Skip to contentSkip to conversation

mm/percpu: Fix possible NOFS/NOIO reclaim recursion

PATCH Applied Kaitao Cheng 2 participants 10 messages Jun 18 – Jun 19

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v4 cover mm/percpu: Fix possible NOFS/NOIO reclaim recursion

Commit 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic") allowed GFP_NOFS and GFP_NOIO percpu allocations to use pcpu_alloc_mutex and the chunk creation slow path. This restored the allocation capability that was lost when those constrained allocations were treated as atomic, but it also makes the percpu slow path visible to callers from constrained reclaim contexts.

There are two related problems.

First, the create and populate slow paths do not fully preserve the caller's allocation constraints. pcpu_alloc_noprof() derives pcpu_gfp from the caller supplied GFP mask and passes it down to the percpu backing page allocator. However, chunk creation calls pcpu_get_vm_areas(), and chunk population can allocate temporary metadata or vmalloc page tables while mapping backing pages. Those internal allocations can still use GFP_KERNEL, so a caller using GFP_NOFS or GFP_NOIO can enter unconstrained FS or IO reclaim while holding pcpu_alloc_mutex.

One possible case is blk-cgroup after commit 5d726c4dbeed ("blk-cgroup: fix possible deadlock while configuring policy"). blkg_conf_prep() now serializes against blkcg_deactivate_policy() with q->blkcg_mutex, and blkg_alloc() uses GFP_NOIO because queue freeze and IO reclaim dependencies can otherwise deadlock. If the percpu slow path loses that GFP_NOIO context, direct reclaim or writeback can issue IO to a frozen queue while q->blkcg_mutex is held.

Second, allowing sleepable GFP_NOFS/GFP_NOIO allocations to take pcpu_alloc_mutex means that unconstrained backing allocations made under the mutex can create an FS/IO reclaim dependency against a constrained caller which already holds an FS or IO lock and then waits for pcpu_alloc_mutex.

This series fixes those issues in three steps:

- pass the caller supplied GFP mask into pcpu_get_vm_areas() and use it

    for vmalloc metadata and KASAN shadow allocations;
  - pass the GFP mask through the chunk population path, including the
    temporary pages array and vmalloc page table allocation scope;

- restrict percpu backing allocations performed while holding pcpu_alloc_mutex to GFP_NOIO, so they cannot recurse into IO or FS reclaim.

This keeps sleepable GFP_NOFS/GFP_NOIO percpu allocations working, while avoiding the reclaim recursion risks introduced by making those allocations eligible for the mutex-protected slow path.

Changes in v4:
- Make cached pages lookup explicit; (Dennis Zhou)
- Remove the changes in v3 and add a comment explaining why __GFP_NOFAIL must not be passed. (Dennis Zhou, Michal Hocko, Andrew Morton)

Changes in v3:
- Allow @gfp to pass __GFP_NOFAIL through.

Changes in v2:
- split the previous first patch into vmalloc-area creation and chunk population changes; (Pedro Falcato)
- pass the GFP mask explicitly to pcpu_get_vm_areas(); (Pedro Falcato)
- apply the corresponding memalloc scope around vmalloc page table allocation during chunk population;
- replace the reclaim recursion avoidance with a GFP_NOIO backing allocation mask instead of only rejecting nested reclaim. (Michal Hocko)

Link to v3:
https://lore.kernel.org/all/[email protected]/

Link to v2:
https://lore.kernel.org/all/[email protected]/

Link to v1:
https://lore.kernel.org/all/[email protected]/

Kaitao Cheng (4):
mm/vmalloc: honor GFP constraints in pcpu_get_vm_areas() mm/percpu: honor GFP constraints when populating chunks mm/percpu: Make cached pages lookup explicit mm/percpu: Avoid IO/FS reclaim in backing allocations

 include/linux/vmalloc.h |  4 ++--
 mm/percpu-vm.c          | 48 +++++++++++++++++++++++++++++------------
 mm/percpu.c             | 20 +++++++++++------
 mm/vmalloc.c            | 23 ++++++++++----------
 4 files changed, 62 insertions(+), 33 deletions(-)
Signature · 1 line
2.50.1 (Apple Git-155)
#1 lore raw

PATCH v4 1/4 mm/vmalloc: honor GFP constraints in pcpu_get_vm_areas()

From: Kaitao Cheng <[email protected]>

pcpu_alloc_noprof() derives pcpu_gfp from the caller supplied GFP mask and passes it down to the backing percpu allocator. However, when the percpu vmalloc allocator has to create a new chunk, pcpu_create_chunk() calls pcpu_get_vm_areas() to allocate the corresponding vmalloc areas.

pcpu_get_vm_areas() currently performs its internal allocations with GFP_KERNEL, including vmap area metadata, vm_struct metadata and KASAN vmalloc shadow population. This means that a caller which deliberately uses GFP_NOFS or GFP_NOIO can still enter FS or IO reclaim while creating the vmalloc areas for a new percpu chunk.

One possible case is blk-cgroup after commit 5d726c4dbeed ("blk-cgroup: fix possible deadlock while configuring policy"). blkg_conf_prep() now serializes against blkcg_deactivate_policy() with q->blkcg_mutex, and blkg_alloc() was changed to GFP_NOIO for that reason:

CPU0: blkg_conf_prep()

    mutex_lock(q->blkcg_mutex)
    blkg_alloc(..., GFP_NOIO)
      alloc_percpu_gfp(..., GFP_NOIO)
        pcpu_alloc_noprof(..., GFP_NOIO)
	  pcpu_create_chunk(GFP_NOIO)
	    pcpu_get_vm_areas()
              -> if percpu chunks are exhausted, chunk create may do
                 internal GFP_KERNEL allocations
              -> direct reclaim / writeback can issue IO to this queue
              -> IO waits because the queue is frozen

CPU1: blkcg_deactivate_policy()

    blk_mq_freeze_queue(q)
    mutex_lock(q->blkcg_mutex)
      -> waits for CPU0
    ... unfreeze only happens after q->blkcg_mutex is acquired/released

So the concern is that the caller deliberately uses GFP_NOIO because it may hold a lock which can be acquired after queue freeze, but the percpu slow path can temporarily lose that allocation context.

Pass the caller supplied GFP mask from pcpu_create_chunk() to pcpu_get_vm_areas(), and use it for the internal vmalloc metadata and KASAN shadow allocations.

Fixes 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic")
---
 include/linux/vmalloc.h |  4 ++--
 mm/percpu-vm.c          |  2 +-
 mm/vmalloc.c            | 23 ++++++++++++-----------
 3 files changed, 15 insertions(+), 14 deletions(-)
include/linux/vmalloc.h +2 -2
index d87dc7f77f4e..e4d8d0a9f30f 100644
@@ -310,14 +310,14 @@ static inline void set_vm_flush_reset_perms(void *addr) {}
 #if defined(CONFIG_MMU) && defined(CONFIG_SMP)
 struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 				     const size_t *sizes, int nr_vms,
-				     size_t align);
+				     size_t align, gfp_t gfp);
 
 void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms);
 # else
 static inline struct vm_struct **
 pcpu_get_vm_areas(const unsigned long *offsets,
 		const size_t *sizes, int nr_vms,
-		size_t align)
+		size_t align, gfp_t gfp)
 {
 	return NULL;
 }
mm/percpu-vm.c +1 -1
index 4f5937090590..69b00741dc68 100644
@@ -340,7 +340,7 @@ static struct pcpu_chunk *pcpu_create_chunk(gfp_t gfp)
 		return NULL;
 
 	vms = pcpu_get_vm_areas(pcpu_group_offsets, pcpu_group_sizes,
-				pcpu_nr_groups, pcpu_atom_size);
+				pcpu_nr_groups, pcpu_atom_size, gfp);
 	if (!vms) {
 		pcpu_free_chunk(chunk);
 		return NULL;
mm/vmalloc.c +12 -11
index 1afca3568b9b..08f468135e4d 100644
@@ -4946,16 +4946,17 @@ pvm_determine_end_from_reverse(struct vmap_area **va, unsigned long align)
  * @sizes: array containing size of each area
  * @nr_vms: the number of areas to allocate
  * @align: alignment, all entries in @offsets and @sizes must be aligned to this
+ * @gfp: allocation flags passed to the underlying memory allocator
  *
  * Returns: kmalloc'd vm_struct pointer array pointing to allocated
  *	    vm_structs on success, %NULL on failure
  *
  * Percpu allocator wants to use congruent vm areas so that it can
  * maintain the offsets among percpu areas.  This function allocates
- * congruent vmalloc areas for it with GFP_KERNEL.  These areas tend to
- * be scattered pretty far, distance between two areas easily going up
- * to gigabytes.  To avoid interacting with regular vmallocs, these
- * areas are allocated from top.
+ * congruent vmalloc areas for it. These areas tend to be scattered
+ * pretty far, distance between two areas easily going up to gigabytes.
+ * To avoid interacting with regular vmallocs, these areas are allocated
+ * from top.
  *
  * Despite its complicated look, this allocator is rather simple. It
  * does everything top-down and scans free blocks from the end looking
@@ -4966,7 +4967,7 @@ pvm_determine_end_from_reverse(struct vmap_area **va, unsigned long align)
  */
 struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 				     const size_t *sizes, int nr_vms,
-				     size_t align)
+				     size_t align, gfp_t gfp)
 {
 	const unsigned long vmalloc_start = ALIGN(VMALLOC_START, align);
 	const unsigned long vmalloc_end = VMALLOC_END & ~(align - 1);
@@ -5004,14 +5005,14 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 		return NULL;
 	}
 
-	vms = kzalloc_objs(vms[0], nr_vms);
-	vas = kzalloc_objs(vas[0], nr_vms);
+	vms = kzalloc_objs(vms[0], nr_vms, gfp);
+	vas = kzalloc_objs(vas[0], nr_vms, gfp);
 	if (!vas || !vms)
 		goto err_free2;
 
 	for (area = 0; area < nr_vms; area++) {
-		vas[area] = kmem_cache_zalloc(vmap_area_cachep, GFP_KERNEL);
-		vms[area] = kzalloc_obj(struct vm_struct);
+		vas[area] = kmem_cache_zalloc(vmap_area_cachep, gfp);
+		vms[area] = kzalloc_obj(struct vm_struct, gfp);
 		if (!vas[area] || !vms[area])
 			goto err_free;
 	}
@@ -5101,7 +5102,7 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 
 	/* populate the kasan shadow space */
 	for (area = 0; area < nr_vms; area++) {
-		if (kasan_populate_vmalloc(vas[area]->va_start, sizes[area], GFP_KERNEL))
+		if (kasan_populate_vmalloc(vas[area]->va_start, sizes[area], gfp))
 			goto err_free_shadow;
 	}
 
@@ -5158,7 +5159,7 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 				continue;
 
 			vas[area] = kmem_cache_zalloc(
-				vmap_area_cachep, GFP_KERNEL);
+				vmap_area_cachep, gfp);
 			if (!vas[area])
 				goto err_free;
 		}
Signature · 1 line
2.50.1 (Apple Git-155)
Acked-by on v4 1/4
#2 lore raw

Re: [PATCH v4 1/4] mm/vmalloc: honor GFP constraints in pcpu_get_vm_areas()

Quoted 44 lines · Kaitao Cheng

From: Kaitao Cheng <[email protected]>

pcpu_alloc_noprof() derives pcpu_gfp from the caller supplied GFP mask and passes it down to the backing percpu allocator. However, when the percpu vmalloc allocator has to create a new chunk, pcpu_create_chunk() calls pcpu_get_vm_areas() to allocate the corresponding vmalloc areas.

pcpu_get_vm_areas() currently performs its internal allocations with GFP_KERNEL, including vmap area metadata, vm_struct metadata and KASAN vmalloc shadow population. This means that a caller which deliberately uses GFP_NOFS or GFP_NOIO can still enter FS or IO reclaim while creating the vmalloc areas for a new percpu chunk.

One possible case is blk-cgroup after commit 5d726c4dbeed ("blk-cgroup: fix possible deadlock while configuring policy"). blkg_conf_prep() now serializes against blkcg_deactivate_policy() with q->blkcg_mutex, and blkg_alloc() was changed to GFP_NOIO for that reason:

CPU0: blkg_conf_prep()

    mutex_lock(q->blkcg_mutex)
    blkg_alloc(..., GFP_NOIO)
      alloc_percpu_gfp(..., GFP_NOIO)
        pcpu_alloc_noprof(..., GFP_NOIO)
	  pcpu_create_chunk(GFP_NOIO)
	    pcpu_get_vm_areas()
              -> if percpu chunks are exhausted, chunk create may do
                 internal GFP_KERNEL allocations
              -> direct reclaim / writeback can issue IO to this queue
              -> IO waits because the queue is frozen

CPU1: blkcg_deactivate_policy()

    blk_mq_freeze_queue(q)
    mutex_lock(q->blkcg_mutex)
      -> waits for CPU0
    ... unfreeze only happens after q->blkcg_mutex is acquired/released

So the concern is that the caller deliberately uses GFP_NOIO because it may hold a lock which can be acquired after queue freeze, but the percpu slow path can temporarily lose that allocation context.

Pass the caller supplied GFP mask from pcpu_create_chunk() to pcpu_get_vm_areas(), and use it for the internal vmalloc metadata and KASAN shadow allocations.

Quoted 5 lines
Fixes 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic")

LGTM

Thanks for fixing this!

Quoted 115 lines
---
 include/linux/vmalloc.h |  4 ++--
 mm/percpu-vm.c          |  2 +-
 mm/vmalloc.c            | 23 ++++++++++++-----------
 3 files changed, 15 insertions(+), 14 deletions(-)
include/linux/vmalloc.h +2 -2
index d87dc7f77f4e..e4d8d0a9f30f 100644
@@ -310,14 +310,14 @@ static inline void set_vm_flush_reset_perms(void *addr) {}
 #if defined(CONFIG_MMU) && defined(CONFIG_SMP)
 struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 				     const size_t *sizes, int nr_vms,
-				     size_t align);
+				     size_t align, gfp_t gfp);
 
 void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms);
 # else
 static inline struct vm_struct **
 pcpu_get_vm_areas(const unsigned long *offsets,
 		const size_t *sizes, int nr_vms,
-		size_t align)
+		size_t align, gfp_t gfp)
 {
 	return NULL;
 }
mm/percpu-vm.c +1 -1
index 4f5937090590..69b00741dc68 100644
@@ -340,7 +340,7 @@ static struct pcpu_chunk *pcpu_create_chunk(gfp_t gfp)
 		return NULL;
 
 	vms = pcpu_get_vm_areas(pcpu_group_offsets, pcpu_group_sizes,
-				pcpu_nr_groups, pcpu_atom_size);
+				pcpu_nr_groups, pcpu_atom_size, gfp);
 	if (!vms) {
 		pcpu_free_chunk(chunk);
 		return NULL;
mm/vmalloc.c +12 -11
index 1afca3568b9b..08f468135e4d 100644
@@ -4946,16 +4946,17 @@ pvm_determine_end_from_reverse(struct vmap_area **va, unsigned long align)
  * @sizes: array containing size of each area
  * @nr_vms: the number of areas to allocate
  * @align: alignment, all entries in @offsets and @sizes must be aligned to this
+ * @gfp: allocation flags passed to the underlying memory allocator
  *
  * Returns: kmalloc'd vm_struct pointer array pointing to allocated
  *	    vm_structs on success, %NULL on failure
  *
  * Percpu allocator wants to use congruent vm areas so that it can
  * maintain the offsets among percpu areas.  This function allocates
- * congruent vmalloc areas for it with GFP_KERNEL.  These areas tend to
- * be scattered pretty far, distance between two areas easily going up
- * to gigabytes.  To avoid interacting with regular vmallocs, these
- * areas are allocated from top.
+ * congruent vmalloc areas for it. These areas tend to be scattered
+ * pretty far, distance between two areas easily going up to gigabytes.
+ * To avoid interacting with regular vmallocs, these areas are allocated
+ * from top.
  *
  * Despite its complicated look, this allocator is rather simple. It
  * does everything top-down and scans free blocks from the end looking
@@ -4966,7 +4967,7 @@ pvm_determine_end_from_reverse(struct vmap_area **va, unsigned long align)
  */
 struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 				     const size_t *sizes, int nr_vms,
-				     size_t align)
+				     size_t align, gfp_t gfp)
 {
 	const unsigned long vmalloc_start = ALIGN(VMALLOC_START, align);
 	const unsigned long vmalloc_end = VMALLOC_END & ~(align - 1);
@@ -5004,14 +5005,14 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 		return NULL;
 	}
 
-	vms = kzalloc_objs(vms[0], nr_vms);
-	vas = kzalloc_objs(vas[0], nr_vms);
+	vms = kzalloc_objs(vms[0], nr_vms, gfp);
+	vas = kzalloc_objs(vas[0], nr_vms, gfp);
 	if (!vas || !vms)
 		goto err_free2;
 
 	for (area = 0; area < nr_vms; area++) {
-		vas[area] = kmem_cache_zalloc(vmap_area_cachep, GFP_KERNEL);
-		vms[area] = kzalloc_obj(struct vm_struct);
+		vas[area] = kmem_cache_zalloc(vmap_area_cachep, gfp);
+		vms[area] = kzalloc_obj(struct vm_struct, gfp);
 		if (!vas[area] || !vms[area])
 			goto err_free;
 	}
@@ -5101,7 +5102,7 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 
 	/* populate the kasan shadow space */
 	for (area = 0; area < nr_vms; area++) {
-		if (kasan_populate_vmalloc(vas[area]->va_start, sizes[area], GFP_KERNEL))
+		if (kasan_populate_vmalloc(vas[area]->va_start, sizes[area], gfp))
 			goto err_free_shadow;
 	}
 
@@ -5158,7 +5159,7 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 				continue;
 
 			vas[area] = kmem_cache_zalloc(
-				vmap_area_cachep, GFP_KERNEL);
+				vmap_area_cachep, gfp);
 			if (!vas[area])
 				goto err_free;
 		}
Signature · 1 line
2.50.1 (Apple Git-155)
Signature · 2 lines
Michal Hocko
SUSE Labs
#3 lore raw

PATCH v4 2/4 mm/percpu: honor GFP constraints when populating chunks

In reply to Kaitao Cheng

From: Kaitao Cheng <[email protected]>

pcpu_alloc_noprof() derives pcpu_gfp from the caller supplied GFP mask and passes it down to pcpu_populate_chunk(). pcpu_alloc_pages() already uses that mask for backing page allocation.

However, the populate slow path still has internal allocations and page table allocations which can lose the caller's allocation context. The temporary pages array is allocated by pcpu_get_pages() with GFP_KERNEL, and pcpu_map_pages() maps the backing pages through vmap_pages_range_noflush() using GFP_KERNEL. The latter can allocate vmalloc page tables implicitly, so a caller which deliberately uses GFP_NOFS or GFP_NOIO can still enter FS or IO reclaim while populating a percpu chunk.

This has the same concern as chunk creation: callers such as blk-cgroup may use GFP_NOIO because they hold locks which can be involved in queue freeze or IO reclaim dependencies. If an allocation reaches the percpu slow path and needs to populate previously unbacked pages, the internal GFP_KERNEL allocations can defeat that context.

One possible case is blk-cgroup after commit 5d726c4dbeed ("blk-cgroup: fix possible deadlock while configuring policy"). blkg_conf_prep() now serializes against blkcg_deactivate_policy() with q->blkcg_mutex, and blkg_alloc() was changed to GFP_NOIO for that reason:

CPU0: blkg_conf_prep()

    mutex_lock(q->blkcg_mutex)
    blkg_alloc(..., GFP_NOIO)
      alloc_percpu_gfp(..., GFP_NOIO)
        pcpu_alloc_noprof(..., GFP_NOIO)
          pcpu_populate_chunk(GFP_NOIO)
            pcpu_get_pages()
	    pcpu_map_pages()
              -> if the selected percpu chunk has unpopulated pages,
	         chunk population may do internal GFP_KERNEL allocations
              -> direct reclaim / writeback can issue IO to this queue
              -> IO waits because the queue is frozen

CPU1: blkcg_deactivate_policy()

    blk_mq_freeze_queue(q)
    mutex_lock(q->blkcg_mutex)
      -> waits for CPU0
    ... unfreeze only happens after q->blkcg_mutex is acquired/released

So the concern is that the caller deliberately uses GFP_NOIO because it may hold a lock which can be acquired after queue freeze, but the percpu slow path can temporarily lose that allocation context.

Pass pcpu_gfp through pcpu_get_pages(), pcpu_map_pages() and __pcpu_map_pages(). Apply the corresponding memalloc scope around vmap_pages_range_noflush(), because vmalloc page table allocation does not pass the GFP mask down explicitly. Keep the first chunk setup path using GFP_KERNEL, matching the previous early-init behavior.

Fixes 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic")
---
 mm/percpu-vm.c | 38 ++++++++++++++++++++++++++------------
 mm/percpu.c    |  2 +-
 2 files changed, 27 insertions(+), 13 deletions(-)
mm/percpu-vm.c +26 -12
index 69b00741dc68..ccd03cc152d4 100644
@@ -21,6 +21,7 @@ static struct page *pcpu_chunk_page(struct pcpu_chunk *chunk,
 
 /**
  * pcpu_get_pages - get temp pages array
+ * @gfp: allocation flags passed to the underlying allocator
  *
  * Returns pointer to array of pointers to struct page which can be indexed
  * with pcpu_page_idx().  Note that there is only one array and accesses
@@ -29,7 +30,7 @@ static struct page *pcpu_chunk_page(struct pcpu_chunk *chunk,
  * RETURNS:
  * Pointer to temp pages array on success.
  */
-static struct page **pcpu_get_pages(void)
+static struct page **pcpu_get_pages(gfp_t gfp)
 {
 	static struct page **pages;
 	size_t pages_size = pcpu_nr_units * pcpu_unit_pages * sizeof(pages[0]);
@@ -37,7 +38,7 @@ static struct page **pcpu_get_pages(void)
 	lockdep_assert_held(&pcpu_alloc_mutex);
 
 	if (!pages)
-		pages = pcpu_mem_zalloc(pages_size, GFP_KERNEL);
+		pages = pcpu_mem_zalloc(pages_size, gfp);
 	return pages;
 }
 
@@ -191,10 +192,22 @@ static void pcpu_post_unmap_tlb_flush(struct pcpu_chunk *chunk,
 }
 
 static int __pcpu_map_pages(unsigned long addr, struct page **pages,
-			    int nr_pages)
+			    int nr_pages, gfp_t gfp)
 {
-	return vmap_pages_range_noflush(addr, addr + (nr_pages << PAGE_SHIFT),
-			PAGE_KERNEL, pages, PAGE_SHIFT, GFP_KERNEL);
+	unsigned int flags;
+	int ret;
+
+	/*
+	 * The vmalloc page table allocation path does not pass @gfp down
+	 * explicitly.  Apply the corresponding memalloc scope so implicit
+	 * page table allocations preserve NOFS/NOIO constraints.
+	 */
+	flags = memalloc_apply_gfp_scope(gfp);
+	ret = vmap_pages_range_noflush(addr, addr + (nr_pages << PAGE_SHIFT),
+				       PAGE_KERNEL, pages, PAGE_SHIFT, gfp);
+	memalloc_restore_scope(flags);
+
+	return ret;
 }
 
 /**
@@ -203,6 +216,7 @@ static int __pcpu_map_pages(unsigned long addr, struct page **pages,
  * @pages: pages array containing pages to be mapped
  * @page_start: page index of the first page to map
  * @page_end: page index of the last page to map + 1
+ * @gfp: allocation flags passed to the underlying allocator
  *
  * For each cpu, map pages [@page_start,@page_end) into @chunk.  The
  * caller is responsible for calling pcpu_post_map_flush() after all
@@ -211,8 +225,8 @@ static int __pcpu_map_pages(unsigned long addr, struct page **pages,
  * This function is responsible for setting up whatever is necessary for
  * reverse lookup (addr -> chunk).
  */
-static int pcpu_map_pages(struct pcpu_chunk *chunk,
-			  struct page **pages, int page_start, int page_end)
+static int pcpu_map_pages(struct pcpu_chunk *chunk, struct page **pages,
+			  int page_start, int page_end, gfp_t gfp)
 {
 	unsigned int cpu, tcpu;
 	int i, err;
@@ -220,7 +234,7 @@ static int pcpu_map_pages(struct pcpu_chunk *chunk,
 	for_each_possible_cpu(cpu) {
 		err = __pcpu_map_pages(pcpu_chunk_addr(chunk, cpu, page_start),
 				       &pages[pcpu_page_idx(cpu, page_start)],
-				       page_end - page_start);
+				       page_end - page_start, gfp);
 		if (err < 0)
 			goto err;
 
@@ -271,21 +285,21 @@ static void pcpu_post_map_flush(struct pcpu_chunk *chunk,
  * @chunk.
  *
  * CONTEXT:
- * pcpu_alloc_mutex, does GFP_KERNEL allocation.
+ * pcpu_alloc_mutex, does @gfp allocation.
  */
 static int pcpu_populate_chunk(struct pcpu_chunk *chunk,
 			       int page_start, int page_end, gfp_t gfp)
 {
 	struct page **pages;
 
-	pages = pcpu_get_pages();
+	pages = pcpu_get_pages(gfp);
 	if (!pages)
 		return -ENOMEM;
 
 	if (pcpu_alloc_pages(chunk, pages, page_start, page_end, gfp))
 		return -ENOMEM;
 
-	if (pcpu_map_pages(chunk, pages, page_start, page_end)) {
+	if (pcpu_map_pages(chunk, pages, page_start, page_end, gfp)) {
 		pcpu_free_pages(chunk, pages, page_start, page_end);
 		return -ENOMEM;
 	}
@@ -319,7 +333,7 @@ static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk,
 	 * successful population attempt so the temp pages array must
 	 * be available now.
 	 */
-	pages = pcpu_get_pages();
+	pages = pcpu_get_pages(GFP_KERNEL);
 	BUG_ON(!pages);
 
 	/* unmap and free */
mm/percpu.c +1 -1
index b0676b8054ed..4d89965cba16 100644
@@ -3256,7 +3256,7 @@ int __init pcpu_page_first_chunk(size_t reserved_size, pcpu_fc_cpu_to_node_fn_t
 
 		/* pte already populated, the following shouldn't fail */
 		rc = __pcpu_map_pages(unit_addr, &pages[unit * unit_pages],
-				      unit_pages);
+				      unit_pages, GFP_KERNEL);
 		if (rc < 0)
 			panic("failed to map percpu area, err=%d\n", rc);
 
Signature · 1 line
2.50.1 (Apple Git-155)
Acked-by on v4 2/4
#4 lore raw

Re: [PATCH v4 2/4] mm/percpu: honor GFP constraints when populating chunks

Quoted 55 lines · Kaitao Cheng

From: Kaitao Cheng <[email protected]>

pcpu_alloc_noprof() derives pcpu_gfp from the caller supplied GFP mask and passes it down to pcpu_populate_chunk(). pcpu_alloc_pages() already uses that mask for backing page allocation.

However, the populate slow path still has internal allocations and page table allocations which can lose the caller's allocation context. The temporary pages array is allocated by pcpu_get_pages() with GFP_KERNEL, and pcpu_map_pages() maps the backing pages through vmap_pages_range_noflush() using GFP_KERNEL. The latter can allocate vmalloc page tables implicitly, so a caller which deliberately uses GFP_NOFS or GFP_NOIO can still enter FS or IO reclaim while populating a percpu chunk.

This has the same concern as chunk creation: callers such as blk-cgroup may use GFP_NOIO because they hold locks which can be involved in queue freeze or IO reclaim dependencies. If an allocation reaches the percpu slow path and needs to populate previously unbacked pages, the internal GFP_KERNEL allocations can defeat that context.

One possible case is blk-cgroup after commit 5d726c4dbeed ("blk-cgroup: fix possible deadlock while configuring policy"). blkg_conf_prep() now serializes against blkcg_deactivate_policy() with q->blkcg_mutex, and blkg_alloc() was changed to GFP_NOIO for that reason:

CPU0: blkg_conf_prep()

    mutex_lock(q->blkcg_mutex)
    blkg_alloc(..., GFP_NOIO)
      alloc_percpu_gfp(..., GFP_NOIO)
        pcpu_alloc_noprof(..., GFP_NOIO)
          pcpu_populate_chunk(GFP_NOIO)
            pcpu_get_pages()
	    pcpu_map_pages()
              -> if the selected percpu chunk has unpopulated pages,
	         chunk population may do internal GFP_KERNEL allocations
              -> direct reclaim / writeback can issue IO to this queue
              -> IO waits because the queue is frozen

CPU1: blkcg_deactivate_policy()

    blk_mq_freeze_queue(q)
    mutex_lock(q->blkcg_mutex)
      -> waits for CPU0
    ... unfreeze only happens after q->blkcg_mutex is acquired/released

So the concern is that the caller deliberately uses GFP_NOIO because it may hold a lock which can be acquired after queue freeze, but the percpu slow path can temporarily lose that allocation context.

Pass pcpu_gfp through pcpu_get_pages(), pcpu_map_pages() and __pcpu_map_pages(). Apply the corresponding memalloc scope around vmap_pages_range_noflush(), because vmalloc page table allocation does not pass the GFP mask down explicitly. Keep the first chunk setup path using GFP_KERNEL, matching the previous early-init behavior.

Quoted 3 lines
Fixes 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic")

OK, seems sensible.

Thanks!

Quoted 138 lines
---
 mm/percpu-vm.c | 38 ++++++++++++++++++++++++++------------
 mm/percpu.c    |  2 +-
 2 files changed, 27 insertions(+), 13 deletions(-)
mm/percpu-vm.c +26 -12
index 69b00741dc68..ccd03cc152d4 100644
@@ -21,6 +21,7 @@ static struct page *pcpu_chunk_page(struct pcpu_chunk *chunk,
 
 /**
  * pcpu_get_pages - get temp pages array
+ * @gfp: allocation flags passed to the underlying allocator
  *
  * Returns pointer to array of pointers to struct page which can be indexed
  * with pcpu_page_idx().  Note that there is only one array and accesses
@@ -29,7 +30,7 @@ static struct page *pcpu_chunk_page(struct pcpu_chunk *chunk,
  * RETURNS:
  * Pointer to temp pages array on success.
  */
-static struct page **pcpu_get_pages(void)
+static struct page **pcpu_get_pages(gfp_t gfp)
 {
 	static struct page **pages;
 	size_t pages_size = pcpu_nr_units * pcpu_unit_pages * sizeof(pages[0]);
@@ -37,7 +38,7 @@ static struct page **pcpu_get_pages(void)
 	lockdep_assert_held(&pcpu_alloc_mutex);
 
 	if (!pages)
-		pages = pcpu_mem_zalloc(pages_size, GFP_KERNEL);
+		pages = pcpu_mem_zalloc(pages_size, gfp);
 	return pages;
 }
 
@@ -191,10 +192,22 @@ static void pcpu_post_unmap_tlb_flush(struct pcpu_chunk *chunk,
 }
 
 static int __pcpu_map_pages(unsigned long addr, struct page **pages,
-			    int nr_pages)
+			    int nr_pages, gfp_t gfp)
 {
-	return vmap_pages_range_noflush(addr, addr + (nr_pages << PAGE_SHIFT),
-			PAGE_KERNEL, pages, PAGE_SHIFT, GFP_KERNEL);
+	unsigned int flags;
+	int ret;
+
+	/*
+	 * The vmalloc page table allocation path does not pass @gfp down
+	 * explicitly.  Apply the corresponding memalloc scope so implicit
+	 * page table allocations preserve NOFS/NOIO constraints.
+	 */
+	flags = memalloc_apply_gfp_scope(gfp);
+	ret = vmap_pages_range_noflush(addr, addr + (nr_pages << PAGE_SHIFT),
+				       PAGE_KERNEL, pages, PAGE_SHIFT, gfp);
+	memalloc_restore_scope(flags);
+
+	return ret;
 }
 
 /**
@@ -203,6 +216,7 @@ static int __pcpu_map_pages(unsigned long addr, struct page **pages,
  * @pages: pages array containing pages to be mapped
  * @page_start: page index of the first page to map
  * @page_end: page index of the last page to map + 1
+ * @gfp: allocation flags passed to the underlying allocator
  *
  * For each cpu, map pages [@page_start,@page_end) into @chunk.  The
  * caller is responsible for calling pcpu_post_map_flush() after all
@@ -211,8 +225,8 @@ static int __pcpu_map_pages(unsigned long addr, struct page **pages,
  * This function is responsible for setting up whatever is necessary for
  * reverse lookup (addr -> chunk).
  */
-static int pcpu_map_pages(struct pcpu_chunk *chunk,
-			  struct page **pages, int page_start, int page_end)
+static int pcpu_map_pages(struct pcpu_chunk *chunk, struct page **pages,
+			  int page_start, int page_end, gfp_t gfp)
 {
 	unsigned int cpu, tcpu;
 	int i, err;
@@ -220,7 +234,7 @@ static int pcpu_map_pages(struct pcpu_chunk *chunk,
 	for_each_possible_cpu(cpu) {
 		err = __pcpu_map_pages(pcpu_chunk_addr(chunk, cpu, page_start),
 				       &pages[pcpu_page_idx(cpu, page_start)],
-				       page_end - page_start);
+				       page_end - page_start, gfp);
 		if (err < 0)
 			goto err;
 
@@ -271,21 +285,21 @@ static void pcpu_post_map_flush(struct pcpu_chunk *chunk,
  * @chunk.
  *
  * CONTEXT:
- * pcpu_alloc_mutex, does GFP_KERNEL allocation.
+ * pcpu_alloc_mutex, does @gfp allocation.
  */
 static int pcpu_populate_chunk(struct pcpu_chunk *chunk,
 			       int page_start, int page_end, gfp_t gfp)
 {
 	struct page **pages;
 
-	pages = pcpu_get_pages();
+	pages = pcpu_get_pages(gfp);
 	if (!pages)
 		return -ENOMEM;
 
 	if (pcpu_alloc_pages(chunk, pages, page_start, page_end, gfp))
 		return -ENOMEM;
 
-	if (pcpu_map_pages(chunk, pages, page_start, page_end)) {
+	if (pcpu_map_pages(chunk, pages, page_start, page_end, gfp)) {
 		pcpu_free_pages(chunk, pages, page_start, page_end);
 		return -ENOMEM;
 	}
@@ -319,7 +333,7 @@ static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk,
 	 * successful population attempt so the temp pages array must
 	 * be available now.
 	 */
-	pages = pcpu_get_pages();
+	pages = pcpu_get_pages(GFP_KERNEL);
 	BUG_ON(!pages);
 
 	/* unmap and free */
mm/percpu.c +1 -1
index b0676b8054ed..4d89965cba16 100644
@@ -3256,7 +3256,7 @@ int __init pcpu_page_first_chunk(size_t reserved_size, pcpu_fc_cpu_to_node_fn_t
 
 		/* pte already populated, the following shouldn't fail */
 		rc = __pcpu_map_pages(unit_addr, &pages[unit * unit_pages],
-				      unit_pages);
+				      unit_pages, GFP_KERNEL);
 		if (rc < 0)
 			panic("failed to map percpu area, err=%d\n", rc);
 
Signature · 1 line
2.50.1 (Apple Git-155)
Signature · 2 lines
Michal Hocko
SUSE Labs
#5 lore raw

PATCH v4 3/4 mm/percpu: Make cached pages lookup explicit

In reply to Kaitao Cheng

From: Kaitao Cheng <[email protected]>

pcpu_depopulate_chunk() only needs the temporary pages array that was already allocated by an earlier successful population attempt. Passing GFP_KERNEL to pcpu_get_pages() in this path is misleading because the depopulation path is not expected to allocate the array.

Teach pcpu_get_pages() to treat a zero gfp mask as a cached-only lookup and add pcpu_get_pages_cached() for that use case. This keeps allocation on the populate path tied to the caller supplied GFP mask while making the depopulate path's dependency on the cached array explicit.

---
 mm/percpu-vm.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
mm/percpu-vm.c +9 -3
index ccd03cc152d4..7ed216192fc0 100644
@@ -21,7 +21,8 @@ static struct page *pcpu_chunk_page(struct pcpu_chunk *chunk,
 
 /**
  * pcpu_get_pages - get temp pages array
- * @gfp: allocation flags passed to the underlying allocator
+ * @gfp: allocation flags passed to the underlying allocator, 0 to only
+ *	 return the cached array
  *
  * Returns pointer to array of pointers to struct page which can be indexed
  * with pcpu_page_idx().  Note that there is only one array and accesses
@@ -37,11 +38,16 @@ static struct page **pcpu_get_pages(gfp_t gfp)
 
 	lockdep_assert_held(&pcpu_alloc_mutex);
 
-	if (!pages)
+	if (!pages && gfp)
 		pages = pcpu_mem_zalloc(pages_size, gfp);
 	return pages;
 }
 
+static struct page **pcpu_get_pages_cached(void)
+{
+	return pcpu_get_pages(0);
+}
+
 /**
  * pcpu_free_pages - free pages which were allocated for @chunk
  * @chunk: chunk pages were allocated for
@@ -333,7 +339,7 @@ static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk,
 	 * successful population attempt so the temp pages array must
 	 * be available now.
 	 */
-	pages = pcpu_get_pages(GFP_KERNEL);
+	pages = pcpu_get_pages_cached();
 	BUG_ON(!pages);
 
 	/* unmap and free */
Signature · 1 line
2.50.1 (Apple Git-155)
#6 lore raw

PATCH v4 4/4 mm/percpu: Avoid IO/FS reclaim in backing allocations

In reply to Kaitao Cheng

From: Kaitao Cheng <[email protected]>

Commit 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic") allows sleepable GFP_NOIO and GFP_NOFS percpu allocations to take pcpu_alloc_mutex. This avoids premature allocation failures, but it also makes the mutex visible to callers from constrained IO/FS contexts.

Thread A calls pcpu_alloc_noprof() with GFP_KERNEL and takes pcpu_alloc_mutex. Since the internal allocation is not constrained by NOFS, it may enter FS reclaim while still holding pcpu_alloc_mutex, creating a dependency like: pcpu_alloc_mutex -> fs_reclaim -> FS lock

At the same time, Thread B may already hold an FS lock and then call pcpu_alloc_noprof() with GFP_NOFS. It will try to acquire pcpu_alloc_mutex and block, creating the reverse dependency:
FS lock -> pcpu_alloc_mutex

This can still form a potential deadlock cycle.

Avoid the dependency by restricting percpu backing allocations to GFP_NOIO. The public allocation still uses the caller's GFP context to decide whether it may block, but the internal memory allocations performed while pcpu_alloc_mutex is held cannot recurse into IO or FS reclaim.

Fixes 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic")
---
 mm/percpu.c | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)
mm/percpu.c +13 -5
index 4d89965cba16..a802d72c116f 100644
@@ -1726,9 +1726,8 @@ static void pcpu_alloc_tag_free_hook(struct pcpu_chunk *chunk, int off, size_t s
  * @gfp: allocation flags
  *
  * Allocate percpu area of @size bytes aligned at @align.  If @gfp doesn't
- * contain %GFP_KERNEL, the allocation is atomic. If @gfp has __GFP_NOWARN
- * then no warning will be triggered on invalid or failed allocation
- * requests.
+ * allow blocking, the allocation is atomic. If @gfp has __GFP_NOWARN then no
+ * warning will be triggered on invalid or failed allocation requests.
  *
  * RETURNS:
  * Percpu pointer to the allocated area on success, NULL on failure.
@@ -1749,8 +1748,17 @@ void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved,
 	size_t bits, bit_align;
 
 	gfp = current_gfp_context(gfp);
-	/* whitelisted flags that can be passed to the backing allocators */
-	pcpu_gfp = gfp & (GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN);
+	/*
+	 * Allowlisted flags that can be passed to the backing allocators.
+	 * Backing allocations under pcpu_alloc_mutex must not recurse into
+	 * IO/FS reclaim.  Otherwise a GFP_KERNEL caller holding the mutex can
+	 * block on reclaim while a GFP_NOIO/NOFS caller holding an IO/FS lock
+	 * waits for the same mutex.
+	 *
+	 * Do not pass __GFP_NOFAIL.  A small percpu allocation may need many
+	 * backing pages, making nofail reclaim too costly under NOIO/NOFS.
+	 */
+	pcpu_gfp = gfp & (GFP_NOIO | __GFP_NORETRY | __GFP_NOWARN);
 	is_atomic = !gfpflags_allow_blocking(gfp);
 	do_warn = !(gfp & __GFP_NOWARN);
 
Signature · 1 line
2.50.1 (Apple Git-155)
Acked-by on v4 4/4
#7 lore raw

Re: [PATCH v4 4/4] mm/percpu: Avoid IO/FS reclaim in backing allocations

Quoted 20 lines · Kaitao Cheng

From: Kaitao Cheng <[email protected]>

Commit 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic") allows sleepable GFP_NOIO and GFP_NOFS percpu allocations to take pcpu_alloc_mutex. This avoids premature allocation failures, but it also makes the mutex visible to callers from constrained IO/FS contexts.

Thread A calls pcpu_alloc_noprof() with GFP_KERNEL and takes pcpu_alloc_mutex. Since the internal allocation is not constrained by NOFS, it may enter FS reclaim while still holding pcpu_alloc_mutex, creating a dependency like: pcpu_alloc_mutex -> fs_reclaim -> FS lock

At the same time, Thread B may already hold an FS lock and then call pcpu_alloc_noprof() with GFP_NOFS. It will try to acquire pcpu_alloc_mutex and block, creating the reverse dependency:
FS lock -> pcpu_alloc_mutex

This can still form a potential deadlock cycle.

Quoted 7 lines

Avoid the dependency by restricting percpu backing allocations to GFP_NOIO. The public allocation still uses the caller's GFP context to decide whether it may block, but the internal memory allocations performed while pcpu_alloc_mutex is held cannot recurse into IO or FS reclaim.

Fixes 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic")

This seems like the only viable short term fix but long term it would be really better to make allocations outside of the lock.

Minor nit

Quoted 7 lines
mm/percpu.c +1 -2
@@ -1749,8 +1748,17 @@ void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved,
 	size_t bits, bit_align;
 
 	gfp = current_gfp_context(gfp);
-	/* whitelisted flags that can be passed to the backing allocators */
-	pcpu_gfp = gfp & (GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN);
+	/*

mm/percpu.c void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved, · line 1761

Quoted 10 lines
mm/percpu.c +10 -0
+	 * Allowlisted flags that can be passed to the backing allocators.
+	 * Backing allocations under pcpu_alloc_mutex must not recurse into
+	 * IO/FS reclaim.  Otherwise a GFP_KERNEL caller holding the mutex can
+	 * block on reclaim while a GFP_NOIO/NOFS caller holding an IO/FS lock
+	 * waits for the same mutex.
+	 *
+	 * Do not pass __GFP_NOFAIL.  A small percpu allocation may need many
+	 * backing pages, making nofail reclaim too costly under NOIO/NOFS.
+	 */
+	pcpu_gfp = gfp & (GFP_NOIO | __GFP_NORETRY | __GFP_NOWARN);

GFP_NOIO, NOFS are negative masks in the sense that that are lacking flags so the overal intention would be more readable IMHO in the following form

	pcpu_gfp = gfp & (GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN)
	pcpu_gfp &= ~(__GFP_IO | __GFP_FS)
Quoted 5 lines
mm/percpu.c +0 -0
 	is_atomic = !gfpflags_allow_blocking(gfp);
 	do_warn = !(gfp & __GFP_NOWARN);
 
Signature · 1 line
2.50.1 (Apple Git-155)
Signature · 2 lines
Michal Hocko
SUSE Labs
#8 lore raw

Re: [PATCH v4 4/4] mm/percpu: Avoid IO/FS reclaim in backing allocations

Quoted 55 lines · Michal Hocko
Quoted 20 lines · Kaitao Cheng

From: Kaitao Cheng <[email protected]>

Commit 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic") allows sleepable GFP_NOIO and GFP_NOFS percpu allocations to take pcpu_alloc_mutex. This avoids premature allocation failures, but it also makes the mutex visible to callers from constrained IO/FS contexts.

Thread A calls pcpu_alloc_noprof() with GFP_KERNEL and takes pcpu_alloc_mutex. Since the internal allocation is not constrained by NOFS, it may enter FS reclaim while still holding pcpu_alloc_mutex, creating a dependency like: pcpu_alloc_mutex -> fs_reclaim -> FS lock

At the same time, Thread B may already hold an FS lock and then call pcpu_alloc_noprof() with GFP_NOFS. It will try to acquire pcpu_alloc_mutex and block, creating the reverse dependency:
FS lock -> pcpu_alloc_mutex

This can still form a potential deadlock cycle.

Quoted 7 lines

Avoid the dependency by restricting percpu backing allocations to GFP_NOIO. The public allocation still uses the caller's GFP context to decide whether it may block, but the internal memory allocations performed while pcpu_alloc_mutex is held cannot recurse into IO or FS reclaim.

Fixes 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic")

This seems like the only viable short term fix but long term it would be really better to make allocations outside of the lock.

Minor nit

Quoted 7 lines
mm/percpu.c +1 -2
@@ -1749,8 +1748,17 @@ void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved,
 	size_t bits, bit_align;
 
 	gfp = current_gfp_context(gfp);
-	/* whitelisted flags that can be passed to the backing allocators */
-	pcpu_gfp = gfp & (GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN);
+	/*

mm/percpu.c void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved, · line 1761

Quoted 10 lines
mm/percpu.c +10 -0
+	 * Allowlisted flags that can be passed to the backing allocators.
+	 * Backing allocations under pcpu_alloc_mutex must not recurse into
+	 * IO/FS reclaim.  Otherwise a GFP_KERNEL caller holding the mutex can
+	 * block on reclaim while a GFP_NOIO/NOFS caller holding an IO/FS lock
+	 * waits for the same mutex.
+	 *
+	 * Do not pass __GFP_NOFAIL.  A small percpu allocation may need many
+	 * backing pages, making nofail reclaim too costly under NOIO/NOFS.
+	 */
+	pcpu_gfp = gfp & (GFP_NOIO | __GFP_NORETRY | __GFP_NOWARN);

GFP_NOIO, NOFS are negative masks in the sense that that are lacking flags so the overal intention would be more readable IMHO in the following form

Quoted 2 lines
	pcpu_gfp = gfp & (GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN)
	pcpu_gfp &= ~(__GFP_IO | __GFP_FS)

This looks a bit redundant. The newly added comment already makes the intent clear, and the extra code seems to serve only as another hint to readers, which is essentially the same role as the comment.

GFP_NOIO already excludes __GFP_IO and __GFP_FS, so its semantics are clear enough. It should not be misleading, and it is also more concise.

Quoted 6 lines
Quoted 5 lines
mm/percpu.c +0 -0
 	is_atomic = !gfpflags_allow_blocking(gfp);
 	do_warn = !(gfp & __GFP_NOWARN);
 
Signature · 1 line
2.50.1 (Apple Git-155)
Signature · 2 lines
Thanks
Kaitao Cheng
#9 lore raw

Re: [PATCH v4 4/4] mm/percpu: Avoid IO/FS reclaim in backing allocations

Quoted 59 lines · Kaitao Cheng
Quoted 55 lines · Michal Hocko
Quoted 20 lines · Kaitao Cheng

From: Kaitao Cheng <[email protected]>

Commit 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic") allows sleepable GFP_NOIO and GFP_NOFS percpu allocations to take pcpu_alloc_mutex. This avoids premature allocation failures, but it also makes the mutex visible to callers from constrained IO/FS contexts.

Thread A calls pcpu_alloc_noprof() with GFP_KERNEL and takes pcpu_alloc_mutex. Since the internal allocation is not constrained by NOFS, it may enter FS reclaim while still holding pcpu_alloc_mutex, creating a dependency like: pcpu_alloc_mutex -> fs_reclaim -> FS lock

At the same time, Thread B may already hold an FS lock and then call pcpu_alloc_noprof() with GFP_NOFS. It will try to acquire pcpu_alloc_mutex and block, creating the reverse dependency:
FS lock -> pcpu_alloc_mutex

This can still form a potential deadlock cycle.

Quoted 7 lines

Avoid the dependency by restricting percpu backing allocations to GFP_NOIO. The public allocation still uses the caller's GFP context to decide whether it may block, but the internal memory allocations performed while pcpu_alloc_mutex is held cannot recurse into IO or FS reclaim.

Fixes 9a5b183941b5 ("mm, percpu: do not consider sleepable allocations atomic")

This seems like the only viable short term fix but long term it would be really better to make allocations outside of the lock.

Minor nit

Quoted 7 lines
mm/percpu.c +1 -2
@@ -1749,8 +1748,17 @@ void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved,
 	size_t bits, bit_align;
 
 	gfp = current_gfp_context(gfp);
-	/* whitelisted flags that can be passed to the backing allocators */
-	pcpu_gfp = gfp & (GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN);
+	/*

mm/percpu.c void __percpu *pcpu_alloc_noprof(size_t size, size_t align, bool reserved, · line 1761

Quoted 10 lines
mm/percpu.c +10 -0
+	 * Allowlisted flags that can be passed to the backing allocators.
+	 * Backing allocations under pcpu_alloc_mutex must not recurse into
+	 * IO/FS reclaim.  Otherwise a GFP_KERNEL caller holding the mutex can
+	 * block on reclaim while a GFP_NOIO/NOFS caller holding an IO/FS lock
+	 * waits for the same mutex.
+	 *
+	 * Do not pass __GFP_NOFAIL.  A small percpu allocation may need many
+	 * backing pages, making nofail reclaim too costly under NOIO/NOFS.
+	 */
+	pcpu_gfp = gfp & (GFP_NOIO | __GFP_NORETRY | __GFP_NOWARN);

GFP_NOIO, NOFS are negative masks in the sense that that are lacking flags so the overal intention would be more readable IMHO in the following form

Quoted 2 lines
	pcpu_gfp = gfp & (GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN)
	pcpu_gfp &= ~(__GFP_IO | __GFP_FS)
Quoted 6 lines

This looks a bit redundant. The newly added comment already makes the intent clear, and the extra code seems to serve only as another hint to readers, which is essentially the same role as the comment.

GFP_NOIO already excludes __GFP_IO and __GFP_FS, so its semantics are clear enough. It should not be misleading, and it is also more concise.

I will certainly not insist, but this is a generally used pattern to drop IO and FS flags. So if you want to grep for the pattern you will not miss this place. Comment _is_ useful but harder to grep for.

Quoted 11 lines
Quoted 7 lines
Quoted 6 lines
mm/percpu.c +0 -0
 	is_atomic = !gfpflags_allow_blocking(gfp);
 	do_warn = !(gfp & __GFP_NOWARN);
 
Signature · 1 line
2.50.1 (Apple Git-155)
Signature · 2 lines
Thanks
Kaitao Cheng
Signature · 2 lines
Michal Hocko
SUSE Labs