Skip to contentSkip to conversation

drm/amdkfd: Fix circular lock in nocpsch path

PATCH In review Amber Lin 3 participants 5 messages Jun 15, 2021 – Jun 16, 2021

Subsystems (3)
lore
Share & tools
lore

Amber Lin

author
#0 lore raw

PATCH 1/1 drm/amdkfd: Fix circular lock in nocpsch path

Calling free_mqd inside of destroy_queue_nocpsch_locked can cause a circular lock. destroy_queue_nocpsch_locked is called under a DQM lock, which is taken in MMU notifiers, potentially in FS reclaim context. Taking another lock, which is BO reservation lock from free_mqd, while causing an FS reclaim inside the DQM lock creates a problematic circular lock dependency. Therefore move free_mqd out of destroy_queue_nocpsch_locked and call it after unlocking DQM.

---
 .../drm/amd/amdkfd/kfd_device_queue_manager.c  | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +13 -5
index 72bea5278add..c069fa259b30 100644
@@ -486,9 +486,6 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm,
 	if (retval == -ETIME)
 		qpd->reset_wavefronts = true;
 
-
-	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
-
 	list_del(&q->list);
 	if (list_empty(&qpd->queues_list)) {
 		if (qpd->reset_wavefronts) {
@@ -523,6 +520,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 	int retval;
 	uint64_t sdma_val = 0;
 	struct kfd_process_device *pdd = qpd_to_pdd(qpd);
+	struct mqd_manager *mqd_mgr =
+		dqm->mqd_mgrs[get_mqd_type_from_queue_type(q->properties.type)];
 
 	/* Get the SDMA queue stats */
 	if ((q->properties.type == KFD_QUEUE_TYPE_SDMA) ||
@@ -540,6 +539,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 		pdd->sdma_past_activity_counter += sdma_val;
 	dqm_unlock(dqm);
 
+	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+
 	return retval;
 }
 
@@ -1629,7 +1630,7 @@ static bool set_cache_memory_policy(struct device_queue_manager *dqm,
 static int process_termination_nocpsch(struct device_queue_manager *dqm,
 		struct qcm_process_device *qpd)
 {
-	struct queue *q, *next;
+	struct queue *q;
 	struct device_process_node *cur, *next_dpn;
 	int retval = 0;
 	bool found = false;
@@ -1637,12 +1638,19 @@ static int process_termination_nocpsch(struct device_queue_manager *dqm,
 	dqm_lock(dqm);
 
 	/* Clear all user mode queues */
-	list_for_each_entry_safe(q, next, &qpd->queues_list, list) {
+	while (!list_empty(&qpd->queues_list)) {
+		struct mqd_manager *mqd_mgr;
 		int ret;
 
+		q = list_first_entry(&qpd->queues_list, struct queue, list);
+		mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type(
+				q->properties.type)];
 		ret = destroy_queue_nocpsch_locked(dqm, qpd, q);
 		if (ret)
 			retval = ret;
+		dqm_unlock(dqm);
+		mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+		dqm_lock(dqm);
 	}
 
 	/* Unregister process */
Signature · 6 lines
2.17.1

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
maintainer
#1 lore raw

[+Xinhui]

Am 2021-06-15 um 1:50 p.m. schrieb Amber Lin:

Quoted 10 lines

Calling free_mqd inside of destroy_queue_nocpsch_locked can cause a circular lock. destroy_queue_nocpsch_locked is called under a DQM lock, which is taken in MMU notifiers, potentially in FS reclaim context. Taking another lock, which is BO reservation lock from free_mqd, while causing an FS reclaim inside the DQM lock creates a problematic circular lock dependency. Therefore move free_mqd out of destroy_queue_nocpsch_locked and call it after unlocking DQM.

Let's submit this patch as is. I'm making some comments inline for things that Xinhui can address in his race condition patch.

Quoted 53 lines
---
 .../drm/amd/amdkfd/kfd_device_queue_manager.c  | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +7 -5
index 72bea5278add..c069fa259b30 100644
@@ -486,9 +486,6 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm,
 	if (retval == -ETIME)
 		qpd->reset_wavefronts = true;
 
-
-	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
-
 	list_del(&q->list);
 	if (list_empty(&qpd->queues_list)) {
 		if (qpd->reset_wavefronts) {
@@ -523,6 +520,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 	int retval;
 	uint64_t sdma_val = 0;
 	struct kfd_process_device *pdd = qpd_to_pdd(qpd);
+	struct mqd_manager *mqd_mgr =
+		dqm->mqd_mgrs[get_mqd_type_from_queue_type(q->properties.type)];
 
 	/* Get the SDMA queue stats */
 	if ((q->properties.type == KFD_QUEUE_TYPE_SDMA) ||
@@ -540,6 +539,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 		pdd->sdma_past_activity_counter += sdma_val;
 	dqm_unlock(dqm);
 
+	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+
 	return retval;
 }
 
@@ -1629,7 +1630,7 @@ static bool set_cache_memory_policy(struct device_queue_manager *dqm,
 static int process_termination_nocpsch(struct device_queue_manager *dqm,
 		struct qcm_process_device *qpd)
 {
-	struct queue *q, *next;
+	struct queue *q;
 	struct device_process_node *cur, *next_dpn;
 	int retval = 0;
 	bool found = false;
@@ -1637,12 +1638,19 @@ static int process_termination_nocpsch(struct device_queue_manager *dqm,
 	dqm_lock(dqm);
 
 	/* Clear all user mode queues */
-	list_for_each_entry_safe(q, next, &qpd->queues_list, list) {
+	while (!list_empty(&qpd->queues_list)) {
+		struct mqd_manager *mqd_mgr;
 		int ret;

drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c static int process_termination_nocpsch(struct device_queue_manager *dqm, · line 1653

Quoted 10 lines
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +6 -0
 
+		q = list_first_entry(&qpd->queues_list, struct queue, list);
+		mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type(
+				q->properties.type)];
 		ret = destroy_queue_nocpsch_locked(dqm, qpd, q);
 		if (ret)
 			retval = ret;
+		dqm_unlock(dqm);
+		mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+		dqm_lock(dqm);

This is the correct way to clean up the list when dropping the dqm-lock in the middle. Xinhui, you can use the same method in process_termination_cpsch.

I believe the swapping of the q->mqd with a temporary variable is not needed. When free_mqd is called, the queue is no longer on the qpd->queues_list, so destroy_queue cannot race with it. If we ensure that queues are always removed from the list before calling free_mqd, and that list-removal happens under the dqm_lock, then there should be no risk of a race condition that causes a double-free.

Regards,
  Felix

drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c static int process_termination_nocpsch(struct device_queue_manager *dqm, · line 1656

Quoted 3 lines
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +0 -0
 	}
 
 	/* Unregister process */

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

#2 lore raw
Quoted 77 lines

2021年6月16日 02:22,Kuehling, Felix <[email protected]> 写道:

[+Xinhui]

Am 2021-06-15 um 1:50 p.m. schrieb Amber Lin:

Quoted 10 lines

Calling free_mqd inside of destroy_queue_nocpsch_locked can cause a circular lock. destroy_queue_nocpsch_locked is called under a DQM lock, which is taken in MMU notifiers, potentially in FS reclaim context. Taking another lock, which is BO reservation lock from free_mqd, while causing an FS reclaim inside the DQM lock creates a problematic circular lock dependency. Therefore move free_mqd out of destroy_queue_nocpsch_locked and call it after unlocking DQM.

Let's submit this patch as is. I'm making some comments inline for things that Xinhui can address in his race condition patch.

Quoted 54 lines

---
.../drm/amd/amdkfd/kfd_device_queue_manager.c | 18 +++++++++++++-----

1 file changed, 13 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +7 -5
index 72bea5278add..c069fa259b30 100644
@@ -486,9 +486,6 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm,
 	if (retval == -ETIME)
 		qpd->reset_wavefronts = true;
 
-
-	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
-
 	list_del(&q->list);
 	if (list_empty(&qpd->queues_list)) {
 		if (qpd->reset_wavefronts) {
@@ -523,6 +520,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 	int retval;
 	uint64_t sdma_val = 0;
 	struct kfd_process_device *pdd = qpd_to_pdd(qpd);
+	struct mqd_manager *mqd_mgr =
+		dqm->mqd_mgrs[get_mqd_type_from_queue_type(q->properties.type)];
 
 	/* Get the SDMA queue stats */
 	if ((q->properties.type == KFD_QUEUE_TYPE_SDMA) ||
@@ -540,6 +539,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 		pdd->sdma_past_activity_counter += sdma_val;
 	dqm_unlock(dqm);
 
+	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+
 	return retval;
 }
 
@@ -1629,7 +1630,7 @@ static bool set_cache_memory_policy(struct device_queue_manager *dqm,
 static int process_termination_nocpsch(struct device_queue_manager *dqm,
 		struct qcm_process_device *qpd)
 {
-	struct queue *q, *next;
+	struct queue *q;
 	struct device_process_node *cur, *next_dpn;
 	int retval = 0;
 	bool found = false;
@@ -1637,12 +1638,19 @@ static int process_termination_nocpsch(struct device_queue_manager *dqm,
 	dqm_lock(dqm);
 
 	/* Clear all user mode queues */
-	list_for_each_entry_safe(q, next, &qpd->queues_list, list) {
+	while (!list_empty(&qpd->queues_list)) {
+		struct mqd_manager *mqd_mgr;
 		int ret;
 
Quoted 14 lines

drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c static int process_termination_nocpsch(struct device_queue_manager *dqm, · line 1653

Quoted 9 lines
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +6 -0
+		q = list_first_entry(&qpd->queues_list, struct queue, list);
+		mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type(
+				q->properties.type)];
 		ret = destroy_queue_nocpsch_locked(dqm, qpd, q);
 		if (ret)
 			retval = ret;
+		dqm_unlock(dqm);
+		mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+		dqm_lock(dqm);

This is the correct way to clean up the list when dropping the dqm-lock in the middle. Xinhui, you can use the same method in process_termination_cpsch.

yes, that is the right way to walk through the list. thanks.

Quoted 7 lines

I believe the swapping of the q->mqd with a temporary variable is not needed. When free_mqd is called, the queue is no longer on the qpd->queues_list, so destroy_queue cannot race with it. If we ensure that queues are always removed from the list before calling free_mqd, and that list-removal happens under the dqm_lock, then there should be no risk of a race condition that causes a double-free.

no, the double free exists because pqm_destroy_queue fetch the queue from qid by get_queue_by_qid() the race is like below.
pqm_destroy_queue

	get_queue_by_qid				process_termination_cpsch
	destroy_queue_cpsch
								lock
								list_for_each_entry_safe
									list_del(q)
								unlock
								free_mqd
	lock
	list_del(q)
	unlock
	free_mqd
Quoted 7 lines

Regards,
Felix

Quoted 3 lines
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +0 -0
 	}
 
 	/* Unregister process */

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

maintainer
#3 lore raw

Am 2021-06-16 um 12:01 a.m. schrieb Pan, Xinhui:

Quoted 100 lines
Quoted 74 lines

2021年6月16日 02:22,Kuehling, Felix <[email protected]> 写道:

[+Xinhui]

Am 2021-06-15 um 1:50 p.m. schrieb Amber Lin:

Quoted 10 lines

Calling free_mqd inside of destroy_queue_nocpsch_locked can cause a circular lock. destroy_queue_nocpsch_locked is called under a DQM lock, which is taken in MMU notifiers, potentially in FS reclaim context. Taking another lock, which is BO reservation lock from free_mqd, while causing an FS reclaim inside the DQM lock creates a problematic circular lock dependency. Therefore move free_mqd out of destroy_queue_nocpsch_locked and call it after unlocking DQM.

Let's submit this patch as is. I'm making some comments inline for things that Xinhui can address in his race condition patch.

Quoted 54 lines

---
.../drm/amd/amdkfd/kfd_device_queue_manager.c | 18 +++++++++++++-----

1 file changed, 13 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +7 -5
index 72bea5278add..c069fa259b30 100644
@@ -486,9 +486,6 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm,
 	if (retval == -ETIME)
 		qpd->reset_wavefronts = true;
 
-
-	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
-
 	list_del(&q->list);
 	if (list_empty(&qpd->queues_list)) {
 		if (qpd->reset_wavefronts) {
@@ -523,6 +520,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 	int retval;
 	uint64_t sdma_val = 0;
 	struct kfd_process_device *pdd = qpd_to_pdd(qpd);
+	struct mqd_manager *mqd_mgr =
+		dqm->mqd_mgrs[get_mqd_type_from_queue_type(q->properties.type)];
 
 	/* Get the SDMA queue stats */
 	if ((q->properties.type == KFD_QUEUE_TYPE_SDMA) ||
@@ -540,6 +539,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 		pdd->sdma_past_activity_counter += sdma_val;
 	dqm_unlock(dqm);
 
+	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+
 	return retval;
 }
 
@@ -1629,7 +1630,7 @@ static bool set_cache_memory_policy(struct device_queue_manager *dqm,
 static int process_termination_nocpsch(struct device_queue_manager *dqm,
 		struct qcm_process_device *qpd)
 {
-	struct queue *q, *next;
+	struct queue *q;
 	struct device_process_node *cur, *next_dpn;
 	int retval = 0;
 	bool found = false;
@@ -1637,12 +1638,19 @@ static int process_termination_nocpsch(struct device_queue_manager *dqm,
 	dqm_lock(dqm);
 
 	/* Clear all user mode queues */
-	list_for_each_entry_safe(q, next, &qpd->queues_list, list) {
+	while (!list_empty(&qpd->queues_list)) {
+		struct mqd_manager *mqd_mgr;
 		int ret;
 
Quoted 13 lines

drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c static int process_termination_nocpsch(struct device_queue_manager *dqm, · line 1653

Quoted 9 lines
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +6 -0
+		q = list_first_entry(&qpd->queues_list, struct queue, list);
+		mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type(
+				q->properties.type)];
 		ret = destroy_queue_nocpsch_locked(dqm, qpd, q);
 		if (ret)
 			retval = ret;
+		dqm_unlock(dqm);
+		mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+		dqm_lock(dqm);

This is the correct way to clean up the list when dropping the dqm-lock in the middle. Xinhui, you can use the same method in process_termination_cpsch.

yes, that is the right way to walk through the list. thanks.

Quoted 7 lines

I believe the swapping of the q->mqd with a temporary variable is not needed. When free_mqd is called, the queue is no longer on the qpd->queues_list, so destroy_queue cannot race with it. If we ensure that queues are always removed from the list before calling free_mqd, and that list-removal happens under the dqm_lock, then there should be no risk of a race condition that causes a double-free.

no, the double free exists because pqm_destroy_queue fetch the queue from qid by get_queue_by_qid() the race is like below.
pqm_destroy_queue

Quoted 11 lines
	get_queue_by_qid				process_termination_cpsch
	destroy_queue_cpsch
								lock
								list_for_each_entry_safe
									list_del(q)
								unlock
								free_mqd
	lock
	list_del(q)
	unlock
	free_mqd

I think if both those threads try to free the same queue, they both need to hold the same process->mutex. For pqm_destroy_queue that happens in kfd_ioctl_destroy_queue. For process_termination_cpsch that happens in kfd_process_notifier_release before it calls kfd_process_dequeue_from_all_devices.

Regards,
  Felix

Quoted 11 lines
Quoted 11 lines

Regards,
Felix

Quoted 3 lines
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +0 -0
 	}
 
 	/* Unregister process */

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

#4 lore raw
Quoted 117 lines

2021年6月16日 12:36,Kuehling, Felix <[email protected]> 写道:

Am 2021-06-16 um 12:01 a.m. schrieb Pan, Xinhui:

Quoted 100 lines
Quoted 74 lines

2021年6月16日 02:22,Kuehling, Felix <[email protected]> 写道:

[+Xinhui]

Am 2021-06-15 um 1:50 p.m. schrieb Amber Lin:

Quoted 10 lines

Calling free_mqd inside of destroy_queue_nocpsch_locked can cause a circular lock. destroy_queue_nocpsch_locked is called under a DQM lock, which is taken in MMU notifiers, potentially in FS reclaim context. Taking another lock, which is BO reservation lock from free_mqd, while causing an FS reclaim inside the DQM lock creates a problematic circular lock dependency. Therefore move free_mqd out of destroy_queue_nocpsch_locked and call it after unlocking DQM.

Let's submit this patch as is. I'm making some comments inline for things that Xinhui can address in his race condition patch.

Quoted 54 lines

---
.../drm/amd/amdkfd/kfd_device_queue_manager.c | 18 +++++++++++++-----

1 file changed, 13 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +7 -5
index 72bea5278add..c069fa259b30 100644
@@ -486,9 +486,6 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm,
 	if (retval == -ETIME)
 		qpd->reset_wavefronts = true;
 
-
-	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
-
 	list_del(&q->list);
 	if (list_empty(&qpd->queues_list)) {
 		if (qpd->reset_wavefronts) {
@@ -523,6 +520,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 	int retval;
 	uint64_t sdma_val = 0;
 	struct kfd_process_device *pdd = qpd_to_pdd(qpd);
+	struct mqd_manager *mqd_mgr =
+		dqm->mqd_mgrs[get_mqd_type_from_queue_type(q->properties.type)];
 
 	/* Get the SDMA queue stats */
 	if ((q->properties.type == KFD_QUEUE_TYPE_SDMA) ||
@@ -540,6 +539,8 @@ static int destroy_queue_nocpsch(struct device_queue_manager *dqm,
 		pdd->sdma_past_activity_counter += sdma_val;
 	dqm_unlock(dqm);
 
+	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+
 	return retval;
 }
 
@@ -1629,7 +1630,7 @@ static bool set_cache_memory_policy(struct device_queue_manager *dqm,
 static int process_termination_nocpsch(struct device_queue_manager *dqm,
 		struct qcm_process_device *qpd)
 {
-	struct queue *q, *next;
+	struct queue *q;
 	struct device_process_node *cur, *next_dpn;
 	int retval = 0;
 	bool found = false;
@@ -1637,12 +1638,19 @@ static int process_termination_nocpsch(struct device_queue_manager *dqm,
 	dqm_lock(dqm);
 
 	/* Clear all user mode queues */
-	list_for_each_entry_safe(q, next, &qpd->queues_list, list) {
+	while (!list_empty(&qpd->queues_list)) {
+		struct mqd_manager *mqd_mgr;
 		int ret;
 
Quoted 13 lines

drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c static int process_termination_nocpsch(struct device_queue_manager *dqm, · line 1653

Quoted 9 lines
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +6 -0
+		q = list_first_entry(&qpd->queues_list, struct queue, list);
+		mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type(
+				q->properties.type)];
 		ret = destroy_queue_nocpsch_locked(dqm, qpd, q);
 		if (ret)
 			retval = ret;
+		dqm_unlock(dqm);
+		mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
+		dqm_lock(dqm);

This is the correct way to clean up the list when dropping the dqm-lock in the middle. Xinhui, you can use the same method in process_termination_cpsch.

yes, that is the right way to walk through the list. thanks.

Quoted 7 lines

I believe the swapping of the q->mqd with a temporary variable is not needed. When free_mqd is called, the queue is no longer on the qpd->queues_list, so destroy_queue cannot race with it. If we ensure that queues are always removed from the list before calling free_mqd, and that list-removal happens under the dqm_lock, then there should be no risk of a race condition that causes a double-free.

no, the double free exists because pqm_destroy_queue fetch the queue from qid by get_queue_by_qid() the race is like below.
pqm_destroy_queue

Quoted 11 lines
	get_queue_by_qid				process_termination_cpsch
	destroy_queue_cpsch
								lock
								list_for_each_entry_safe
									list_del(q)
								unlock
								free_mqd
	lock
	list_del(q)
	unlock
	free_mqd
Quoted 5 lines

I think if both those threads try to free the same queue, they both need to hold the same process->mutex. For pqm_destroy_queue that happens in kfd_ioctl_destroy_queue. For process_termination_cpsch that happens in kfd_process_notifier_release before it calls kfd_process_dequeue_from_all_devices.

oh, yes, you are right.
So the double free I am seeing has different root cause. :(

Quoted 16 lines

Regards,
Felix

Quoted 11 lines
Quoted 11 lines

Regards,
Felix

Quoted 3 lines
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +0 -0
 	}
 
 	/* Unregister process */

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx