Hi Dave,
Is it easy to reproduce this backtrace? If it is easy, would you give
this patch a try?
From 3c83e1f05352f4795ddc2a8c9acca65b4b58ded9 Mon Sep 17 00:00:00 2001
From: Alex Xie <[email protected]>
Date: Wed, 19 Jul 2017 21:51:56 -0400
Subject: [PATCH] drm/amdgpu: Fix a warning on suspicious RCU usage
From Dave Airlie
[ 141.965723] =============================
[ 141.965724] WARNING: suspicious RCU usage
[ 141.965726] 4.12.0-rc7 #221 Not tainted
[ 141.965727] -----------------------------
[ 141.965728]
/home/airlied/devel/kernel/linux-2.6/include/linux/rcupdate.h:531
Illegal context switch in RCU read-side critical section!
[ 141.965730]
other info that might help us debug this:
[ 141.965731]
rcu_scheduler_active = 2, debug_locks = 0
[ 141.965732] 1 lock held by amdgpu_cs:0/1332:
[ 141.965733] #0: (rcu_read_lock){......}, at: [<ffffffffa01a0d07>]
amdgpu_bo_list_get+0x0/0x109 [amdgpu]
[ 141.965774]
stack backtrace:
[ 141.965776] CPU: 6 PID: 1332 Comm: amdgpu_cs:0 Not tainted 4.12.0-rc7
#221
[ 141.965777] Hardware name: To be filled by O.E.M. To be filled by
O.E.M./M5A97 R2.0, BIOS 2603 06/26/2015
[ 141.965778] Call Trace:
[ 141.965782] dump_stack+0x68/0x92
[ 141.965785] lockdep_rcu_suspicious+0xf7/0x100
[ 141.965788] ___might_sleep+0x56/0x1fc
[ 141.965790] __might_sleep+0x68/0x6f
[ 141.965793] __mutex_lock+0x4e/0x7b5
[ 141.965817] ? amdgpu_bo_list_get+0xa4/0x109 [amdgpu]
[ 141.965820] ? lock_acquire+0x125/0x1b9
[ 141.965844] ? amdgpu_bo_list_set+0x464/0x464 [amdgpu]
[ 141.965846] mutex_lock_nested+0x16/0x18
[ 141.965848] ? mutex_lock_nested+0x16/0x18
[ 141.965872] amdgpu_bo_list_get+0xa4/0x109 [amdgpu]
[ 141.965895] amdgpu_cs_ioctl+0x4a0/0x17dd [amdgpu]
[ 141.965898] ? radix_tree_node_alloc.constprop.11+0x77/0xab
[ 141.965916] drm_ioctl+0x264/0x393 [drm]
[ 141.965939] ? amdgpu_cs_find_mapping+0x83/0x83 [amdgpu]
[ 141.965942] ? trace_hardirqs_on_caller+0x16a/0x186
Signed-off-by: Alex Xie <[email protected]>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index dc87962..565ca90 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -198,12 +198,15 @@ amdgpu_bo_list_get(struct amdgpu_fpriv *fpriv, int id)
result = idr_find(&fpriv->bo_list_handles, id);
if (result) {
- if (kref_get_unless_zero(&result->refcount))
+ if (kref_get_unless_zero(&result->refcount)) {
+ rcu_read_unlock();
mutex_lock(&result->lock);
- else
+ }
+ else {
+ rcu_read_unlock();
result = NULL;
+ }
}
- rcu_read_unlock();
return result;
}
--
2.7.4
Thanks,
Alex Bin Xie
On 2017-07-19 09:21 PM, Dave Airlie wrote:
On 16 June 2017 at 23:08, Alex Xie <[email protected]> wrote:
In original function amdgpu_bo_list_get, the waiting
for result->lock can be quite long while mutex
bo_list_lock was holding. It can make other tasks
waiting for bo_list_lock for long period.
Secondly, this patch allows several tasks(readers of idr)
to proceed at the same time.
v2: use rcu and kref (Dave Airlie and Christian König)
v3: update v1 commit message (Michel Dänzer)
Got this, I'm guessing due to this patch.
Dave.
[ 141.965723] =============================
[ 141.965724] WARNING: suspicious RCU usage
[ 141.965726] 4.12.0-rc7 #221 Not tainted
[ 141.965727] -----------------------------
[ 141.965728] /home/airlied/devel/kernel/linux-2.6/include/linux/rcupdate.h:531
Illegal context switch in RCU read-side critical section!
[ 141.965730]
other info that might help us debug this:
[ 141.965731]
rcu_scheduler_active = 2, debug_locks = 0
[ 141.965732] 1 lock held by amdgpu_cs:0/1332:
[ 141.965733] #0: (rcu_read_lock){......}, at: [<ffffffffa01a0d07>]
amdgpu_bo_list_get+0x0/0x109 [amdgpu]
[ 141.965774]
stack backtrace:
[ 141.965776] CPU: 6 PID: 1332 Comm: amdgpu_cs:0 Not tainted 4.12.0-rc7 #221
[ 141.965777] Hardware name: To be filled by O.E.M. To be filled by
O.E.M./M5A97 R2.0, BIOS 2603 06/26/2015
[ 141.965778] Call Trace:
[ 141.965782] dump_stack+0x68/0x92
[ 141.965785] lockdep_rcu_suspicious+0xf7/0x100
[ 141.965788] ___might_sleep+0x56/0x1fc
[ 141.965790] __might_sleep+0x68/0x6f
[ 141.965793] __mutex_lock+0x4e/0x7b5
[ 141.965817] ? amdgpu_bo_list_get+0xa4/0x109 [amdgpu]
[ 141.965820] ? lock_acquire+0x125/0x1b9
[ 141.965844] ? amdgpu_bo_list_set+0x464/0x464 [amdgpu]
[ 141.965846] mutex_lock_nested+0x16/0x18
[ 141.965848] ? mutex_lock_nested+0x16/0x18
[ 141.965872] amdgpu_bo_list_get+0xa4/0x109 [amdgpu]
[ 141.965895] amdgpu_cs_ioctl+0x4a0/0x17dd [amdgpu]
[ 141.965898] ? radix_tree_node_alloc.constprop.11+0x77/0xab
[ 141.965916] drm_ioctl+0x264/0x393 [drm]
[ 141.965939] ? amdgpu_cs_find_mapping+0x83/0x83 [amdgpu]
[ 141.965942] ? trace_hardirqs_on_caller+0x16a/0x186
Signed-off-by: Alex Xie <[email protected]>
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 40 ++++++++++++++++++++---------
2 files changed, 30 insertions(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 063fc73..e9b3981 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -871,6 +871,8 @@ struct amdgpu_fpriv {
struct amdgpu_bo_list {
struct mutex lock;
+ struct rcu_head rhead;
+ struct kref refcount;
struct amdgpu_bo *gds_obj;
struct amdgpu_bo *gws_obj;
struct amdgpu_bo *oa_obj;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index 5af956f..efa6903 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -41,6 +41,20 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev,
struct drm_amdgpu_bo_list_entry *info,
unsigned num_entries);
+static void amdgpu_bo_list_release_rcu(struct kref *ref)
+{
+ unsigned i;
+ struct amdgpu_bo_list *list = container_of(ref, struct amdgpu_bo_list,
+ refcount);
+
+ for (i = 0; i < list->num_entries; ++i)
+ amdgpu_bo_unref(&list->array[i].robj);
+
+ mutex_destroy(&list->lock);
+ drm_free_large(list->array);
+ kfree_rcu(list, rhead);
+}
+
static int amdgpu_bo_list_create(struct amdgpu_device *adev,
struct drm_file *filp,
struct drm_amdgpu_bo_list_entry *info,
@@ -57,7 +71,7 @@ static int amdgpu_bo_list_create(struct amdgpu_device *adev,
/* initialize bo list*/
mutex_init(&list->lock);
-
+ kref_init(&list->refcount);
r = amdgpu_bo_list_set(adev, filp, list, info, num_entries);
if (r) {
kfree(list);
@@ -83,14 +97,9 @@ static void amdgpu_bo_list_destroy(struct amdgpu_fpriv
*fpriv, int id)
mutex_lock(&fpriv->bo_list_lock);
list = idr_remove(&fpriv->bo_list_handles, id);
- if (list) {
- /* Another user may have a reference to this list still */
- mutex_lock(&list->lock);
- mutex_unlock(&list->lock);
- amdgpu_bo_list_free(list);
- }
-
mutex_unlock(&fpriv->bo_list_lock);
+ if (list)
+ kref_put(&list->refcount, amdgpu_bo_list_release_rcu);
}
static int amdgpu_bo_list_set(struct amdgpu_device *adev,
@@ -185,11 +194,17 @@ amdgpu_bo_list_get(struct amdgpu_fpriv *fpriv, int id)
{
struct amdgpu_bo_list *result;
- mutex_lock(&fpriv->bo_list_lock);
+ rcu_read_lock();
result = idr_find(&fpriv->bo_list_handles, id);
- if (result)
- mutex_lock(&result->lock);
- mutex_unlock(&fpriv->bo_list_lock);
+
+ if (result) {
+ if (kref_get_unless_zero(&result->refcount))
+ mutex_lock(&result->lock);
+ else
+ result = NULL;
+ }
+ rcu_read_unlock();
+
return result;
}
@@ -227,6 +242,7 @@ void amdgpu_bo_list_get_list(struct amdgpu_bo_list *list,
void amdgpu_bo_list_put(struct amdgpu_bo_list *list)
{
mutex_unlock(&list->lock);
+ kref_put(&list->refcount, amdgpu_bo_list_release_rcu);
}
void amdgpu_bo_list_free(struct amdgpu_bo_list *list)
--
2.7.4
_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
