Skip to contentSkip to conversation

benchmarks/gem_wsim: explicit VM create/assign + LR mode

PATCH Superseded Marcin Bernatowicz 5 participants 13 messages Jul 14 – Aug 3

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover benchmarks/gem_wsim: explicit VM create/assign + LR mode

This series extends gem_wsim VM management for Xe.

Patch 1 adds explicit VM create/assign steps in workload descriptors:
V.N to create VM N and v.N.M to bind context M to VM N.

Patch 2 adds c.N to enable compute/LR mode on selected VMs. README is updated with syntax and VM usage examples.

Marcin Bernatowicz (2):
benchmarks/gem_wsim: add explicit VM create/assign steps benchmarks/gem_wsim: add compute/LR mode support for Xe VMs

 benchmarks/gem_wsim.c  | 310 ++++++++++++++++++++++++++++++++++++-----
 benchmarks/wsim/README |  32 +++++
 2 files changed, 311 insertions(+), 31 deletions(-)
Signature · 1 line
2.43.0
#1 lore raw

PATCH 1/2 benchmarks/gem_wsim: add explicit VM create/assign steps

Add explicit VM management to workload descriptors so Xe runs can model VM topology directly.

Introduce two new descriptor steps:
- V.N create VM id N (1-based)
- v.N.M assign VM N to context M

Implementation details:
- Add VM_CREATE and CTX_VM step types and vm_id storage in w_step.
- Parse V/v syntax with validation and clear error messages.
- Add allocate_vms() pre-scan to size and allocate vm_list from V steps.
- Update xe_prepare_contexts() to support explicit VM mode:
- create all explicit VMs
- default contexts to VM 0
- apply per-context VM assignments from v steps
- retain implicit single-VM fallback when no V steps are present
- Make get_vm() use per-context VM linkage.

Documentation:
- Add VM management step documentation into benchmarks/wsim/README.

This keeps i915 behavior unchanged while enabling explicit VM topology control for Xe workloads.

---
 benchmarks/gem_wsim.c  | 143 +++++++++++++++++++++++++++++++++++++----
 benchmarks/wsim/README |  19 ++++++
 2 files changed, 150 insertions(+), 12 deletions(-)
benchmarks/gem_wsim.c +131 -12
index 8be44d0c8..b3c46772f 100644
@@ -130,6 +130,8 @@ enum w_type {
 	TERMINATE,
 	SSEU,
 	WORKINGSET,
+	VM_CREATE,
+	CTX_VM,
 };
 
 struct dep_entry {
@@ -210,6 +212,7 @@ struct w_step {
 		struct bond bond;
 		int sseu;
 		struct working_set working_set;
+		int vm_id;
 	};
 
 	/* Implementation details */
@@ -241,6 +244,7 @@ struct w_step {
 
 struct vm {
 	uint32_t id;
+	bool declared;
 	bool compute_mode;
 	uint64_t ahnd;
 };
@@ -1240,6 +1244,55 @@ parse_workload(struct w_arg *arg, unsigned int flags, double scale_dur,
 
 				step.type = WORKINGSET;
 				goto add_step;
+			} else if (!strcmp(field, "V")) {
+				/*
+				 * V.N  - create VM with id N (xe only, ignored on i915)
+				 * VM ids are 1-based in the descriptor.
+				 */
+				field = strtok_r(fstart, ".", &fctx);
+				check_arg(!field, "Missing VM id at step %u!\n",
+					  nr_steps);
+				tmp = atoi(field);
+				check_arg(tmp <= 0, "Invalid VM id at step %u!\n",
+					  nr_steps);
+				step.vm_id = tmp;
+				step.type = VM_CREATE;
+				goto add_step;
+			} else if (!strcmp(field, "v")) {
+				/*
+				 * v.N.M  - assign VM N to ctx M
+				 * (xe only, ignored on i915)
+				 */
+				unsigned int nr = 0;
+				int vm = 0, ctx_id = 0;
+
+				while ((field = strtok_r(fstart, ".", &fctx))) {
+					fstart = NULL;
+					tmp = atoi(field);
+					if (nr == 0) {
+						check_arg(tmp <= 0,
+							  "Invalid VM id at step %u!\n",
+							  nr_steps);
+						vm = tmp;
+					} else if (nr == 1) {
+						check_arg(tmp <= 0,
+							  "Invalid ctx id at step %u!\n",
+							  nr_steps);
+						ctx_id = tmp;
+					} else {
+						check_arg(1,
+							  "Too many fields in v step at step %u!\n",
+							  nr_steps);
+					}
+					nr++;
+				}
+				check_arg(nr != 2,
+					  "v step requires VM id and ctx id at step %u!\n",
+					  nr_steps);
+				step.vm_id = vm;
+				step.context = ctx_id;
+				step.type = CTX_VM;
+				goto add_step;
 			}
 
 			if (!field) {
@@ -1665,6 +1718,14 @@ xe_get_eq(struct workload *wrk, const struct w_step *w)
 static struct vm *
 get_vm(struct workload *wrk, const struct w_step *w)
 {
+	struct ctx *ctx = __get_ctx(wrk, w);
+
+	/* If the ctx has been linked to an explicit VM, use it; otherwise
+	 * fall back to the implicit single vm_list[0].
+	 */
+	if (ctx->vm)
+		return ctx->vm;
+
 	return wrk->vm_list;
 }
 
@@ -2087,6 +2148,33 @@ static void xe_exec_queue_create_(struct ctx *ctx, struct xe_exec_queue *eq)
 	eq->id = create.exec_queue_id;
 }
 
+static void allocate_vms(struct workload *wrk)
+{
+	int max_vm = 0;
+	struct w_step *w;
+
+	/*
+	 * Slot 0 is always the implicit/default VM. Descriptor VM ids are
+	 * 1-based and map directly to vm_list[N] for explicit V.N steps.
+	 */
+	for_each_w_step(w, wrk) {
+		if (w->type != VM_CREATE)
+			continue;
+		if (w->vm_id > max_vm)
+			max_vm = w->vm_id;
+	}
+
+	wrk->nr_vms = max_vm + 1;
+	wrk->vm_list = calloc(wrk->nr_vms, sizeof(*wrk->vm_list));
+	igt_assert(wrk->vm_list);
+
+	for_each_w_step(w, wrk) {
+		if (w->type == VM_CREATE)
+			/* Mark only explicitly declared VMs; slot 0 stays implicit. */
+			wrk->vm_list[w->vm_id].declared = true;
+	}
+}
+
 static void allocate_contexts(unsigned int id, struct workload *wrk)
 {
 	int max_ctx = -1;
@@ -2094,13 +2182,19 @@ static void allocate_contexts(unsigned int id, struct workload *wrk)
 
 	/*
 	 * Pre-scan workload steps to allocate context list storage.
+	 * Skip VM management steps whose context field is unused.
 	 */
 	for_each_w_step(w, wrk) {
-		int ctx = w->context + 1;
+		int ctx;
 		int delta;
 
 		w->wrk = wrk;
 
+		if (w->type == VM_CREATE)
+			continue;
+
+		ctx = w->context + 1;
+
 		if (ctx <= max_ctx)
 			continue;
 
@@ -2333,18 +2427,38 @@ static int xe_prepare_contexts(unsigned int id, struct workload *wrk)
 	struct ctx *ctx;
 	unsigned int i;
 
-	/* shortcut, create one vm */
-	wrk->nr_vms = 1;
-	wrk->vm_list = calloc(wrk->nr_vms, sizeof(struct vm));
-	igt_assert(wrk->vm_list);
-	wrk->vm_list->compute_mode = false;
-	xe_vm_create_(wrk->vm_list);
-	wrk->vm_list->ahnd = intel_allocator_open(fd, wrk->vm_list->id,
-						  INTEL_ALLOCATOR_RELOC);
+	/* Slot 0 is the implicit/default VM. */
+	xe_vm_create_(&wrk->vm_list[0]);
+	wrk->vm_list[0].ahnd = intel_allocator_open(fd, wrk->vm_list[0].id,
+						    INTEL_ALLOCATOR_RELOC);
+
+	for (i = 1; i < wrk->nr_vms; i++) {
+		if (!wrk->vm_list[i].declared)
+			continue;
+
+		xe_vm_create_(&wrk->vm_list[i]);
+		wrk->vm_list[i].ahnd =
+			intel_allocator_open(fd, wrk->vm_list[i].id,
+					     INTEL_ALLOCATOR_RELOC);
+	}
+
+	/*
+	 * Process CTX_VM steps to link each ctx to an explicit VM. Contexts
+	 * without a v.N.M mapping remain on the implicit VM in slot 0.
+	 */
+	__for_each_ctx(ctx, wrk, ctx_idx)
+		ctx->vm = &wrk->vm_list[0];
+
+	for_each_w_step(w, wrk) {
+		if (w->type != CTX_VM)
+			continue;
+		igt_assert_lt((unsigned int)w->vm_id, wrk->nr_vms);
+		igt_assert(wrk->vm_list[w->vm_id].declared);
+		igt_assert_lt((unsigned int)w->context, wrk->nr_ctxs);
+		wrk->ctx_list[w->context].vm = &wrk->vm_list[w->vm_id];
+	}
 
 	__for_each_ctx(ctx, wrk, ctx_idx) {
-		/* link with vm */
-		ctx->vm = wrk->vm_list;
 		for_each_w_step(w, wrk) {
 			if (w->context != ctx_idx)
 				continue;
@@ -2537,6 +2651,9 @@ static int prepare_workload(unsigned int id, struct workload *wrk)
 
 	allocate_contexts(id, wrk);
 
+	if (is_xe)
+		allocate_vms(wrk);
+
 	if (is_xe)
 		ret = xe_prepare_contexts(id, wrk);
 	else
@@ -2857,7 +2974,9 @@ static void *run_workload(void *data)
 				   w->type == ENGINE_MAP ||
 				   w->type == LOAD_BALANCE ||
 				   w->type == BOND ||
-				   w->type == WORKINGSET) {
+				   w->type == WORKINGSET ||
+				   w->type == VM_CREATE ||
+				   w->type == CTX_VM) {
 				   /* No action for these at execution time. */
 				continue;
 			}
benchmarks/wsim/README +19 -0
index e43813335..909c1f039 100644
@@ -11,6 +11,8 @@ P|S|X.<uint>.<int>
 d|p|s|t|q|a|T.<int>,...
 b.<uint>.<str>[|<str>].<str>
 w|W.<uint>.<str>[/<str>]...
+V.<uint>
+v.<uint>.<uint>
 f
 
 For duration a range can be given from which a random value will be picked
@@ -39,6 +41,11 @@ Additional workload steps are also supported:
  'W' - Shared working set.
  'X' - Context preemption control.
 
+VM management steps (Xe: active; i915: ignored):
+
+ 'V' - Create VM with id N (V.N, VM ids are 1-based).
+ 'v' - Assign VM N to context/queue M (v.N.M).
+
 Engine ids: DEFAULT, RCS, BCS, VCS, VCS1, VCS2, VECS
 
 Example (leading spaces must not be present in the actual file):
@@ -88,6 +95,18 @@ Batch durations can also be specified as infinite by using the '*' in the
 duration field. Such batches must be ended by the terminate command ('T')
 otherwise they will cause a GPU hang to be reported.
 
+VM Example:
+
+  V.1
+  M.1.VCS
+  v.1.1
+  B.1
+  1.VCS.1000.0.0
+
+Creates VM 1, configures context 1 with a VCS engine map, assigns context 1
+to VM 1, enables load balancing for context 1, and runs a single 1ms VCS
+batch.
+
 Xe and i915 differences
 ------------------------
 
Signature · 1 line
2.43.0
#2 lore raw

Re: [PATCH i-g-t 1/2] benchmarks/gem_wsim: add explicit VM create/assign steps

Hi Marcin,

Quoted 1 line · Marcin Bernatowicz

Add explicit VM management to workload descriptors so Xe runs can model

Please write explicitly what VM means, my guess: Virtual Memory

Quoted 17 lines

VM topology directly.

Introduce two new descriptor steps:
- V.N create VM id N (1-based)
- v.N.M assign VM N to context M

Implementation details:
- Add VM_CREATE and CTX_VM step types and vm_id storage in w_step.
- Parse V/v syntax with validation and clear error messages.
- Add allocate_vms() pre-scan to size and allocate vm_list from V steps.
- Update xe_prepare_contexts() to support explicit VM mode:
- create all explicit VMs
- default contexts to VM 0
- apply per-context VM assignments from v steps
- retain implicit single-VM fallback when no V steps are present
- Make get_vm() use per-context VM linkage.

Quoted 2 lines

Documentation:
- Add VM management step documentation into benchmarks/wsim/README.

Same in documentation, add there at least once a full name for VM.

Regards,
Kamil

Quoted 14 lines

This keeps i915 behavior unchanged while enabling explicit VM topology control for Xe workloads.

---
 benchmarks/gem_wsim.c  | 143 +++++++++++++++++++++++++++++++++++++----
 benchmarks/wsim/README |  19 ++++++
 2 files changed, 150 insertions(+), 12 deletions(-)

[cut]

Reviewed-by on v1 1/2
#3 lore raw

Re: [PATCH i-g-t 1/2] benchmarks/gem_wsim: add explicit VM create/assign steps

In reply to Marcin Bernatowicz
Quoted 293 lines · Marcin Bernatowicz

Add explicit VM management to workload descriptors so Xe runs can model VM topology directly.

Introduce two new descriptor steps:
- V.N create VM id N (1-based)
- v.N.M assign VM N to context M

Implementation details:
- Add VM_CREATE and CTX_VM step types and vm_id storage in w_step.
- Parse V/v syntax with validation and clear error messages.
- Add allocate_vms() pre-scan to size and allocate vm_list from V steps.
- Update xe_prepare_contexts() to support explicit VM mode:
- create all explicit VMs
- default contexts to VM 0
- apply per-context VM assignments from v steps
- retain implicit single-VM fallback when no V steps are present
- Make get_vm() use per-context VM linkage.

Documentation:
- Add VM management step documentation into benchmarks/wsim/README.

This keeps i915 behavior unchanged while enabling explicit VM topology control for Xe workloads.

---
  benchmarks/gem_wsim.c  | 143 +++++++++++++++++++++++++++++++++++++----
  benchmarks/wsim/README |  19 ++++++
  2 files changed, 150 insertions(+), 12 deletions(-)
benchmarks/gem_wsim.c +131 -12
index 8be44d0c8..b3c46772f 100644
@@ -130,6 +130,8 @@ enum w_type {
  	TERMINATE,
  	SSEU,
  	WORKINGSET,
+	VM_CREATE,
+	CTX_VM,
  };
 
  struct dep_entry {
@@ -210,6 +212,7 @@ struct w_step {
  		struct bond bond;
  		int sseu;
  		struct working_set working_set;
+		int vm_id;
  	};
 
  	/* Implementation details */
@@ -241,6 +244,7 @@ struct w_step {
 
  struct vm {
  	uint32_t id;
+	bool declared;
  	bool compute_mode;
  	uint64_t ahnd;
  };
@@ -1240,6 +1244,55 @@ parse_workload(struct w_arg *arg, unsigned int flags, double scale_dur,
 
  				step.type = WORKINGSET;
  				goto add_step;
+			} else if (!strcmp(field, "V")) {
+				/*
+				 * V.N  - create VM with id N (xe only, ignored on i915)
+				 * VM ids are 1-based in the descriptor.
+				 */
+				field = strtok_r(fstart, ".", &fctx);
+				check_arg(!field, "Missing VM id at step %u!\n",
+					  nr_steps);
+				tmp = atoi(field);
+				check_arg(tmp <= 0, "Invalid VM id at step %u!\n",
+					  nr_steps);
+				step.vm_id = tmp;
+				step.type = VM_CREATE;
+				goto add_step;
+			} else if (!strcmp(field, "v")) {
+				/*
+				 * v.N.M  - assign VM N to ctx M
+				 * (xe only, ignored on i915)
+				 */
+				unsigned int nr = 0;
+				int vm = 0, ctx_id = 0;
+
+				while ((field = strtok_r(fstart, ".", &fctx))) {
+					fstart = NULL;
+					tmp = atoi(field);
+					if (nr == 0) {
+						check_arg(tmp <= 0,
+							  "Invalid VM id at step %u!\n",
+							  nr_steps);
+						vm = tmp;
+					} else if (nr == 1) {
+						check_arg(tmp <= 0,
+							  "Invalid ctx id at step %u!\n",
+							  nr_steps);
+						ctx_id = tmp;
+					} else {
+						check_arg(1,
+							  "Too many fields in v step at step %u!\n",
+							  nr_steps);
+					}
+					nr++;
+				}
+				check_arg(nr != 2,
+					  "v step requires VM id and ctx id at step %u!\n",
+					  nr_steps);
+				step.vm_id = vm;
+				step.context = ctx_id;
+				step.type = CTX_VM;
+				goto add_step;
  			}
 
  			if (!field) {
@@ -1665,6 +1718,14 @@ xe_get_eq(struct workload *wrk, const struct w_step *w)
  static struct vm *
  get_vm(struct workload *wrk, const struct w_step *w)
  {
+	struct ctx *ctx = __get_ctx(wrk, w);
+
+	/* If the ctx has been linked to an explicit VM, use it; otherwise
+	 * fall back to the implicit single vm_list[0].
+	 */
+	if (ctx->vm)
+		return ctx->vm;
+
  	return wrk->vm_list;
  }
 
@@ -2087,6 +2148,33 @@ static void xe_exec_queue_create_(struct ctx *ctx, struct xe_exec_queue *eq)
  	eq->id = create.exec_queue_id;
  }
 
+static void allocate_vms(struct workload *wrk)
+{
+	int max_vm = 0;
+	struct w_step *w;
+
+	/*
+	 * Slot 0 is always the implicit/default VM. Descriptor VM ids are
+	 * 1-based and map directly to vm_list[N] for explicit V.N steps.
+	 */
+	for_each_w_step(w, wrk) {
+		if (w->type != VM_CREATE)
+			continue;
+		if (w->vm_id > max_vm)
+			max_vm = w->vm_id;
+	}
+
+	wrk->nr_vms = max_vm + 1;
+	wrk->vm_list = calloc(wrk->nr_vms, sizeof(*wrk->vm_list));
+	igt_assert(wrk->vm_list);
+
+	for_each_w_step(w, wrk) {
+		if (w->type == VM_CREATE)
+			/* Mark only explicitly declared VMs; slot 0 stays implicit. */
+			wrk->vm_list[w->vm_id].declared = true;
+	}
+}
+
  static void allocate_contexts(unsigned int id, struct workload *wrk)
  {
  	int max_ctx = -1;
@@ -2094,13 +2182,19 @@ static void allocate_contexts(unsigned int id, struct workload *wrk)
 
  	/*
  	 * Pre-scan workload steps to allocate context list storage.
+	 * Skip VM management steps whose context field is unused.
  	 */
  	for_each_w_step(w, wrk) {
-		int ctx = w->context + 1;
+		int ctx;
  		int delta;
 
  		w->wrk = wrk;
 
+		if (w->type == VM_CREATE)
+			continue;
+
+		ctx = w->context + 1;
+
  		if (ctx <= max_ctx)
  			continue;
 
@@ -2333,18 +2427,38 @@ static int xe_prepare_contexts(unsigned int id, struct workload *wrk)
  	struct ctx *ctx;
  	unsigned int i;
 
-	/* shortcut, create one vm */
-	wrk->nr_vms = 1;
-	wrk->vm_list = calloc(wrk->nr_vms, sizeof(struct vm));
-	igt_assert(wrk->vm_list);
-	wrk->vm_list->compute_mode = false;
-	xe_vm_create_(wrk->vm_list);
-	wrk->vm_list->ahnd = intel_allocator_open(fd, wrk->vm_list->id,
-						  INTEL_ALLOCATOR_RELOC);
+	/* Slot 0 is the implicit/default VM. */
+	xe_vm_create_(&wrk->vm_list[0]);
+	wrk->vm_list[0].ahnd = intel_allocator_open(fd, wrk->vm_list[0].id,
+						    INTEL_ALLOCATOR_RELOC);
+
+	for (i = 1; i < wrk->nr_vms; i++) {
+		if (!wrk->vm_list[i].declared)
+			continue;
+
+		xe_vm_create_(&wrk->vm_list[i]);
+		wrk->vm_list[i].ahnd =
+			intel_allocator_open(fd, wrk->vm_list[i].id,
+					     INTEL_ALLOCATOR_RELOC);
+	}
+
+	/*
+	 * Process CTX_VM steps to link each ctx to an explicit VM. Contexts
+	 * without a v.N.M mapping remain on the implicit VM in slot 0.
+	 */
+	__for_each_ctx(ctx, wrk, ctx_idx)
+		ctx->vm = &wrk->vm_list[0];
+
+	for_each_w_step(w, wrk) {
+		if (w->type != CTX_VM)
+			continue;
+		igt_assert_lt((unsigned int)w->vm_id, wrk->nr_vms);
+		igt_assert(wrk->vm_list[w->vm_id].declared);
+		igt_assert_lt((unsigned int)w->context, wrk->nr_ctxs);
+		wrk->ctx_list[w->context].vm = &wrk->vm_list[w->vm_id];
+	}
 
  	__for_each_ctx(ctx, wrk, ctx_idx) {
-		/* link with vm */
-		ctx->vm = wrk->vm_list;
  		for_each_w_step(w, wrk) {
  			if (w->context != ctx_idx)
  				continue;
@@ -2537,6 +2651,9 @@ static int prepare_workload(unsigned int id, struct workload *wrk)
 
  	allocate_contexts(id, wrk);
 
+	if (is_xe)
+		allocate_vms(wrk);
+
  	if (is_xe)
  		ret = xe_prepare_contexts(id, wrk);
  	else
@@ -2857,7 +2974,9 @@ static void *run_workload(void *data)
  				   w->type == ENGINE_MAP ||
  				   w->type == LOAD_BALANCE ||
  				   w->type == BOND ||
-				   w->type == WORKINGSET) {
+				   w->type == WORKINGSET ||
+				   w->type == VM_CREATE ||
+				   w->type == CTX_VM) {
  				   /* No action for these at execution time. */
  				continue;
  			}
benchmarks/wsim/README +12 -0
index e43813335..909c1f039 100644
@@ -11,6 +11,8 @@ P|S|X.<uint>.<int>
  d|p|s|t|q|a|T.<int>,...
  b.<uint>.<str>[|<str>].<str>
  w|W.<uint>.<str>[/<str>]...
+V.<uint>
+v.<uint>.<uint>
  f
 
  For duration a range can be given from which a random value will be picked
@@ -39,6 +41,11 @@ Additional workload steps are also supported:
   'W' - Shared working set.
   'X' - Context preemption control.
 
+VM management steps (Xe: active; i915: ignored):
+
+ 'V' - Create VM with id N (V.N, VM ids are 1-based).
+ 'v' - Assign VM N to context/queue M (v.N.M).
+
  Engine ids: DEFAULT, RCS, BCS, VCS, VCS1, VCS2, VECS
 
  Example (leading spaces must not be present in the actual file):
@@ -88,6 +95,18 @@ Batch durations can also be specified as infinite by using the '*' in the
  duration field. Such batches must be ended by the terminate command ('T')
  otherwise they will cause a GPU hang to be reported.
 
+VM Example:
+
+  V.1
+  M.1.VCS
+  v.1.1

benchmarks/wsim/README Batch durations can also be specified as infinite by using the '*' in the · line 112

Quoted 10 lines
benchmarks/wsim/README +7 -0
+  B.1
+  1.VCS.1000.0.0
+
+Creates VM 1, configures context 1 with a VCS engine map, assigns context 1
+to VM 1, enables load balancing for context 1, and runs a single 1ms VCS
+batch.
+
  Xe and i915 differences
  ------------------------
 

Agree with Kamil, please at least once reference as VM (Virtual Memory) to avoid potential confusion with virtualization VM (Virtual Machine). Other than that LGTM:

Regards,
Adam

#4 lore raw

PATCH 2/2 benchmarks/gem_wsim: add compute/LR mode support for Xe VMs

In reply to Marcin Bernatowicz

Introduce new descriptor step:
'c' - Enable compute/LR mode for VM N (c.N).

Example:

gem_wsim -w "V.1,c.1,v.1.1,1.BCS.100.0.0,1.CCS.8000000.-1.0,1.BCS.100.-1.1" 8.009s elapsed (0.125 workloads/s)

Creates VM 1, enables compute/LR mode on it, assigns context 1 to the VM, then runs BCS and CCS work with cross-step dependencies (8s CCS depends on the first BCS step, and the final BCS step depends on the CCS step).

---
 benchmarks/gem_wsim.c  | 171 ++++++++++++++++++++++++++++++++++++-----
 benchmarks/wsim/README |  17 +++-
 2 files changed, 165 insertions(+), 23 deletions(-)
benchmarks/gem_wsim.c +150 -21
index b3c46772f..d6638bcdf 100644
@@ -132,6 +132,7 @@ enum w_type {
 	WORKINGSET,
 	VM_CREATE,
 	CTX_VM,
+	VM_COMPUTE_MODE,
 };
 
 struct dep_entry {
@@ -246,6 +247,7 @@ struct vm {
 	uint32_t id;
 	bool declared;
 	bool compute_mode;
+	uint32_t bind_exec_queue;
 	uint64_t ahnd;
 };
 
@@ -323,6 +325,7 @@ static unsigned int master_prng;
 static int verbose = 1;
 static int fd;
 static bool is_xe;
+static const uint64_t user_fence_value = 0xdeadbeefdeadbeefull;
 static struct drm_i915_gem_context_param_sseu device_sseu = {
 	.slice_mask = -1 /* Force read on first use. */
 };
@@ -334,10 +337,18 @@ static struct drm_i915_gem_context_param_sseu device_sseu = {
 
 static void w_step_sync(struct w_step *w)
 {
-	if (is_xe)
-		igt_assert(syncobj_wait(fd, &w->xe.syncs[0].handle, 1, INT64_MAX, 0, NULL));
-	else
+	if (is_xe) {
+		if (w->xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE) {
+			xe_wait_ufence(fd, &w->xe.data->spin.exec_sync,
+				       w->xe.syncs[0].timeline_value, 0,
+				       INT64_MAX);
+		} else {
+			igt_assert(syncobj_wait(fd, &w->xe.syncs[0].handle, 1,
+						INT64_MAX, 0, NULL));
+		}
+	} else {
 		gem_sync(fd, w->i915.obj[0].handle);
+	}
 }
 
 static int read_timestamp_frequency(int i915)
@@ -1293,6 +1304,20 @@ parse_workload(struct w_arg *arg, unsigned int flags, double scale_dur,
 				step.context = ctx_id;
 				step.type = CTX_VM;
 				goto add_step;
+			} else if (!strcmp(field, "c")) {
+				/*
+				 * c.N  - enable compute/LR mode on VM N
+				 * (xe only, ignored on i915)
+				 */
+				field = strtok_r(fstart, ".", &fctx);
+				check_arg(!field,
+					  "Missing VM id at step %u!\n", nr_steps);
+				tmp = atoi(field);
+				check_arg(tmp <= 0,
+					  "Invalid VM id at step %u!\n", nr_steps);
+				step.vm_id = tmp;
+				step.type = VM_COMPUTE_MODE;
+				goto add_step;
 			}
 
 			if (!field) {
@@ -1855,10 +1880,27 @@ xe_alloc_step_batch(struct workload *wrk, struct w_step *w)
 				    vram_if_possible(fd, eq->hwe_list[0].gt_id),
 				    DRM_XE_GEM_CREATE_FLAG_NEEDS_VISIBLE_VRAM);
 	w->xe.data = xe_bo_map(fd, w->bb_handle, w->bb_size);
+	w->xe.data->spin.exec_sync = 0;
+	w->xe.data->vm_sync = 0;
 	w->xe.exec.address =
 		intel_allocator_alloc_with_strategy(vm->ahnd, w->bb_handle, w->bb_size,
 						    0, ALLOC_STRATEGY_LOW_TO_HIGH);
-	xe_vm_bind_sync(fd, vm->id, w->bb_handle, 0, w->xe.exec.address, w->bb_size);
+	if (vm->compute_mode) {
+		struct drm_xe_sync sync = {
+			.type = DRM_XE_SYNC_TYPE_USER_FENCE,
+			.flags = DRM_XE_SYNC_FLAG_SIGNAL,
+			.addr = to_user_pointer(&w->xe.data->vm_sync),
+			.timeline_value = user_fence_value,
+		};
+
+		xe_vm_bind_async(fd, vm->id, vm->bind_exec_queue, w->bb_handle, 0,
+				 w->xe.exec.address, w->bb_size, &sync, 1);
+		xe_wait_ufence(fd, &w->xe.data->vm_sync, user_fence_value,
+			       vm->bind_exec_queue, 10 * NSEC_PER_SEC);
+		w->xe.data->vm_sync = 0;
+	} else {
+		xe_vm_bind_sync(fd, vm->id, w->bb_handle, 0, w->xe.exec.address, w->bb_size);
+	}
 	w->duration.requested_ticks = xe_spin_nsec_to_ticks(fd, eq->hwe_list[0].gt_id,
 							    1000LL * get_duration(wrk, w));
 	xe_spin_init_opts(&w->xe.data->spin,
@@ -1875,8 +1917,15 @@ xe_alloc_step_batch(struct workload *wrk, struct w_step *w)
 
 		igt_assert(dep_idx >= 0 && dep_idx < w->idx);
 		igt_assert(wrk->steps[dep_idx].type == BATCH);
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
 
-		w->xe.exec.num_syncs++;
+		/*
+		 * USER_FENCE is valid as an out-fence in LR mode, but cannot be
+		 * consumed as an in-fence in xe_exec. Those deps are resolved in
+		 * userspace before submission.
+		 */
+		if (wrk->steps[dep_idx].xe.syncs[0].type != DRM_XE_SYNC_TYPE_USER_FENCE)
+			w->xe.exec.num_syncs++;
 	}
 	for_each_dep(dep, w->fence_deps) {
 		int dep_idx = w->idx + dep->target;
@@ -1884,34 +1933,70 @@ xe_alloc_step_batch(struct workload *wrk, struct w_step *w)
 		igt_assert(dep_idx >= 0 && dep_idx < w->idx);
 		igt_assert(wrk->steps[dep_idx].type == SW_FENCE ||
 			   wrk->steps[dep_idx].type == BATCH);
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
 
-		w->xe.exec.num_syncs++;
+		if (wrk->steps[dep_idx].xe.syncs[0].type != DRM_XE_SYNC_TYPE_USER_FENCE)
+			w->xe.exec.num_syncs++;
 	}
 	w->xe.syncs = calloc(w->xe.exec.num_syncs, sizeof(*w->xe.syncs));
 	/* fill syncs */
 	i = 0;
 	/* out fence */
-	w->xe.syncs[i].handle = syncobj_create(fd, 0);
-	w->xe.syncs[i].type = DRM_XE_SYNC_TYPE_SYNCOBJ;
-	w->xe.syncs[i++].flags = DRM_XE_SYNC_FLAG_SIGNAL;
+	if (vm->compute_mode) {
+		w->xe.syncs[i].type = DRM_XE_SYNC_TYPE_USER_FENCE;
+		w->xe.syncs[i].flags = DRM_XE_SYNC_FLAG_SIGNAL;
+		w->xe.syncs[i].addr = w->xe.exec.address + offsetof(struct xe_spin, exec_sync);
+		w->xe.syncs[i++].timeline_value = user_fence_value;
+	} else {
+		w->xe.syncs[i].handle = syncobj_create(fd, 0);
+		w->xe.syncs[i].type = DRM_XE_SYNC_TYPE_SYNCOBJ;
+		w->xe.syncs[i++].flags = DRM_XE_SYNC_FLAG_SIGNAL;
+	}
 	/* in fence(s) */
 	for_each_dep(dep, w->data_deps) {
 		int dep_idx = w->idx + dep->target;
 
-		igt_assert(wrk->steps[dep_idx].xe.syncs && wrk->steps[dep_idx].xe.syncs[0].handle);
-		w->xe.syncs[i].handle = wrk->steps[dep_idx].xe.syncs[0].handle;
-		w->xe.syncs[i++].type = DRM_XE_SYNC_TYPE_SYNCOBJ;
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
+		if (wrk->steps[dep_idx].xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE)
+			continue;
+		w->xe.syncs[i] = wrk->steps[dep_idx].xe.syncs[0];
+		w->xe.syncs[i++].flags &= ~DRM_XE_SYNC_FLAG_SIGNAL;
 	}
 	for_each_dep(dep, w->fence_deps) {
 		int dep_idx = w->idx + dep->target;
 
-		igt_assert(wrk->steps[dep_idx].xe.syncs && wrk->steps[dep_idx].xe.syncs[0].handle);
-		w->xe.syncs[i].handle = wrk->steps[dep_idx].xe.syncs[0].handle;
-		w->xe.syncs[i++].type = DRM_XE_SYNC_TYPE_SYNCOBJ;
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
+		if (wrk->steps[dep_idx].xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE)
+			continue;
+		w->xe.syncs[i] = wrk->steps[dep_idx].xe.syncs[0];
+		w->xe.syncs[i++].flags &= ~DRM_XE_SYNC_FLAG_SIGNAL;
 	}
 	w->xe.exec.syncs = to_user_pointer(w->xe.syncs);
 }
 
+static void xe_sync_user_fence_deps(struct workload *wrk, struct w_step *w)
+{
+	struct dep_entry *dep;
+
+	for_each_dep(dep, w->data_deps) {
+		int dep_idx = w->idx + dep->target;
+
+		igt_assert(dep_idx >= 0 && dep_idx < w->idx);
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
+		if (wrk->steps[dep_idx].xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE)
+			w_step_sync(&wrk->steps[dep_idx]);
+	}
+
+	for_each_dep(dep, w->fence_deps) {
+		int dep_idx = w->idx + dep->target;
+
+		igt_assert(dep_idx >= 0 && dep_idx < w->idx);
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
+		if (wrk->steps[dep_idx].xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE)
+			w_step_sync(&wrk->steps[dep_idx]);
+	}
+}
+
 static bool set_priority(uint32_t ctx_id, int prio)
 {
 	struct drm_i915_gem_context_param param = {
@@ -2190,7 +2275,7 @@ static void allocate_contexts(unsigned int id, struct workload *wrk)
 
 		w->wrk = wrk;
 
-		if (w->type == VM_CREATE)
+		if (w->type == VM_CREATE || w->type == VM_COMPUTE_MODE)
 			continue;
 
 		ctx = w->context + 1;
@@ -2427,6 +2512,14 @@ static int xe_prepare_contexts(unsigned int id, struct workload *wrk)
 	struct ctx *ctx;
 	unsigned int i;
 
+	for_each_w_step(w, wrk) {
+		if (w->type == VM_COMPUTE_MODE) {
+			igt_assert_lt((unsigned int)w->vm_id, wrk->nr_vms);
+			igt_assert(wrk->vm_list[w->vm_id].declared);
+			wrk->vm_list[w->vm_id].compute_mode = true;
+		}
+	}
+
 	/* Slot 0 is the implicit/default VM. */
 	xe_vm_create_(&wrk->vm_list[0]);
 	wrk->vm_list[0].ahnd = intel_allocator_open(fd, wrk->vm_list[0].id,
@@ -2437,6 +2530,9 @@ static int xe_prepare_contexts(unsigned int id, struct workload *wrk)
 			continue;
 
 		xe_vm_create_(&wrk->vm_list[i]);
+		if (wrk->vm_list[i].compute_mode)
+			wrk->vm_list[i].bind_exec_queue =
+				xe_bind_exec_queue_create(fd, wrk->vm_list[i].id, 0);
 		wrk->vm_list[i].ahnd =
 			intel_allocator_open(fd, wrk->vm_list[i].id,
 					     INTEL_ALLOCATOR_RELOC);
@@ -2767,11 +2863,22 @@ static void w_sync_to(struct workload *wrk, struct w_step *w, int target)
 static void do_xe_exec(struct workload *wrk, struct w_step *w)
 {
 	struct xe_exec_queue *eq = xe_get_eq(wrk, w);
+	struct vm *vm = get_vm(wrk, w);
 
 	igt_assert(w->emit_fence <= 0);
 	if (w->emit_fence == -1)
 		syncobj_reset(fd, &w->xe.syncs[0].handle, 1);
 
+	/*
+	 * LR mode reuses the same user-fence location for completion.
+	 * Re-arm it before every submit so waiters do not observe the
+	 * previous iteration's signaled value.
+	 */
+	if (vm->compute_mode)
+		w->xe.data->spin.exec_sync = 0;
+
+	xe_sync_user_fence_deps(wrk, w);
+
 	/* update duration if random */
 	if (w->duration.max != w->duration.min) {
 		w->duration.requested_ticks = xe_spin_nsec_to_ticks(fd, eq->hwe_list[0].gt_id,
@@ -2976,7 +3083,8 @@ static void *run_workload(void *data)
 				   w->type == BOND ||
 				   w->type == WORKINGSET ||
 				   w->type == VM_CREATE ||
-				   w->type == CTX_VM) {
+				   w->type == CTX_VM ||
+				   w->type == VM_COMPUTE_MODE) {
 				   /* No action for these at execution time. */
 				continue;
 			}
@@ -3066,14 +3174,35 @@ static void *run_workload(void *data)
 		for_each_w_step(w, wrk) {
 			if (w->type == BATCH) {
 				w_step_sync(w);
-				syncobj_destroy(fd, w->xe.syncs[0].handle);
+				if (w->xe.syncs[0].type == DRM_XE_SYNC_TYPE_SYNCOBJ)
+					syncobj_destroy(fd, w->xe.syncs[0].handle);
 				free(w->xe.syncs);
-				xe_vm_unbind_sync(fd, get_vm(wrk, w)->id, 0, w->xe.exec.address,
-						  w->bb_size);
+				if (get_vm(wrk, w)->compute_mode) {
+					struct vm *vm = get_vm(wrk, w);
+					struct drm_xe_sync sync = {
+						.type = DRM_XE_SYNC_TYPE_USER_FENCE,
+						.flags = DRM_XE_SYNC_FLAG_SIGNAL,
+						.addr = to_user_pointer(&w->xe.data->vm_sync),
+						.timeline_value = user_fence_value,
+					};
+
+					xe_vm_unbind_async(fd, vm->id, vm->bind_exec_queue,
+							   0, w->xe.exec.address, w->bb_size,
+							   &sync, 1);
+					xe_wait_ufence(fd, &w->xe.data->vm_sync,
+						       user_fence_value,
+						       vm->bind_exec_queue,
+						       10 * NSEC_PER_SEC);
+					w->xe.data->vm_sync = 0;
+				} else {
+					xe_vm_unbind_sync(fd, get_vm(wrk, w)->id, 0,
+							  w->xe.exec.address, w->bb_size);
+				}
 				gem_munmap(w->xe.data, w->bb_size);
 				gem_close(fd, w->bb_handle);
 			} else if (w->type == SW_FENCE) {
-				syncobj_destroy(fd, w->xe.syncs[0].handle);
+				if (w->xe.syncs[0].type == DRM_XE_SYNC_TYPE_SYNCOBJ)
+					syncobj_destroy(fd, w->xe.syncs[0].handle);
 				free(w->xe.syncs);
 			}
 		}
benchmarks/wsim/README +15 -2
index 909c1f039..7c180cfb5 100644
@@ -11,7 +11,7 @@ P|S|X.<uint>.<int>
 d|p|s|t|q|a|T.<int>,...
 b.<uint>.<str>[|<str>].<str>
 w|W.<uint>.<str>[/<str>]...
-V.<uint>
+V|c.<uint>
 v.<uint>.<uint>
 f
 
@@ -45,6 +45,7 @@ VM management steps (Xe: active; i915: ignored):
 
  'V' - Create VM with id N (V.N, VM ids are 1-based).
  'v' - Assign VM N to context/queue M (v.N.M).
+ 'c' - Enable compute/LR mode for VM N (c.N).
 
 Engine ids: DEFAULT, RCS, BCS, VCS, VCS1, VCS2, VECS
 
@@ -95,7 +96,7 @@ Batch durations can also be specified as infinite by using the '*' in the
 duration field. Such batches must be ended by the terminate command ('T')
 otherwise they will cause a GPU hang to be reported.
 
-VM Example:
+VM Examples:
 
   V.1
   M.1.VCS
@@ -107,6 +108,18 @@ Creates VM 1, configures context 1 with a VCS engine map, assigns context 1
 to VM 1, enables load balancing for context 1, and runs a single 1ms VCS
 batch.
 
+  V.1
+  c.1
+  v.1.1
+  v.1.2
+  1.BCS.5000.0.0
+  2.CCS.7000000.-1.0
+  1.BCS.5000.-1.1
+
+Creates VM 1, enables compute/LR mode on it, assigns contexts 1 and 2 to the
+same VM, then runs BCS and CCS work with cross-step dependencies (CCS depends
+on the first BCS step, and the final BCS step depends on the CCS step).
+
 Xe and i915 differences
 ------------------------
 
Signature · 1 line
2.43.0
Reviewed-by on v1 2/2
#5 lore raw

Re: [PATCH i-g-t 2/2] benchmarks/gem_wsim: add compute/LR mode support for Xe VMs

Quoted 366 lines · Marcin Bernatowicz

Introduce new descriptor step:
'c' - Enable compute/LR mode for VM N (c.N).

Example:

gem_wsim -w "V.1,c.1,v.1.1,1.BCS.100.0.0,1.CCS.8000000.-1.0,1.BCS.100.-1.1" 8.009s elapsed (0.125 workloads/s)

Creates VM 1, enables compute/LR mode on it, assigns context 1 to the VM, then runs BCS and CCS work with cross-step dependencies (8s CCS depends on the first BCS step, and the final BCS step depends on the CCS step).

---
  benchmarks/gem_wsim.c  | 171 ++++++++++++++++++++++++++++++++++++-----
  benchmarks/wsim/README |  17 +++-
  2 files changed, 165 insertions(+), 23 deletions(-)
benchmarks/gem_wsim.c +150 -21
index b3c46772f..d6638bcdf 100644
@@ -132,6 +132,7 @@ enum w_type {
  	WORKINGSET,
  	VM_CREATE,
  	CTX_VM,
+	VM_COMPUTE_MODE,
  };
 
  struct dep_entry {
@@ -246,6 +247,7 @@ struct vm {
  	uint32_t id;
  	bool declared;
  	bool compute_mode;
+	uint32_t bind_exec_queue;
  	uint64_t ahnd;
  };
 
@@ -323,6 +325,7 @@ static unsigned int master_prng;
  static int verbose = 1;
  static int fd;
  static bool is_xe;
+static const uint64_t user_fence_value = 0xdeadbeefdeadbeefull;
  static struct drm_i915_gem_context_param_sseu device_sseu = {
  	.slice_mask = -1 /* Force read on first use. */
  };
@@ -334,10 +337,18 @@ static struct drm_i915_gem_context_param_sseu device_sseu = {
 
  static void w_step_sync(struct w_step *w)
  {
-	if (is_xe)
-		igt_assert(syncobj_wait(fd, &w->xe.syncs[0].handle, 1, INT64_MAX, 0, NULL));
-	else
+	if (is_xe) {
+		if (w->xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE) {
+			xe_wait_ufence(fd, &w->xe.data->spin.exec_sync,
+				       w->xe.syncs[0].timeline_value, 0,
+				       INT64_MAX);
+		} else {
+			igt_assert(syncobj_wait(fd, &w->xe.syncs[0].handle, 1,
+						INT64_MAX, 0, NULL));
+		}
+	} else {
  		gem_sync(fd, w->i915.obj[0].handle);
+	}
  }
 
  static int read_timestamp_frequency(int i915)
@@ -1293,6 +1304,20 @@ parse_workload(struct w_arg *arg, unsigned int flags, double scale_dur,
  				step.context = ctx_id;
  				step.type = CTX_VM;
  				goto add_step;
+			} else if (!strcmp(field, "c")) {
+				/*
+				 * c.N  - enable compute/LR mode on VM N
+				 * (xe only, ignored on i915)
+				 */
+				field = strtok_r(fstart, ".", &fctx);
+				check_arg(!field,
+					  "Missing VM id at step %u!\n", nr_steps);
+				tmp = atoi(field);
+				check_arg(tmp <= 0,
+					  "Invalid VM id at step %u!\n", nr_steps);
+				step.vm_id = tmp;
+				step.type = VM_COMPUTE_MODE;
+				goto add_step;
  			}
 
  			if (!field) {
@@ -1855,10 +1880,27 @@ xe_alloc_step_batch(struct workload *wrk, struct w_step *w)
  				    vram_if_possible(fd, eq->hwe_list[0].gt_id),
  				    DRM_XE_GEM_CREATE_FLAG_NEEDS_VISIBLE_VRAM);
  	w->xe.data = xe_bo_map(fd, w->bb_handle, w->bb_size);
+	w->xe.data->spin.exec_sync = 0;
+	w->xe.data->vm_sync = 0;
  	w->xe.exec.address =
  		intel_allocator_alloc_with_strategy(vm->ahnd, w->bb_handle, w->bb_size,
  						    0, ALLOC_STRATEGY_LOW_TO_HIGH);
-	xe_vm_bind_sync(fd, vm->id, w->bb_handle, 0, w->xe.exec.address, w->bb_size);
+	if (vm->compute_mode) {
+		struct drm_xe_sync sync = {
+			.type = DRM_XE_SYNC_TYPE_USER_FENCE,
+			.flags = DRM_XE_SYNC_FLAG_SIGNAL,
+			.addr = to_user_pointer(&w->xe.data->vm_sync),
+			.timeline_value = user_fence_value,
+		};
+
+		xe_vm_bind_async(fd, vm->id, vm->bind_exec_queue, w->bb_handle, 0,
+				 w->xe.exec.address, w->bb_size, &sync, 1);
+		xe_wait_ufence(fd, &w->xe.data->vm_sync, user_fence_value,
+			       vm->bind_exec_queue, 10 * NSEC_PER_SEC);
+		w->xe.data->vm_sync = 0;
+	} else {
+		xe_vm_bind_sync(fd, vm->id, w->bb_handle, 0, w->xe.exec.address, w->bb_size);
+	}
  	w->duration.requested_ticks = xe_spin_nsec_to_ticks(fd, eq->hwe_list[0].gt_id,
  							    1000LL * get_duration(wrk, w));
  	xe_spin_init_opts(&w->xe.data->spin,
@@ -1875,8 +1917,15 @@ xe_alloc_step_batch(struct workload *wrk, struct w_step *w)
 
  		igt_assert(dep_idx >= 0 && dep_idx < w->idx);
  		igt_assert(wrk->steps[dep_idx].type == BATCH);
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
 
-		w->xe.exec.num_syncs++;
+		/*
+		 * USER_FENCE is valid as an out-fence in LR mode, but cannot be
+		 * consumed as an in-fence in xe_exec. Those deps are resolved in
+		 * userspace before submission.
+		 */
+		if (wrk->steps[dep_idx].xe.syncs[0].type != DRM_XE_SYNC_TYPE_USER_FENCE)
+			w->xe.exec.num_syncs++;
  	}
  	for_each_dep(dep, w->fence_deps) {
  		int dep_idx = w->idx + dep->target;
@@ -1884,34 +1933,70 @@ xe_alloc_step_batch(struct workload *wrk, struct w_step *w)
  		igt_assert(dep_idx >= 0 && dep_idx < w->idx);
  		igt_assert(wrk->steps[dep_idx].type == SW_FENCE ||
  			   wrk->steps[dep_idx].type == BATCH);
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
 
-		w->xe.exec.num_syncs++;
+		if (wrk->steps[dep_idx].xe.syncs[0].type != DRM_XE_SYNC_TYPE_USER_FENCE)
+			w->xe.exec.num_syncs++;
  	}
  	w->xe.syncs = calloc(w->xe.exec.num_syncs, sizeof(*w->xe.syncs));
  	/* fill syncs */
  	i = 0;
  	/* out fence */
-	w->xe.syncs[i].handle = syncobj_create(fd, 0);
-	w->xe.syncs[i].type = DRM_XE_SYNC_TYPE_SYNCOBJ;
-	w->xe.syncs[i++].flags = DRM_XE_SYNC_FLAG_SIGNAL;
+	if (vm->compute_mode) {
+		w->xe.syncs[i].type = DRM_XE_SYNC_TYPE_USER_FENCE;
+		w->xe.syncs[i].flags = DRM_XE_SYNC_FLAG_SIGNAL;
+		w->xe.syncs[i].addr = w->xe.exec.address + offsetof(struct xe_spin, exec_sync);
+		w->xe.syncs[i++].timeline_value = user_fence_value;
+	} else {
+		w->xe.syncs[i].handle = syncobj_create(fd, 0);
+		w->xe.syncs[i].type = DRM_XE_SYNC_TYPE_SYNCOBJ;
+		w->xe.syncs[i++].flags = DRM_XE_SYNC_FLAG_SIGNAL;
+	}
  	/* in fence(s) */
  	for_each_dep(dep, w->data_deps) {
  		int dep_idx = w->idx + dep->target;
 
-		igt_assert(wrk->steps[dep_idx].xe.syncs && wrk->steps[dep_idx].xe.syncs[0].handle);
-		w->xe.syncs[i].handle = wrk->steps[dep_idx].xe.syncs[0].handle;
-		w->xe.syncs[i++].type = DRM_XE_SYNC_TYPE_SYNCOBJ;
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
+		if (wrk->steps[dep_idx].xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE)
+			continue;
+		w->xe.syncs[i] = wrk->steps[dep_idx].xe.syncs[0];
+		w->xe.syncs[i++].flags &= ~DRM_XE_SYNC_FLAG_SIGNAL;
  	}
  	for_each_dep(dep, w->fence_deps) {
  		int dep_idx = w->idx + dep->target;
 
-		igt_assert(wrk->steps[dep_idx].xe.syncs && wrk->steps[dep_idx].xe.syncs[0].handle);
-		w->xe.syncs[i].handle = wrk->steps[dep_idx].xe.syncs[0].handle;
-		w->xe.syncs[i++].type = DRM_XE_SYNC_TYPE_SYNCOBJ;
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
+		if (wrk->steps[dep_idx].xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE)
+			continue;
+		w->xe.syncs[i] = wrk->steps[dep_idx].xe.syncs[0];
+		w->xe.syncs[i++].flags &= ~DRM_XE_SYNC_FLAG_SIGNAL;
  	}
  	w->xe.exec.syncs = to_user_pointer(w->xe.syncs);
  }
 
+static void xe_sync_user_fence_deps(struct workload *wrk, struct w_step *w)
+{
+	struct dep_entry *dep;
+
+	for_each_dep(dep, w->data_deps) {
+		int dep_idx = w->idx + dep->target;
+
+		igt_assert(dep_idx >= 0 && dep_idx < w->idx);
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
+		if (wrk->steps[dep_idx].xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE)
+			w_step_sync(&wrk->steps[dep_idx]);
+	}
+
+	for_each_dep(dep, w->fence_deps) {
+		int dep_idx = w->idx + dep->target;
+
+		igt_assert(dep_idx >= 0 && dep_idx < w->idx);
+		igt_assert(wrk->steps[dep_idx].xe.syncs);
+		if (wrk->steps[dep_idx].xe.syncs[0].type == DRM_XE_SYNC_TYPE_USER_FENCE)
+			w_step_sync(&wrk->steps[dep_idx]);
+	}
+}
+
  static bool set_priority(uint32_t ctx_id, int prio)
  {
  	struct drm_i915_gem_context_param param = {
@@ -2190,7 +2275,7 @@ static void allocate_contexts(unsigned int id, struct workload *wrk)
 
  		w->wrk = wrk;
 
-		if (w->type == VM_CREATE)
+		if (w->type == VM_CREATE || w->type == VM_COMPUTE_MODE)
  			continue;
 
  		ctx = w->context + 1;
@@ -2427,6 +2512,14 @@ static int xe_prepare_contexts(unsigned int id, struct workload *wrk)
  	struct ctx *ctx;
  	unsigned int i;
 
+	for_each_w_step(w, wrk) {
+		if (w->type == VM_COMPUTE_MODE) {
+			igt_assert_lt((unsigned int)w->vm_id, wrk->nr_vms);
+			igt_assert(wrk->vm_list[w->vm_id].declared);
+			wrk->vm_list[w->vm_id].compute_mode = true;
+		}
+	}
+
  	/* Slot 0 is the implicit/default VM. */
  	xe_vm_create_(&wrk->vm_list[0]);
  	wrk->vm_list[0].ahnd = intel_allocator_open(fd, wrk->vm_list[0].id,
@@ -2437,6 +2530,9 @@ static int xe_prepare_contexts(unsigned int id, struct workload *wrk)
  			continue;
 
  		xe_vm_create_(&wrk->vm_list[i]);
+		if (wrk->vm_list[i].compute_mode)
+			wrk->vm_list[i].bind_exec_queue =
+				xe_bind_exec_queue_create(fd, wrk->vm_list[i].id, 0);
  		wrk->vm_list[i].ahnd =
  			intel_allocator_open(fd, wrk->vm_list[i].id,
  					     INTEL_ALLOCATOR_RELOC);
@@ -2767,11 +2863,22 @@ static void w_sync_to(struct workload *wrk, struct w_step *w, int target)
  static void do_xe_exec(struct workload *wrk, struct w_step *w)
  {
  	struct xe_exec_queue *eq = xe_get_eq(wrk, w);
+	struct vm *vm = get_vm(wrk, w);
 
  	igt_assert(w->emit_fence <= 0);
  	if (w->emit_fence == -1)
  		syncobj_reset(fd, &w->xe.syncs[0].handle, 1);
 
+	/*
+	 * LR mode reuses the same user-fence location for completion.
+	 * Re-arm it before every submit so waiters do not observe the
+	 * previous iteration's signaled value.
+	 */
+	if (vm->compute_mode)
+		w->xe.data->spin.exec_sync = 0;
+
+	xe_sync_user_fence_deps(wrk, w);
+
  	/* update duration if random */
  	if (w->duration.max != w->duration.min) {
  		w->duration.requested_ticks = xe_spin_nsec_to_ticks(fd, eq->hwe_list[0].gt_id,
@@ -2976,7 +3083,8 @@ static void *run_workload(void *data)
  				   w->type == BOND ||
  				   w->type == WORKINGSET ||
  				   w->type == VM_CREATE ||
-				   w->type == CTX_VM) {
+				   w->type == CTX_VM ||
+				   w->type == VM_COMPUTE_MODE) {
  				   /* No action for these at execution time. */
  				continue;
  			}
@@ -3066,14 +3174,35 @@ static void *run_workload(void *data)
  		for_each_w_step(w, wrk) {
  			if (w->type == BATCH) {
  				w_step_sync(w);
-				syncobj_destroy(fd, w->xe.syncs[0].handle);
+				if (w->xe.syncs[0].type == DRM_XE_SYNC_TYPE_SYNCOBJ)
+					syncobj_destroy(fd, w->xe.syncs[0].handle);
  				free(w->xe.syncs);
-				xe_vm_unbind_sync(fd, get_vm(wrk, w)->id, 0, w->xe.exec.address,
-						  w->bb_size);
+				if (get_vm(wrk, w)->compute_mode) {
+					struct vm *vm = get_vm(wrk, w);
+					struct drm_xe_sync sync = {
+						.type = DRM_XE_SYNC_TYPE_USER_FENCE,
+						.flags = DRM_XE_SYNC_FLAG_SIGNAL,
+						.addr = to_user_pointer(&w->xe.data->vm_sync),
+						.timeline_value = user_fence_value,
+					};
+
+					xe_vm_unbind_async(fd, vm->id, vm->bind_exec_queue,
+							   0, w->xe.exec.address, w->bb_size,
+							   &sync, 1);
+					xe_wait_ufence(fd, &w->xe.data->vm_sync,
+						       user_fence_value,
+						       vm->bind_exec_queue,
+						       10 * NSEC_PER_SEC);
+					w->xe.data->vm_sync = 0;
+				} else {
+					xe_vm_unbind_sync(fd, get_vm(wrk, w)->id, 0,
+							  w->xe.exec.address, w->bb_size);
+				}
  				gem_munmap(w->xe.data, w->bb_size);
  				gem_close(fd, w->bb_handle);
  			} else if (w->type == SW_FENCE) {
-				syncobj_destroy(fd, w->xe.syncs[0].handle);
+				if (w->xe.syncs[0].type == DRM_XE_SYNC_TYPE_SYNCOBJ)
+					syncobj_destroy(fd, w->xe.syncs[0].handle);
  				free(w->xe.syncs);
  			}
  		}
benchmarks/wsim/README +8 -2
index 909c1f039..7c180cfb5 100644
@@ -11,7 +11,7 @@ P|S|X.<uint>.<int>
  d|p|s|t|q|a|T.<int>,...
  b.<uint>.<str>[|<str>].<str>
  w|W.<uint>.<str>[/<str>]...
-V.<uint>
+V|c.<uint>
  v.<uint>.<uint>
  f
 
@@ -45,6 +45,7 @@ VM management steps (Xe: active; i915: ignored):
 
   'V' - Create VM with id N (V.N, VM ids are 1-based).
   'v' - Assign VM N to context/queue M (v.N.M).
+ 'c' - Enable compute/LR mode for VM N (c.N).
 
  Engine ids: DEFAULT, RCS, BCS, VCS, VCS1, VCS2, VECS
 
@@ -95,7 +96,7 @@ Batch durations can also be specified as infinite by using the '*' in the
  duration field. Such batches must be ended by the terminate command ('T')
  otherwise they will cause a GPU hang to be reported.
 
-VM Example:
+VM Examples:
 
    V.1
    M.1.VCS
@@ -107,6 +108,18 @@ Creates VM 1, configures context 1 with a VCS engine map, assigns context 1
  to VM 1, enables load balancing for context 1, and runs a single 1ms VCS
  batch.
 
+  V.1
+  c.1
+  v.1.1
+  v.1.2
+  1.BCS.5000.0.0

benchmarks/wsim/README Creates VM 1, configures context 1 with a VCS engine map, assigns context 1 · line 125

Quoted 10 lines
benchmarks/wsim/README +7 -0
+  2.CCS.7000000.-1.0
+  1.BCS.5000.-1.1
+
+Creates VM 1, enables compute/LR mode on it, assigns contexts 1 and 2 to the
+same VM, then runs BCS and CCS work with cross-step dependencies (CCS depends
+on the first BCS step, and the final BCS step depends on the CCS step).
+
  Xe and i915 differences
  ------------------------
 

LGTM,

Regards,
Adam

#6 lore raw

✗ Xe.CI.BAT: failure for benchmarks/gem_wsim: explicit VM create/assign + LR mode

In reply to Marcin Bernatowicz
== Series Details ==

== Series Details ==

Series: benchmarks/gem_wsim: explicit VM create/assign + LR mode
URL : https://patchwork.freedesktop.org/series/170422/
State : failure

== Summary ==

CI Bug Log - changes from XEIGT_9006_BAT -> XEIGTPW_15527_BAT
====================================================

Summary
-------

**FAILURE**

Serious unknown changes coming with XEIGTPW_15527_BAT absolutely need to be verified manually.

If you think the reported changes have nothing to do with the changes
introduced in XEIGTPW_15527_BAT, please notify your bug team ([email protected]) to allow them to document this new failure mode, which will reduce false positives in CI.

Participating hosts (12 -> 12)
------------------------------

Additional (1): bat-bmg-2
Missing (1): bat-bmg-vm

Possible new issues
-------------------

Here are the unknown changes that may have been introduced in XEIGTPW_15527_BAT:

### IGT changes ###

#### Possible regressions ####

* igt@xe_module_load@load:
- bat-ptl-2: [PASS][1] -> [ABORT][2]
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-ptl-2/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-ptl-2/igt@[email protected]
- bat-wcl-1: [PASS][3] -> [ABORT][4]
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-wcl-1/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-wcl-1/igt@[email protected]
- bat-ptl-1: [PASS][5] -> [ABORT][6]
[5]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-ptl-1/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-ptl-1/igt@[email protected]
- bat-wcl-2: [PASS][7] -> [ABORT][8]
[7]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-wcl-2/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-wcl-2/igt@[email protected]
- bat-lnl-1: [PASS][9] -> [ABORT][10]
[9]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-lnl-1/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-lnl-1/igt@[email protected]
- bat-bmg-2: NOTRUN -> [ABORT][11]
[11]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-bmg-2/igt@[email protected]
- bat-bmg-3: [PASS][12] -> [ABORT][13]
[12]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-bmg-3/igt@[email protected]
[13]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-bmg-3/igt@[email protected]
- bat-bmg-1: [PASS][14] -> [ABORT][15]
[14]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-bmg-1/igt@[email protected]
[15]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-bmg-1/igt@[email protected]
- bat-adlp-7: [PASS][16] -> [ABORT][17]
[16]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-adlp-7/igt@[email protected]
[17]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-adlp-7/igt@[email protected]
- bat-lnl-2: [PASS][18] -> [ABORT][19]
[18]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/bat-lnl-2/igt@[email protected]
[19]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/bat-lnl-2/igt@[email protected]

Build changes
-------------

* IGT: IGT_9006 -> IGTPW_15527
* Linux: xe-5397-4df02bb6acd1307de4de001699112cbdaf5afdfc -> xe-5402-9cf91140eff27765d402aedeea89505d52db121e

IGTPW_15527: 15527
IGT_9006: 6380a8af26359dd222e22679442272ded836c463 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-5397-4df02bb6acd1307de4de001699112cbdaf5afdfc: 4df02bb6acd1307de4de001699112cbdaf5afdfc xe-5402-9cf91140eff27765d402aedeea89505d52db121e: 9cf91140eff27765d402aedeea89505d52db121e

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/index.html

#7 lore raw

✗ i915.CI.BAT: failure for benchmarks/gem_wsim: explicit VM create/assign + LR mode

In reply to Marcin Bernatowicz
== Series Details ==

== Series Details ==

Series: benchmarks/gem_wsim: explicit VM create/assign + LR mode
URL : https://patchwork.freedesktop.org/series/170422/
State : failure

== Summary ==

CI Bug Log - changes from IGT_9006 -> IGTPW_15527
====================================================

Summary
-------

**FAILURE**

Serious unknown changes coming with IGTPW_15527 absolutely need to be verified manually.

If you think the reported changes have nothing to do with the changes
introduced in IGTPW_15527, please notify your bug team ([email protected]) to allow them to document this new failure mode, which will reduce false positives in CI.

External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/index.html

Participating hosts (42 -> 40)
------------------------------

Missing (2): bat-dg2-13 fi-snb-2520m

Possible new issues
-------------------

Here are the unknown changes that may have been introduced in IGTPW_15527:

### IGT changes ###

#### Possible regressions ####

* igt@i915_module_load@load:
- fi-ilk-650: [PASS][1] -> [ABORT][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-ilk-650/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-ilk-650/igt@[email protected]
- fi-bsw-n3050: [PASS][3] -> [ABORT][4]
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-bsw-n3050/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-bsw-n3050/igt@[email protected]
- bat-adlp-6: [PASS][5] -> [ABORT][6]
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-adlp-6/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-adlp-6/igt@[email protected]
- bat-arlh-2: [PASS][7] -> [ABORT][8]
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-arlh-2/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-arlh-2/igt@[email protected]
- fi-rkl-11600: [PASS][9] -> [ABORT][10]
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-rkl-11600/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-rkl-11600/igt@[email protected]
- fi-skl-6600u: [PASS][11] -> [ABORT][12]
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-skl-6600u/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-skl-6600u/igt@[email protected]
- bat-arlh-3: [PASS][13] -> [ABORT][14]
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-arlh-3/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-arlh-3/igt@[email protected]
- fi-pnv-d510: [PASS][15] -> [ABORT][16]
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-pnv-d510/igt@[email protected]
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-pnv-d510/igt@[email protected]
- bat-dg1-7: [PASS][17] -> [ABORT][18]
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-dg1-7/igt@[email protected]
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-dg1-7/igt@[email protected]
- fi-glk-j4005: [PASS][19] -> [ABORT][20]
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-glk-j4005/igt@[email protected]
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-glk-j4005/igt@[email protected]
- bat-adlp-9: [PASS][21] -> [ABORT][22]
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-adlp-9/igt@[email protected]
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-adlp-9/igt@[email protected]
- bat-twl-2: [PASS][23] -> [ABORT][24]
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-twl-2/igt@[email protected]
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-twl-2/igt@[email protected]
- bat-rpls-4: [PASS][25] -> [ABORT][26]
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-rpls-4/igt@[email protected]
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-rpls-4/igt@[email protected]
- fi-kbl-7567u: [PASS][27] -> [ABORT][28]
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-kbl-7567u/igt@[email protected]
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-kbl-7567u/igt@[email protected]
- fi-cfl-8700k: [PASS][29] -> [ABORT][30]
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-cfl-8700k/igt@[email protected]
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-cfl-8700k/igt@[email protected]
- bat-twl-1: [PASS][31] -> [ABORT][32]
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-twl-1/igt@[email protected]
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-twl-1/igt@[email protected]
- fi-kbl-8809g: [PASS][33] -> [ABORT][34]
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-kbl-8809g/igt@[email protected]
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-kbl-8809g/igt@[email protected]
- bat-jsl-5: [PASS][35] -> [ABORT][36]
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-jsl-5/igt@[email protected]
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-jsl-5/igt@[email protected]
- bat-apl-1: [PASS][37] -> [ABORT][38]
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-apl-1/igt@[email protected]
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-apl-1/igt@[email protected]
- bat-dg2-14: [PASS][39] -> [ABORT][40]
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-dg2-14/igt@[email protected]
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-dg2-14/igt@[email protected]
- fi-elk-e7500: [PASS][41] -> [ABORT][42]
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-elk-e7500/igt@[email protected]
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-elk-e7500/igt@[email protected]
- fi-bsw-nick: [PASS][43] -> [ABORT][44]
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-bsw-nick/igt@[email protected]
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-bsw-nick/igt@[email protected]
- bat-kbl-2: [PASS][45] -> [ABORT][46]
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-kbl-2/igt@[email protected]
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-kbl-2/igt@[email protected]
- bat-arls-5: [PASS][47] -> [ABORT][48]
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-arls-5/igt@[email protected]
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-arls-5/igt@[email protected]
- bat-rplp-1: [PASS][49] -> [ABORT][50]
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-rplp-1/igt@[email protected]
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-rplp-1/igt@[email protected]
- fi-tgl-1115g4: [PASS][51] -> [ABORT][52]
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-tgl-1115g4/igt@[email protected]
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-tgl-1115g4/igt@[email protected]
- fi-cfl-guc: [PASS][53] -> [ABORT][54]
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-cfl-guc/igt@[email protected]
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-cfl-guc/igt@[email protected]
- bat-mtlp-9: [PASS][55] -> [ABORT][56]
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-mtlp-9/igt@[email protected]
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-mtlp-9/igt@[email protected]
- bat-arls-6: [PASS][57] -> [ABORT][58]
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-arls-6/igt@[email protected]
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-arls-6/igt@[email protected]
- bat-dg2-9: [PASS][59] -> [ABORT][60]
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-dg2-9/igt@[email protected]
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-dg2-9/igt@[email protected]
- fi-kbl-x1275: [PASS][61] -> [ABORT][62]
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-kbl-x1275/igt@[email protected]
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-kbl-x1275/igt@[email protected]
- bat-adlp-11: [PASS][63] -> [ABORT][64]
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-adlp-11/igt@[email protected]
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-adlp-11/igt@[email protected]
- fi-hsw-4770: [PASS][65] -> [ABORT][66]
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-hsw-4770/igt@[email protected]
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-hsw-4770/igt@[email protected]
- fi-cfl-8109u: [PASS][67] -> [ABORT][68]
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-cfl-8109u/igt@[email protected]
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-cfl-8109u/igt@[email protected]
- fi-ivb-3770: [PASS][69] -> [ABORT][70]
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/fi-ivb-3770/igt@[email protected]
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/fi-ivb-3770/igt@[email protected]
- bat-mtlp-8: [PASS][71] -> [ABORT][72]
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-mtlp-8/igt@[email protected]
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-mtlp-8/igt@[email protected]
- bat-dg1-6: [PASS][73] -> [ABORT][74]
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-dg1-6/igt@[email protected]
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-dg1-6/igt@[email protected]
- bat-dg2-8: [PASS][75] -> [ABORT][76]
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-dg2-8/igt@[email protected]
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-dg2-8/igt@[email protected]
- bat-adls-6: [PASS][77] -> [ABORT][78]
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_9006/bat-adls-6/igt@[email protected]
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/bat-adls-6/igt@[email protected]

Build changes
-------------

* CI: CI-20190529 -> None
* IGT: IGT_9006 -> IGTPW_15527
* Linux: CI_DRM_18816 -> CI_DRM_18821

CI-20190529: 20190529
CI_DRM_18816: 4df02bb6acd1307de4de001699112cbdaf5afdfc @ git://anongit.freedesktop.org/gfx-ci/linux
CI_DRM_18821: 9cf91140eff27765d402aedeea89505d52db121e @ git://anongit.freedesktop.org/gfx-ci/linux
IGTPW_15527: 15527
IGT_9006: 6380a8af26359dd222e22679442272ded836c463 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_15527/index.html

#8 lore raw

✗ Xe.CI.FULL: failure for benchmarks/gem_wsim: explicit VM create/assign + LR mode

In reply to Marcin Bernatowicz
== Series Details ==

== Series Details ==

Series: benchmarks/gem_wsim: explicit VM create/assign + LR mode
URL : https://patchwork.freedesktop.org/series/170422/
State : failure

== Summary ==

CI Bug Log - changes from XEIGT_9006_FULL -> XEIGTPW_15527_FULL
====================================================

Summary
-------

**WARNING**

Minor unknown changes coming with XEIGTPW_15527_FULL need to be verified manually.

If you think the reported changes have nothing to do with the changes
introduced in XEIGTPW_15527_FULL, please notify your bug team ([email protected]) to allow them to document this new failure mode, which will reduce false positives in CI.

Participating hosts (2 -> 2)
------------------------------

No changes in participating hosts

Possible new issues
-------------------

Here are the unknown changes that may have been introduced in XEIGTPW_15527_FULL:

### IGT changes ###

#### Warnings ####

* igt@xe_module_load@load:
- shard-lnl: ([PASS][1], [PASS][2], [PASS][3], [PASS][4], [SKIP][5], [PASS][6], [PASS][7], [PASS][8], [PASS][9], [PASS][10], [PASS][11], [PASS][12], [PASS][13], [PASS][14], [PASS][15], [PASS][16], [PASS][17], [PASS][18], [PASS][19], [PASS][20], [PASS][21], [PASS][22], [PASS][23], [PASS][24], [PASS][25], [PASS][26]) ([Intel XE#378] / [Intel XE#7405]) -> ([DMESG-WARN][27], [DMESG-WARN][28], [DMESG-WARN][29], [DMESG-WARN][30], [DMESG-WARN][31], [DMESG-WARN][32], [DMESG-WARN][33], [DMESG-WARN][34], [DMESG-WARN][35], [DMESG-WARN][36], [DMESG-WARN][37], [DMESG-WARN][38], [DMESG-WARN][39], [DMESG-WARN][40], [DMESG-WARN][41], [DMESG-WARN][42], [DMESG-WARN][43], [DMESG-WARN][44], [DMESG-WARN][45], [DMESG-WARN][46], [DMESG-WARN][47], [DMESG-WARN][48], [DMESG-WARN][49], [DMESG-WARN][50], [DMESG-WARN][51])
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-2/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-6/igt@[email protected]
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-5/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-5/igt@[email protected]
[5]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-5/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-6/igt@[email protected]
[7]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-6/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-5/igt@[email protected]
[9]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-3/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-3/igt@[email protected]
[11]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-3/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-2/igt@[email protected]
[13]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-2/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-4/igt@[email protected]
[15]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-7/igt@[email protected]
[16]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-7/igt@[email protected]
[17]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-7/igt@[email protected]
[18]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-8/igt@[email protected]
[19]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-8/igt@[email protected]
[20]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-8/igt@[email protected]
[21]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-4/igt@[email protected]
[22]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-4/igt@[email protected]
[23]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-1/igt@[email protected]
[24]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-1/igt@[email protected]
[25]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-1/igt@[email protected]
[26]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-lnl-4/igt@[email protected]
[27]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-1/igt@[email protected]
[28]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-1/igt@[email protected]
[29]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-1/igt@[email protected]
[30]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-8/igt@[email protected]
[31]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-8/igt@[email protected]
[32]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-8/igt@[email protected]
[33]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-4/igt@[email protected]
[34]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-4/igt@[email protected]
[35]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-4/igt@[email protected]
[36]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-6/igt@[email protected]
[37]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-6/igt@[email protected]
[38]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-6/igt@[email protected]
[39]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-3/igt@[email protected]
[40]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-3/igt@[email protected]
[41]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-3/igt@[email protected]
[42]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-2/igt@[email protected]
[43]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-2/igt@[email protected]
[44]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-2/igt@[email protected]
[45]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-7/igt@[email protected]
[46]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-7/igt@[email protected]
[47]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-7/igt@[email protected]
[48]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-5/igt@[email protected]
[49]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-5/igt@[email protected]
[50]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-5/igt@[email protected]
[51]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-lnl-5/igt@[email protected]
- shard-bmg: ([PASS][52], [PASS][53], [PASS][54], [PASS][55], [PASS][56], [PASS][57], [PASS][58], [PASS][59], [PASS][60], [PASS][61], [PASS][62], [PASS][63], [SKIP][64], [PASS][65], [PASS][66], [PASS][67], [PASS][68], [PASS][69], [PASS][70], [PASS][71], [PASS][72], [PASS][73], [PASS][74], [PASS][75], [PASS][76], [PASS][77]) ([Intel XE#2457] / [Intel XE#7405]) -> ([DMESG-WARN][78], [DMESG-WARN][79], [DMESG-WARN][80], [DMESG-WARN][81], [DMESG-WARN][82], [DMESG-WARN][83], [DMESG-WARN][84], [DMESG-WARN][85], [DMESG-WARN][86], [DMESG-WARN][87], [DMESG-WARN][88], [DMESG-WARN][89], [DMESG-WARN][90], [DMESG-WARN][91], [DMESG-WARN][92], [DMESG-WARN][93], [DMESG-WARN][94], [DMESG-WARN][95], [DMESG-WARN][96], [DMESG-WARN][97], [DMESG-WARN][98], [DMESG-WARN][99], [DMESG-WARN][100], [DMESG-WARN][101], [DMESG-WARN][102])
[52]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-4/igt@[email protected]
[53]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-4/igt@[email protected]
[54]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-4/igt@[email protected]
[55]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-2/igt@[email protected]
[56]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-2/igt@[email protected]
[57]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-2/igt@[email protected]
[58]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-5/igt@[email protected]
[59]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-3/igt@[email protected]
[60]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-8/igt@[email protected]
[61]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-5/igt@[email protected]
[62]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-8/igt@[email protected]
[63]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-6/igt@[email protected]
[64]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-2/igt@[email protected]
[65]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-3/igt@[email protected]
[66]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-3/igt@[email protected]
[67]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-9/igt@[email protected]
[68]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-9/igt@[email protected]
[69]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-5/igt@[email protected]
[70]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-10/igt@[email protected]
[71]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-7/igt@[email protected]
[72]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-7/igt@[email protected]
[73]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-7/igt@[email protected]
[74]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-6/igt@[email protected]
[75]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-1/igt@[email protected]
[76]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-1/igt@[email protected]
[77]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_9006/shard-bmg-10/igt@[email protected]
[78]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-4/igt@[email protected]
[79]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-4/igt@[email protected]
[80]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-4/igt@[email protected]
[81]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-8/igt@[email protected]
[82]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-8/igt@[email protected]
[83]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-10/igt@[email protected]
[84]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-10/igt@[email protected]
[85]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-10/igt@[email protected]
[86]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-1/igt@[email protected]
[87]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-1/igt@[email protected]
[88]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-1/igt@[email protected]
[89]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-7/igt@[email protected]
[90]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-7/igt@[email protected]
[91]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-9/igt@[email protected]
[92]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-9/igt@[email protected]
[93]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-2/igt@[email protected]
[94]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-2/igt@[email protected]
[95]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-2/igt@[email protected]
[96]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-6/igt@[email protected]
[97]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-6/igt@[email protected]
[98]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-3/igt@[email protected]
[99]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-3/igt@[email protected]
[100]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-3/igt@[email protected]
[101]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-5/igt@[email protected]
[102]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/shard-bmg-5/igt@[email protected]

[Intel XE#2457]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2457
[Intel XE#378]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/378
[Intel XE#7405]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7405

Build changes
-------------

* IGT: IGT_9006 -> IGTPW_15527
* Linux: xe-5397-4df02bb6acd1307de4de001699112cbdaf5afdfc -> xe-5402-9cf91140eff27765d402aedeea89505d52db121e

IGTPW_15527: 15527
IGT_9006: 6380a8af26359dd222e22679442272ded836c463 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-5397-4df02bb6acd1307de4de001699112cbdaf5afdfc: 4df02bb6acd1307de4de001699112cbdaf5afdfc xe-5402-9cf91140eff27765d402aedeea89505d52db121e: 9cf91140eff27765d402aedeea89505d52db121e

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_15527/index.html

#9 lore raw
In reply to Marcin Bernatowicz
Quoted 11 lines · Marcin Bernatowicz

This series extends gem_wsim VM management for Xe.

Patch 1 adds explicit VM create/assign steps in workload descriptors:
V.N to create VM N and v.N.M to bind context M to VM N.

Patch 2 adds c.N to enable compute/LR mode on selected VMs. README is updated with syntax and VM usage examples.

Marcin Bernatowicz (2):
benchmarks/gem_wsim: add explicit VM create/assign steps benchmarks/gem_wsim: add compute/LR mode support for Xe VMs

I see these two patches have my Suggested-by but to be honest I do not remember any of it! :)

Can I use the opportunity to ask how is the status of the tool on xe? Do working set and working set dependencies work for example?

Regards,

Tvrtko

Quoted 4 lines
  benchmarks/gem_wsim.c  | 310 ++++++++++++++++++++++++++++++++++++-----
  benchmarks/wsim/README |  32 +++++
  2 files changed, 311 insertions(+), 31 deletions(-)
#10 lore raw
Quoted 7 lines · Tvrtko Ursulin
Quoted 5 lines · Marcin Bernatowicz

This series extends gem_wsim VM management for Xe.

Patch 1 adds explicit VM create/assign steps in workload descriptors:
V.N to create VM N and v.N.M to bind context M to VM N.

Quoted 8 lines
Quoted 6 lines

Patch 2 adds c.N to enable compute/LR mode on selected VMs. README is updated with syntax and VM usage examples.

Marcin Bernatowicz (2):
   benchmarks/gem_wsim: add explicit VM create/assign steps benchmarks/gem_wsim: add compute/LR mode support for Xe VMs

I see these two patches have my Suggested-by but to be honest I do not remember any of it! :)

That's a very old thread on the first patches adjusting the tool to Xe: https://lore.kernel.org/all/[email protected]/ ;)

Quoted 2 lines

Can I use the opportunity to ask how is the status of the tool on xe? Do working set and working set dependencies work for example?

It's used for basic/simple workloads on VMs (Virtual Machines) with passed-through SR-IOV VF devices in regular runs (we use the expected duration of repeated workloads).

Lately, it helped me investigate an RCS/CCS dependency issue: https://lore.kernel.org/all/[email protected]/

So, I find it very useful for quick scheduling dependency checks without having to craft or adjust IGT tests.

I haven't looked at working set / working set dependencies on Xe yet :/

Regards,

marcin

Quoted 10 lines

Regards,

Tvrtko

Quoted 4 lines

  benchmarks/gem_wsim.c  | 310 ++++++++++++++++++++++++++++++++++++----- benchmarks/wsim/README |  32 +++++

  2 files changed, 311 insertions(+), 31 deletions(-)
#11 lore raw
Quoted 21 lines · Marcin
Quoted 7 lines · Tvrtko Ursulin
Quoted 5 lines · Marcin Bernatowicz

This series extends gem_wsim VM management for Xe.

Patch 1 adds explicit VM create/assign steps in workload descriptors:
V.N to create VM N and v.N.M to bind context M to VM N.

Quoted 8 lines
Quoted 6 lines

Patch 2 adds c.N to enable compute/LR mode on selected VMs. README is updated with syntax and VM usage examples.

Marcin Bernatowicz (2):
   benchmarks/gem_wsim: add explicit VM create/assign steps benchmarks/gem_wsim: add compute/LR mode support for Xe VMs

I see these two patches have my Suggested-by but to be honest I do not remember any of it! :)

That's a very old thread on the first patches adjusting the tool to Xe: https://lore.kernel.org/all/[email protected]/ ;)

Quoted 1 line
Quoted 8 lines
Quoted 1 line

Can I use the opportunity to ask how is the status of the tool on xe? Do working set and working set dependencies work for example?

It's used for basic/simple workloads on VMs (Virtual Machines) with passed-through SR-IOV VF devices in regular runs (we use the expected duration of repeated workloads).

Lately, it helped me investigate an RCS/CCS dependency issue: https://lore.kernel.org/all/[email protected]/

So, I find it very useful for quick scheduling dependency checks without having to craft or adjust IGT tests.

I haven't looked at working set / working set dependencies on Xe yet :/

I'm planning to send a follow-up series for working sets on Xe soon.

Quoted 16 lines

Regards,

marcin

Quoted 11 lines

Regards,

Tvrtko

Quoted 5 lines

  benchmarks/gem_wsim.c  | 310
++++++++++++++++++++++++++++++++++++-----
  benchmarks/wsim/README |  32 +++++

  2 files changed, 311 insertions(+), 31 deletions(-)
#12 lore raw
Quoted 26 lines · Marcin
Quoted 24 lines · Marcin
Quoted 7 lines · Tvrtko Ursulin
Quoted 5 lines · Marcin Bernatowicz

This series extends gem_wsim VM management for Xe.

Patch 1 adds explicit VM create/assign steps in workload descriptors:
V.N to create VM N and v.N.M to bind context M to VM N.

Quoted 8 lines
Quoted 6 lines

Patch 2 adds c.N to enable compute/LR mode on selected VMs. README is updated with syntax and VM usage examples.

Marcin Bernatowicz (2):
   benchmarks/gem_wsim: add explicit VM create/assign steps benchmarks/gem_wsim: add compute/LR mode support for Xe VMs

I see these two patches have my Suggested-by but to be honest I do not remember any of it! :)

That's a very old thread on the first patches adjusting the tool to Xe: https://lore.kernel.org/all/431b9135-1c7c-cbf7-cc3a- [email protected]/ ;)

Quoted 2 lines

Can I use the opportunity to ask how is the status of the tool on xe? Do working set and working set dependencies work for example?

It's used for basic/simple workloads on VMs (Virtual Machines) with passed-through SR-IOV VF devices in regular runs (we use the expected duration of repeated workloads).

Quoted 7 lines
Quoted 5 lines

Lately, it helped me investigate an RCS/CCS dependency issue: https:// lore.kernel.org/all/20260708100616.1050537-1- [email protected]/

So, I find it very useful for quick scheduling dependency checks without having to craft or adjust IGT tests.

I haven't looked at working set / working set dependencies on Xe yet :/

I'm planning to send a follow-up series for working sets on Xe soon.

For the record I don't personally need it, but was just curious given a colleague is looking into TTM eviction behaviour so this reminded me how
back in the day, when I added working set support to wsim, it was
exactly for improving the eviction behaviour of i915 as we we adding discrete GPU support I believe. Details escape me.. But 61d49ef6f274 ("gem_wsim: Snippet of a workload extracted from carchase") suggests this demo was useful back then and I think I had a script which converted the trace log to this wsim. Plan being to make some more mirroring interesting games, with the benefit of not needing to install the actual game, etc, but if the system can run IGT that's it, done, it can be replayed to oblivion and kernel tweaks tested quickly and easily.

Anyway, I'll copy Cascardo just as a curiosity/visibility on crazy/interesting tools. If you go to read it, you will sure get confused by the .wsim syntax.

Essentially the tools allows to write a workload defining arbitrary dummy GPU jobs, each with arbitrary buffers and/or fence dependencies, loop them, run N in parallel or not etc. It could be useful for TTM improvements once Marcin brings the xe backend to feature parity with i915.

Regards,

Tvrtko

Quoted 17 lines
Quoted 17 lines

Regards,

marcin

Quoted 11 lines

Regards,

Tvrtko

Quoted 5 lines

  benchmarks/gem_wsim.c  | 310 +++++++++++++++++++++++++++++++++++

+-----
  benchmarks/wsim/README |  32 +++++
  2 files changed, 311 insertions(+), 31 deletions(-)