drm: mm always protect change to unused_nodes with unused_lock spinlock
Subsystems (2)
PATCH 2/2 drm: mm always protect change to unused_nodes with unused_lock spinlock
unused_nodes modification needs to be protected by unused_lock spinlock. Here is an example of an usage where there is no such protection without this patch.
Process 1: 1-drm_mm_pre_get(this function modify unused_nodes list)
2-spin_lock(spinlock protecting mm struct) 3-drm_mm_put_block(this function might modify unused_nodes
list but doesn't protect modification with unused_lock)
4-spin_unlock(spinlock protecting mm struct)
Process2: 1-drm_mm_pre_get(this function modify unused_nodes list)
At this point Process1 & Process2 might both be doing modification to unused_nodes list. This patch add unused_lock protection into drm_mm_put_block to avoid such issue.
--- drivers/gpu/drm/drm_mm.c | 9 +++++++++ 1 files changed, 9 insertions(+), 0 deletions(-)
drivers/gpu/drm/drm_mm.c +9 -0
@@ -103,6 +103,11 @@ static struct drm_mm_node *drm_mm_kmalloc(struct drm_mm *mm, int atomic)return child; } +/* drm_mm_pre_get() - pre allocate drm_mm_node structure +* drm_mm: memory manager struct we are pre-allocating for +* +* Returns 0 on success or -ENOMEM if allocation fails. +*/ int drm_mm_pre_get (struct drm_mm *mm) {struct drm_mm_node *node;
@@ -253,12 +258,14 @@ void drm_mm_put_block(struct drm_mm_node *cur) prev_node->size += next_node->size;list_del (&next_node->ml_entry);list_del (&next_node->fl_entry); +spin_lock (&mm->unused_lock);if (mm->num_unused <MM_UNUSED_TARGET ) {list_add (&next_node->fl_entry, &mm->unused_nodes); ++mm->num_unused; }else kfree (next_node); +spin_unlock (&mm->unused_lock); }else { next_node->size += cur->size; next_node->start = cur->start;
@@ -271,11 +278,13 @@ void drm_mm_put_block(struct drm_mm_node *cur)list_add (&cur->fl_entry, &mm->fl_entry); }else {list_del (&cur->ml_entry); +spin_lock (&mm->unused_lock);if (mm->num_unused <MM_UNUSED_TARGET ) {list_add (&cur->fl_entry, &mm->unused_nodes); ++mm->num_unused; }else kfree (cur); +spin_unlock (&mm->unused_lock); } }
Signature · 1 line
1.6.5.2
Dave Airlie
maintainerQuoted 4 lines · Jerome Glisse
unused_nodesmodification needs to be protected byunused_lockspinlock. Here is an example of an usage where there is no such protection without this patch.
Quoted 9 lines
Process 1: 1-drm_mm_pre_get(this function modify
unused_nodeslist) 2-spin_lock(spinlock protecting mm struct) 3-drm_mm_put_block(this function might modifyunused_nodeslist but doesn't protect modification withunused_lock) 4-spin_unlock(spinlock protecting mm struct)
Process2: 1-drm_mm_pre_get(this function modifyunused_nodeslist) At this point Process1 & Process2 might both be doing modification tounused_nodeslist. This patch addunused_lockprotection intodrm_mm_put_blockto avoid such issue.
Have we got a bug number or reproducer for this?
I've cc'ed Thomas and Chris who were last ppl to touch drm_mm.c for some sort of acks.
Dave.
Quoted 55 lines
Signed-off-by Jerome Glisse <[email protected]>---
drivers/gpu/drm/drm_mm.c | 9 +++++++++1 files changed, 9 insertions(+), 0 deletions(-)drivers/gpu/drm/drm_mm.c +9
-1@@ -103,6 +103,11 @@ static struct drm_mm_node *drm_mm_kmalloc(struct drm_mm *mm, int atomic) return child; } +/* drm_mm_pre_get() - pre allocate drm_mm_node structure +* drm_mm: memory manager struct we are pre-allocating for +* +* Returns 0 on success or -ENOMEM if allocation fails. +*/ int drm_mm_pre_get(struct drm_mm *mm) { struct drm_mm_node *node;@@ -253,12 +258,14 @@ void drm_mm_put_block(struct drm_mm_node *cur) prev_node->size += next_node->size; list_del(&next_node->ml_entry); list_del(&next_node->fl_entry); + spin_lock(&mm->unused_lock); if (mm->num_unused < MM_UNUSED_TARGET) { list_add(&next_node->fl_entry, &mm->unused_nodes); ++mm->num_unused; } else kfree(next_node); + spin_unlock(&mm->unused_lock); } else { next_node->size += cur->size; next_node->start = cur->start;@@ -271,11 +278,13 @@ void drm_mm_put_block(struct drm_mm_node *cur) list_add(&cur->fl_entry, &mm->fl_entry); } else { list_del(&cur->ml_entry); + spin_lock(&mm->unused_lock); if (mm->num_unused < MM_UNUSED_TARGET) { list_add(&cur->fl_entry, &mm->unused_nodes); ++mm->num_unused; } else kfree(cur); + spin_unlock(&mm->unused_lock); } } --1.6.5.2
Quoted 15 lines · Dave Airlie
Quoted 4 lines · Jerome Glisse
unused_nodesmodification needs to be protected byunused_lockspinlock. Here is an example of an usage where there is no such protection without this patch.Quoted 9 lines
Process 1: 1-drm_mm_pre_get(this function modify
unused_nodeslist)
2-spin_lock(spinlock protecting mm struct) 3-drm_mm_put_block(this function might modifyunused_nodes
list but doesn't protect modification withunused_lock)
4-spin_unlock(spinlock protecting mm struct)
Process2: 1-drm_mm_pre_get(this function modifyunused_nodeslist)
At this point Process1 & Process2 might both be doing modification tounused_nodeslist. This patch addunused_lockprotection intodrm_mm_put_blockto avoid such issue.
Quoted 6 lines
Have we got a bug number or reproducer for this?
I've cc'ed Thomas and Chris who were last ppl to touch
drm_mm.cfor some sort of acks.Dave.
No bug, this comes from code review while working on TTM. I think my analysis is correct.
Cheers,
Jerome
Quoted 75 lines · Dave Airlie
Quoted 4 lines · Jerome Glisse
unused_nodesmodification needs to be protected byunused_lockspinlock. Here is an example of an usage where there is no such protection without this patch.Quoted 10 lines
Process 1: 1-drm_mm_pre_get(this function modify
unused_nodeslist)
2-spin_lock(spinlock protecting mm struct) 3-drm_mm_put_block(this function might modifyunused_nodes
list but doesn't protect modification withunused_lock)
4-spin_unlock(spinlock protecting mm struct)
Process2: 1-drm_mm_pre_get(this function modifyunused_nodeslist)
At this point Process1 & Process2 might both be doing modification tounused_nodeslist. This patch addunused_lockprotection intodrm_mm_put_blockto avoid such issue.Have we got a bug number or reproducer for this?
I've cc'ed Thomas and Chris who were last ppl to touch
drm_mm.cfor some sort of acks.Dave.
Quoted 50 lines
Signed-off-by Jerome Glisse <[email protected]>--- drivers/gpu/drm/drm_mm.c | 9 +++++++++ 1 files changed, 9 insertions(+), 0 deletions(-)drivers/gpu/drm/drm_mm.c +9
-0@@ -103,6 +103,11 @@ static struct drm_mm_node *drm_mm_kmalloc(struct drm_mm *mm, int atomic)return child; } +/* drm_mm_pre_get() - pre allocate drm_mm_node structure +* drm_mm: memory manager struct we are pre-allocating for +* +* Returns 0 on success or -ENOMEM if allocation fails. +*/ int drm_mm_pre_get (struct drm_mm *mm) {struct drm_mm_node *node;@@ -253,12 +258,14 @@ void drm_mm_put_block(struct drm_mm_node *cur) prev_node->size += next_node->size;list_del (&next_node->ml_entry);list_del (&next_node->fl_entry); +spin_lock (&mm->unused_lock);if (mm->num_unused <MM_UNUSED_TARGET ) {list_add (&next_node->fl_entry, &mm->unused_nodes); ++mm->num_unused; }else kfree (next_node); +spin_unlock (&mm->unused_lock); }else { next_node->size += cur->size; next_node->start = cur->start;@@ -271,11 +278,13 @@ void drm_mm_put_block(struct drm_mm_node *cur)list_add (&cur->fl_entry, &mm->fl_entry); }else {list_del (&cur->ml_entry); +spin_lock (&mm->unused_lock);if (mm->num_unused <MM_UNUSED_TARGET ) {list_add (&cur->fl_entry, &mm->unused_nodes); ++mm->num_unused; }else kfree (cur); +spin_unlock (&mm->unused_lock); } }
Quoted 5 lines
Quoted 5 lines
--
1.6.5.2
Hmm. Ouch. The patch looks correct, although I'm not 100% sure it's OK to kfree() within a spinlocked region? Perhaps better to take it out.
/Thomas
Quoted 76 lines · Thomas Hellstrom
Quoted 75 lines · Dave Airlie
Quoted 4 lines · Jerome Glisse
unused_nodesmodification needs to be protected byunused_lockspinlock. Here is an example of an usage where there is no such protection without this patch.Quoted 10 lines
Process 1: 1-drm_mm_pre_get(this function modify
unused_nodeslist)
2-spin_lock(spinlock protecting mm struct) 3-drm_mm_put_block(this function might modifyunused_nodes
list but doesn't protect modification withunused_lock)
4-spin_unlock(spinlock protecting mm struct)
Process2: 1-drm_mm_pre_get(this function modifyunused_nodeslist)
At this point Process1 & Process2 might both be doing modification tounused_nodeslist. This patch addunused_lockprotection intodrm_mm_put_blockto avoid such issue.Have we got a bug number or reproducer for this?
I've cc'ed Thomas and Chris who were last ppl to touch
drm_mm.cfor some sort of acks.Dave.
Quoted 50 lines
Signed-off-by Jerome Glisse <[email protected]>--- drivers/gpu/drm/drm_mm.c | 9 +++++++++ 1 files changed, 9 insertions(+), 0 deletions(-)drivers/gpu/drm/drm_mm.c +9
-0@@ -103,6 +103,11 @@ static struct drm_mm_node *drm_mm_kmalloc(struct drm_mm *mm, int atomic)return child; } +/* drm_mm_pre_get() - pre allocate drm_mm_node structure +* drm_mm: memory manager struct we are pre-allocating for +* +* Returns 0 on success or -ENOMEM if allocation fails. +*/ int drm_mm_pre_get (struct drm_mm *mm) {struct drm_mm_node *node;@@ -253,12 +258,14 @@ void drm_mm_put_block(struct drm_mm_node *cur) prev_node->size += next_node->size;list_del (&next_node->ml_entry);list_del (&next_node->fl_entry); +spin_lock (&mm->unused_lock);if (mm->num_unused <MM_UNUSED_TARGET ) {list_add (&next_node->fl_entry, &mm->unused_nodes); ++mm->num_unused; }else kfree (next_node); +spin_unlock (&mm->unused_lock); }else { next_node->size += cur->size; next_node->start = cur->start;@@ -271,11 +278,13 @@ void drm_mm_put_block(struct drm_mm_node *cur)list_add (&cur->fl_entry, &mm->fl_entry); }else {list_del (&cur->ml_entry); +spin_lock (&mm->unused_lock);if (mm->num_unused <MM_UNUSED_TARGET ) {list_add (&cur->fl_entry, &mm->unused_nodes); ++mm->num_unused; }else kfree (cur); +spin_unlock (&mm->unused_lock); } }
Quoted 7 lines
Quoted 5 lines
Quoted 5 lines
--
1.6.5.2Hmm. Ouch. The patch looks correct, although I'm not 100% sure it's OK to
kfree()within a spinlocked region? Perhaps better to take it out.
Would kfree() possibly sleep? I wouldn't think so, if not it should be safe.
robert.
Quoted 15 lines
/Thomas
------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel
Signature · 2 lines
Robert Noland <[email protected]> 2Hip Networks