Note that the HWSP and ring context are not included. +* +* Note also that the storage must be pinned in the GGTT, so that the +* address won't change after GuC has been told where to find it. The +* GuC will also validate that the LRC base + size fall within the +* allowed GGTT range. +*/ +if (blob) { + offset =guc_ads_golden_ctxt_offset (guc); + addr_ggtt =intel_guc_ggtt_offset (guc, guc->ads_vma) + offset; + info = &blob->system_info; + }else { +memset (&local_info,0 ,sizeof (local_info)); + info = &local_info; +fill_engine_enable_masks (gt, info); + } + +for (engine_class =0 ; engine_class <=MAX_ENGINE_CLASS ; ++engine_class) { +if (engine_class ==OTHER_CLASS ) +continue ; + + guc_class =engine_class_to_guc_class (engine_class); + +if (!info->engine_enabled_masks[guc_class]) +continue ; + + real_size =intel_engine_context_size (gt, engine_class); + alloc_size =PAGE_ALIGN (real_size); + total_size += alloc_size; + +if (!blob) +continue ; + + blob->ads.eng_state_size[guc_class] = real_size; + blob->ads.golden_context_lrca[guc_class] = addr_ggtt; + addr_ggtt += alloc_size; + } + +if (!blob) +return total_size; + +GEM_BUG_ON (guc->ads_golden_ctxt_size != total_size); +return total_size; +} + +static struct intel_engine_cs *find_engine_state (struct intel_gt *gt,u8 engine_class) +{ +struct intel_engine_cs *engine; +enum intel_engine_id id; + +for_each_engine (engine, gt, id) { +if (engine->class != engine_class) +continue ; + +if (!engine->default_state) +continue ; + +return engine; + } + +return NULL ; +} + +static void guc_init_golden_context (struct intel_guc *guc) +{struct __guc_ads_blob *blob = guc->ads_blob; -const u32 skipped_size =LRC_PPHWSP_SZ *PAGE_SIZE +LR_HW_CONTEXT_SIZE ; -u32 base; +struct intel_engine_cs *engine; +struct intel_gt *gt =guc_to_gt (guc); +u32 addr_ggtt, offset; +u32 total_size =0 , alloc_size, real_size;u8 engine_class, guc_class; +u8 *ptr; -/* GuC scheduling policies */ -guc_policies_init (guc, &blob->policies); +if (!intel_uc_uses_guc_submission (>->uc)) +return ; + +GEM_BUG_ON (!blob);/* -* GuC expects a per-engine-class context image and size -* (minus hwsp and ring context). The context image will be -* used to reinitialize engines after a reset. It must exist -* and be pinned in the GGTT, so that the address won't change after -* we have told GuC where to find it. The context size will be used -* to validate that the LRC base + size fall within allowed GGTT. +* Go back and fill in the golden context data now that it is +* available. */ + offset =guc_ads_golden_ctxt_offset (guc); + addr_ggtt =intel_guc_ggtt_offset (guc, guc->ads_vma) + offset; + ptr = ((u8 *) blob) + offset; +for (engine_class =0 ; engine_class <=MAX_ENGINE_CLASS ; ++engine_class) {if (engine_class ==OTHER_CLASS )continue ; guc_class =engine_class_to_guc_class (engine_class); -/* -* TODO: Set context pointer to default state to allow -* GuC to re-init guilty contexts after internal reset. -*/ - blob->ads.golden_context_lrca[guc_class] =0 ; - blob->ads.eng_state_size[guc_class] = -intel_engine_context_size (gt, engine_class) - - skipped_size; +if (!blob->system_info.engine_enabled_masks[guc_class]) +continue ; + + real_size =intel_engine_context_size (gt, engine_class); + alloc_size =PAGE_ALIGN (real_size); + total_size += alloc_size; + + engine =find_engine_state (gt, engine_class); +if (!engine) { +drm_err (>->i915->drm,"No engine state recorded for class %d!\n" , engine_class);drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c static void guc_mmio_reg_state_init(struct intel_guc *guc, · line 475
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +10
-0+ blob->ads.eng_state_size[guc_class] =0 ; + blob->ads.golden_context_lrca[guc_class] =0 ; +continue ; + } + +GEM_BUG_ON (blob->ads.eng_state_size[guc_class] != real_size); +GEM_BUG_ON (blob->ads.golden_context_lrca[guc_class] != addr_ggtt); + addr_ggtt += alloc_size; + +shmem_read (engine->default_state,SKIP_SIZE , ptr +SKIP_SIZE , real_size);There is a confirmed memory corruption here, should be:
shmem_read(engine->default_state, SKIP_SIZE, ptr + SKIP_SIZE, real_size - SKIP_SIZE);Also a bit of nit but I think s/SKIP_SIZE/GOLDEN_CONTEXT_SKIP_SIZE/g
Quoted 8 lines
Quoted 3 lines
As long you agree John, I'll fix both of these in the next rev.
Matt
Yup. Good catch.
Rather than ballooning the name out, maybe make it and the
LR_HW_define local enums within this function? I don't think they are used anywhere else.
I like this better too as you say they are specific to a single function.
With those changes:
Quoted 114 lines
John.
Quoted 110 lines
Quoted 106 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +37
-4+ ptr += alloc_size; } +GEM_BUG_ON (guc->ads_golden_ctxt_size != total_size); +} + +static void __guc_ads_init (struct intel_guc *guc) +{ +struct intel_gt *gt =guc_to_gt (guc); +struct drm_i915_private *i915 = gt->i915; +struct __guc_ads_blob *blob = guc->ads_blob; +u32 base; + +/* GuC scheduling policies */ +guc_policies_init (guc, &blob->policies); +/* System info */ - blob->system_info.engine_enabled_masks[GUC_RENDER_CLASS ] =1 ; - blob->system_info.engine_enabled_masks[GUC_BLITTER_CLASS ] =1 ; - blob->system_info.engine_enabled_masks[GUC_VIDEO_CLASS ] =VDBOX_MASK (gt); - blob->system_info.engine_enabled_masks[GUC_VIDEOENHANCE_CLASS ] =VEBOX_MASK (gt); +fill_engine_enable_masks (gt, &blob->system_info); blob->system_info.generic_gt_sysinfo[GUC_GENERIC_GT_SYSINFO_SLICE_ENABLED ] =hweight8 (gt->info.sseu.slice_mask);@@ -380,6 +511,9 @@ static void __guc_ads_init(struct intel_guc *guc) GEN12_DOORBELLS_PER_SQIDI) + 1; } +/* Golden contexts for re-initialising after a watchdog reset */ +guc_prep_golden_context (guc, blob); +guc_mapping_table_init (guc_to_gt (guc), &blob->system_info); base =intel_guc_ggtt_offset (guc, guc->ads_vma);@@ -417,6 +551,13 @@ int intel_guc_ads_create(struct intel_guc *guc)return ret; guc->ads_regset_size = ret; +/* Likewise the golden contexts: */ + ret =guc_prep_golden_context (guc,NULL ); +if (ret <0 ) +return ret; + guc->ads_golden_ctxt_size = ret; + +/* Now the total size can be determined: */ size =guc_ads_blob_size (guc); ret = intel_guc_allocate_and_map_vma(guc, size, &guc->ads_vma,@@ -429,6 +570,18 @@ int intel_guc_ads_create(struct intel_guc *guc)return 0 ; } +void intel_guc_ads_init_late (struct intel_guc *guc) +{ +/* +* The golden context setup requires the saved engine state from +* __engines_record_defaults(). However, that requires engines to be +* operational which means the ADS must already have been configured. +* Fortunately, the golden context state is not needed until a hang +* occurs, so it can be filled in during this late init phase. +*/ +guc_init_golden_context (guc); +} +void intel_guc_ads_destroy (struct intel_guc *guc) {i915_vma_unpin_and_release (&guc->ads_vma,I915_VMA_RELEASE_MAP );drivers/gpu/drm/i915/gt/uc/intel_guc_ads.h +1
-0@@ -11,6 +11,7 @@ struct drm_printer;int intel_guc_ads_create (struct intel_guc *guc);void intel_guc_ads_destroy (struct intel_guc *guc); +void intel_guc_ads_init_late (struct intel_guc *guc);void intel_guc_ads_reset (struct intel_guc *guc);void intel_guc_ads_print_policy_info (struct intel_guc *guc,struct drm_printer *p);drivers/gpu/drm/i915/gt/uc/intel_uc.c +5
-0@@ -120,6 +120,11 @@ void intel_uc_init_early(struct intel_uc *uc) uc->ops = &uc_ops_off; } +void intel_uc_init_late (struct intel_uc *uc) +{ +intel_guc_init_late (&uc->guc); +} +void intel_uc_driver_late_release (struct intel_uc *uc) { }drivers/gpu/drm/i915/gt/uc/intel_uc.h +1
-0@@ -35,6 +35,7 @@ struct intel_uc { };void intel_uc_init_early (struct intel_uc *uc); +void intel_uc_init_late (struct intel_uc *uc);void intel_uc_driver_late_release (struct intel_uc *uc);void intel_uc_driver_remove (struct intel_uc *uc);void intel_uc_init_mmio (struct intel_uc *uc);Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 42/51 drm/i915/guc: Implement banned contexts for GuC submission
In reply to Matthew BrostWhen using GuC submission, if a context gets banned disable scheduling and mark all inflight requests as complete.
--- drivers/gpu/drm/i915/gem/i915_gem_context.c | 2 +- drivers/gpu/drm/i915/gt/intel_context.h | 13 ++ drivers/gpu/drm/i915/gt/intel_context_types.h | 2 + drivers/gpu/drm/i915/gt/intel_reset.c | 32 +--- .../gpu/drm/i915/gt/intel_ring_submission.c | 20 +++ drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 151 ++++++++++++++++-- drivers/gpu/drm/i915/i915_trace.h | 10 ++ 8 files changed, 195 insertions(+), 37 deletions(-)
drivers/gpu/drm/i915/gem/i915_gem_context.c +1 -1
@@ -1084,7 +1084,7 @@ static void kill_engines(struct i915_gem_engines *engines, bool ban)for_each_gem_engine (ce, engines, it) {struct intel_engine_cs *engine; -if (ban &&intel_context_set_banned (ce)) +if (ban &&intel_context_ban (ce,NULL ))continue ;/*
drivers/gpu/drm/i915/gt/intel_context.h +13 -0
@@ -16,6 +16,7 @@#include "intel_engine_types.h" #include "intel_ring_types.h" #include "intel_timeline_types.h" +#include "i915_trace.h" #define CE_TRACE (ce, fmt, ...) do { \ const struct intel_context *ce__ = (ce); \
@@ -243,6 +244,18 @@ static inline bool intel_context_set_banned(struct intel_context *ce)return test_and_set_bit (CONTEXT_BANNED , &ce->flags); } +static inline bool intel_context_ban (struct intel_context *ce, +struct i915_request *rq) +{ +bool ret =intel_context_set_banned (ce); + +trace_intel_context_ban (ce); +if (ce->ops->ban) + ce->ops->ban (ce, rq); + +return ret; +} +static inline bool intel_context_force_single_submission (const struct intel_context *ce) {
drivers/gpu/drm/i915/gt/intel_context_types.h +2 -0
@@ -35,6 +35,8 @@ struct intel_context_ops {int (*alloc)(struct intel_context *ce); +void (*ban)(struct intel_context *ce,struct i915_request *rq); +int (*pre_pin)(struct intel_context *ce,struct i915_gem_ww_ctx *ww,void **vaddr);int (*pin)(struct intel_context *ce,void *vaddr);void (*unpin)(struct intel_context *ce);
drivers/gpu/drm/i915/gt/intel_reset.c +8 -24
@@ -22,7 +22,6 @@#include "intel_reset.h" #include "uc/intel_guc.h" -#include "uc/intel_guc_submission.h" #define RESET_MAX_RETRIES 3
@@ -39,21 +38,6 @@ static void rmw_clear_fw(struct intel_uncore *uncore, i915_reg_t reg, u32 clr)intel_uncore_rmw_fw (uncore, reg, clr,0 ); } -static void skip_context (struct i915_request *rq) -{ -struct intel_context *hung_ctx = rq->context; - -list_for_each_entry_from_rcu (rq, &hung_ctx->timeline->requests, link) { -if (!i915_request_is_active (rq)) -return ; - -if (rq->context == hung_ctx) { -i915_request_set_error_once (rq, -EIO ); -__i915_request_skip (rq); - } - } -} -static void client_mark_guilty (struct i915_gem_context *ctx,bool banned) {struct drm_i915_file_private *file_priv = ctx->file_priv;
@@ -88,10 +72,8 @@ static bool mark_guilty(struct i915_request *rq)bool banned;int i; -if (intel_context_is_closed (rq->context)) { -intel_context_set_banned (rq->context); +if (intel_context_is_closed (rq->context))return true; - }rcu_read_lock (); ctx =rcu_dereference (rq->context->gem_context);
@@ -123,11 +105,9 @@ static bool mark_guilty(struct i915_request *rq) banned = !i915_gem_context_is_recoverable (ctx);if (time_before (jiffies, prev_hang +CONTEXT_FAST_HANG_JIFFIES )) banned = true; -if (banned) { +if (banned)drm_dbg (&ctx->i915->drm,"context %s: guilty %d, banned\n" , ctx->name,atomic_read (&ctx->guilty_count)); -intel_context_set_banned (rq->context); - }client_mark_guilty (ctx, banned);
@@ -149,6 +129,8 @@ static void mark_innocent(struct i915_request *rq)void __i915_request_reset (struct i915_request *rq,bool guilty) { +bool banned = false; +RQ_TRACE (rq,"guilty? %s\n" ,yesno (guilty));GEM_BUG_ON (__i915_request_is_complete (rq));
@@ -156,13 +138,15 @@ void __i915_request_reset(struct i915_request *rq, bool guilty)if (guilty) {i915_request_set_error_once (rq, -EIO );__i915_request_skip (rq); -if (mark_guilty (rq) && !intel_engine_uses_guc (rq->engine)) -skip_context (rq); + banned =mark_guilty (rq); }else {i915_request_set_error_once (rq, -EAGAIN );mark_innocent (rq); }rcu_read_unlock (); + +if (banned) +intel_context_ban (rq->context, rq); }static bool i915_in_reset (struct pci_dev *pdev)
drivers/gpu/drm/i915/gt/intel_ring_submission.c +20 -0
@@ -586,9 +586,29 @@ static void ring_context_reset(struct intel_context *ce)clear_bit (CONTEXT_VALID_BIT , &ce->flags); } +static void ring_context_ban (struct intel_context *ce, +struct i915_request *rq) +{ +struct intel_engine_cs *engine; + +if (!rq || !i915_request_is_active (rq)) +return ; + + engine = rq->engine; +lockdep_assert_held (&engine->sched_engine->lock); +list_for_each_entry_continue (rq, &engine->sched_engine->requests, + sched.link) +if (rq->context == ce) { +i915_request_set_error_once (rq, -EIO ); +__i915_request_skip (rq); + } +} +static const struct intel_context_ops ring_context_ops = { .alloc = ring_context_alloc, + .ban = ring_context_ban, + .pre_pin = ring_context_pre_pin, .pin = ring_context_pin, .unpin = ring_context_unpin,
drivers/gpu/drm/i915/gt/uc/intel_guc.h +2 -0
@@ -274,6 +274,8 @@ void intel_guc_find_hung_context(struct intel_engine_cs *engine);int intel_guc_global_policies_update (struct intel_guc *guc); +void intel_guc_context_ban (struct intel_context *ce,struct i915_request *rq); +void intel_guc_submission_reset_prepare (struct intel_guc *guc);void intel_guc_submission_reset (struct intel_guc *guc,bool stalled);void intel_guc_submission_reset_finish (struct intel_guc *guc);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +139 -12
289 lines not on this page · raw message
drivers/gpu/drm/i915/i915_trace.h +10 -0
@@ -934,6 +934,11 @@ DEFINE_EVENT(intel_context, intel_context_reset,TP_ARGS (ce) ); +DEFINE_EVENT (intel_context, intel_context_ban, + TP_PROTO(struct intel_context *ce), +TP_ARGS (ce) +); +DEFINE_EVENT (intel_context, intel_context_register, TP_PROTO(struct intel_context *ce), TP_ARGS(ce)
@@ -1036,6 +1041,11 @@ trace_intel_context_reset(struct intel_context *ce) { } +static inline void +trace_intel_context_ban (struct intel_context *ce) +{ +} +static inline void trace_intel_context_register (struct intel_context *ce) {
Signature · 1 line
2.28.0
Re: [PATCH 42/51] drm/i915/guc: Implement banned contexts for GuC submission
Quoted 5 lines · Matthew Brost
When using GuC submission, if a context gets banned disable scheduling and mark all inflight requests as complete.
Cc John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>
Quoted 531 lines
--- drivers/gpu/drm/i915/gem/i915_gem_context.c | 2 +- drivers/gpu/drm/i915/gt/intel_context.h | 13 ++ drivers/gpu/drm/i915/gt/intel_context_types.h | 2 + drivers/gpu/drm/i915/gt/intel_reset.c | 32 +--- .../gpu/drm/i915/gt/intel_ring_submission.c | 20 +++ drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 + .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 151 ++++++++++++++++-- drivers/gpu/drm/i915/i915_trace.h | 10 ++ 8 files changed, 195 insertions(+), 37 deletions(-)drivers/gpu/drm/i915/gem/i915_gem_context.c +1
-1@@ -1084,7 +1084,7 @@ static void kill_engines(struct i915_gem_engines *engines, bool ban)for_each_gem_engine (ce, engines, it) {struct intel_engine_cs *engine; -if (ban &&intel_context_set_banned (ce)) +if (ban &&intel_context_ban (ce,NULL ))continue ;/* drivers/gpu/drm/i915/gt/intel_context.h +13
-0@@ -16,6 +16,7 @@#include "intel_engine_types.h" #include "intel_ring_types.h" #include "intel_timeline_types.h" +#include "i915_trace.h" #define CE_TRACE (ce, fmt, ...) do { \ const struct intel_context *ce__ = (ce); \@@ -243,6 +244,18 @@ static inline bool intel_context_set_banned(struct intel_context *ce)return test_and_set_bit (CONTEXT_BANNED , &ce->flags); } +static inline bool intel_context_ban (struct intel_context *ce, +struct i915_request *rq) +{ +bool ret =intel_context_set_banned (ce); + +trace_intel_context_ban (ce); +if (ce->ops->ban) + ce->ops->ban (ce, rq); + +return ret; +} +static inline bool intel_context_force_single_submission (const struct intel_context *ce) {drivers/gpu/drm/i915/gt/intel_context_types.h +2
-0@@ -35,6 +35,8 @@ struct intel_context_ops {int (*alloc)(struct intel_context *ce); +void (*ban)(struct intel_context *ce,struct i915_request *rq); +int (*pre_pin)(struct intel_context *ce,struct i915_gem_ww_ctx *ww,void **vaddr);int (*pin)(struct intel_context *ce,void *vaddr);void (*unpin)(struct intel_context *ce);drivers/gpu/drm/i915/gt/intel_reset.c +8
-24@@ -22,7 +22,6 @@#include "intel_reset.h" #include "uc/intel_guc.h" -#include "uc/intel_guc_submission.h" #define RESET_MAX_RETRIES 3@@ -39,21 +38,6 @@ static void rmw_clear_fw(struct intel_uncore *uncore, i915_reg_t reg, u32 clr)intel_uncore_rmw_fw (uncore, reg, clr,0 ); } -static void skip_context (struct i915_request *rq) -{ -struct intel_context *hung_ctx = rq->context; - -list_for_each_entry_from_rcu (rq, &hung_ctx->timeline->requests, link) { -if (!i915_request_is_active (rq)) -return ; - -if (rq->context == hung_ctx) { -i915_request_set_error_once (rq, -EIO ); -__i915_request_skip (rq); - } - } -} -static void client_mark_guilty (struct i915_gem_context *ctx,bool banned) {struct drm_i915_file_private *file_priv = ctx->file_priv;@@ -88,10 +72,8 @@ static bool mark_guilty(struct i915_request *rq)bool banned;int i; -if (intel_context_is_closed (rq->context)) { -intel_context_set_banned (rq->context); +if (intel_context_is_closed (rq->context))return true; - }rcu_read_lock (); ctx =rcu_dereference (rq->context->gem_context);@@ -123,11 +105,9 @@ static bool mark_guilty(struct i915_request *rq) banned = !i915_gem_context_is_recoverable (ctx);if (time_before (jiffies, prev_hang +CONTEXT_FAST_HANG_JIFFIES )) banned = true; -if (banned) { +if (banned)drm_dbg (&ctx->i915->drm,"context %s: guilty %d, banned\n" , ctx->name,atomic_read (&ctx->guilty_count)); -intel_context_set_banned (rq->context); - }client_mark_guilty (ctx, banned);@@ -149,6 +129,8 @@ static void mark_innocent(struct i915_request *rq)void __i915_request_reset (struct i915_request *rq,bool guilty) { +bool banned = false; +RQ_TRACE (rq,"guilty? %s\n" ,yesno (guilty));GEM_BUG_ON (__i915_request_is_complete (rq));@@ -156,13 +138,15 @@ void __i915_request_reset(struct i915_request *rq, bool guilty)if (guilty) {i915_request_set_error_once (rq, -EIO );__i915_request_skip (rq); -if (mark_guilty (rq) && !intel_engine_uses_guc (rq->engine)) -skip_context (rq); + banned =mark_guilty (rq); }else {i915_request_set_error_once (rq, -EAGAIN );mark_innocent (rq); }rcu_read_unlock (); + +if (banned) +intel_context_ban (rq->context, rq); }static bool i915_in_reset (struct pci_dev *pdev)drivers/gpu/drm/i915/gt/intel_ring_submission.c +20
-0@@ -586,9 +586,29 @@ static void ring_context_reset(struct intel_context *ce)clear_bit (CONTEXT_VALID_BIT , &ce->flags); } +static void ring_context_ban (struct intel_context *ce, +struct i915_request *rq) +{ +struct intel_engine_cs *engine; + +if (!rq || !i915_request_is_active (rq)) +return ; + + engine = rq->engine; +lockdep_assert_held (&engine->sched_engine->lock); +list_for_each_entry_continue (rq, &engine->sched_engine->requests, + sched.link) +if (rq->context == ce) { +i915_request_set_error_once (rq, -EIO ); +__i915_request_skip (rq); + } +} +static const struct intel_context_ops ring_context_ops = { .alloc = ring_context_alloc, + .ban = ring_context_ban, + .pre_pin = ring_context_pre_pin, .pin = ring_context_pin, .unpin = ring_context_unpin,drivers/gpu/drm/i915/gt/uc/intel_guc.h +2
-0@@ -274,6 +274,8 @@ void intel_guc_find_hung_context(struct intel_engine_cs *engine);int intel_guc_global_policies_update (struct intel_guc *guc); +void intel_guc_context_ban (struct intel_context *ce,struct i915_request *rq); +void intel_guc_submission_reset_prepare (struct intel_guc *guc);void intel_guc_submission_reset (struct intel_guc *guc,bool stalled);void intel_guc_submission_reset_finish (struct intel_guc *guc);drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +139
-12@@ -125,6 +125,7 @@ static inline void clr_context_pending_enable(struct intel_context *ce)#define SCHED_STATE_WAIT_FOR_DEREGISTER_TO_REGISTER BIT(0)#define SCHED_STATE_DESTROYED BIT(1)#define SCHED_STATE_PENDING_DISABLE BIT(2) +#define SCHED_STATE_BANNED BIT(3)static inline void init_sched_state (struct intel_context *ce) {/* Only should be called from guc_lrc_desc_pin() */ @@ -187,6 +188,23 @@ static inline void clr_context_pending_disable(struct intel_context *ce) (ce->guc_state.sched_state & ~SCHED_STATE_PENDING_DISABLE ); } +static inline bool context_banned (struct intel_context *ce) +{ +return (ce->guc_state.sched_state &SCHED_STATE_BANNED ); +} + +static inline void set_context_banned (struct intel_context *ce) +{ +lockdep_assert_held (&ce->guc_state.lock); + ce->guc_state.sched_state |=SCHED_STATE_BANNED ; +} + +static inline void clr_context_banned (struct intel_context *ce) +{ +lockdep_assert_held (&ce->guc_state.lock); + ce->guc_state.sched_state &= ~SCHED_STATE_BANNED ; +} +static inline bool context_guc_id_invalid (struct intel_context *ce) {return (ce->guc_id ==GUC_INVALID_LRC_ID );@@ -356,13 +374,23 @@ static int guc_lrc_desc_pin(struct intel_context *ce, bool loop);static int guc_add_request (struct intel_guc *guc,struct i915_request *rq) { -int err; +int err =0 ;struct intel_context *ce = rq->context;u32 action[3 ];int len =0 ;u32 g2h_len_dw =0 ;bool enabled; +/* +* Corner case where requests were sitting in the priority list or a +* request resubmitted after the context was banned. +*/ +if (unlikely (intel_context_is_banned (ce))) { +i915_request_put (i915_request_mark_eio (rq)); +intel_engine_signal_breadcrumbs (ce->engine); + goto out; + } +GEM_BUG_ON (!atomic_read (&ce->guc_id_ref));GEM_BUG_ON (context_guc_id_invalid (ce));@@ -398,6 +426,8 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)clr_context_pending_enable (ce);intel_context_put (ce); } +if (likely (!err)) +trace_i915_request_guc_submit (rq); out:return err;@@ -462,7 +492,6 @@ static int guc_dequeue_one_context(struct intel_guc *guc) guc->stalled_request = last;return false; } -trace_i915_request_guc_submit (last); } guc->stalled_request =NULL ;@@ -501,12 +530,13 @@ static void cs_irq_handler(struct intel_engine_cs *engine, u16 iir)static void __guc_context_destroy (struct intel_context *ce);static void release_guc_id (struct intel_guc *guc,struct intel_context *ce);static void guc_signal_context_fence (struct intel_context *ce); +static void guc_cancel_context_requests (struct intel_context *ce);static void scrub_guc_desc_for_outstanding_g2h (struct intel_guc *guc) {struct intel_context *ce;unsigned long index, flags; -bool pending_disable, pending_enable, deregister, destroyed; +bool pending_disable, pending_enable, deregister, destroyed, banned;xa_for_each (&guc->context_lookup, index, ce) {/* Flush context */ @@ -524,6 +554,7 @@ static void scrub_guc_desc_for_outstanding_g2h(struct intel_guc *guc) pending_enable =context_pending_enable (ce); pending_disable =context_pending_disable (ce); deregister =context_wait_for_deregister_to_register (ce); + banned =context_banned (ce);init_sched_state (ce);if (pending_enable || destroyed || deregister) {@@ -541,6 +572,10 @@ static void scrub_guc_desc_for_outstanding_g2h(struct intel_guc *guc)/* Not mutualy exclusive with above if statement. */ if (pending_disable) {guc_signal_context_fence (ce); +if (banned) { +guc_cancel_context_requests (ce); +intel_engine_signal_breadcrumbs (ce->engine); + }intel_context_sched_disable_unpin (ce);atomic_dec (&guc->outstanding_submission_g2h);intel_context_put (ce);@@ -659,6 +694,9 @@ static void guc_reset_state(struct intel_context *ce, u32 head, bool scrub) {struct intel_engine_cs *engine =__context_to_physical_engine (ce); +if (intel_context_is_banned (ce)) +return ; +GEM_BUG_ON (!intel_context_is_pinned (ce));/* @@ -729,6 +767,8 @@ static void __guc_reset_context(struct intel_context *ce, bool stalled)struct i915_request *rq;u32 head; +intel_context_get (ce); +/* * GuC will implicitly mark the context as non-schedulable * when it sends the reset notification. Make sure our state @@ -754,6 +794,7 @@ static void __guc_reset_context(struct intel_context *ce, bool stalled) out_replay:guc_reset_state (ce, head, stalled);__unwind_incomplete_requests (ce); +intel_context_put (ce); }void intel_guc_submission_reset (struct intel_guc *guc,bool stalled)@@ -936,8 +977,6 @@ static int guc_bypass_tasklet_submit(struct intel_guc *guc, ret =guc_add_request (guc, rq);if (ret == -EBUSY ) guc->stalled_request = rq; -else -trace_i915_request_guc_submit (rq);if (unlikely (ret == -EPIPE ))disable_submission (guc);@@ -1329,13 +1368,77 @@ static u16 prep_context_pending_disable(struct intel_context *ce)return ce->guc_id; } +static void __guc_context_set_preemption_timeout (struct intel_guc *guc, +u16 guc_id, +u32 preemption_timeout) +{ +u32 action [] = { +INTEL_GUC_ACTION_SET_CONTEXT_PREEMPTION_TIMEOUT , + guc_id, + preemption_timeout + }; + +intel_guc_send_busy_loop (guc, action,ARRAY_SIZE (action),0 , true); +} + +static void guc_context_ban (struct intel_context *ce,struct i915_request *rq) +{ +struct intel_guc *guc =ce_to_guc (ce); +struct intel_runtime_pm *runtime_pm = + &ce->engine->gt->i915->runtime_pm; +intel_wakeref_t wakeref; +unsigned long flags; + +guc_flush_submissions (guc); + +spin_lock_irqsave (&ce->guc_state.lock, flags); +set_context_banned (ce); + +if (submission_disabled (guc) || (!context_enabled (ce) && + !context_pending_disable (ce))) { +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + +guc_cancel_context_requests (ce); +intel_engine_signal_breadcrumbs (ce->engine); + }else if (!context_pending_disable (ce)) { +u16 guc_id; + +/* +* We add +2 here as the schedule disable complete CTB handler +* calls intel_context_sched_disable_unpin (-2 to pin_count). +*/ +atomic_add (2 , &ce->pin_count); + + guc_id =prep_context_pending_disable (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + +/* +* In addition to disabling scheduling, set the preemption +* timeout to the minimum value (1 us) so the banned context +* gets kicked off the HW ASAP. +*/ + with_intel_runtime_pm(runtime_pm, wakeref) { +__guc_context_set_preemption_timeout (guc, guc_id,1 ); +__guc_context_sched_disable (guc, ce, guc_id); + } + }else { +if (!context_guc_id_invalid (ce)) + with_intel_runtime_pm(runtime_pm, wakeref) +__guc_context_set_preemption_timeout (guc, + ce->guc_id, +1 ); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + } +} +static void guc_context_sched_disable (struct intel_context *ce) {struct intel_guc *guc =ce_to_guc (ce); -struct intel_runtime_pm *runtime_pm = &ce->engine->gt->i915->runtime_pm;unsigned long flags; -u16 guc_id; +struct intel_runtime_pm *runtime_pm = &ce->engine->gt->i915->runtime_pm;intel_wakeref_t wakeref; +u16 guc_id; +bool enabled;if (submission_disabled (guc) ||context_guc_id_invalid (ce) || !lrc_desc_registered (guc, ce->guc_id)) {@@ -1349,14 +1452,22 @@ static void guc_context_sched_disable(struct intel_context *ce)spin_lock_irqsave (&ce->guc_state.lock, flags);/* -* We have to check if the context has been pinned again as another pin -* operation is allowed to pass this function. Checking the pin count, -* within ce->guc_state.lock, synchronizes this function with +* We have to check if the context has been disabled by another thread. +* We also have to check if the context has been pinned again as another +* pin operation is allowed to pass this function. Checking the pin +* count, within ce->guc_state.lock, synchronizes this function with * guc_request_alloc ensuring a request doesn't slip through the * 'context_pending_disable' fence. Checking within the spin lock (can't * sleep) ensures another process doesn't pin this context and generate * a request before we set the 'context_pending_disable' flag here. */ + enabled =context_enabled (ce); +if (unlikely (!enabled ||submission_disabled (guc))) { +if (enabled) +clr_context_enabled (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + goto unpin; + }if (unlikely (atomic_add_unless (&ce->pin_count,-2 ,2 ))) {spin_unlock_irqrestore (&ce->guc_state.lock, flags);return ;@@ -1513,6 +1624,8 @@ static const struct intel_context_ops guc_context_ops = { .unpin = guc_context_unpin, .post_unpin = guc_context_post_unpin, + .ban = guc_context_ban, + .enter = intel_context_enter_engine, .exit = intel_context_exit_engine,@@ -1706,6 +1819,8 @@ static const struct intel_context_ops virtual_guc_context_ops = { .unpin = guc_context_unpin, .post_unpin = guc_context_post_unpin, + .ban = guc_context_ban, + .enter = guc_virtual_context_enter, .exit = guc_virtual_context_exit,@@ -2159,6 +2274,8 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc,if (context_pending_enable (ce)) {clr_context_pending_enable (ce); }else if (context_pending_disable (ce)) { +bool banned; +/* * Unpin must be done before __guc_signal_context_fence, * otherwise a race exists between the requests getting @@ -2169,9 +2286,16 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc,intel_context_sched_disable_unpin (ce);spin_lock_irqsave (&ce->guc_state.lock, flags); + banned =context_banned (ce); +clr_context_banned (ce);clr_context_pending_disable (ce);__guc_signal_context_fence (ce);spin_unlock_irqrestore (&ce->guc_state.lock, flags); + +if (banned) { +guc_cancel_context_requests (ce); +intel_engine_signal_breadcrumbs (ce->engine); + } }decr_outstanding_submission_g2h (guc);@@ -2206,8 +2330,11 @@ static void guc_handle_context_reset(struct intel_guc *guc, struct intel_context *ce) {trace_intel_context_reset (ce); -capture_error_state (guc, ce); -guc_context_replay (ce); + +if (likely (!intel_context_is_banned (ce))) { +capture_error_state (guc, ce); +guc_context_replay (ce); + } } int intel_guc_context_reset_process_msg(struct intel_guc *guc,drivers/gpu/drm/i915/i915_trace.h +5
-0@@ -934,6 +934,11 @@ DEFINE_EVENT(intel_context, intel_context_reset,TP_ARGS (ce) ); +DEFINE_EVENT (intel_context, intel_context_ban, + TP_PROTO(struct intel_context *ce), +TP_ARGS (ce) +); +DEFINE_EVENT (intel_context, intel_context_register, TP_PROTO(struct intel_context *ce), TP_ARGS(ce)@@ -1036,6 +1041,11 @@ trace_intel_context_reset(struct intel_context *ce) {
drivers/gpu/drm/i915/i915_trace.h trace_intel_context_reset(struct intel_context *ce) · line 1051
Quoted 10 lines
drivers/gpu/drm/i915/i915_trace.h +5
-0} +static inline void +trace_intel_context_ban (struct intel_context *ce) +{ +} +static inline void trace_intel_context_register (struct intel_context *ce) {
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 43/51 drm/i915/guc: Support request cancellation
In reply to Matthew BrostThis adds GuC backend support for i915_request_cancel(), which in turn makes CONFIG_DRM_I915_REQUEST_TIMEOUT work.
--- drivers/gpu/drm/i915/gt/intel_context.c | 9 + drivers/gpu/drm/i915/gt/intel_context.h | 7 + drivers/gpu/drm/i915/gt/intel_context_types.h | 7 + .../drm/i915/gt/intel_execlists_submission.c | 18 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 169 ++++++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 14 +- 6 files changed, 211 insertions(+), 13 deletions(-)
drivers/gpu/drm/i915/gt/intel_context.c +9 -0
@@ -366,6 +366,12 @@ static int __intel_context_active(struct i915_active *active)return 0 ; } +static int sw_fence_dummy_notify (struct i915_sw_fence *sf, +enum i915_sw_fence_notify state) +{ +return NOTIFY_DONE ; +} +void intel_context_init (struct intel_context *ce,struct intel_engine_cs *engine) {
@@ -399,6 +405,9 @@ intel_context_init(struct intel_context *ce, struct intel_engine_cs *engine) ce->guc_id =GUC_INVALID_LRC_ID ;INIT_LIST_HEAD (&ce->guc_id_link); +i915_sw_fence_init (&ce->guc_blocked, sw_fence_dummy_notify); +i915_sw_fence_commit (&ce->guc_blocked); +i915_active_init (&ce->active, __intel_context_active, __intel_context_retire,0 ); }
drivers/gpu/drm/i915/gt/intel_context.h +7 -0
@@ -70,6 +70,13 @@ intel_context_is_pinned(struct intel_context *ce)return atomic_read (&ce->pin_count); } +static inline void intel_context_cancel_request (struct intel_context *ce, +struct i915_request *rq) +{ +GEM_BUG_ON (!ce->ops->cancel_request); +return ce->ops->cancel_request (ce, rq); +} +/** * intel_context_unlock_pinned - Releases the earlier locking of 'pinned' status * @ce - the context
drivers/gpu/drm/i915/gt/intel_context_types.h +7 -0
@@ -13,6 +13,7 @@#include <linux/types.h> #include "i915_active_types.h" +#include "i915_sw_fence.h" #include "i915_utils.h" #include "intel_engine_types.h" #include "intel_sseu.h"
@@ -42,6 +43,9 @@ struct intel_context_ops {void (*unpin)(struct intel_context *ce);void (*post_unpin)(struct intel_context *ce); +void (*cancel_request)(struct intel_context *ce, +struct i915_request *rq); +void (*enter)(struct intel_context *ce);void (*exit)(struct intel_context *ce);
@@ -184,6 +188,9 @@ struct intel_context {* GuC ID link - in list when unpinned but guc_id still valid in GuC */ struct list_head guc_id_link; + +/* GuC context blocked fence */ + struct i915_sw_fence guc_blocked; };#endif /* __INTEL_CONTEXT_TYPES__ */
drivers/gpu/drm/i915/gt/intel_execlists_submission.c +18 -0
@@ -114,6 +114,7 @@#include "gen8_engine_cs.h" #include "intel_breadcrumbs.h" #include "intel_context.h" +#include "intel_engine_heartbeat.h" #include "intel_engine_pm.h" #include "intel_engine_stats.h" #include "intel_execlists_submission.h"
@@ -2536,11 +2537,26 @@ static int execlists_context_alloc(struct intel_context *ce)return lrc_alloc (ce, ce->engine); } +static void execlists_context_cancel_request (struct intel_context *ce, +struct i915_request *rq) +{ +struct intel_engine_cs *engine =NULL ; + +i915_request_active_engine (rq, &engine); + +if (engine &&intel_engine_pulse (engine)) +intel_gt_handle_error (engine->gt, engine->mask,0 , +"request cancellation by %s" , + current->comm); +} +static const struct intel_context_ops execlists_context_ops = { .flags =COPS_HAS_INFLIGHT , .alloc = execlists_context_alloc, + .cancel_request = execlists_context_cancel_request, + .pre_pin = execlists_context_pre_pin, .pin = execlists_context_pin, .unpin = lrc_unpin,
@@ -3558,6 +3574,8 @@ static const struct intel_context_ops virtual_context_ops = { .alloc = virtual_context_alloc, + .cancel_request = execlists_context_cancel_request, + .pre_pin = virtual_context_pre_pin, .pin = virtual_context_pin, .unpin = lrc_unpin,
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +169 -0
240 lines not on this page · raw message
drivers/gpu/drm/i915/i915_request.c +1 -13
@@ -708,18 +708,6 @@ void i915_request_unsubmit(struct i915_request *request)spin_unlock_irqrestore (&engine->sched_engine->lock, flags); } -static void __cancel_request (struct i915_request *rq) -{ -struct intel_engine_cs *engine =NULL ; - -i915_request_active_engine (rq, &engine); - -if (engine &&intel_engine_pulse (engine)) -intel_gt_handle_error (engine->gt, engine->mask,0 , -"request cancellation by %s" , - current->comm); -} -void i915_request_cancel (struct i915_request *rq,int error) {if (!i915_request_set_error_once (rq, error))
@@ -727,7 +715,7 @@ void i915_request_cancel(struct i915_request *rq, int error)set_bit (I915_FENCE_FLAG_SENTINEL , &rq->fence.flags); -__cancel_request (rq); +intel_context_cancel_request (rq->context, rq); }static int __i915_sw_fence_call
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 43/51] drm/i915/guc: Support request cancellation
Quoted 2 lines · Matthew Brost
This adds GuC backend support for
i915_request_cancel(), which in turn makesCONFIG_DRM_I915_REQUEST_TIMEOUTwork.
This needs a bit of explanation on why we're using fences for this instead of other simpler options.
Quoted 29 lines
Signed-off-by Matthew Brost <[email protected]>Cc Tvrtko Ursulin <[email protected]>--- drivers/gpu/drm/i915/gt/intel_context.c | 9 + drivers/gpu/drm/i915/gt/intel_context.h | 7 + drivers/gpu/drm/i915/gt/intel_context_types.h | 7 + .../drm/i915/gt/intel_execlists_submission.c | 18 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 169 ++++++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 14 +- 6 files changed, 211 insertions(+), 13 deletions(-)drivers/gpu/drm/i915/gt/intel_context.c +6
-0@@ -366,6 +366,12 @@ static int __intel_context_active(struct i915_active *active)return 0 ; } +static int sw_fence_dummy_notify (struct i915_sw_fence *sf, +enum i915_sw_fence_notify state) +{ +return NOTIFY_DONE ; +} +void intel_context_init (struct intel_context *ce,struct intel_engine_cs *engine) {
drivers/gpu/drm/i915/gt/intel_context.c intel_context_init(struct intel_context *ce, struct intel_engine_cs *engine) · line 409
Quoted 5 lines
drivers/gpu/drm/i915/gt/intel_context.c +2
-0@@ -399,6 +405,9 @@ intel_context_init(struct intel_context *ce, struct intel_engine_cs *engine) ce->guc_id =GUC_INVALID_LRC_ID ;INIT_LIST_HEAD (&ce->guc_id_link); +i915_sw_fence_init (&ce->guc_blocked, sw_fence_dummy_notify); +i915_sw_fence_commit (&ce->guc_blocked);
We need a comment somewhere to explain how we use this blocked fence, I.e. that fence starts signaled to indicate unblocked and we re-init it to unsignaled status when we need to mark something as blocked.
Quoted 35 lines
drivers/gpu/drm/i915/gt/intel_context.c +1
-0+i915_active_init (&ce->active, __intel_context_active, __intel_context_retire,0 ); }drivers/gpu/drm/i915/gt/intel_context.h +7
-0@@ -70,6 +70,13 @@ intel_context_is_pinned(struct intel_context *ce)return atomic_read (&ce->pin_count); } +static inline void intel_context_cancel_request (struct intel_context *ce, +struct i915_request *rq) +{ +GEM_BUG_ON (!ce->ops->cancel_request); +return ce->ops->cancel_request (ce, rq); +} +/** * intel_context_unlock_pinned - Releases the earlier locking of 'pinned' status * @ce - the context drivers/gpu/drm/i915/gt/intel_context_types.h +1
-0@@ -13,6 +13,7 @@#include <linux/types.h> #include "i915_active_types.h" +#include "i915_sw_fence.h" #include "i915_utils.h" #include "intel_engine_types.h" #include "intel_sseu.h"
drivers/gpu/drm/i915/gt/intel_context_types.h struct intel_context_ops { · line 47
Quoted 5 lines
drivers/gpu/drm/i915/gt/intel_context_types.h +2
-0@@ -42,6 +43,9 @@ struct intel_context_ops {void (*unpin)(struct intel_context *ce);void (*post_unpin)(struct intel_context *ce); +void (*cancel_request)(struct intel_context *ce, +struct i915_request *rq);
I don't see an implementation for this for the ringbuffer backend.
Quoted 82 lines
drivers/gpu/drm/i915/gt/intel_context_types.h +4
-0+void (*enter)(struct intel_context *ce);void (*exit)(struct intel_context *ce);@@ -184,6 +188,9 @@ struct intel_context {* GuC ID link - in list when unpinned but guc_id still valid in GuC */ struct list_head guc_id_link; + +/* GuC context blocked fence */ + struct i915_sw_fence guc_blocked; };#endif /* __INTEL_CONTEXT_TYPES__ */ drivers/gpu/drm/i915/gt/intel_execlists_submission.c +18
-0@@ -114,6 +114,7 @@#include "gen8_engine_cs.h" #include "intel_breadcrumbs.h" #include "intel_context.h" +#include "intel_engine_heartbeat.h" #include "intel_engine_pm.h" #include "intel_engine_stats.h" #include "intel_execlists_submission.h" @@ -2536,11 +2537,26 @@ static int execlists_context_alloc(struct intel_context *ce)return lrc_alloc (ce, ce->engine); } +static void execlists_context_cancel_request (struct intel_context *ce, +struct i915_request *rq) +{ +struct intel_engine_cs *engine =NULL ; + +i915_request_active_engine (rq, &engine); + +if (engine &&intel_engine_pulse (engine)) +intel_gt_handle_error (engine->gt, engine->mask,0 , +"request cancellation by %s" , + current->comm); +} +static const struct intel_context_ops execlists_context_ops = { .flags =COPS_HAS_INFLIGHT , .alloc = execlists_context_alloc, + .cancel_request = execlists_context_cancel_request, + .pre_pin = execlists_context_pre_pin, .pin = execlists_context_pin, .unpin = lrc_unpin,@@ -3558,6 +3574,8 @@ static const struct intel_context_ops virtual_context_ops = { .alloc = virtual_context_alloc, + .cancel_request = execlists_context_cancel_request, + .pre_pin = virtual_context_pre_pin, .pin = virtual_context_pin, .unpin = lrc_unpin,drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +5
-0@@ -81,6 +81,11 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count);*/ #define SCHED_STATE_NO_LOCK_ENABLED BIT(0)#define SCHED_STATE_NO_LOCK_PENDING_ENABLE BIT(1) +#define SCHED_STATE_NO_LOCK_BLOCKED_SHIFT 2 +#define SCHED_STATE_NO_LOCK_BLOCKED \ + BIT(SCHED_STATE_NO_LOCK_BLOCKED_SHIFT) +#define SCHED_STATE_NO_LOCK_BLOCKED_MASK \ + (0xffff << SCHED_STATE_NO_LOCK_BLOCKED_SHIFT)static inline bool context_enabled (struct intel_context *ce) { return (atomic_read(&ce->guc_sched_state_no_lock) &@@ -116,6 +121,27 @@ static inline void clr_context_pending_enable(struct intel_context *ce) &ce->guc_sched_state_no_lock); }
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static inline void clr_context_pending_enable(struct intel_context *ce) · line 133
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +10
-0+static inline u32 context_blocked (struct intel_context *ce) +{ +return (atomic_read (&ce->guc_sched_state_no_lock) & +SCHED_STATE_NO_LOCK_BLOCKED_MASK ) >> +SCHED_STATE_NO_LOCK_BLOCKED_SHIFT ; +} + +static inline void incr_context_blocked (struct intel_context *ce) +{ +lockdep_assert_held (&ce->engine->sched_engine->lock);
It's a bit weird requiring a lock for a variable that is purposely called no_lock, but I do get it is not the GuC lock. Can you explain which race you're trying to guard against?
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static inline void clr_context_pending_enable(struct intel_context *ce) · line 135
Quoted 2 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +2
-0+atomic_add (SCHED_STATE_NO_LOCK_BLOCKED , + &ce->guc_sched_state_no_lock);
Do we need an overflow check, or are we guaranteed that the count will stay within a certain range?
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static inline void clr_context_pending_enable(struct intel_context *ce) · line 140
Quoted 5 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +5
-0+} + +static inline void decr_context_blocked (struct intel_context *ce) +{ +lockdep_assert_held (&ce->engine->sched_engine->lock);
GEM_BUG_ON(!context_blocked(ce)) ?
Quoted 8 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +4
-0+atomic_sub (SCHED_STATE_NO_LOCK_BLOCKED , + &ce->guc_sched_state_no_lock); +} +/* * Below is a set of functions which control the GuC scheduling state which * require a lock, aside from the special case where the functions are called
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static int guc_add_request(struct intel_guc *guc, struct i915_request *rq) · line 434
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +3
-0@@ -403,6 +429,10 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)if (unlikely (err)) goto out; } + +if (unlikely (context_blocked (ce))) + goto out;
You're not setting any error state here for this aborted request. Will the request be automatically re-submitted on unblock? could use a comment if that's the case.
Quoted 24 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0+ enabled =context_enabled (ce);if (!enabled) {@@ -531,6 +561,7 @@ static void __guc_context_destroy(struct intel_context *ce);static void release_guc_id (struct intel_guc *guc,struct intel_context *ce);static void guc_signal_context_fence (struct intel_context *ce);static void guc_cancel_context_requests (struct intel_context *ce); +static void guc_blocked_fence_complete (struct intel_context *ce);static void scrub_guc_desc_for_outstanding_g2h (struct intel_guc *guc) {@@ -578,6 +609,10 @@ static void scrub_guc_desc_for_outstanding_g2h(struct intel_guc *guc) }intel_context_sched_disable_unpin (ce);atomic_dec (&guc->outstanding_submission_g2h); +spin_lock_irqsave (&ce->guc_state.lock, flags); +guc_blocked_fence_complete (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); +intel_context_put (ce); } }
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void guc_context_post_unpin(struct intel_context *ce) · line 1377
Quoted 4 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +1
-0@@ -1339,6 +1374,21 @@ static void guc_context_post_unpin(struct intel_context *ce)lrc_post_unpin (ce); } +static void __guc_context_sched_enable(struct intel_guc *guc,
Why void? this can fail
Quoted 47 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +33
-0+ struct intel_context *ce) +{ +u32 action[] = { +INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_SET , + ce->guc_id, +GUC_CONTEXT_ENABLE + }; + +trace_intel_context_sched_enable (ce); + +guc_submission_send_busy_loop (guc, action,ARRAY_SIZE (action), +G2H_LEN_DW_SCHED_CONTEXT_MODE_SET , true); +} +static void __guc_context_sched_disable (struct intel_guc *guc,struct intel_context *ce,u16 guc_id)@@ -1357,17 +1407,131 @@ static void __guc_context_sched_disable(struct intel_guc *guc, G2H_LEN_DW_SCHED_CONTEXT_MODE_SET, true); } +static void guc_blocked_fence_complete (struct intel_context *ce) +{ +lockdep_assert_held (&ce->guc_state.lock); + +if (!i915_sw_fence_done (&ce->guc_blocked)) +i915_sw_fence_complete (&ce->guc_blocked); +} + +static void guc_blocked_fence_reinit (struct intel_context *ce) +{ +lockdep_assert_held (&ce->guc_state.lock); +GEM_BUG_ON (!i915_sw_fence_done (&ce->guc_blocked)); +i915_sw_fence_fini (&ce->guc_blocked); +i915_sw_fence_reinit (&ce->guc_blocked); +i915_sw_fence_await (&ce->guc_blocked); +i915_sw_fence_commit (&ce->guc_blocked); +} +static u16 prep_context_pending_disable (struct intel_context *ce) {lockdep_assert_held (&ce->guc_state.lock);set_context_pending_disable (ce);clr_context_enabled (ce); +guc_blocked_fence_reinit (ce);intel_context_get (ce);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void __guc_context_sched_disable(struct intel_guc *guc, · line 1445
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0return ce->guc_id; } +static struct i915_sw_fence *guc_context_block (struct intel_context *ce) +{ +struct intel_guc *guc =ce_to_guc (ce); +struct i915_sched_engine *sched_engine = ce->engine->sched_engine; +unsigned long flags; +struct intel_runtime_pm *runtime_pm = &ce->engine->gt->i915->runtime_pm;
engine->uncore->rpm
Quoted 28 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +28
-0+intel_wakeref_t wakeref; +u16 guc_id; +bool enabled; + +spin_lock_irqsave (&sched_engine->lock, flags); +incr_context_blocked (ce); +spin_unlock_irqrestore (&sched_engine->lock, flags); + +spin_lock_irqsave (&ce->guc_state.lock, flags); + enabled =context_enabled (ce); +if (unlikely (!enabled ||submission_disabled (guc))) { +if (enabled) +clr_context_enabled (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); +return &ce->guc_blocked; + } + +/* +* We add +2 here as the schedule disable complete CTB handler calls +* intel_context_sched_disable_unpin (-2 to pin_count). +*/ +atomic_add (2 , &ce->pin_count); + + guc_id =prep_context_pending_disable (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + + with_intel_runtime_pm(runtime_pm, wakeref) +__guc_context_sched_disable (guc, ce, guc_id);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void __guc_context_sched_disable(struct intel_guc *guc, · line 1483
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +10
-0+ +return &ce->guc_blocked; +} + +static void guc_context_unblock (struct intel_context *ce) +{ +struct intel_guc *guc =ce_to_guc (ce); +struct i915_sched_engine *sched_engine = ce->engine->sched_engine; +unsigned long flags; +struct intel_runtime_pm *runtime_pm = &ce->engine->gt->i915->runtime_pm;
engine->uncore->rpm
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0+intel_wakeref_t wakeref; + +GEM_BUG_ON (context_enabled (ce)); + +if (unlikely (context_blocked (ce) >1 )) { +spin_lock_irqsave (&sched_engine->lock, flags);
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void __guc_context_sched_disable(struct intel_guc *guc, · line 1499
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +10
-0+if (likely (context_blocked (ce) >1 )) + goto decrement; +spin_unlock_irqrestore (&sched_engine->lock, flags); + } + +spin_lock_irqsave (&ce->guc_state.lock, flags); +if (unlikely (submission_disabled (guc) || + !intel_context_is_pinned (ce) || +context_pending_disable (ce) || +context_blocked (ce) >1 )) {
you've already checked context_blocked > 1 twice above. If you can't trust the value to remain stable, maybe keep the spinlock locked for more of the flow?
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void __guc_context_sched_disable(struct intel_guc *guc, · line 1505
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0+spin_unlock_irqrestore (&ce->guc_state.lock, flags); + goto out; + } + +set_context_pending_enable (ce); +set_context_enabled (ce);
Shouldn't we set this to enabled only after the H2G has succeeded?
Daniele
Quoted 90 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +34
-0+intel_context_get (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + + with_intel_runtime_pm(runtime_pm, wakeref) +__guc_context_sched_enable (guc, ce); + +out: +spin_lock_irqsave (&sched_engine->lock, flags); +decrement: +decr_context_blocked (ce); +spin_unlock_irqrestore (&sched_engine->lock, flags); +} + +static void guc_context_cancel_request (struct intel_context *ce, +struct i915_request *rq) +{ +if (i915_sw_fence_signaled (&rq->submit)) { +struct i915_sw_fence *fence =guc_context_block (ce); + +i915_sw_fence_wait (fence); +if (!i915_request_completed (rq)) { +__i915_request_skip (rq); +guc_reset_state (ce,intel_ring_wrap (ce->ring, rq->head), + true); + } +guc_context_unblock (ce); + } +} +static void __guc_context_set_preemption_timeout (struct intel_guc *guc,u16 guc_id,u32 preemption_timeout)@@ -1626,6 +1790,8 @@ static const struct intel_context_ops guc_context_ops = { .ban = guc_context_ban, + .cancel_request = guc_context_cancel_request, + .enter = intel_context_enter_engine, .exit = intel_context_exit_engine,@@ -1821,6 +1987,8 @@ static const struct intel_context_ops virtual_guc_context_ops = { .ban = guc_context_ban, + .cancel_request = guc_context_cancel_request, + .enter = guc_virtual_context_enter, .exit = guc_virtual_context_exit,@@ -2290,6 +2458,7 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc,clr_context_banned (ce);clr_context_pending_disable (ce);__guc_signal_context_fence (ce); +guc_blocked_fence_complete (ce);spin_unlock_irqrestore (&ce->guc_state.lock, flags);if (banned) {drivers/gpu/drm/i915/i915_request.c +1
-13@@ -708,18 +708,6 @@ void i915_request_unsubmit(struct i915_request *request)spin_unlock_irqrestore (&engine->sched_engine->lock, flags); } -static void __cancel_request (struct i915_request *rq) -{ -struct intel_engine_cs *engine =NULL ; - -i915_request_active_engine (rq, &engine); - -if (engine &&intel_engine_pulse (engine)) -intel_gt_handle_error (engine->gt, engine->mask,0 , -"request cancellation by %s" , - current->comm); -} -void i915_request_cancel (struct i915_request *rq,int error) {if (!i915_request_set_error_once (rq, error))@@ -727,7 +715,7 @@ void i915_request_cancel(struct i915_request *rq, int error)set_bit (I915_FENCE_FLAG_SENTINEL , &rq->fence.flags); -__cancel_request (rq); +intel_context_cancel_request (rq->context, rq); }static int __i915_sw_fence_call
Re: [PATCH 43/51] drm/i915/guc: Support request cancellation
Quoted 9 lines · Daniele Ceraolo Spurio
Quoted 2 lines · Matthew Brost
This adds GuC backend support for
i915_request_cancel(), which in turn makesCONFIG_DRM_I915_REQUEST_TIMEOUTwork.This needs a bit of explanation on why we're using fences for this instead of other simpler options.
Yep, agree. It doesn't really make sense unless I explain an upcoming patch leverages guc_context_block and waits on the returned fence.
Quoted 28 lines
Quoted 28 lines
Signed-off-by Matthew Brost <[email protected]>Cc Tvrtko Ursulin <[email protected]>--- drivers/gpu/drm/i915/gt/intel_context.c | 9 + drivers/gpu/drm/i915/gt/intel_context.h | 7 + drivers/gpu/drm/i915/gt/intel_context_types.h | 7 + .../drm/i915/gt/intel_execlists_submission.c | 18 ++ .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 169 ++++++++++++++++++ drivers/gpu/drm/i915/i915_request.c | 14 +- 6 files changed, 211 insertions(+), 13 deletions(-)drivers/gpu/drm/i915/gt/intel_context.c +6
-0@@ -366,6 +366,12 @@ static int __intel_context_active(struct i915_active *active)return 0 ; } +static int sw_fence_dummy_notify (struct i915_sw_fence *sf, +enum i915_sw_fence_notify state) +{ +return NOTIFY_DONE ; +} +void intel_context_init (struct intel_context *ce,struct intel_engine_cs *engine) {@@ -399,6 +405,9 @@ intel_context_init(struct intel_context *ce, struct intel_engine_cs *engine)
Quoted 9 lines
drivers/gpu/drm/i915/gt/intel_context.c intel_context_init(struct intel_context *ce, struct intel_engine_cs *engine) · line 409
Quoted 4 lines
drivers/gpu/drm/i915/gt/intel_context.c +2
-0ce->guc_id =GUC_INVALID_LRC_ID ;INIT_LIST_HEAD (&ce->guc_id_link); +i915_sw_fence_init (&ce->guc_blocked, sw_fence_dummy_notify); +i915_sw_fence_commit (&ce->guc_blocked);We need a comment somewhere to explain how we use this blocked fence, I.e. that fence starts signaled to indicate unblocked and we re-init it to unsignaled status when we need to mark something as blocked.
Sure.
Quoted 31 lines
Quoted 31 lines
drivers/gpu/drm/i915/gt/intel_context.c +1
-0+i915_active_init (&ce->active, __intel_context_active, __intel_context_retire,0 ); }drivers/gpu/drm/i915/gt/intel_context.h +7
-0@@ -70,6 +70,13 @@ intel_context_is_pinned(struct intel_context *ce)return atomic_read (&ce->pin_count); } +static inline void intel_context_cancel_request (struct intel_context *ce, +struct i915_request *rq) +{ +GEM_BUG_ON (!ce->ops->cancel_request); +return ce->ops->cancel_request (ce, rq); +} +/** * intel_context_unlock_pinned - Releases the earlier locking of 'pinned' status * @ce - the context drivers/gpu/drm/i915/gt/intel_context_types.h +1
-0@@ -13,6 +13,7 @@#include <linux/types.h> #include "i915_active_types.h" +#include "i915_sw_fence.h" #include "i915_utils.h" #include "intel_engine_types.h"
Quoted 9 lines
drivers/gpu/drm/i915/gt/intel_context_types.h struct intel_context_ops { · line 46
Quoted 6 lines
drivers/gpu/drm/i915/gt/intel_context_types.h +2
-0#include "intel_sseu.h" @@ -42,6 +43,9 @@ struct intel_context_ops {void (*unpin)(struct intel_context *ce);void (*post_unpin)(struct intel_context *ce); +void (*cancel_request)(struct intel_context *ce, +struct i915_request *rq);I don't see an implementation for this for the ringbuffer backend.
Yea, I probably need to add that.
Quoted 80 lines
Quoted 80 lines
drivers/gpu/drm/i915/gt/intel_context_types.h +4
-0+void (*enter)(struct intel_context *ce);void (*exit)(struct intel_context *ce);@@ -184,6 +188,9 @@ struct intel_context {* GuC ID link - in list when unpinned but guc_id still valid in GuC */ struct list_head guc_id_link; + +/* GuC context blocked fence */ + struct i915_sw_fence guc_blocked; };#endif /* __INTEL_CONTEXT_TYPES__ */ drivers/gpu/drm/i915/gt/intel_execlists_submission.c +18
-0@@ -114,6 +114,7 @@#include "gen8_engine_cs.h" #include "intel_breadcrumbs.h" #include "intel_context.h" +#include "intel_engine_heartbeat.h" #include "intel_engine_pm.h" #include "intel_engine_stats.h" #include "intel_execlists_submission.h" @@ -2536,11 +2537,26 @@ static int execlists_context_alloc(struct intel_context *ce)return lrc_alloc (ce, ce->engine); } +static void execlists_context_cancel_request (struct intel_context *ce, +struct i915_request *rq) +{ +struct intel_engine_cs *engine =NULL ; + +i915_request_active_engine (rq, &engine); + +if (engine &&intel_engine_pulse (engine)) +intel_gt_handle_error (engine->gt, engine->mask,0 , +"request cancellation by %s" , + current->comm); +} +static const struct intel_context_ops execlists_context_ops = { .flags =COPS_HAS_INFLIGHT , .alloc = execlists_context_alloc, + .cancel_request = execlists_context_cancel_request, + .pre_pin = execlists_context_pre_pin, .pin = execlists_context_pin, .unpin = lrc_unpin,@@ -3558,6 +3574,8 @@ static const struct intel_context_ops virtual_context_ops = { .alloc = virtual_context_alloc, + .cancel_request = execlists_context_cancel_request, + .pre_pin = virtual_context_pre_pin, .pin = virtual_context_pin, .unpin = lrc_unpin,drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +11
-0@@ -81,6 +81,11 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count);*/ #define SCHED_STATE_NO_LOCK_ENABLED BIT(0)#define SCHED_STATE_NO_LOCK_PENDING_ENABLE BIT(1) +#define SCHED_STATE_NO_LOCK_BLOCKED_SHIFT 2 +#define SCHED_STATE_NO_LOCK_BLOCKED \ + BIT(SCHED_STATE_NO_LOCK_BLOCKED_SHIFT) +#define SCHED_STATE_NO_LOCK_BLOCKED_MASK \ + (0xffff << SCHED_STATE_NO_LOCK_BLOCKED_SHIFT)static inline bool context_enabled (struct intel_context *ce) { return (atomic_read(&ce->guc_sched_state_no_lock) &@@ -116,6 +121,27 @@ static inline void clr_context_pending_enable(struct intel_context *ce) &ce->guc_sched_state_no_lock); } +static inline u32 context_blocked (struct intel_context *ce) +{ +return (atomic_read (&ce->guc_sched_state_no_lock) & +SCHED_STATE_NO_LOCK_BLOCKED_MASK ) >> +SCHED_STATE_NO_LOCK_BLOCKED_SHIFT ; +}
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static inline void clr_context_pending_enable(struct intel_context *ce) · line 133
Quoted 4 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +4
-0+ +static inline void incr_context_blocked (struct intel_context *ce) +{ +lockdep_assert_held (&ce->engine->sched_engine->lock);It's a bit weird requiring a lock for a variable that is purposely called
no_lock, but I do get it is not the GuC lock. Can you explain which race you're trying to guard against?
This is tricky, basically it is to make the 'context_block' visible to the submission path. I'll add a comment.
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static inline void clr_context_pending_enable(struct intel_context *ce) · line 134
Quoted 2 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +2
-0+atomic_add (SCHED_STATE_NO_LOCK_BLOCKED , + &ce->guc_sched_state_no_lock);Do we need an overflow check, or are we guaranteed that the count will stay within a certain range?
I should add a GEM_BUG_ON here.
Quoted 8 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static inline void clr_context_pending_enable(struct intel_context *ce) · line 140
Quoted 5 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +5
-0+} + +static inline void decr_context_blocked (struct intel_context *ce) +{ +lockdep_assert_held (&ce->engine->sched_engine->lock);GEM_BUG_ON(!context_blocked(ce)) ?
Yep.
Quoted 10 lines
Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +4
-0+atomic_sub (SCHED_STATE_NO_LOCK_BLOCKED , + &ce->guc_sched_state_no_lock); +} +/* * Below is a set of functions which control the GuC scheduling state which * require a lock, aside from the special case where the functions are called @@ -403,6 +429,10 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)if (unlikely (err)) goto out;
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static int guc_add_request(struct intel_guc *guc, struct i915_request *rq) · line 434
Quoted 4 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +3
-0} + +if (unlikely (context_blocked (ce))) + goto out;You're not setting any error state here for this aborted request. Will the request be automatically re-submitted on unblock? could use a comment if that's the case.
Yes, when scheduling gets enabled by guc_context_unblock the request can start running. Can add a comment explaining that.
Quoted 19 lines
Quoted 19 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0+ enabled =context_enabled (ce);if (!enabled) {@@ -531,6 +561,7 @@ static void __guc_context_destroy(struct intel_context *ce);static void release_guc_id (struct intel_guc *guc,struct intel_context *ce);static void guc_signal_context_fence (struct intel_context *ce);static void guc_cancel_context_requests (struct intel_context *ce); +static void guc_blocked_fence_complete (struct intel_context *ce);static void scrub_guc_desc_for_outstanding_g2h (struct intel_guc *guc) {@@ -578,6 +609,10 @@ static void scrub_guc_desc_for_outstanding_g2h(struct intel_guc *guc) }intel_context_sched_disable_unpin (ce);atomic_dec (&guc->outstanding_submission_g2h); +spin_lock_irqsave (&ce->guc_state.lock, flags); +guc_blocked_fence_complete (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); +intel_context_put (ce);
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void guc_context_post_unpin(struct intel_context *ce) · line 1376
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +1
-0} }@@ -1339,6 +1374,21 @@ static void guc_context_post_unpin(struct intel_context *ce)lrc_post_unpin (ce); } +static void __guc_context_sched_enable(struct intel_guc *guc,Why void? this can fail
This is a case where we don't care it it fails as that means a full GT reset will happen. This isn't the H2G that we do this with (e.g. __guc_context_sched_disable is the same).
Quoted 47 lines
Quoted 47 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +33
-0+ struct intel_context *ce) +{ +u32 action[] = { +INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_SET , + ce->guc_id, +GUC_CONTEXT_ENABLE + }; + +trace_intel_context_sched_enable (ce); + +guc_submission_send_busy_loop (guc, action,ARRAY_SIZE (action), +G2H_LEN_DW_SCHED_CONTEXT_MODE_SET , true); +} +static void __guc_context_sched_disable (struct intel_guc *guc,struct intel_context *ce,u16 guc_id)@@ -1357,17 +1407,131 @@ static void __guc_context_sched_disable(struct intel_guc *guc, G2H_LEN_DW_SCHED_CONTEXT_MODE_SET, true); } +static void guc_blocked_fence_complete (struct intel_context *ce) +{ +lockdep_assert_held (&ce->guc_state.lock); + +if (!i915_sw_fence_done (&ce->guc_blocked)) +i915_sw_fence_complete (&ce->guc_blocked); +} + +static void guc_blocked_fence_reinit (struct intel_context *ce) +{ +lockdep_assert_held (&ce->guc_state.lock); +GEM_BUG_ON (!i915_sw_fence_done (&ce->guc_blocked)); +i915_sw_fence_fini (&ce->guc_blocked); +i915_sw_fence_reinit (&ce->guc_blocked); +i915_sw_fence_await (&ce->guc_blocked); +i915_sw_fence_commit (&ce->guc_blocked); +} +static u16 prep_context_pending_disable (struct intel_context *ce) {lockdep_assert_held (&ce->guc_state.lock);set_context_pending_disable (ce);clr_context_enabled (ce); +guc_blocked_fence_reinit (ce);intel_context_get (ce);return ce->guc_id; }
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void __guc_context_sched_disable(struct intel_guc *guc, · line 1445
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0+static struct i915_sw_fence *guc_context_block (struct intel_context *ce) +{ +struct intel_guc *guc =ce_to_guc (ce); +struct i915_sched_engine *sched_engine = ce->engine->sched_engine; +unsigned long flags; +struct intel_runtime_pm *runtime_pm = &ce->engine->gt->i915->runtime_pm;
engine->uncore->rpm
Yep.
Quoted 32 lines
Quoted 32 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +32
-0+intel_wakeref_t wakeref; +u16 guc_id; +bool enabled; + +spin_lock_irqsave (&sched_engine->lock, flags); +incr_context_blocked (ce); +spin_unlock_irqrestore (&sched_engine->lock, flags); + +spin_lock_irqsave (&ce->guc_state.lock, flags); + enabled =context_enabled (ce); +if (unlikely (!enabled ||submission_disabled (guc))) { +if (enabled) +clr_context_enabled (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); +return &ce->guc_blocked; + } + +/* +* We add +2 here as the schedule disable complete CTB handler calls +* intel_context_sched_disable_unpin (-2 to pin_count). +*/ +atomic_add (2 , &ce->pin_count); + + guc_id =prep_context_pending_disable (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + + with_intel_runtime_pm(runtime_pm, wakeref) +__guc_context_sched_disable (guc, ce, guc_id); + +return &ce->guc_blocked; +} +
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void __guc_context_sched_disable(struct intel_guc *guc, · line 1483
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0+static void guc_context_unblock (struct intel_context *ce) +{ +struct intel_guc *guc =ce_to_guc (ce); +struct i915_sched_engine *sched_engine = ce->engine->sched_engine; +unsigned long flags; +struct intel_runtime_pm *runtime_pm = &ce->engine->gt->i915->runtime_pm;
engine->uncore->rpm
Yep.
Quoted 12 lines
Quoted 12 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +12
-0+intel_wakeref_t wakeref; + +GEM_BUG_ON (context_enabled (ce)); + +if (unlikely (context_blocked (ce) >1 )) { +spin_lock_irqsave (&sched_engine->lock, flags); +if (likely (context_blocked (ce) >1 )) + goto decrement; +spin_unlock_irqrestore (&sched_engine->lock, flags); + } + +spin_lock_irqsave (&ce->guc_state.lock, flags);
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void __guc_context_sched_disable(struct intel_guc *guc, · line 1499
Quoted 4 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +4
-0+if (unlikely (submission_disabled (guc) || + !intel_context_is_pinned (ce) || +context_pending_disable (ce) || +context_blocked (ce) >1 )) {you've already checked
context_blocked> 1 twice above. If you can't trust the value to remain stable, maybe keep the spinlock locked for more of the flow?
I really don't know what I was thinking with the checks above this, maybe trying to optimize this to avoid taking the spin lock? The above checks can / should be deleted.
Quoted 9 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c static void __guc_context_sched_disable(struct intel_guc *guc, · line 1505
Quoted 6 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +6
-0+spin_unlock_irqrestore (&ce->guc_state.lock, flags); + goto out; + } + +set_context_pending_enable (ce); +set_context_enabled (ce);Shouldn't we set this to enabled only after the H2G has succeeded?
No. We always enable immediately - see the submission path. It is enabled from the point of view if you submit something after this it is enabled as the schedule enable is in the queue. Again if this fails for some reason it means we to do a full GT reset and don't care.
Matt
Quoted 73 lines
Daniele
Quoted 71 lines
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +34
-0+intel_context_get (ce); +spin_unlock_irqrestore (&ce->guc_state.lock, flags); + + with_intel_runtime_pm(runtime_pm, wakeref) +__guc_context_sched_enable (guc, ce); + +out: +spin_lock_irqsave (&sched_engine->lock, flags); +decrement: +decr_context_blocked (ce); +spin_unlock_irqrestore (&sched_engine->lock, flags); +} + +static void guc_context_cancel_request (struct intel_context *ce, +struct i915_request *rq) +{ +if (i915_sw_fence_signaled (&rq->submit)) { +struct i915_sw_fence *fence =guc_context_block (ce); + +i915_sw_fence_wait (fence); +if (!i915_request_completed (rq)) { +__i915_request_skip (rq); +guc_reset_state (ce,intel_ring_wrap (ce->ring, rq->head), + true); + } +guc_context_unblock (ce); + } +} +static void __guc_context_set_preemption_timeout (struct intel_guc *guc,u16 guc_id,u32 preemption_timeout)@@ -1626,6 +1790,8 @@ static const struct intel_context_ops guc_context_ops = { .ban = guc_context_ban, + .cancel_request = guc_context_cancel_request, + .enter = intel_context_enter_engine, .exit = intel_context_exit_engine,@@ -1821,6 +1987,8 @@ static const struct intel_context_ops virtual_guc_context_ops = { .ban = guc_context_ban, + .cancel_request = guc_context_cancel_request, + .enter = guc_virtual_context_enter, .exit = guc_virtual_context_exit,@@ -2290,6 +2458,7 @@ int intel_guc_sched_done_process_msg(struct intel_guc *guc,clr_context_banned (ce);clr_context_pending_disable (ce);__guc_signal_context_fence (ce); +guc_blocked_fence_complete (ce);spin_unlock_irqrestore (&ce->guc_state.lock, flags);if (banned) {drivers/gpu/drm/i915/i915_request.c +0
-12@@ -708,18 +708,6 @@ void i915_request_unsubmit(struct i915_request *request)spin_unlock_irqrestore (&engine->sched_engine->lock, flags); } -static void __cancel_request (struct i915_request *rq) -{ -struct intel_engine_cs *engine =NULL ; - -i915_request_active_engine (rq, &engine); - -if (engine &&intel_engine_pulse (engine)) -intel_gt_handle_error (engine->gt, engine->mask,0 , -"request cancellation by %s" , - current->comm); -} -void i915_request_cancel (struct i915_request *rq,int error)
Quoted 9 lines
drivers/gpu/drm/i915/i915_request.c void i915_request_cancel(struct i915_request *rq, int error) · line 718
Quoted 8 lines
drivers/gpu/drm/i915/i915_request.c +1
-1{if (!i915_request_set_error_once (rq, error))@@ -727,7 +715,7 @@ void i915_request_cancel(struct i915_request *rq, int error)set_bit (I915_FENCE_FLAG_SENTINEL , &rq->fence.flags); -__cancel_request (rq); +intel_context_cancel_request (rq->context, rq); }static int __i915_sw_fence_call
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 44/51 drm/i915/selftest: Better error reporting from hangcheck selftest
In reply to Matthew BrostFrom: John Harrison <[email protected]>
There are many ways in which the hangcheck selftest can fail. Very few of them actually printed an error message to say what happened. So, fill in the missing messages.
--- drivers/gpu/drm/i915/gt/selftest_hangcheck.c | 89 ++++++++++++++++---- 1 file changed, 72 insertions(+), 17 deletions(-)
drivers/gpu/drm/i915/gt/selftest_hangcheck.c +72 -17
@@ -378,6 +378,7 @@ static int igt_reset_nop(void *arg) ce =intel_context_create (engine);if (IS_ERR (ce)) { err =PTR_ERR (ce); +pr_err ("[%s] Create context failed: %d!\n" , engine->name, err);break ; }
@@ -387,6 +388,7 @@ static int igt_reset_nop(void *arg) rq =intel_context_create_request (ce);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create request failed: %d!\n" , engine->name, err);break ; }
@@ -401,24 +403,31 @@ static int igt_reset_nop(void *arg)igt_global_reset_unlock (gt);if (intel_gt_is_wedged (gt)) { +pr_err ("[%s] GT is wedged!\n" , engine->name); err = -EIO ;break ; }if (i915_reset_count (global) != reset_count + ++count) { -pr_err ("Full GPU reset not recorded!\n" ); +pr_err ("[%s] Reset not recorded: %d vs %d + %d!\n" , + engine->name,i915_reset_count (global), reset_count, count); err = -EINVAL ;break ; } err =igt_flush_test (gt->i915); -if (err) +if (err) { +pr_err ("[%s] Flush failed: %d!\n" , engine->name, err);break ; + } }while (time_before (jiffies, end_time));pr_info ("%s: %d resets\n" , __func__, count); -if (igt_flush_test (gt->i915)) +if (igt_flush_test (gt->i915)) { +pr_err ("Post flush failed: %d!\n" , err); err = -EIO ; + } +return err; }
@@ -441,8 +450,10 @@ static int igt_reset_nop_engine(void *arg)int err; ce =intel_context_create (engine); -if (IS_ERR (ce)) +if (IS_ERR (ce)) { +pr_err ("[%s] Create context failed: %d!\n" , engine->name, err);return PTR_ERR (ce); + } reset_count =i915_reset_count (global); reset_engine_count =i915_reset_engine_count (global, engine);
@@ -550,8 +561,10 @@ static int igt_reset_fail_engine(void *arg)int err; ce =intel_context_create (engine); -if (IS_ERR (ce)) +if (IS_ERR (ce)) { +pr_err ("[%s] Create context failed: %d!\n" , engine->name, err);return PTR_ERR (ce); + }st_engine_heartbeat_disable (engine);set_bit (I915_RESET_ENGINE + id, >->reset.flags);
@@ -711,6 +724,7 @@ static int __igt_reset_engine(struct intel_gt *gt, bool active) rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err);break ; }
@@ -765,12 +779,16 @@ static int __igt_reset_engine(struct intel_gt *gt, bool active)break ; err =igt_flush_test (gt->i915); -if (err) +if (err) { +pr_err ("[%s] Flush failed: %d!\n" , engine->name, err);break ; + } } -if (intel_gt_is_wedged (gt)) +if (intel_gt_is_wedged (gt)) { +pr_err ("GT is wedged!\n" ); err = -EIO ; + }if (active)hang_fini (&h);
@@ -837,6 +855,7 @@ static int active_engine(void *data) ce[count] =intel_context_create (engine);if (IS_ERR (ce[count])) { err =PTR_ERR (ce[count]); +pr_err ("[%s] Create context #%ld failed: %d!\n" , engine->name, count, err);while (--count)intel_context_put (ce[count]);return err;
@@ -852,6 +871,7 @@ static int active_engine(void *data) new =intel_context_create_request (ce[idx]);if (IS_ERR (new)) { err =PTR_ERR (new); +pr_err ("[%s] Create request #%d failed: %d!\n" , engine->name, idx, err);break ; }
@@ -867,8 +887,10 @@ static int active_engine(void *data) } err =active_request_put (old); -if (err) +if (err) { +pr_err ("[%s] Request put failed: %d!\n" , engine->name, err);break ; + }cond_resched (); }
@@ -876,6 +898,9 @@ static int active_engine(void *data)for (count =0 ; count <ARRAY_SIZE (rq); count++) {int err__ =active_request_put (rq[count]); +if (err) +pr_err ("[%s] Request put #%ld failed: %d!\n" , engine->name, count, err); +/* Keep the first error */ if (!err) err = err__;
@@ -949,6 +974,7 @@ static int __igt_reset_engines(struct intel_gt *gt, "igt/%s", other->name);if (IS_ERR (tsk)) { err =PTR_ERR (tsk); +pr_err ("[%s] Thread spawn failed: %d!\n" , engine->name, err); goto unwind; }
@@ -967,6 +993,7 @@ static int __igt_reset_engines(struct intel_gt *gt, rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err);break ; }
@@ -999,10 +1026,10 @@ static int __igt_reset_engines(struct intel_gt *gt,if (rq) {if (rq->fence.error != -EIO ) {pr_err ("i915_reset_engine(%s:%s):" -" failed to reset request %llx:%lld\n" , +" failed to reset request %lld:%lld [0x%04X]\n" , engine->name, test_name, rq->fence.context, - rq->fence.seqno); + rq->fence.seqno, rq->context->guc_id);i915_request_put (rq);GEM_TRACE_DUMP ();
@@ -1101,8 +1128,10 @@ static int __igt_reset_engines(struct intel_gt *gt,break ; err =igt_flush_test (gt->i915); -if (err) +if (err) { +pr_err ("[%s] Flush failed: %d!\n" , engine->name, err);break ; + } }if (intel_gt_is_wedged (gt))
@@ -1180,12 +1209,15 @@ static int igt_reset_wait(void *arg)igt_global_reset_lock (gt); err =hang_init (&h, gt); -if (err) +if (err) { +pr_err ("[%s] Hang init failed: %d!\n" , engine->name, err); goto unlock; + } rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto fini; }
@@ -1310,12 +1342,15 @@ static int __igt_reset_evict_vma(struct intel_gt *gt,/* Check that we can recover an unbind stuck on a hanging request */ err =hang_init (&h, gt); -if (err) +if (err) { +pr_err ("[%s] Hang init failed: %d!\n" , engine->name, err);return err; + } obj =i915_gem_object_create_internal (gt->i915,SZ_1M );if (IS_ERR (obj)) { err =PTR_ERR (obj); +pr_err ("[%s] Create object failed: %d!\n" , engine->name, err); goto fini; }
@@ -1330,12 +1365,14 @@ static int __igt_reset_evict_vma(struct intel_gt *gt, arg.vma =i915_vma_instance (obj, vm,NULL );if (IS_ERR (arg.vma)) { err =PTR_ERR (arg.vma); +pr_err ("[%s] VMA instance failed: %d!\n" , engine->name, err); goto out_obj; } rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto out_obj; }
@@ -1347,6 +1384,7 @@ static int __igt_reset_evict_vma(struct intel_gt *gt, err =i915_vma_pin (arg.vma,0 ,0 , pin_flags);if (err) {i915_request_add (rq); +pr_err ("[%s] VMA pin failed: %d!\n" , engine->name, err); goto out_obj; }
@@ -1363,8 +1401,14 @@ static int __igt_reset_evict_vma(struct intel_gt *gt,i915_vma_lock (arg.vma); err =i915_request_await_object (rq, arg.vma->obj, flags &EXEC_OBJECT_WRITE ); -if (err ==0 ) +if (err ==0 ) { err =i915_vma_move_to_active (arg.vma, rq, flags); +if (err) +pr_err ("[%s] Move to active failed: %d!\n" , engine->name, err); + }else { +pr_err ("[%s] Request await failed: %d!\n" , engine->name, err); + } +i915_vma_unlock (arg.vma);if (flags &EXEC_OBJECT_NEEDS_FENCE )
@@ -1392,6 +1436,7 @@ static int __igt_reset_evict_vma(struct intel_gt *gt, tsk =kthread_run (fn, &arg,"igt/evict_vma" );if (IS_ERR (tsk)) { err =PTR_ERR (tsk); +pr_err ("[%s] Thread spawn failed: %d!\n" , engine->name, err); tsk =NULL ; goto out_reset; }
@@ -1518,6 +1563,7 @@ static int igt_reset_queue(void *arg) prev =hang_create_request (&h, engine);if (IS_ERR (prev)) { err =PTR_ERR (prev); +pr_err ("[%s] Create 'prev' hang request failed: %d!\n" , engine->name, err); goto fini; }
@@ -1532,6 +1578,7 @@ static int igt_reset_queue(void *arg) rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto fini; }
@@ -1619,8 +1666,10 @@ static int igt_reset_queue(void *arg)i915_request_put (prev); err =igt_flush_test (gt->i915); -if (err) +if (err) { +pr_err ("[%s] Flush failed: %d!\n" , engine->name, err);break ; + } } fini:
@@ -1653,12 +1702,15 @@ static int igt_handle_error(void *arg)return 0 ; err =hang_init (&h, gt); -if (err) +if (err) { +pr_err ("[%s] Hang init failed: %d!\n" , engine->name, err);return err; + } rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto err_fini; }
@@ -1743,12 +1795,15 @@ static int igt_atomic_reset_engine(struct intel_engine_cs *engine,return err; err =hang_init (&h, engine->gt); -if (err) +if (err) { +pr_err ("[%s] Hang init failed: %d!\n" , engine->name, err);return err; + } rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto out; }
Signature · 1 line
2.28.0
Re: [Intel-gfx] [PATCH 44/51] drm/i915/selftest: Better error reporting from hangcheck selftest
Quoted 9 lines · Matthew Brost
From: John Harrison <[email protected]>
There are many ways in which the hangcheck selftest can fail. Very few of them actually printed an error message to say what happened. So, fill in the missing messages.
Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>Cc Daniele Ceraolo Spurio <[email protected]>
Quoted 339 lines
--- drivers/gpu/drm/i915/gt/selftest_hangcheck.c | 89 ++++++++++++++++---- 1 file changed, 72 insertions(+), 17 deletions(-)drivers/gpu/drm/i915/gt/selftest_hangcheck.c +72
-17@@ -378,6 +378,7 @@ static int igt_reset_nop(void *arg) ce =intel_context_create (engine);if (IS_ERR (ce)) { err =PTR_ERR (ce); +pr_err ("[%s] Create context failed: %d!\n" , engine->name, err);break ; }@@ -387,6 +388,7 @@ static int igt_reset_nop(void *arg) rq =intel_context_create_request (ce);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create request failed: %d!\n" , engine->name, err);break ; }@@ -401,24 +403,31 @@ static int igt_reset_nop(void *arg)igt_global_reset_unlock (gt);if (intel_gt_is_wedged (gt)) { +pr_err ("[%s] GT is wedged!\n" , engine->name); err = -EIO ;break ; }if (i915_reset_count (global) != reset_count + ++count) { -pr_err ("Full GPU reset not recorded!\n" ); +pr_err ("[%s] Reset not recorded: %d vs %d + %d!\n" , + engine->name,i915_reset_count (global), reset_count, count); err = -EINVAL ;break ; } err =igt_flush_test (gt->i915); -if (err) +if (err) { +pr_err ("[%s] Flush failed: %d!\n" , engine->name, err);break ; + } }while (time_before (jiffies, end_time));pr_info ("%s: %d resets\n" , __func__, count); -if (igt_flush_test (gt->i915)) +if (igt_flush_test (gt->i915)) { +pr_err ("Post flush failed: %d!\n" , err); err = -EIO ; + } +return err; }@@ -441,8 +450,10 @@ static int igt_reset_nop_engine(void *arg)int err; ce =intel_context_create (engine); -if (IS_ERR (ce)) +if (IS_ERR (ce)) { +pr_err ("[%s] Create context failed: %d!\n" , engine->name, err);return PTR_ERR (ce); + } reset_count =i915_reset_count (global); reset_engine_count =i915_reset_engine_count (global, engine);@@ -550,8 +561,10 @@ static int igt_reset_fail_engine(void *arg)int err; ce =intel_context_create (engine); -if (IS_ERR (ce)) +if (IS_ERR (ce)) { +pr_err ("[%s] Create context failed: %d!\n" , engine->name, err);return PTR_ERR (ce); + }st_engine_heartbeat_disable (engine);set_bit (I915_RESET_ENGINE + id, >->reset.flags);@@ -711,6 +724,7 @@ static int __igt_reset_engine(struct intel_gt *gt, bool active) rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err);break ; }@@ -765,12 +779,16 @@ static int __igt_reset_engine(struct intel_gt *gt, bool active)break ; err =igt_flush_test (gt->i915); -if (err) +if (err) { +pr_err ("[%s] Flush failed: %d!\n" , engine->name, err);break ; + } } -if (intel_gt_is_wedged (gt)) +if (intel_gt_is_wedged (gt)) { +pr_err ("GT is wedged!\n" ); err = -EIO ; + }if (active)hang_fini (&h);@@ -837,6 +855,7 @@ static int active_engine(void *data) ce[count] =intel_context_create (engine);if (IS_ERR (ce[count])) { err =PTR_ERR (ce[count]); +pr_err ("[%s] Create context #%ld failed: %d!\n" , engine->name, count, err);while (--count)intel_context_put (ce[count]);return err;@@ -852,6 +871,7 @@ static int active_engine(void *data) new =intel_context_create_request (ce[idx]);if (IS_ERR (new)) { err =PTR_ERR (new); +pr_err ("[%s] Create request #%d failed: %d!\n" , engine->name, idx, err);break ; }@@ -867,8 +887,10 @@ static int active_engine(void *data) } err =active_request_put (old); -if (err) +if (err) { +pr_err ("[%s] Request put failed: %d!\n" , engine->name, err);break ; + }cond_resched (); }@@ -876,6 +898,9 @@ static int active_engine(void *data)for (count =0 ; count <ARRAY_SIZE (rq); count++) {int err__ =active_request_put (rq[count]); +if (err) +pr_err ("[%s] Request put #%ld failed: %d!\n" , engine->name, count, err); +/* Keep the first error */ if (!err) err = err__;@@ -949,6 +974,7 @@ static int __igt_reset_engines(struct intel_gt *gt, "igt/%s", other->name);if (IS_ERR (tsk)) { err =PTR_ERR (tsk); +pr_err ("[%s] Thread spawn failed: %d!\n" , engine->name, err); goto unwind; }@@ -967,6 +993,7 @@ static int __igt_reset_engines(struct intel_gt *gt, rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err);break ; }@@ -999,10 +1026,10 @@ static int __igt_reset_engines(struct intel_gt *gt,if (rq) {if (rq->fence.error != -EIO ) {pr_err ("i915_reset_engine(%s:%s):" -" failed to reset request %llx:%lld\n" , +" failed to reset request %lld:%lld [0x%04X]\n" , engine->name, test_name, rq->fence.context, - rq->fence.seqno); + rq->fence.seqno, rq->context->guc_id);i915_request_put (rq);GEM_TRACE_DUMP ();@@ -1101,8 +1128,10 @@ static int __igt_reset_engines(struct intel_gt *gt,break ; err =igt_flush_test (gt->i915); -if (err) +if (err) { +pr_err ("[%s] Flush failed: %d!\n" , engine->name, err);break ; + } }if (intel_gt_is_wedged (gt))@@ -1180,12 +1209,15 @@ static int igt_reset_wait(void *arg)igt_global_reset_lock (gt); err =hang_init (&h, gt); -if (err) +if (err) { +pr_err ("[%s] Hang init failed: %d!\n" , engine->name, err); goto unlock; + } rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto fini; }@@ -1310,12 +1342,15 @@ static int __igt_reset_evict_vma(struct intel_gt *gt,/* Check that we can recover an unbind stuck on a hanging request */ err =hang_init (&h, gt); -if (err) +if (err) { +pr_err ("[%s] Hang init failed: %d!\n" , engine->name, err);return err; + } obj =i915_gem_object_create_internal (gt->i915,SZ_1M );if (IS_ERR (obj)) { err =PTR_ERR (obj); +pr_err ("[%s] Create object failed: %d!\n" , engine->name, err); goto fini; }@@ -1330,12 +1365,14 @@ static int __igt_reset_evict_vma(struct intel_gt *gt, arg.vma =i915_vma_instance (obj, vm,NULL );if (IS_ERR (arg.vma)) { err =PTR_ERR (arg.vma); +pr_err ("[%s] VMA instance failed: %d!\n" , engine->name, err); goto out_obj; } rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto out_obj; }@@ -1347,6 +1384,7 @@ static int __igt_reset_evict_vma(struct intel_gt *gt, err =i915_vma_pin (arg.vma,0 ,0 , pin_flags);if (err) {i915_request_add (rq); +pr_err ("[%s] VMA pin failed: %d!\n" , engine->name, err); goto out_obj; }@@ -1363,8 +1401,14 @@ static int __igt_reset_evict_vma(struct intel_gt *gt,i915_vma_lock (arg.vma); err =i915_request_await_object (rq, arg.vma->obj, flags &EXEC_OBJECT_WRITE ); -if (err ==0 ) +if (err ==0 ) { err =i915_vma_move_to_active (arg.vma, rq, flags); +if (err) +pr_err ("[%s] Move to active failed: %d!\n" , engine->name, err); + }else { +pr_err ("[%s] Request await failed: %d!\n" , engine->name, err); + } +i915_vma_unlock (arg.vma);if (flags &EXEC_OBJECT_NEEDS_FENCE )@@ -1392,6 +1436,7 @@ static int __igt_reset_evict_vma(struct intel_gt *gt, tsk =kthread_run (fn, &arg,"igt/evict_vma" );if (IS_ERR (tsk)) { err =PTR_ERR (tsk); +pr_err ("[%s] Thread spawn failed: %d!\n" , engine->name, err); tsk =NULL ; goto out_reset; }@@ -1518,6 +1563,7 @@ static int igt_reset_queue(void *arg) prev =hang_create_request (&h, engine);if (IS_ERR (prev)) { err =PTR_ERR (prev); +pr_err ("[%s] Create 'prev' hang request failed: %d!\n" , engine->name, err); goto fini; }@@ -1532,6 +1578,7 @@ static int igt_reset_queue(void *arg) rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto fini; }@@ -1619,8 +1666,10 @@ static int igt_reset_queue(void *arg)i915_request_put (prev); err =igt_flush_test (gt->i915); -if (err) +if (err) { +pr_err ("[%s] Flush failed: %d!\n" , engine->name, err);break ; + } } fini:@@ -1653,12 +1702,15 @@ static int igt_handle_error(void *arg)return 0 ; err =hang_init (&h, gt); -if (err) +if (err) { +pr_err ("[%s] Hang init failed: %d!\n" , engine->name, err);return err; + } rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto err_fini; }@@ -1743,12 +1795,15 @@ static int igt_atomic_reset_engine(struct intel_engine_cs *engine,return err; err =hang_init (&h, engine->gt); -if (err) +if (err) { +pr_err ("[%s] Hang init failed: %d!\n" , engine->name, err);return err; + } rq =hang_create_request (&h, engine);if (IS_ERR (rq)) { err =PTR_ERR (rq); +pr_err ("[%s] Create hang request failed: %d!\n" , engine->name, err); goto out; }Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 45/51 drm/i915/selftest: Fix workarounds selftest for GuC submission
In reply to Matthew BrostFrom: Rahul Kumar Singh <[email protected]>
When GuC submission is enabled, the GuC controls engine resets. Rather than explicitly triggering a reset, the driver must submit a hanging context to GuC and wait for the reset to occur.
--- drivers/gpu/drm/i915/Makefile | 1 + .../gpu/drm/i915/gt/selftest_workarounds.c | 130 +++++++++++++----- .../i915/selftests/intel_scheduler_helpers.c | 76 ++++++++++ .../i915/selftests/intel_scheduler_helpers.h | 28 ++++ 4 files changed, 201 insertions(+), 34 deletions(-) create mode 100644 drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.c create mode 100644 drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.h
drivers/gpu/drm/i915/Makefile +1 -0
@@ -280,6 +280,7 @@ i915-$(CONFIG_DRM_I915_CAPTURE_ERROR) += i915_gpu_error.o i915-$(CONFIG_DRM_I915_SELFTEST) += \ gem/selftests/i915_gem_client_blt.o \ gem/selftests/igt_gem_utils.o \ + selftests/intel_scheduler_helpers.o \ selftests/i915_random.o \ selftests/i915_selftest.o \ selftests/igt_atomic.o \
drivers/gpu/drm/i915/gt/selftest_workarounds.c +96 -34
214 lines not on this page · raw message
drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.c new file +76 -0
76 lines not on this page · raw message
drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.h new file +28 -0
@@ -0,0 +1,28 @@ +/* SPDX-License-Identifier: MIT */ +/* +* Copyright © 2014-2019 Intel Corporation +*/ + +#ifndef _INTEL_SELFTEST_SCHEDULER_HELPERS_H_ +#define _INTEL_SELFTEST_SCHEDULER_HELPERS_H_ + +#include <linux/types.h> + +struct i915_request ; +struct intel_engine_cs ; + +struct intel_selftest_saved_policy +{ +u32 flags; +u32 reset; +u64 timeslice; +u64 preempt_timeout; +}; + +int intel_selftest_modify_policy (struct intel_engine_cs *engine, +struct intel_selftest_saved_policy *saved); +int intel_selftest_restore_policy (struct intel_engine_cs *engine, +struct intel_selftest_saved_policy *saved); +int intel_selftest_wait_for_rq (struct i915_request *rq); + +#endif
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 45/51] drm/i915/selftest: Fix workarounds selftest for GuC submission
Quoted 11 lines · Matthew Brost
From: Rahul Kumar Singh <[email protected]>
When GuC submission is enabled, the GuC controls engine resets. Rather than explicitly triggering a reset, the driver must submit a hanging context to GuC and wait for the reset to occur.
Signed-off-by Rahul Kumar Singh <[email protected]>Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>Cc Daniele Ceraolo Spurio <[email protected]>Cc Matthew Brost <[email protected]>
Quoted 369 lines
--- drivers/gpu/drm/i915/Makefile | 1 + .../gpu/drm/i915/gt/selftest_workarounds.c | 130 +++++++++++++----- .../i915/selftests/intel_scheduler_helpers.c | 76 ++++++++++ .../i915/selftests/intel_scheduler_helpers.h | 28 ++++ 4 files changed, 201 insertions(+), 34 deletions(-) create mode 100644 drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.c create mode 100644 drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.hdrivers/gpu/drm/i915/Makefile +1
-0@@ -280,6 +280,7 @@ i915-$(CONFIG_DRM_I915_CAPTURE_ERROR) += i915_gpu_error.o i915-$(CONFIG_DRM_I915_SELFTEST) += \ gem/selftests/i915_gem_client_blt.o \ gem/selftests/igt_gem_utils.o \ + selftests/intel_scheduler_helpers.o \ selftests/i915_random.o \ selftests/i915_selftest.o \ selftests/igt_atomic.o \drivers/gpu/drm/i915/gt/selftest_workarounds.c +96
-34@@ -12,6 +12,7 @@#include "selftests/igt_flush_test.h" #include "selftests/igt_reset.h" #include "selftests/igt_spinner.h" +#include "selftests/intel_scheduler_helpers.h" #include "selftests/mock_drm.h" #include "gem/selftests/igt_gem_utils.h" @@ -261,28 +262,34 @@ static int do_engine_reset(struct intel_engine_cs *engine)return intel_engine_reset (engine,"live_workarounds" ); } +static int do_guc_reset (struct intel_engine_cs *engine) +{ +/* Currently a no-op as the reset is handled by GuC */ +return 0 ; +} +static int switch_to_scratch_context (struct intel_engine_cs *engine, -struct igt_spinner *spin) +struct igt_spinner *spin, +struct i915_request **rq) {struct intel_context *ce; -struct i915_request *rq;int err =0 ; ce =intel_context_create (engine);if (IS_ERR (ce))return PTR_ERR (ce); - rq =igt_spinner_create_request (spin, ce,MI_NOOP ); + *rq =igt_spinner_create_request (spin, ce,MI_NOOP );intel_context_put (ce); -if (IS_ERR (rq)) { +if (IS_ERR (*rq)) { spin =NULL ; - err =PTR_ERR (rq); + err =PTR_ERR (*rq); goto err; } - err =request_add_spin (rq, spin); + err =request_add_spin (*rq, spin); err:if (err && spin)igt_spinner_end (spin);@@ -296,6 +303,7 @@ static int check_whitelist_across_reset(struct intel_engine_cs *engine, {struct intel_context *ce, *tmp;struct igt_spinner spin; +struct i915_request *rq;intel_wakeref_t wakeref;int err;@@ -316,13 +324,24 @@ static int check_whitelist_across_reset(struct intel_engine_cs *engine, goto out_spin; } - err =switch_to_scratch_context (engine, &spin); + err =switch_to_scratch_context (engine, &spin, &rq);if (err) goto out_spin; +/* Ensure the spinner hasn't aborted */ +if (i915_request_completed (rq)) { +pr_err ("%s spinner failed to start\n" , name); + err = -ETIMEDOUT ; + goto out_spin; + } +with_intel_runtime_pm (engine->uncore->rpm, wakeref) err = reset(engine); +/* Ensure the reset happens and kills the engine */ +if (err ==0 ) + err =intel_selftest_wait_for_rq (rq); +igt_spinner_end (&spin);if (err) {@@ -787,9 +806,26 @@ static int live_reset_whitelist(void *arg)continue ;if (intel_has_reset_engine (gt)) { - err =check_whitelist_across_reset (engine, - do_engine_reset, -"engine" ); +if (intel_engine_uses_guc (engine)) { +struct intel_selftest_saved_policy saved; +int err2; + + err =intel_selftest_modify_policy (engine, &saved); +if (err) + goto out; + + err =check_whitelist_across_reset (engine, + do_guc_reset, +"guc" ); + + err2 =intel_selftest_restore_policy (engine, &saved); +if (err ==0 ) + err = err2; + }else + err =check_whitelist_across_reset (engine, + do_engine_reset, +"engine" ); +if (err) goto out; }@@ -1226,31 +1262,41 @@ live_engine_reset_workarounds(void *arg)reference_lists_init (gt, &lists);for_each_engine (engine, gt, id) { +struct intel_selftest_saved_policy saved; +bool using_guc =intel_engine_uses_guc (engine);bool ok; +int ret2;pr_info ("Verifying after %s reset...\n" , engine->name); + ret =intel_selftest_modify_policy (engine, &saved); +if (ret) +break ; + + ce =intel_context_create (engine);if (IS_ERR (ce)) { ret =PTR_ERR (ce); -break ; + goto restore; } - ok =verify_wa_lists (gt, &lists,"before reset" ); -if (!ok) { - ret = -ESRCH ; - goto err; - } +if (!using_guc) { + ok =verify_wa_lists (gt, &lists,"before reset" ); +if (!ok) { + ret = -ESRCH ; + goto err; + } - ret =intel_engine_reset (engine,"live_workarounds:idle" ); -if (ret) { -pr_err ("%s: Reset failed while idle\n" , engine->name); - goto err; - } + ret =intel_engine_reset (engine,"live_workarounds:idle" ); +if (ret) { +pr_err ("%s: Reset failed while idle\n" , engine->name); + goto err; + } - ok =verify_wa_lists (gt, &lists,"after idle reset" ); -if (!ok) { - ret = -ESRCH ; - goto err; + ok =verify_wa_lists (gt, &lists,"after idle reset" ); +if (!ok) { + ret = -ESRCH ; + goto err; + } } ret =igt_spinner_init (&spin, engine->gt);@@ -1271,25 +1317,41 @@ live_engine_reset_workarounds(void *arg) goto err; } - ret =intel_engine_reset (engine,"live_workarounds:active" ); -if (ret) { -pr_err ("%s: Reset failed on an active spinner\n" , - engine->name); -igt_spinner_fini (&spin); - goto err; +/* Ensure the spinner hasn't aborted */ +if (i915_request_completed (rq)) { + ret = -ETIMEDOUT ; + goto skip; + } + +if (!using_guc) { + ret =intel_engine_reset (engine,"live_workarounds:active" ); +if (ret) { +pr_err ("%s: Reset failed on an active spinner\n" , + engine->name); +igt_spinner_fini (&spin); + goto err; + } } +/* Ensure the reset happens and kills the engine */ +if (ret ==0 ) + ret =intel_selftest_wait_for_rq (rq); + +skip:igt_spinner_end (&spin);igt_spinner_fini (&spin); ok =verify_wa_lists (gt, &lists,"after busy reset" ); -if (!ok) { +if (!ok) ret = -ESRCH ; - goto err; - } err:intel_context_put (ce); + +restore: + ret2 =intel_selftest_restore_policy (engine, &saved); +if (ret ==0 ) + ret = ret2;if (ret)break ; }drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.c new file +76
-0@@ -0,0 +1,76 @@ +/* +* SPDX-License-Identifier: MIT +* +* Copyright © 2018 Intel Corporation +*/ + +//#include "gt/intel_engine_user.h" +#include "gt/intel_gt.h" +#include "i915_drv.h" +#include "i915_selftest.h" + +#include "selftests/intel_scheduler_helpers.h" + +#define REDUCED_TIMESLICE 5 +#define REDUCED_PREEMPT 10 +#define WAIT_FOR_RESET_TIME 1000 + +int intel_selftest_modify_policy (struct intel_engine_cs *engine, +struct intel_selftest_saved_policy *saved) + +{ +int err; + + saved->reset = engine->i915->params.reset; + saved->flags = engine->flags; + saved->timeslice = engine->props.timeslice_duration_ms; + saved->preempt_timeout = engine->props.preempt_timeout_ms; + +/* +* Enable force pre-emption on time slice expiration +* together with engine reset on pre-emption timeout. +* This is required to make the GuC notice and reset +* the single hanging context. +* Also, reduce the preemption timeout to something +* small to speed the test up. +*/ + engine->i915->params.reset =2 ; + engine->flags |=I915_ENGINE_WANT_FORCED_PREEMPTION ; + engine->props.timeslice_duration_ms =REDUCED_TIMESLICE ; + engine->props.preempt_timeout_ms =REDUCED_PREEMPT ; + +if (!intel_engine_uses_guc (engine)) +return 0 ; + + err =intel_guc_global_policies_update (&engine->gt->uc.guc); +if (err) +intel_selftest_restore_policy (engine, saved); + +return err; +} + +int intel_selftest_restore_policy (struct intel_engine_cs *engine, +struct intel_selftest_saved_policy *saved) +{ +/* Restore the original policies */ + engine->i915->params.reset = saved->reset; + engine->flags = saved->flags; + engine->props.timeslice_duration_ms = saved->timeslice; + engine->props.preempt_timeout_ms = saved->preempt_timeout; + +if (!intel_engine_uses_guc (engine)) +return 0 ; + +return intel_guc_global_policies_update (&engine->gt->uc.guc); +} + +int intel_selftest_wait_for_rq (struct i915_request *rq) +{ +long ret; + + ret =i915_request_wait (rq,0 ,WAIT_FOR_RESET_TIME ); +if (ret <0 ) +return ret; + +return 0 ; +}drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.h new file +28
-0@@ -0,0 +1,28 @@ +/* SPDX-License-Identifier: MIT */ +/* +* Copyright © 2014-2019 Intel Corporation +*/ + +#ifndef _INTEL_SELFTEST_SCHEDULER_HELPERS_H_ +#define _INTEL_SELFTEST_SCHEDULER_HELPERS_H_ + +#include <linux/types.h> + +struct i915_request ; +struct intel_engine_cs ; + +struct intel_selftest_saved_policy +{ +u32 flags; +u32 reset; +u64 timeslice; +u64 preempt_timeout; +}; + +int intel_selftest_modify_policy (struct intel_engine_cs *engine, +struct intel_selftest_saved_policy *saved); +int intel_selftest_restore_policy (struct intel_engine_cs *engine, +struct intel_selftest_saved_policy *saved); +int intel_selftest_wait_for_rq (struct i915_request *rq); + +#endif Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 46/51 drm/i915/selftest: Fix MOCS selftest for GuC submission
In reply to Matthew BrostFrom: Rahul Kumar Singh <[email protected]>
When GuC submission is enabled, the GuC controls engine resets. Rather than explicitly triggering a reset, the driver must submit a hanging context to GuC and wait for the reset to occur.
--- drivers/gpu/drm/i915/gt/selftest_mocs.c | 49 ++++++++++++++++++------- 1 file changed, 35 insertions(+), 14 deletions(-)
drivers/gpu/drm/i915/gt/selftest_mocs.c +35 -14
@@ -10,6 +10,7 @@#include "gem/selftests/mock_context.h" #include "selftests/igt_reset.h" #include "selftests/igt_spinner.h" +#include "selftests/intel_scheduler_helpers.h" struct live_mocs {struct drm_i915_mocs_table table;
@@ -318,7 +319,8 @@ static int live_mocs_clean(void *arg) }static int active_engine_reset (struct intel_context *ce, -const char *reason) +const char *reason, +bool using_guc) {struct igt_spinner spin;struct i915_request *rq;
@@ -335,9 +337,13 @@ static int active_engine_reset(struct intel_context *ce, } err =request_add_spin (rq, &spin); -if (err ==0 ) +if (err ==0 && !using_guc) err =intel_engine_reset (ce->engine, reason); +/* Ensure the reset happens and kills the engine */ +if (err ==0 ) + err =intel_selftest_wait_for_rq (rq); +igt_spinner_end (&spin);igt_spinner_fini (&spin);
@@ -345,21 +351,23 @@ static int active_engine_reset(struct intel_context *ce, }static int __live_mocs_reset (struct live_mocs *mocs, -struct intel_context *ce) +struct intel_context *ce,bool using_guc) {struct intel_gt *gt = ce->engine->gt;int err;if (intel_has_reset_engine (gt)) { - err =intel_engine_reset (ce->engine,"mocs" ); -if (err) -return err; - - err =check_mocs_engine (mocs, ce); -if (err) -return err; +if (!using_guc) { + err =intel_engine_reset (ce->engine,"mocs" ); +if (err) +return err; + + err =check_mocs_engine (mocs, ce); +if (err) +return err; + } - err =active_engine_reset (ce,"mocs" ); + err =active_engine_reset (ce,"mocs" , using_guc);if (err)return err;
@@ -395,19 +403,32 @@ static int live_mocs_reset(void *arg)igt_global_reset_lock (gt);for_each_engine (engine, gt, id) { +bool using_guc =intel_engine_uses_guc (engine); +struct intel_selftest_saved_policy saved;struct intel_context *ce; +int err2; + + err =intel_selftest_modify_policy (engine, &saved); +if (err) +break ; ce =mocs_context_create (engine);if (IS_ERR (ce)) { err =PTR_ERR (ce); -break ; + goto restore; }intel_engine_pm_get (engine); - err =__live_mocs_reset (&mocs, ce); -intel_engine_pm_put (engine); + err =__live_mocs_reset (&mocs, ce, using_guc); + +intel_engine_pm_put (engine);intel_context_put (ce); + +restore: + err2 =intel_selftest_restore_policy (engine, &saved); +if (err ==0 ) + err = err2;if (err)break ; }
Signature · 1 line
2.28.0
Re: [PATCH 46/51] drm/i915/selftest: Fix MOCS selftest for GuC submission
Quoted 11 lines · Matthew Brost
From: Rahul Kumar Singh <[email protected]>
When GuC submission is enabled, the GuC controls engine resets. Rather than explicitly triggering a reset, the driver must submit a hanging context to GuC and wait for the reset to occur.
Signed-off-by Rahul Kumar Singh <[email protected]>Signed-off-by John Harrison <[email protected]>Signed-off-by Matthew Brost <[email protected]>Cc Daniele Ceraolo Spurio <[email protected]>Cc Matthew Brost <[email protected]>
Quoted 113 lines
--- drivers/gpu/drm/i915/gt/selftest_mocs.c | 49 ++++++++++++++++++------- 1 file changed, 35 insertions(+), 14 deletions(-)drivers/gpu/drm/i915/gt/selftest_mocs.c +35
-14@@ -10,6 +10,7 @@#include "gem/selftests/mock_context.h" #include "selftests/igt_reset.h" #include "selftests/igt_spinner.h" +#include "selftests/intel_scheduler_helpers.h" struct live_mocs {struct drm_i915_mocs_table table;@@ -318,7 +319,8 @@ static int live_mocs_clean(void *arg) }static int active_engine_reset (struct intel_context *ce, -const char *reason) +const char *reason, +bool using_guc) {struct igt_spinner spin;struct i915_request *rq;@@ -335,9 +337,13 @@ static int active_engine_reset(struct intel_context *ce, } err =request_add_spin (rq, &spin); -if (err ==0 ) +if (err ==0 && !using_guc) err =intel_engine_reset (ce->engine, reason); +/* Ensure the reset happens and kills the engine */ +if (err ==0 ) + err =intel_selftest_wait_for_rq (rq); +igt_spinner_end (&spin);igt_spinner_fini (&spin);@@ -345,21 +351,23 @@ static int active_engine_reset(struct intel_context *ce, }static int __live_mocs_reset (struct live_mocs *mocs, -struct intel_context *ce) +struct intel_context *ce,bool using_guc) {struct intel_gt *gt = ce->engine->gt;int err;if (intel_has_reset_engine (gt)) { - err =intel_engine_reset (ce->engine,"mocs" ); -if (err) -return err; - - err =check_mocs_engine (mocs, ce); -if (err) -return err; +if (!using_guc) { + err =intel_engine_reset (ce->engine,"mocs" ); +if (err) +return err; + + err =check_mocs_engine (mocs, ce); +if (err) +return err; + } - err =active_engine_reset (ce,"mocs" ); + err =active_engine_reset (ce,"mocs" , using_guc);if (err)return err;@@ -395,19 +403,32 @@ static int live_mocs_reset(void *arg)igt_global_reset_lock (gt);for_each_engine (engine, gt, id) { +bool using_guc =intel_engine_uses_guc (engine); +struct intel_selftest_saved_policy saved;struct intel_context *ce; +int err2; + + err =intel_selftest_modify_policy (engine, &saved); +if (err) +break ; ce =mocs_context_create (engine);if (IS_ERR (ce)) { err =PTR_ERR (ce); -break ; + goto restore; }intel_engine_pm_get (engine); - err =__live_mocs_reset (&mocs, ce); -intel_engine_pm_put (engine); + err =__live_mocs_reset (&mocs, ce, using_guc); + +intel_engine_pm_put (engine);intel_context_put (ce); + +restore: + err2 =intel_selftest_restore_policy (engine, &saved); +if (err ==0 ) + err = err2;if (err)break ; }Signature · 1 line
2.28.0
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 47/51 drm/i915/selftest: Increase some timeouts in live_requests
In reply to Matthew BrostRequests may take slightly longer with GuC submission, let's increase the timeouts in live_requests.
--- drivers/gpu/drm/i915/selftests/i915_request.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/selftests/i915_request.c +2 -2
@@ -1313,7 +1313,7 @@ static int __live_parallel_engine1(void *arg)i915_request_add (rq); err =0 ; -if (i915_request_wait (rq,0 ,HZ /5 ) <0 ) +if (i915_request_wait (rq,0 ,HZ ) <0 ) err = -ETIME ;i915_request_put (rq);if (err)
@@ -1419,7 +1419,7 @@ static int __live_parallel_spin(void *arg) }igt_spinner_end (&spin); -if (err ==0 &&i915_request_wait (rq,0 ,HZ /5 ) <0 ) +if (err ==0 &&i915_request_wait (rq,0 ,HZ ) <0 ) err = -EIO ;i915_request_put (rq);
Signature · 6 lines
2.28.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 47/51] drm/i915/selftest: Increase some timeouts in live_requests
Quoted 4 lines · Matthew Brost
Requests may take slightly longer with GuC submission, let's increase the timeouts in
live_requests.Signed-off-by Matthew Brost <[email protected]>
Quoted 18 lines
--- drivers/gpu/drm/i915/selftests/i915_request.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)drivers/gpu/drm/i915/selftests/i915_request.c +1
-1@@ -1313,7 +1313,7 @@ static int __live_parallel_engine1(void *arg)i915_request_add (rq); err =0 ; -if (i915_request_wait (rq,0 ,HZ /5 ) <0 ) +if (i915_request_wait (rq,0 ,HZ ) <0 ) err = -ETIME ;i915_request_put (rq);if (err)
drivers/gpu/drm/i915/selftests/i915_request.c static int __live_parallel_spin(void *arg) · line 1425
Quoted 8 lines
drivers/gpu/drm/i915/selftests/i915_request.c +1
-1@@ -1419,7 +1419,7 @@ static int __live_parallel_spin(void *arg) }igt_spinner_end (&spin); -if (err ==0 &&i915_request_wait (rq,0 ,HZ /5 ) <0 ) +if (err ==0 &&i915_request_wait (rq,0 ,HZ ) <0 ) err = -EIO ;i915_request_put (rq);
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 47/51] drm/i915/selftest: Increase some timeouts in live_requests
In reply to Matthew BrostQuoted 2 lines · Matthew Brost
Requests may take slightly longer with GuC submission, let's increase the timeouts in
live_requests.
Hm "slightly" ends up 5x longer here and one second feels a lot.
Out of curiosity, given this is about a simple submit of a no-op batches in a tight loop, where does the huge amount of extra latency come from? Is it single H2G channel shared for all engines, the firmware, or GuC speed itself?
Regards,
Tvrtko
Quoted 20 lines
Signed-off-by Matthew Brost <[email protected]>--- drivers/gpu/drm/i915/selftests/i915_request.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)drivers/gpu/drm/i915/selftests/i915_request.c +1
-1@@ -1313,7 +1313,7 @@ static int __live_parallel_engine1(void *arg)i915_request_add (rq); err =0 ; -if (i915_request_wait (rq,0 ,HZ /5 ) <0 ) +if (i915_request_wait (rq,0 ,HZ ) <0 ) err = -ETIME ;i915_request_put (rq);if (err)
drivers/gpu/drm/i915/selftests/i915_request.c static int __live_parallel_spin(void *arg) · line 1425
Quoted 8 lines
drivers/gpu/drm/i915/selftests/i915_request.c +1
-1@@ -1419,7 +1419,7 @@ static int __live_parallel_spin(void *arg) }igt_spinner_end (&spin); -if (err ==0 &&i915_request_wait (rq,0 ,HZ /5 ) <0 ) +if (err ==0 &&i915_request_wait (rq,0 ,HZ ) <0 ) err = -EIO ;i915_request_put (rq);
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 48/51 drm/i915/selftest: Fix hangcheck self test for GuC submission
In reply to Matthew BrostFrom: John Harrison <[email protected]>
When GuC submission is enabled, the GuC controls engine resets. Rather than explicitly triggering a reset, the driver must submit a hanging context to GuC and wait for the reset to occur.
Conversely, one of the tests specifically sends hanging batches to the engines but wants them to sit around until a manual reset of the full GT (including GuC itself). That means disabling GuC based engine resets to prevent those from killing the hanging batch too soon. So, add support to the scheduling policy helper for disabling resets as well as making them quicker!
In GuC submission mode, the 'is engine idle' test basically turns into 'is engine PM wakelock held'. Independently, there is a heartbeat disable helper function that the tests use. For unexplained reasons, this acquires the engine wakelock before disabling the heartbeat and only releases it when re-enabling the heartbeat. As one of the tests tries to do a wait for idle in the middle of a heartbeat disabled section, it is therefore guaranteed to always fail. Added a 'no_pm' variant of the heartbeat helper that allows the engine to be asleep while also having heartbeats disabled.
--- drivers/gpu/drm/i915/gt/intel_engine_types.h | 1 + .../drm/i915/gt/selftest_engine_heartbeat.c | 22 ++ .../drm/i915/gt/selftest_engine_heartbeat.h | 2 + drivers/gpu/drm/i915/gt/selftest_hangcheck.c | 223 +++++++++++++----- drivers/gpu/drm/i915/gt/selftest_mocs.c | 3 +- .../gpu/drm/i915/gt/selftest_workarounds.c | 6 +- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 3 + .../i915/selftests/intel_scheduler_helpers.c | 39 ++- .../i915/selftests/intel_scheduler_helpers.h | 9 +- 9 files changed, 237 insertions(+), 71 deletions(-)
drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
@@ -449,6 +449,7 @@ struct intel_engine_cs {#define I915_ENGINE_IS_VIRTUAL BIT(5)#define I915_ENGINE_HAS_RELATIVE_MMIO BIT(6)#define I915_ENGINE_REQUIRES_CMD_PARSER BIT(7) +#define I915_ENGINE_WANT_FORCED_PREEMPTION BIT(8)unsigned int flags;/*
drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c +22 -0
@@ -405,3 +405,25 @@ void st_engine_heartbeat_enable(struct intel_engine_cs *engine) engine->props.heartbeat_interval_ms = engine->defaults.heartbeat_interval_ms; } + +void st_engine_heartbeat_disable_no_pm (struct intel_engine_cs *engine) +{ + engine->props.heartbeat_interval_ms =0 ; + +/* +* Park the heartbeat but without holding the PM lock as that +* makes the engines appear not-idle. Note that if/when unpark +* is called due to the PM lock being acquired later the +* heartbeat still won't be enabled because of the above = 0. +*/ +if (intel_engine_pm_get_if_awake (engine)) { +intel_engine_park_heartbeat (engine); +intel_engine_pm_put (engine); + } +} + +void st_engine_heartbeat_enable_no_pm (struct intel_engine_cs *engine) +{ + engine->props.heartbeat_interval_ms = + engine->defaults.heartbeat_interval_ms; +}
drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.h +2 -0
@@ -9,6 +9,8 @@struct intel_engine_cs ;void st_engine_heartbeat_disable (struct intel_engine_cs *engine); +void st_engine_heartbeat_disable_no_pm (struct intel_engine_cs *engine);void st_engine_heartbeat_enable (struct intel_engine_cs *engine); +void st_engine_heartbeat_enable_no_pm (struct intel_engine_cs *engine);#endif /* SELFTEST_ENGINE_HEARTBEAT_H */
drivers/gpu/drm/i915/gt/selftest_hangcheck.c +169 -54
407 lines not on this page · raw message
drivers/gpu/drm/i915/gt/selftest_mocs.c +2 -1
@@ -408,7 +408,8 @@ static int live_mocs_reset(void *arg)struct intel_context *ce;int err2; - err =intel_selftest_modify_policy (engine, &saved); + err =intel_selftest_modify_policy (engine, &saved, +SELFTEST_SCHEDULER_MODIFY_FAST_RESET );if (err)break ;
drivers/gpu/drm/i915/gt/selftest_workarounds.c +4 -2
@@ -810,7 +810,8 @@ static int live_reset_whitelist(void *arg)struct intel_selftest_saved_policy saved;int err2; - err =intel_selftest_modify_policy (engine, &saved); + err =intel_selftest_modify_policy (engine, &saved, +SELFTEST_SCHEDULER_MODIFY_FAST_RESET );if (err) goto out;
@@ -1268,7 +1269,8 @@ live_engine_reset_workarounds(void *arg)int ret2;pr_info ("Verifying after %s reset...\n" , engine->name); - ret =intel_selftest_modify_policy (engine, &saved); + ret =intel_selftest_modify_policy (engine, &saved, +SELFTEST_SCHEDULER_MODIFY_FAST_RESET );if (ret)break ;
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +3 -0
@@ -1235,6 +1235,9 @@ static void guc_context_policy_init(struct intel_engine_cs *engine, { desc->policy_flags =0 ; +if (engine->flags &I915_ENGINE_WANT_FORCED_PREEMPTION ) + desc->policy_flags |=CONTEXT_POLICY_FLAG_PREEMPT_TO_IDLE ; +/* NB: For both of these, zero means disabled. */ desc->execution_quantum = engine->props.timeslice_duration_ms *1000 ; desc->preemption_timeout = engine->props.preempt_timeout_ms *1000 ;
drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.c +26 -13
@@ -16,7 +16,8 @@#define WAIT_FOR_RESET_TIME 1000int intel_selftest_modify_policy (struct intel_engine_cs *engine, -struct intel_selftest_saved_policy *saved) +struct intel_selftest_saved_policy *saved, +u32 modify_type) {int err;
@@ -26,18 +27,30 @@ int intel_selftest_modify_policy(struct intel_engine_cs *engine, saved->timeslice = engine->props.timeslice_duration_ms; saved->preempt_timeout = engine->props.preempt_timeout_ms; -/* -* Enable force pre-emption on time slice expiration -* together with engine reset on pre-emption timeout. -* This is required to make the GuC notice and reset -* the single hanging context. -* Also, reduce the preemption timeout to something -* small to speed the test up. -*/ - engine->i915->params.reset =2 ; - engine->flags |=I915_ENGINE_WANT_FORCED_PREEMPTION ; - engine->props.timeslice_duration_ms =REDUCED_TIMESLICE ; - engine->props.preempt_timeout_ms =REDUCED_PREEMPT ; +switch (modify_type) { +case SELFTEST_SCHEDULER_MODIFY_FAST_RESET : +/* +* Enable force pre-emption on time slice expiration +* together with engine reset on pre-emption timeout. +* This is required to make the GuC notice and reset +* the single hanging context. +* Also, reduce the preemption timeout to something +* small to speed the test up. +*/ + engine->i915->params.reset =2 ; + engine->flags |=I915_ENGINE_WANT_FORCED_PREEMPTION ; + engine->props.timeslice_duration_ms =REDUCED_TIMESLICE ; + engine->props.preempt_timeout_ms =REDUCED_PREEMPT ; +break ; + +case SELFTEST_SCHEDULER_MODIFY_NO_HANGCHECK : + engine->props.preempt_timeout_ms =0 ; +break ; + +default : +pr_err ("Invalid scheduler policy modification type: %d!\n" , modify_type); +return -EINVAL ; + }if (!intel_engine_uses_guc (engine))return 0 ;
drivers/gpu/drm/i915/selftests/intel_scheduler_helpers.h +8 -1
@@ -19,8 +19,15 @@ struct intel_selftest_saved_policyu64 preempt_timeout; }; +enum selftest_scheduler_modify +{ +SELFTEST_SCHEDULER_MODIFY_NO_HANGCHECK =0 , +SELFTEST_SCHEDULER_MODIFY_FAST_RESET , +}; +int intel_selftest_modify_policy (struct intel_engine_cs *engine, -struct intel_selftest_saved_policy *saved); +struct intel_selftest_saved_policy *saved, +enum selftest_scheduler_modify modify_type);int intel_selftest_restore_policy (struct intel_engine_cs *engine,struct intel_selftest_saved_policy *saved);int intel_selftest_wait_for_rq (struct i915_request *rq);
Signature · 1 line
2.28.0