Recently, we have observed some janky issues caused by unpleasantly long
contention on mmap_lock which is held by smaps_rollup when probing large
processes. To address the problem, we let smaps_rollup detect if anyone
wants to acquire mmap_lock for write attempts. If yes, just release the
lock temporarily to ease the contention.
smaps_rollup is a procfs interface which allows users to summarize the
process's memory usage without the overhead of seq_* calls. Android uses it
to sample the memory usage of various processes to balance its memory pool
sizes. If no one wants to take the lock for write requests, smaps_rollup
with this patch will behave like the original one.
Although there are on-going mmap_lock optimizations like range-based locks,
the lock applied to smaps_rollup would be the coarse one, which is hard to
avoid the occurrence of aforementioned issues. So the detection and
temporary release for write attempts on mmap_lock in smaps_rollup is still
necessary.
Change since v1:
- If current VMA is freed after dropping the lock, it will return
- incomplete result. To fix this issue, refine the code flow as
- suggested by Steve. [1]
Change since v2:
- When getting back the mmap lock, the address where you stopped last
- time could now be in the middle of a vma. Add one more check to handle
- this case as suggested by Michel. [2]
Change since v3:
- last_stopped is easily confused with last_vma_end. Replace it with
- a direct call to smap_gather_stats(vma, &mss, last_vma_end) as
- suggested by Steve. [3]
[1] https://lore.kernel.org/lkml/[email protected]/
[2] https://lore.kernel.org/lkml/CANN689FtCsC71cjAjs0GPspOhgo_HRj+diWsoU1wr98YPktgWg@mail.gmail.com/
[3] https://lore.kernel.org/lkml/[email protected]/
Chinwen Chang (3):
mmap locking API: add mmap_lock_is_contended()
mm: smaps*: extend smap_gather_stats to support specified beginning
mm: proc: smaps_rollup: do not stall write attempts on mmap_lock
fs/proc/task_mmu.c | 96 +++++++++++++++++++++++++++++++++++----
include/linux/mmap_lock.h | 5 ++
2 files changed, 92 insertions(+), 9 deletions(-)
Add new API to query if someone wants to acquire mmap_lock
for write attempts.
Using this instead of rwsem_is_contended makes it more tolerant
of future changes to the lock type.
Change-Id: Idb21478bb0580ba72b9926aba3bbc4b1f75deec2
Signed-off-by: Chinwen Chang <[email protected]>
Reviewed-by: Steven Price <[email protected]>
Acked-by: Michel Lespinasse <[email protected]>
---
include/linux/mmap_lock.h | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/include/linux/mmap_lock.h b/include/linux/mmap_lock.h
index 0707671..18e7eae 100644
--- a/include/linux/mmap_lock.h
+++ b/include/linux/mmap_lock.h
@@ -87,4 +87,9 @@ static inline void mmap_assert_write_locked(struct mm_struct *mm)
VM_BUG_ON_MM(!rwsem_is_locked(&mm->mmap_lock), mm);
}
+static inline int mmap_lock_is_contended(struct mm_struct *mm)
+{
+ return rwsem_is_contended(&mm->mmap_lock);
+}
+
#endif /* _LINUX_MMAP_LOCK_H */
--
1.9.1
Extend smap_gather_stats to support indicated beginning address at
which it should start gathering. To achieve the goal, we add a new
parameter @start assigned by the caller and try to refactor it for
simplicity.
If @start is 0, it will use the range of @vma for gathering.
Change since v2:
- This is a new change to make the retry behavior of smaps_rollup
- more complete as suggested by Michel [1]
[1] https://lore.kernel.org/lkml/CANN689FtCsC71cjAjs0GPspOhgo_HRj+diWsoU1wr98YPktgWg@mail.gmail.com/
Change-Id: I8652e0ee6c5e93fb56376a68d71ed6cdd8ac10e8
Signed-off-by: Chinwen Chang <[email protected]>
CC: Michel Lespinasse <[email protected]>
Reviewed-by: Steven Price <[email protected]>
---
fs/proc/task_mmu.c | 30 ++++++++++++++++++++++--------
1 file changed, 22 insertions(+), 8 deletions(-)
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index dbda449..76e623a 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -723,9 +723,21 @@ static int smaps_hugetlb_range(pte_t *pte, unsigned long hmask,
.pte_hole = smaps_pte_hole,
};
+/*
+ * Gather mem stats from @vma with the indicated beginning
+ * address @start, and keep them in @mss.
+ *
+ * Use vm_start of @vma as the beginning address if @start is 0.
+ */
static void smap_gather_stats(struct vm_area_struct *vma,
- struct mem_size_stats *mss)
+ struct mem_size_stats *mss, unsigned long start)
{
+ const struct mm_walk_ops *ops = &smaps_walk_ops;
+
+ /* Invalid start */
+ if (start >= vma->vm_end)
+ return;
+
#ifdef CONFIG_SHMEM
/* In case of smaps_rollup, reset the value from previous vma */
mss->check_shmem_swap = false;
@@ -742,18 +754,20 @@ static void smap_gather_stats(struct vm_area_struct *vma,
*/
unsigned long shmem_swapped = shmem_swap_usage(vma);
- if (!shmem_swapped || (vma->vm_flags & VM_SHARED) ||
- !(vma->vm_flags & VM_WRITE)) {
+ if (!start && (!shmem_swapped || (vma->vm_flags & VM_SHARED) ||
+ !(vma->vm_flags & VM_WRITE))) {
mss->swap += shmem_swapped;
} else {
mss->check_shmem_swap = true;
- walk_page_vma(vma, &smaps_shmem_walk_ops, mss);
- return;
+ ops = &smaps_shmem_walk_ops;
}
}
#endif
/* mmap_lock is held in m_start */
- walk_page_vma(vma, &smaps_walk_ops, mss);
+ if (!start)
+ walk_page_vma(vma, ops, mss);
+ else
+ walk_page_range(vma->vm_mm, start, vma->vm_end, ops, mss);
}
#define SEQ_PUT_DEC(str, val) \
@@ -805,7 +819,7 @@ static int show_smap(struct seq_file *m, void *v)
memset(&mss, 0, sizeof(mss));
- smap_gather_stats(vma, &mss);
+ smap_gather_stats(vma, &mss, 0);
show_map_vma(m, vma);
@@ -854,7 +868,7 @@ static int show_smaps_rollup(struct seq_file *m, void *v)
hold_task_mempolicy(priv);
for (vma = priv->mm->mmap; vma; vma = vma->vm_next) {
- smap_gather_stats(vma, &mss);
+ smap_gather_stats(vma, &mss, 0);
last_vma_end = vma->vm_end;
}
--
1.9.1
smaps_rollup will try to grab mmap_lock and go through the whole vma
list until it finishes the iterating. When encountering large processes,
the mmap_lock will be held for a longer time, which may block other
write requests like mmap and munmap from progressing smoothly.
There are upcoming mmap_lock optimizations like range-based locks, but
the lock applied to smaps_rollup would be the coarse type, which doesn't
avoid the occurrence of unpleasant contention.
To solve aforementioned issue, we add a check which detects whether
anyone wants to grab mmap_lock for write attempts.
Change since v1:
- If current VMA is freed after dropping the lock, it will return
- incomplete result. To fix this issue, refine the code flow as
- suggested by Steve. [1]
Change since v2:
- When getting back the mmap lock, the address where you stopped last
- time could now be in the middle of a vma. Add one more check to handle
- this case as suggested by Michel. [2]
Change since v3:
- last_stopped is easily confused with last_vma_end. Replace it with
- a direct call to smap_gather_stats(vma, &mss, last_vma_end) as
- suggested by Steve. [3]
[1] https://lore.kernel.org/lkml/[email protected]/
[2] https://lore.kernel.org/lkml/CANN689FtCsC71cjAjs0GPspOhgo_HRj+diWsoU1wr98YPktgWg@mail.gmail.com/
[3] https://lore.kernel.org/lkml/[email protected]/
Change-Id: Idcdb6478ccd06a9e5edd4eda9285378e961a6b94
Signed-off-by: Chinwen Chang <[email protected]>
Reviewed-by: Steven Price <[email protected]>
CC: Michel Lespinasse <[email protected]>
---
fs/proc/task_mmu.c | 66 +++++++++++++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 65 insertions(+), 1 deletion(-)
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 76e623a..1a80624 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -867,9 +867,73 @@ static int show_smaps_rollup(struct seq_file *m, void *v)
hold_task_mempolicy(priv);
- for (vma = priv->mm->mmap; vma; vma = vma->vm_next) {
+ for (vma = priv->mm->mmap; vma;) {
smap_gather_stats(vma, &mss, 0);
last_vma_end = vma->vm_end;
+
+ /*
+ * Release mmap_lock temporarily if someone wants to
+ * access it for write request.
+ */
+ if (mmap_lock_is_contended(mm)) {
+ mmap_read_unlock(mm);
+ ret = mmap_read_lock_killable(mm);
+ if (ret) {
+ release_task_mempolicy(priv);
+ goto out_put_mm;
+ }
+
+ /*
+ * After dropping the lock, there are four cases to
+ * consider. See the following example for explanation.
+ *
+ * +------+------+-----------+
+ * | VMA1 | VMA2 | VMA3 |
+ * +------+------+-----------+
+ * | | | |
+ * 4k 8k 16k 400k
+ *
+ * Suppose we drop the lock after reading VMA2 due to
+ * contention, then we get:
+ *
+ * last_vma_end = 16k
+ *
+ * 1) VMA2 is freed, but VMA3 exists:
+ *
+ * find_vma(mm, 16k - 1) will return VMA3.
+ * In this case, just continue from VMA3.
+ *
+ * 2) VMA2 still exists:
+ *
+ * find_vma(mm, 16k - 1) will return VMA2.
+ * Iterate the loop like the original one.
+ *
+ * 3) No more VMAs can be found:
+ *
+ * find_vma(mm, 16k - 1) will return NULL.
+ * No more things to do, just break.
+ *
+ * 4) (last_vma_end - 1) is the middle of a vma (VMA'):
+ *
+ * find_vma(mm, 16k - 1) will return VMA' whose range
+ * contains last_vma_end.
+ * Iterate VMA' from last_vma_end.
+ */
+ vma = find_vma(mm, last_vma_end - 1);
+ /* Case 3 above */
+ if (!vma)
+ break;
+
+ /* Case 1 above */
+ if (vma->vm_start >= last_vma_end)
+ continue;
+
+ /* Case 4 above */
+ if (vma->vm_end > last_vma_end)
+ smap_gather_stats(vma, &mss, last_vma_end);
+ }
+ /* Case 2 above */
+ vma = vma->vm_next;
}
show_vma_header_prefix(m, priv->mm->mmap->vm_start,
--
1.9.1