* [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin
@ 2025-06-09 17:17 Kairui Song
2025-06-09 17:24 ` Kairui Song
2025-06-09 18:16 ` Nhat Pham
0 siblings, 2 replies; 7+ messages in thread
From: Kairui Song @ 2025-06-09 17:17 UTC (permalink / raw)
To: linux-mm
Cc: Andrew Morton, Hugh Dickins, Baolin Wang, Kemeng Shi, Chris Li,
Nhat Pham, Baoquan He, Barry Song, Usama Arif, linux-kernel,
Kairui Song
From: Kairui Song <kasong@tencent.com>
Following softlockup can be easily reproduced on my test machine with:
echo always > /sys/kernel/mm/transparent_hugepage/hugepages-64kB/enabled
swapon /dev/zram0 # zram0 is a 48G swap device
mkdir -p /sys/fs/cgroup/memory/test
echo 1G > /sys/fs/cgroup/test/memory.max
echo $BASHPID > /sys/fs/cgroup/test/cgroup.procs
while true; do
dd if=/dev/zero of=/tmp/test.img bs=1M count=5120
cat /tmp/test.img > /dev/null
rm /tmp/test.img
done
Then after a while:
watchdog: BUG: soft lockup - CPU#0 stuck for 763s! [cat:5787]
Modules linked in: zram virtiofs
CPU: 0 UID: 0 PID: 5787 Comm: cat Kdump: loaded Tainted: G L 6.15.0.orig-gf3021d9246bc-dirty #118 PREEMPT(voluntary)·
Tainted: [L]=SOFTLOCKUP
Hardware name: Red Hat KVM/RHEL-AV, BIOS 0.0.0 02/06/2015
RIP: 0010:mpol_shared_policy_lookup+0xd/0x70
Code: e9 b8 b4 ff ff 31 c0 c3 cc cc cc cc 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 41 54 55 53 <48> 8b 1f 48 85 db 74 41 4c 8d 67 08 48 89 fb 48 89 f5 4c 89 e7 e8
RSP: 0018:ffffc90002b1fc28 EFLAGS: 00000202
RAX: 00000000001c20ca RBX: 0000000000724e1e RCX: 0000000000000001
RDX: ffff888118e214c8 RSI: 0000000000057d42 RDI: ffff888118e21518
RBP: 000000000002bec8 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000bf4 R11: 0000000000000000 R12: 0000000000000001
R13: 00000000001c20ca R14: 00000000001c20ca R15: 0000000000000000
FS: 00007f03f995c740(0000) GS:ffff88a07ad9a000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f03f98f1000 CR3: 0000000144626004 CR4: 0000000000770eb0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
PKRU: 55555554
Call Trace:
<TASK>
shmem_alloc_folio+0x31/0xc0
shmem_swapin_folio+0x309/0xcf0
? filemap_get_entry+0x117/0x1e0
? xas_load+0xd/0xb0
? filemap_get_entry+0x101/0x1e0
shmem_get_folio_gfp+0x2ed/0x5b0
shmem_file_read_iter+0x7f/0x2e0
vfs_read+0x252/0x330
ksys_read+0x68/0xf0
do_syscall_64+0x4c/0x1c0
entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x7f03f9a46991
Code: 00 48 8b 15 81 14 10 00 f7 d8 64 89 02 b8 ff ff ff ff eb bd e8 20 ad 01 00 f3 0f 1e fa 80 3d 35 97 10 00 00 74 13 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 4f c3 66 0f 1f 44 00 00 55 48 89 e5 48 83 ec
RSP: 002b:00007fff3c52bd28 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
RAX: ffffffffffffffda RBX: 0000000000040000 RCX: 00007f03f9a46991
RDX: 0000000000040000 RSI: 00007f03f98ba000 RDI: 0000000000000003
RBP: 00007fff3c52bd50 R08: 0000000000000000 R09: 00007f03f9b9a380
R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000040000
R13: 00007f03f98ba000 R14: 0000000000000003 R15: 0000000000000000
</TASK>
The reason is simple, readahead brought some order 0 folio in swap
cache, and the swapin mTHP folio being allocated is in confict with it,
so swapcache_prepare fails and causes shmem_swap_alloc_folio to return
-EEXIST, and shmem simply retries again and again causing this loop.
Fix it by applying a similar fix for anon mTHP swapin.
The performance change is very slight, time of swapin 10g zero folios
with shmem (test for 12 times):
Before: 2.47s
After: 2.48s
Fixes: 1dd44c0af4fa1 ("mm: shmem: skip swapcache for swapin of synchronous swap device")
Signed-off-by: Kairui Song <kasong@tencent.com>
---
V1: https://lore.kernel.org/linux-mm/20250608192713.95875-1-ryncsn@gmail.com/
Updates:
- Move non_swapcache_batch check before swapcache_prepare, I was
expecting this could improve the performance, turns out it barely
helps and may even cause more overhead in some cases. [ Barry Song ]
- Remove zero map check, no need to do that for shmem [ Barry Song,
Baolin Wang ]
- Fix build bot error.
mm/memory.c | 20 --------------------
mm/shmem.c | 4 +++-
mm/swap.h | 23 +++++++++++++++++++++++
3 files changed, 26 insertions(+), 21 deletions(-)
diff --git a/mm/memory.c b/mm/memory.c
index 9ead7ab07e8e..3845ed068d74 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -4313,26 +4313,6 @@ static struct folio *__alloc_swap_folio(struct vm_fault *vmf)
}
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
-{
- struct swap_info_struct *si = swp_swap_info(entry);
- pgoff_t offset = swp_offset(entry);
- int i;
-
- /*
- * While allocating a large folio and doing swap_read_folio, which is
- * the case the being faulted pte doesn't have swapcache. We need to
- * ensure all PTEs have no cache as well, otherwise, we might go to
- * swap devices while the content is in swapcache.
- */
- for (i = 0; i < max_nr; i++) {
- if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
- return i;
- }
-
- return i;
-}
-
/*
* Check if the PTEs within a range are contiguous swap entries
* and have consistent swapcache, zeromap.
diff --git a/mm/shmem.c b/mm/shmem.c
index 73182e904f9c..a4fdfbd086f1 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -2256,6 +2256,7 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
folio = swap_cache_get_folio(swap, NULL, 0);
order = xa_get_order(&mapping->i_pages, index);
if (!folio) {
+ int nr_pages = 1 << order;
bool fallback_order0 = false;
/* Or update major stats only when swapin succeeds?? */
@@ -2271,7 +2272,8 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
* to swapin order-0 folio, as well as for zswap case.
*/
if (order > 0 && ((vma && unlikely(userfaultfd_armed(vma))) ||
- !zswap_never_enabled()))
+ !zswap_never_enabled() ||
+ non_swapcache_batch(swap, nr_pages) != nr_pages))
fallback_order0 = true;
/* Skip swapcache for synchronous device. */
diff --git a/mm/swap.h b/mm/swap.h
index e87a0f19a0ee..911ad5ff0f89 100644
--- a/mm/swap.h
+++ b/mm/swap.h
@@ -108,6 +108,25 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
return find_next_bit(sis->zeromap, end, start) - start;
}
+static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
+{
+ struct swap_info_struct *si = swp_swap_info(entry);
+ pgoff_t offset = swp_offset(entry);
+ int i;
+
+ /*
+ * While allocating a large folio and doing mTHP swapin, we need to
+ * ensure all entries are not cached, otherwise, the mTHP folio will
+ * be in conflict with the folio in swap cache.
+ */
+ for (i = 0; i < max_nr; i++) {
+ if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
+ return i;
+ }
+
+ return i;
+}
+
#else /* CONFIG_SWAP */
struct swap_iocb;
static inline void swap_read_folio(struct folio *folio, struct swap_iocb **plug)
@@ -202,6 +221,10 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
return 0;
}
+static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
+{
+ return 0;
+}
#endif /* CONFIG_SWAP */
/**
--
2.49.0
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin
2025-06-09 17:17 [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin Kairui Song
@ 2025-06-09 17:24 ` Kairui Song
2025-06-09 21:16 ` Barry Song
2025-06-09 18:16 ` Nhat Pham
1 sibling, 1 reply; 7+ messages in thread
From: Kairui Song @ 2025-06-09 17:24 UTC (permalink / raw)
To: linux-mm, stable
Cc: Andrew Morton, Hugh Dickins, Baolin Wang, Kemeng Shi, Chris Li,
Nhat Pham, Baoquan He, Barry Song, Usama Arif, linux-kernel
On Tue, Jun 10, 2025 at 1:18 AM Kairui Song <ryncsn@gmail.com> wrote:
>
> From: Kairui Song <kasong@tencent.com>
>
> Following softlockup can be easily reproduced on my test machine with:
>
> echo always > /sys/kernel/mm/transparent_hugepage/hugepages-64kB/enabled
> swapon /dev/zram0 # zram0 is a 48G swap device
> mkdir -p /sys/fs/cgroup/memory/test
> echo 1G > /sys/fs/cgroup/test/memory.max
> echo $BASHPID > /sys/fs/cgroup/test/cgroup.procs
> while true; do
> dd if=/dev/zero of=/tmp/test.img bs=1M count=5120
> cat /tmp/test.img > /dev/null
> rm /tmp/test.img
> done
>
> Then after a while:
> watchdog: BUG: soft lockup - CPU#0 stuck for 763s! [cat:5787]
> Modules linked in: zram virtiofs
> CPU: 0 UID: 0 PID: 5787 Comm: cat Kdump: loaded Tainted: G L 6.15.0.orig-gf3021d9246bc-dirty #118 PREEMPT(voluntary)·
> Tainted: [L]=SOFTLOCKUP
> Hardware name: Red Hat KVM/RHEL-AV, BIOS 0.0.0 02/06/2015
> RIP: 0010:mpol_shared_policy_lookup+0xd/0x70
> Code: e9 b8 b4 ff ff 31 c0 c3 cc cc cc cc 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 41 54 55 53 <48> 8b 1f 48 85 db 74 41 4c 8d 67 08 48 89 fb 48 89 f5 4c 89 e7 e8
> RSP: 0018:ffffc90002b1fc28 EFLAGS: 00000202
> RAX: 00000000001c20ca RBX: 0000000000724e1e RCX: 0000000000000001
> RDX: ffff888118e214c8 RSI: 0000000000057d42 RDI: ffff888118e21518
> RBP: 000000000002bec8 R08: 0000000000000001 R09: 0000000000000000
> R10: 0000000000000bf4 R11: 0000000000000000 R12: 0000000000000001
> R13: 00000000001c20ca R14: 00000000001c20ca R15: 0000000000000000
> FS: 00007f03f995c740(0000) GS:ffff88a07ad9a000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007f03f98f1000 CR3: 0000000144626004 CR4: 0000000000770eb0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> PKRU: 55555554
> Call Trace:
> <TASK>
> shmem_alloc_folio+0x31/0xc0
> shmem_swapin_folio+0x309/0xcf0
> ? filemap_get_entry+0x117/0x1e0
> ? xas_load+0xd/0xb0
> ? filemap_get_entry+0x101/0x1e0
> shmem_get_folio_gfp+0x2ed/0x5b0
> shmem_file_read_iter+0x7f/0x2e0
> vfs_read+0x252/0x330
> ksys_read+0x68/0xf0
> do_syscall_64+0x4c/0x1c0
> entry_SYSCALL_64_after_hwframe+0x76/0x7e
> RIP: 0033:0x7f03f9a46991
> Code: 00 48 8b 15 81 14 10 00 f7 d8 64 89 02 b8 ff ff ff ff eb bd e8 20 ad 01 00 f3 0f 1e fa 80 3d 35 97 10 00 00 74 13 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 4f c3 66 0f 1f 44 00 00 55 48 89 e5 48 83 ec
> RSP: 002b:00007fff3c52bd28 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
> RAX: ffffffffffffffda RBX: 0000000000040000 RCX: 00007f03f9a46991
> RDX: 0000000000040000 RSI: 00007f03f98ba000 RDI: 0000000000000003
> RBP: 00007fff3c52bd50 R08: 0000000000000000 R09: 00007f03f9b9a380
> R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000040000
> R13: 00007f03f98ba000 R14: 0000000000000003 R15: 0000000000000000
> </TASK>
>
> The reason is simple, readahead brought some order 0 folio in swap
> cache, and the swapin mTHP folio being allocated is in confict with it,
> so swapcache_prepare fails and causes shmem_swap_alloc_folio to return
> -EEXIST, and shmem simply retries again and again causing this loop.
>
> Fix it by applying a similar fix for anon mTHP swapin.
>
> The performance change is very slight, time of swapin 10g zero folios
> with shmem (test for 12 times):
> Before: 2.47s
> After: 2.48s
>
> Fixes: 1dd44c0af4fa1 ("mm: shmem: skip swapcache for swapin of synchronous swap device")
> Signed-off-by: Kairui Song <kasong@tencent.com>
>
> ---
>
> V1: https://lore.kernel.org/linux-mm/20250608192713.95875-1-ryncsn@gmail.com/
> Updates:
> - Move non_swapcache_batch check before swapcache_prepare, I was
> expecting this could improve the performance, turns out it barely
> helps and may even cause more overhead in some cases. [ Barry Song ]
> - Remove zero map check, no need to do that for shmem [ Barry Song,
> Baolin Wang ]
> - Fix build bot error.
>
> mm/memory.c | 20 --------------------
> mm/shmem.c | 4 +++-
> mm/swap.h | 23 +++++++++++++++++++++++
> 3 files changed, 26 insertions(+), 21 deletions(-)
>
> diff --git a/mm/memory.c b/mm/memory.c
> index 9ead7ab07e8e..3845ed068d74 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -4313,26 +4313,6 @@ static struct folio *__alloc_swap_folio(struct vm_fault *vmf)
> }
>
> #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> -static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> -{
> - struct swap_info_struct *si = swp_swap_info(entry);
> - pgoff_t offset = swp_offset(entry);
> - int i;
> -
> - /*
> - * While allocating a large folio and doing swap_read_folio, which is
> - * the case the being faulted pte doesn't have swapcache. We need to
> - * ensure all PTEs have no cache as well, otherwise, we might go to
> - * swap devices while the content is in swapcache.
> - */
> - for (i = 0; i < max_nr; i++) {
> - if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> - return i;
> - }
> -
> - return i;
> -}
> -
> /*
> * Check if the PTEs within a range are contiguous swap entries
> * and have consistent swapcache, zeromap.
> diff --git a/mm/shmem.c b/mm/shmem.c
> index 73182e904f9c..a4fdfbd086f1 100644
> --- a/mm/shmem.c
> +++ b/mm/shmem.c
> @@ -2256,6 +2256,7 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> folio = swap_cache_get_folio(swap, NULL, 0);
> order = xa_get_order(&mapping->i_pages, index);
> if (!folio) {
> + int nr_pages = 1 << order;
> bool fallback_order0 = false;
>
> /* Or update major stats only when swapin succeeds?? */
> @@ -2271,7 +2272,8 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> * to swapin order-0 folio, as well as for zswap case.
> */
> if (order > 0 && ((vma && unlikely(userfaultfd_armed(vma))) ||
> - !zswap_never_enabled()))
> + !zswap_never_enabled() ||
> + non_swapcache_batch(swap, nr_pages) != nr_pages))
> fallback_order0 = true;
>
> /* Skip swapcache for synchronous device. */
> diff --git a/mm/swap.h b/mm/swap.h
> index e87a0f19a0ee..911ad5ff0f89 100644
> --- a/mm/swap.h
> +++ b/mm/swap.h
> @@ -108,6 +108,25 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
> return find_next_bit(sis->zeromap, end, start) - start;
> }
>
> +static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> +{
> + struct swap_info_struct *si = swp_swap_info(entry);
> + pgoff_t offset = swp_offset(entry);
> + int i;
> +
> + /*
> + * While allocating a large folio and doing mTHP swapin, we need to
> + * ensure all entries are not cached, otherwise, the mTHP folio will
> + * be in conflict with the folio in swap cache.
> + */
> + for (i = 0; i < max_nr; i++) {
> + if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> + return i;
> + }
> +
> + return i;
> +}
> +
> #else /* CONFIG_SWAP */
> struct swap_iocb;
> static inline void swap_read_folio(struct folio *folio, struct swap_iocb **plug)
> @@ -202,6 +221,10 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
> return 0;
> }
>
> +static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> +{
> + return 0;
> +}
> #endif /* CONFIG_SWAP */
>
> /**
> --
> 2.49.0
>
I really should Cc stable for this, sorry I forgot it.
Cc: stable@vger.kernel.org # 6.14
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin
2025-06-09 17:17 [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin Kairui Song
2025-06-09 17:24 ` Kairui Song
@ 2025-06-09 18:16 ` Nhat Pham
2025-06-09 21:19 ` Barry Song
2025-06-10 12:07 ` Kairui Song
1 sibling, 2 replies; 7+ messages in thread
From: Nhat Pham @ 2025-06-09 18:16 UTC (permalink / raw)
To: Kairui Song
Cc: linux-mm, Andrew Morton, Hugh Dickins, Baolin Wang, Kemeng Shi,
Chris Li, Baoquan He, Barry Song, Usama Arif, linux-kernel
On Mon, Jun 9, 2025 at 10:18 AM Kairui Song <ryncsn@gmail.com> wrote:
>
> From: Kairui Song <kasong@tencent.com>
>
> Following softlockup can be easily reproduced on my test machine with:
>
> echo always > /sys/kernel/mm/transparent_hugepage/hugepages-64kB/enabled
> swapon /dev/zram0 # zram0 is a 48G swap device
> mkdir -p /sys/fs/cgroup/memory/test
> echo 1G > /sys/fs/cgroup/test/memory.max
> echo $BASHPID > /sys/fs/cgroup/test/cgroup.procs
> while true; do
> dd if=/dev/zero of=/tmp/test.img bs=1M count=5120
> cat /tmp/test.img > /dev/null
> rm /tmp/test.img
> done
>
> Then after a while:
> watchdog: BUG: soft lockup - CPU#0 stuck for 763s! [cat:5787]
> Modules linked in: zram virtiofs
> CPU: 0 UID: 0 PID: 5787 Comm: cat Kdump: loaded Tainted: G L 6.15.0.orig-gf3021d9246bc-dirty #118 PREEMPT(voluntary)·
> Tainted: [L]=SOFTLOCKUP
> Hardware name: Red Hat KVM/RHEL-AV, BIOS 0.0.0 02/06/2015
> RIP: 0010:mpol_shared_policy_lookup+0xd/0x70
> Code: e9 b8 b4 ff ff 31 c0 c3 cc cc cc cc 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 41 54 55 53 <48> 8b 1f 48 85 db 74 41 4c 8d 67 08 48 89 fb 48 89 f5 4c 89 e7 e8
> RSP: 0018:ffffc90002b1fc28 EFLAGS: 00000202
> RAX: 00000000001c20ca RBX: 0000000000724e1e RCX: 0000000000000001
> RDX: ffff888118e214c8 RSI: 0000000000057d42 RDI: ffff888118e21518
> RBP: 000000000002bec8 R08: 0000000000000001 R09: 0000000000000000
> R10: 0000000000000bf4 R11: 0000000000000000 R12: 0000000000000001
> R13: 00000000001c20ca R14: 00000000001c20ca R15: 0000000000000000
> FS: 00007f03f995c740(0000) GS:ffff88a07ad9a000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007f03f98f1000 CR3: 0000000144626004 CR4: 0000000000770eb0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> PKRU: 55555554
> Call Trace:
> <TASK>
> shmem_alloc_folio+0x31/0xc0
> shmem_swapin_folio+0x309/0xcf0
> ? filemap_get_entry+0x117/0x1e0
> ? xas_load+0xd/0xb0
> ? filemap_get_entry+0x101/0x1e0
> shmem_get_folio_gfp+0x2ed/0x5b0
> shmem_file_read_iter+0x7f/0x2e0
> vfs_read+0x252/0x330
> ksys_read+0x68/0xf0
> do_syscall_64+0x4c/0x1c0
> entry_SYSCALL_64_after_hwframe+0x76/0x7e
> RIP: 0033:0x7f03f9a46991
> Code: 00 48 8b 15 81 14 10 00 f7 d8 64 89 02 b8 ff ff ff ff eb bd e8 20 ad 01 00 f3 0f 1e fa 80 3d 35 97 10 00 00 74 13 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 4f c3 66 0f 1f 44 00 00 55 48 89 e5 48 83 ec
> RSP: 002b:00007fff3c52bd28 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
> RAX: ffffffffffffffda RBX: 0000000000040000 RCX: 00007f03f9a46991
> RDX: 0000000000040000 RSI: 00007f03f98ba000 RDI: 0000000000000003
> RBP: 00007fff3c52bd50 R08: 0000000000000000 R09: 00007f03f9b9a380
> R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000040000
> R13: 00007f03f98ba000 R14: 0000000000000003 R15: 0000000000000000
> </TASK>
>
> The reason is simple, readahead brought some order 0 folio in swap
> cache, and the swapin mTHP folio being allocated is in confict with it,
> so swapcache_prepare fails and causes shmem_swap_alloc_folio to return
> -EEXIST, and shmem simply retries again and again causing this loop.
>
> Fix it by applying a similar fix for anon mTHP swapin.
>
> The performance change is very slight, time of swapin 10g zero folios
> with shmem (test for 12 times):
> Before: 2.47s
> After: 2.48s
>
> Fixes: 1dd44c0af4fa1 ("mm: shmem: skip swapcache for swapin of synchronous swap device")
> Signed-off-by: Kairui Song <kasong@tencent.com>
LGTM.
Acked-by: Nhat Pham <nphamcs@gmail.com>
>
> ---
>
> V1: https://lore.kernel.org/linux-mm/20250608192713.95875-1-ryncsn@gmail.com/
> Updates:
> - Move non_swapcache_batch check before swapcache_prepare, I was
> expecting this could improve the performance, turns out it barely
> helps and may even cause more overhead in some cases. [ Barry Song ]
> - Remove zero map check, no need to do that for shmem [ Barry Song,
> Baolin Wang ]
> - Fix build bot error.
>
> mm/memory.c | 20 --------------------
> mm/shmem.c | 4 +++-
> mm/swap.h | 23 +++++++++++++++++++++++
> 3 files changed, 26 insertions(+), 21 deletions(-)
>
> diff --git a/mm/memory.c b/mm/memory.c
> index 9ead7ab07e8e..3845ed068d74 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -4313,26 +4313,6 @@ static struct folio *__alloc_swap_folio(struct vm_fault *vmf)
> }
>
> #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> -static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> -{
> - struct swap_info_struct *si = swp_swap_info(entry);
> - pgoff_t offset = swp_offset(entry);
> - int i;
> -
> - /*
> - * While allocating a large folio and doing swap_read_folio, which is
> - * the case the being faulted pte doesn't have swapcache. We need to
> - * ensure all PTEs have no cache as well, otherwise, we might go to
> - * swap devices while the content is in swapcache.
> - */
> - for (i = 0; i < max_nr; i++) {
> - if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> - return i;
> - }
> -
> - return i;
> -}
> -
> /*
> * Check if the PTEs within a range are contiguous swap entries
> * and have consistent swapcache, zeromap.
> diff --git a/mm/shmem.c b/mm/shmem.c
> index 73182e904f9c..a4fdfbd086f1 100644
> --- a/mm/shmem.c
> +++ b/mm/shmem.c
> @@ -2256,6 +2256,7 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> folio = swap_cache_get_folio(swap, NULL, 0);
> order = xa_get_order(&mapping->i_pages, index);
> if (!folio) {
> + int nr_pages = 1 << order;
> bool fallback_order0 = false;
>
> /* Or update major stats only when swapin succeeds?? */
> @@ -2271,7 +2272,8 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> * to swapin order-0 folio, as well as for zswap case.
nit: maybe update the above comment?
> */
> if (order > 0 && ((vma && unlikely(userfaultfd_armed(vma))) ||
> - !zswap_never_enabled()))
> + !zswap_never_enabled() ||
> + non_swapcache_batch(swap, nr_pages) != nr_pages))
> fallback_order0 = true;
>
> /* Skip swapcache for synchronous device. */
> diff --git a/mm/swap.h b/mm/swap.h
> index e87a0f19a0ee..911ad5ff0f89 100644
> --- a/mm/swap.h
> +++ b/mm/swap.h
> @@ -108,6 +108,25 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
> return find_next_bit(sis->zeromap, end, start) - start;
> }
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin
2025-06-09 17:24 ` Kairui Song
@ 2025-06-09 21:16 ` Barry Song
2025-06-10 12:09 ` Kairui Song
0 siblings, 1 reply; 7+ messages in thread
From: Barry Song @ 2025-06-09 21:16 UTC (permalink / raw)
To: Kairui Song
Cc: linux-mm, stable, Andrew Morton, Hugh Dickins, Baolin Wang,
Kemeng Shi, Chris Li, Nhat Pham, Baoquan He, Usama Arif,
linux-kernel
On Tue, Jun 10, 2025 at 5:24 AM Kairui Song <ryncsn@gmail.com> wrote:
>
> On Tue, Jun 10, 2025 at 1:18 AM Kairui Song <ryncsn@gmail.com> wrote:
> >
> > From: Kairui Song <kasong@tencent.com>
> >
> > Following softlockup can be easily reproduced on my test machine with:
> >
> > echo always > /sys/kernel/mm/transparent_hugepage/hugepages-64kB/enabled
> > swapon /dev/zram0 # zram0 is a 48G swap device
> > mkdir -p /sys/fs/cgroup/memory/test
> > echo 1G > /sys/fs/cgroup/test/memory.max
> > echo $BASHPID > /sys/fs/cgroup/test/cgroup.procs
> > while true; do
> > dd if=/dev/zero of=/tmp/test.img bs=1M count=5120
> > cat /tmp/test.img > /dev/null
> > rm /tmp/test.img
> > done
> >
> > Then after a while:
> > watchdog: BUG: soft lockup - CPU#0 stuck for 763s! [cat:5787]
> > Modules linked in: zram virtiofs
> > CPU: 0 UID: 0 PID: 5787 Comm: cat Kdump: loaded Tainted: G L 6.15.0.orig-gf3021d9246bc-dirty #118 PREEMPT(voluntary)·
> > Tainted: [L]=SOFTLOCKUP
> > Hardware name: Red Hat KVM/RHEL-AV, BIOS 0.0.0 02/06/2015
> > RIP: 0010:mpol_shared_policy_lookup+0xd/0x70
> > Code: e9 b8 b4 ff ff 31 c0 c3 cc cc cc cc 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 41 54 55 53 <48> 8b 1f 48 85 db 74 41 4c 8d 67 08 48 89 fb 48 89 f5 4c 89 e7 e8
> > RSP: 0018:ffffc90002b1fc28 EFLAGS: 00000202
> > RAX: 00000000001c20ca RBX: 0000000000724e1e RCX: 0000000000000001
> > RDX: ffff888118e214c8 RSI: 0000000000057d42 RDI: ffff888118e21518
> > RBP: 000000000002bec8 R08: 0000000000000001 R09: 0000000000000000
> > R10: 0000000000000bf4 R11: 0000000000000000 R12: 0000000000000001
> > R13: 00000000001c20ca R14: 00000000001c20ca R15: 0000000000000000
> > FS: 00007f03f995c740(0000) GS:ffff88a07ad9a000(0000) knlGS:0000000000000000
> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: 00007f03f98f1000 CR3: 0000000144626004 CR4: 0000000000770eb0
> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > PKRU: 55555554
> > Call Trace:
> > <TASK>
> > shmem_alloc_folio+0x31/0xc0
> > shmem_swapin_folio+0x309/0xcf0
> > ? filemap_get_entry+0x117/0x1e0
> > ? xas_load+0xd/0xb0
> > ? filemap_get_entry+0x101/0x1e0
> > shmem_get_folio_gfp+0x2ed/0x5b0
> > shmem_file_read_iter+0x7f/0x2e0
> > vfs_read+0x252/0x330
> > ksys_read+0x68/0xf0
> > do_syscall_64+0x4c/0x1c0
> > entry_SYSCALL_64_after_hwframe+0x76/0x7e
> > RIP: 0033:0x7f03f9a46991
> > Code: 00 48 8b 15 81 14 10 00 f7 d8 64 89 02 b8 ff ff ff ff eb bd e8 20 ad 01 00 f3 0f 1e fa 80 3d 35 97 10 00 00 74 13 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 4f c3 66 0f 1f 44 00 00 55 48 89 e5 48 83 ec
> > RSP: 002b:00007fff3c52bd28 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
> > RAX: ffffffffffffffda RBX: 0000000000040000 RCX: 00007f03f9a46991
> > RDX: 0000000000040000 RSI: 00007f03f98ba000 RDI: 0000000000000003
> > RBP: 00007fff3c52bd50 R08: 0000000000000000 R09: 00007f03f9b9a380
> > R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000040000
> > R13: 00007f03f98ba000 R14: 0000000000000003 R15: 0000000000000000
> > </TASK>
> >
> > The reason is simple, readahead brought some order 0 folio in swap
> > cache, and the swapin mTHP folio being allocated is in confict with it,
> > so swapcache_prepare fails and causes shmem_swap_alloc_folio to return
> > -EEXIST, and shmem simply retries again and again causing this loop.
> >
> > Fix it by applying a similar fix for anon mTHP swapin.
> >
> > The performance change is very slight, time of swapin 10g zero folios
> > with shmem (test for 12 times):
> > Before: 2.47s
> > After: 2.48s
> >
> > Fixes: 1dd44c0af4fa1 ("mm: shmem: skip swapcache for swapin of synchronous swap device")
> > Signed-off-by: Kairui Song <kasong@tencent.com>
Reviewed-by: Barry Song <baohua@kernel.org>
> >
> > ---
> >
> > V1: https://lore.kernel.org/linux-mm/20250608192713.95875-1-ryncsn@gmail.com/
> > Updates:
> > - Move non_swapcache_batch check before swapcache_prepare, I was
> > expecting this could improve the performance, turns out it barely
> > helps and may even cause more overhead in some cases. [ Barry Song ]
> > - Remove zero map check, no need to do that for shmem [ Barry Song,
> > Baolin Wang ]
> > - Fix build bot error.
> >
> > mm/memory.c | 20 --------------------
> > mm/shmem.c | 4 +++-
> > mm/swap.h | 23 +++++++++++++++++++++++
> > 3 files changed, 26 insertions(+), 21 deletions(-)
> >
> > diff --git a/mm/memory.c b/mm/memory.c
> > index 9ead7ab07e8e..3845ed068d74 100644
> > --- a/mm/memory.c
> > +++ b/mm/memory.c
> > @@ -4313,26 +4313,6 @@ static struct folio *__alloc_swap_folio(struct vm_fault *vmf)
> > }
> >
> > #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> > -static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> > -{
> > - struct swap_info_struct *si = swp_swap_info(entry);
> > - pgoff_t offset = swp_offset(entry);
> > - int i;
> > -
> > - /*
> > - * While allocating a large folio and doing swap_read_folio, which is
> > - * the case the being faulted pte doesn't have swapcache. We need to
> > - * ensure all PTEs have no cache as well, otherwise, we might go to
> > - * swap devices while the content is in swapcache.
> > - */
> > - for (i = 0; i < max_nr; i++) {
> > - if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> > - return i;
> > - }
> > -
> > - return i;
> > -}
> > -
> > /*
> > * Check if the PTEs within a range are contiguous swap entries
> > * and have consistent swapcache, zeromap.
> > diff --git a/mm/shmem.c b/mm/shmem.c
> > index 73182e904f9c..a4fdfbd086f1 100644
> > --- a/mm/shmem.c
> > +++ b/mm/shmem.c
> > @@ -2256,6 +2256,7 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> > folio = swap_cache_get_folio(swap, NULL, 0);
> > order = xa_get_order(&mapping->i_pages, index);
> > if (!folio) {
> > + int nr_pages = 1 << order;
> > bool fallback_order0 = false;
> >
> > /* Or update major stats only when swapin succeeds?? */
> > @@ -2271,7 +2272,8 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> > * to swapin order-0 folio, as well as for zswap case.
> > */
> > if (order > 0 && ((vma && unlikely(userfaultfd_armed(vma))) ||
> > - !zswap_never_enabled()))
> > + !zswap_never_enabled() ||
> > + non_swapcache_batch(swap, nr_pages) != nr_pages))
> > fallback_order0 = true;
> >
> > /* Skip swapcache for synchronous device. */
> > diff --git a/mm/swap.h b/mm/swap.h
> > index e87a0f19a0ee..911ad5ff0f89 100644
> > --- a/mm/swap.h
> > +++ b/mm/swap.h
> > @@ -108,6 +108,25 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
> > return find_next_bit(sis->zeromap, end, start) - start;
> > }
> >
> > +static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> > +{
> > + struct swap_info_struct *si = swp_swap_info(entry);
> > + pgoff_t offset = swp_offset(entry);
> > + int i;
> > +
> > + /*
> > + * While allocating a large folio and doing mTHP swapin, we need to
> > + * ensure all entries are not cached, otherwise, the mTHP folio will
> > + * be in conflict with the folio in swap cache.
> > + */
> > + for (i = 0; i < max_nr; i++) {
> > + if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> > + return i;
> > + }
> > +
> > + return i;
> > +}
Nit:
This was previously under CONFIG_TRANSPARENT_HUGEPAGE, but is now
guarded by CONFIG_SWAP. I assume there's no performance impact, since
if (order > 0) would never be true on platforms where THP is
disabled, meaning non_swapcache_batch() would never be called?
> > +
> > #else /* CONFIG_SWAP */
> > struct swap_iocb;
> > static inline void swap_read_folio(struct folio *folio, struct swap_iocb **plug)
> > @@ -202,6 +221,10 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
> > return 0;
> > }
> >
> > +static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> > +{
> > + return 0;
> > +}
> > #endif /* CONFIG_SWAP */
> >
> > /**
> > --
> > 2.49.0
> >
>
> I really should Cc stable for this, sorry I forgot it.
>
> Cc: stable@vger.kernel.org # 6.14
Thanks
Barry
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin
2025-06-09 18:16 ` Nhat Pham
@ 2025-06-09 21:19 ` Barry Song
2025-06-10 12:07 ` Kairui Song
1 sibling, 0 replies; 7+ messages in thread
From: Barry Song @ 2025-06-09 21:19 UTC (permalink / raw)
To: Nhat Pham
Cc: Kairui Song, linux-mm, Andrew Morton, Hugh Dickins, Baolin Wang,
Kemeng Shi, Chris Li, Baoquan He, Usama Arif, linux-kernel
On Tue, Jun 10, 2025 at 6:16 AM Nhat Pham <nphamcs@gmail.com> wrote:
>
> On Mon, Jun 9, 2025 at 10:18 AM Kairui Song <ryncsn@gmail.com> wrote:
> >
> > From: Kairui Song <kasong@tencent.com>
> >
> > Following softlockup can be easily reproduced on my test machine with:
> >
> > echo always > /sys/kernel/mm/transparent_hugepage/hugepages-64kB/enabled
> > swapon /dev/zram0 # zram0 is a 48G swap device
> > mkdir -p /sys/fs/cgroup/memory/test
> > echo 1G > /sys/fs/cgroup/test/memory.max
> > echo $BASHPID > /sys/fs/cgroup/test/cgroup.procs
> > while true; do
> > dd if=/dev/zero of=/tmp/test.img bs=1M count=5120
> > cat /tmp/test.img > /dev/null
> > rm /tmp/test.img
> > done
> >
> > Then after a while:
> > watchdog: BUG: soft lockup - CPU#0 stuck for 763s! [cat:5787]
> > Modules linked in: zram virtiofs
> > CPU: 0 UID: 0 PID: 5787 Comm: cat Kdump: loaded Tainted: G L 6.15.0.orig-gf3021d9246bc-dirty #118 PREEMPT(voluntary)·
> > Tainted: [L]=SOFTLOCKUP
> > Hardware name: Red Hat KVM/RHEL-AV, BIOS 0.0.0 02/06/2015
> > RIP: 0010:mpol_shared_policy_lookup+0xd/0x70
> > Code: e9 b8 b4 ff ff 31 c0 c3 cc cc cc cc 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 41 54 55 53 <48> 8b 1f 48 85 db 74 41 4c 8d 67 08 48 89 fb 48 89 f5 4c 89 e7 e8
> > RSP: 0018:ffffc90002b1fc28 EFLAGS: 00000202
> > RAX: 00000000001c20ca RBX: 0000000000724e1e RCX: 0000000000000001
> > RDX: ffff888118e214c8 RSI: 0000000000057d42 RDI: ffff888118e21518
> > RBP: 000000000002bec8 R08: 0000000000000001 R09: 0000000000000000
> > R10: 0000000000000bf4 R11: 0000000000000000 R12: 0000000000000001
> > R13: 00000000001c20ca R14: 00000000001c20ca R15: 0000000000000000
> > FS: 00007f03f995c740(0000) GS:ffff88a07ad9a000(0000) knlGS:0000000000000000
> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: 00007f03f98f1000 CR3: 0000000144626004 CR4: 0000000000770eb0
> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > PKRU: 55555554
> > Call Trace:
> > <TASK>
> > shmem_alloc_folio+0x31/0xc0
> > shmem_swapin_folio+0x309/0xcf0
> > ? filemap_get_entry+0x117/0x1e0
> > ? xas_load+0xd/0xb0
> > ? filemap_get_entry+0x101/0x1e0
> > shmem_get_folio_gfp+0x2ed/0x5b0
> > shmem_file_read_iter+0x7f/0x2e0
> > vfs_read+0x252/0x330
> > ksys_read+0x68/0xf0
> > do_syscall_64+0x4c/0x1c0
> > entry_SYSCALL_64_after_hwframe+0x76/0x7e
> > RIP: 0033:0x7f03f9a46991
> > Code: 00 48 8b 15 81 14 10 00 f7 d8 64 89 02 b8 ff ff ff ff eb bd e8 20 ad 01 00 f3 0f 1e fa 80 3d 35 97 10 00 00 74 13 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 4f c3 66 0f 1f 44 00 00 55 48 89 e5 48 83 ec
> > RSP: 002b:00007fff3c52bd28 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
> > RAX: ffffffffffffffda RBX: 0000000000040000 RCX: 00007f03f9a46991
> > RDX: 0000000000040000 RSI: 00007f03f98ba000 RDI: 0000000000000003
> > RBP: 00007fff3c52bd50 R08: 0000000000000000 R09: 00007f03f9b9a380
> > R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000040000
> > R13: 00007f03f98ba000 R14: 0000000000000003 R15: 0000000000000000
> > </TASK>
> >
> > The reason is simple, readahead brought some order 0 folio in swap
> > cache, and the swapin mTHP folio being allocated is in confict with it,
> > so swapcache_prepare fails and causes shmem_swap_alloc_folio to return
> > -EEXIST, and shmem simply retries again and again causing this loop.
> >
> > Fix it by applying a similar fix for anon mTHP swapin.
> >
> > The performance change is very slight, time of swapin 10g zero folios
> > with shmem (test for 12 times):
> > Before: 2.47s
> > After: 2.48s
> >
> > Fixes: 1dd44c0af4fa1 ("mm: shmem: skip swapcache for swapin of synchronous swap device")
> > Signed-off-by: Kairui Song <kasong@tencent.com>
>
> LGTM.
> Acked-by: Nhat Pham <nphamcs@gmail.com>
>
> >
> > ---
> >
> > V1: https://lore.kernel.org/linux-mm/20250608192713.95875-1-ryncsn@gmail.com/
> > Updates:
> > - Move non_swapcache_batch check before swapcache_prepare, I was
> > expecting this could improve the performance, turns out it barely
> > helps and may even cause more overhead in some cases. [ Barry Song ]
> > - Remove zero map check, no need to do that for shmem [ Barry Song,
> > Baolin Wang ]
> > - Fix build bot error.
> >
> > mm/memory.c | 20 --------------------
> > mm/shmem.c | 4 +++-
> > mm/swap.h | 23 +++++++++++++++++++++++
> > 3 files changed, 26 insertions(+), 21 deletions(-)
> >
> > diff --git a/mm/memory.c b/mm/memory.c
> > index 9ead7ab07e8e..3845ed068d74 100644
> > --- a/mm/memory.c
> > +++ b/mm/memory.c
> > @@ -4313,26 +4313,6 @@ static struct folio *__alloc_swap_folio(struct vm_fault *vmf)
> > }
> >
> > #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> > -static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> > -{
> > - struct swap_info_struct *si = swp_swap_info(entry);
> > - pgoff_t offset = swp_offset(entry);
> > - int i;
> > -
> > - /*
> > - * While allocating a large folio and doing swap_read_folio, which is
> > - * the case the being faulted pte doesn't have swapcache. We need to
> > - * ensure all PTEs have no cache as well, otherwise, we might go to
> > - * swap devices while the content is in swapcache.
> > - */
> > - for (i = 0; i < max_nr; i++) {
> > - if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> > - return i;
> > - }
> > -
> > - return i;
> > -}
> > -
> > /*
> > * Check if the PTEs within a range are contiguous swap entries
> > * and have consistent swapcache, zeromap.
> > diff --git a/mm/shmem.c b/mm/shmem.c
> > index 73182e904f9c..a4fdfbd086f1 100644
> > --- a/mm/shmem.c
> > +++ b/mm/shmem.c
> > @@ -2256,6 +2256,7 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> > folio = swap_cache_get_folio(swap, NULL, 0);
> > order = xa_get_order(&mapping->i_pages, index);
> > if (!folio) {
> > + int nr_pages = 1 << order;
> > bool fallback_order0 = false;
> >
> > /* Or update major stats only when swapin succeeds?? */
> > @@ -2271,7 +2272,8 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> > * to swapin order-0 folio, as well as for zswap case.
>
> nit: maybe update the above comment?
+1
>
> > */
> > if (order > 0 && ((vma && unlikely(userfaultfd_armed(vma))) ||
> > - !zswap_never_enabled()))
> > + !zswap_never_enabled() ||
> > + non_swapcache_batch(swap, nr_pages) != nr_pages))
> > fallback_order0 = true;
> >
> > /* Skip swapcache for synchronous device. */
> > diff --git a/mm/swap.h b/mm/swap.h
> > index e87a0f19a0ee..911ad5ff0f89 100644
> > --- a/mm/swap.h
> > +++ b/mm/swap.h
> > @@ -108,6 +108,25 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
> > return find_next_bit(sis->zeromap, end, start) - start;
> > }
> >
Thanks
Barry
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin
2025-06-09 18:16 ` Nhat Pham
2025-06-09 21:19 ` Barry Song
@ 2025-06-10 12:07 ` Kairui Song
1 sibling, 0 replies; 7+ messages in thread
From: Kairui Song @ 2025-06-10 12:07 UTC (permalink / raw)
To: Nhat Pham
Cc: linux-mm, Andrew Morton, Hugh Dickins, Baolin Wang, Kemeng Shi,
Chris Li, Baoquan He, Barry Song, Usama Arif, linux-kernel
On Tue, Jun 10, 2025 at 2:16 AM Nhat Pham <nphamcs@gmail.com> wrote:
>
> On Mon, Jun 9, 2025 at 10:18 AM Kairui Song <ryncsn@gmail.com> wrote:
> >
> > From: Kairui Song <kasong@tencent.com>
> >
> > Following softlockup can be easily reproduced on my test machine with:
> >
> > echo always > /sys/kernel/mm/transparent_hugepage/hugepages-64kB/enabled
> > swapon /dev/zram0 # zram0 is a 48G swap device
> > mkdir -p /sys/fs/cgroup/memory/test
> > echo 1G > /sys/fs/cgroup/test/memory.max
> > echo $BASHPID > /sys/fs/cgroup/test/cgroup.procs
> > while true; do
> > dd if=/dev/zero of=/tmp/test.img bs=1M count=5120
> > cat /tmp/test.img > /dev/null
> > rm /tmp/test.img
> > done
> >
> > Then after a while:
> > watchdog: BUG: soft lockup - CPU#0 stuck for 763s! [cat:5787]
> > Modules linked in: zram virtiofs
> > CPU: 0 UID: 0 PID: 5787 Comm: cat Kdump: loaded Tainted: G L 6.15.0.orig-gf3021d9246bc-dirty #118 PREEMPT(voluntary)·
> > Tainted: [L]=SOFTLOCKUP
> > Hardware name: Red Hat KVM/RHEL-AV, BIOS 0.0.0 02/06/2015
> > RIP: 0010:mpol_shared_policy_lookup+0xd/0x70
> > Code: e9 b8 b4 ff ff 31 c0 c3 cc cc cc cc 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 41 54 55 53 <48> 8b 1f 48 85 db 74 41 4c 8d 67 08 48 89 fb 48 89 f5 4c 89 e7 e8
> > RSP: 0018:ffffc90002b1fc28 EFLAGS: 00000202
> > RAX: 00000000001c20ca RBX: 0000000000724e1e RCX: 0000000000000001
> > RDX: ffff888118e214c8 RSI: 0000000000057d42 RDI: ffff888118e21518
> > RBP: 000000000002bec8 R08: 0000000000000001 R09: 0000000000000000
> > R10: 0000000000000bf4 R11: 0000000000000000 R12: 0000000000000001
> > R13: 00000000001c20ca R14: 00000000001c20ca R15: 0000000000000000
> > FS: 00007f03f995c740(0000) GS:ffff88a07ad9a000(0000) knlGS:0000000000000000
> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: 00007f03f98f1000 CR3: 0000000144626004 CR4: 0000000000770eb0
> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > PKRU: 55555554
> > Call Trace:
> > <TASK>
> > shmem_alloc_folio+0x31/0xc0
> > shmem_swapin_folio+0x309/0xcf0
> > ? filemap_get_entry+0x117/0x1e0
> > ? xas_load+0xd/0xb0
> > ? filemap_get_entry+0x101/0x1e0
> > shmem_get_folio_gfp+0x2ed/0x5b0
> > shmem_file_read_iter+0x7f/0x2e0
> > vfs_read+0x252/0x330
> > ksys_read+0x68/0xf0
> > do_syscall_64+0x4c/0x1c0
> > entry_SYSCALL_64_after_hwframe+0x76/0x7e
> > RIP: 0033:0x7f03f9a46991
> > Code: 00 48 8b 15 81 14 10 00 f7 d8 64 89 02 b8 ff ff ff ff eb bd e8 20 ad 01 00 f3 0f 1e fa 80 3d 35 97 10 00 00 74 13 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 4f c3 66 0f 1f 44 00 00 55 48 89 e5 48 83 ec
> > RSP: 002b:00007fff3c52bd28 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
> > RAX: ffffffffffffffda RBX: 0000000000040000 RCX: 00007f03f9a46991
> > RDX: 0000000000040000 RSI: 00007f03f98ba000 RDI: 0000000000000003
> > RBP: 00007fff3c52bd50 R08: 0000000000000000 R09: 00007f03f9b9a380
> > R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000040000
> > R13: 00007f03f98ba000 R14: 0000000000000003 R15: 0000000000000000
> > </TASK>
> >
> > The reason is simple, readahead brought some order 0 folio in swap
> > cache, and the swapin mTHP folio being allocated is in confict with it,
> > so swapcache_prepare fails and causes shmem_swap_alloc_folio to return
> > -EEXIST, and shmem simply retries again and again causing this loop.
> >
> > Fix it by applying a similar fix for anon mTHP swapin.
> >
> > The performance change is very slight, time of swapin 10g zero folios
> > with shmem (test for 12 times):
> > Before: 2.47s
> > After: 2.48s
> >
> > Fixes: 1dd44c0af4fa1 ("mm: shmem: skip swapcache for swapin of synchronous swap device")
> > Signed-off-by: Kairui Song <kasong@tencent.com>
>
> LGTM.
> Acked-by: Nhat Pham <nphamcs@gmail.com>
>
> >
> > ---
> >
> > V1: https://lore.kernel.org/linux-mm/20250608192713.95875-1-ryncsn@gmail.com/
> > Updates:
> > - Move non_swapcache_batch check before swapcache_prepare, I was
> > expecting this could improve the performance, turns out it barely
> > helps and may even cause more overhead in some cases. [ Barry Song ]
> > - Remove zero map check, no need to do that for shmem [ Barry Song,
> > Baolin Wang ]
> > - Fix build bot error.
> >
> > mm/memory.c | 20 --------------------
> > mm/shmem.c | 4 +++-
> > mm/swap.h | 23 +++++++++++++++++++++++
> > 3 files changed, 26 insertions(+), 21 deletions(-)
> >
> > diff --git a/mm/memory.c b/mm/memory.c
> > index 9ead7ab07e8e..3845ed068d74 100644
> > --- a/mm/memory.c
> > +++ b/mm/memory.c
> > @@ -4313,26 +4313,6 @@ static struct folio *__alloc_swap_folio(struct vm_fault *vmf)
> > }
> >
> > #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> > -static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> > -{
> > - struct swap_info_struct *si = swp_swap_info(entry);
> > - pgoff_t offset = swp_offset(entry);
> > - int i;
> > -
> > - /*
> > - * While allocating a large folio and doing swap_read_folio, which is
> > - * the case the being faulted pte doesn't have swapcache. We need to
> > - * ensure all PTEs have no cache as well, otherwise, we might go to
> > - * swap devices while the content is in swapcache.
> > - */
> > - for (i = 0; i < max_nr; i++) {
> > - if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> > - return i;
> > - }
> > -
> > - return i;
> > -}
> > -
> > /*
> > * Check if the PTEs within a range are contiguous swap entries
> > * and have consistent swapcache, zeromap.
> > diff --git a/mm/shmem.c b/mm/shmem.c
> > index 73182e904f9c..a4fdfbd086f1 100644
> > --- a/mm/shmem.c
> > +++ b/mm/shmem.c
> > @@ -2256,6 +2256,7 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> > folio = swap_cache_get_folio(swap, NULL, 0);
> > order = xa_get_order(&mapping->i_pages, index);
> > if (!folio) {
> > + int nr_pages = 1 << order;
> > bool fallback_order0 = false;
> >
> > /* Or update major stats only when swapin succeeds?? */
> > @@ -2271,7 +2272,8 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> > * to swapin order-0 folio, as well as for zswap case.
>
> nit: maybe update the above comment?
Good idea. There are already some comments in non_swapcache_batch and
I've updated it for this commit, but mentioning it here definitely
helps more.
>
> > */
> > if (order > 0 && ((vma && unlikely(userfaultfd_armed(vma))) ||
> > - !zswap_never_enabled()))
> > + !zswap_never_enabled() ||
> > + non_swapcache_batch(swap, nr_pages) != nr_pages))
> > fallback_order0 = true;
> >
> > /* Skip swapcache for synchronous device. */
> > diff --git a/mm/swap.h b/mm/swap.h
> > index e87a0f19a0ee..911ad5ff0f89 100644
> > --- a/mm/swap.h
> > +++ b/mm/swap.h
> > @@ -108,6 +108,25 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
> > return find_next_bit(sis->zeromap, end, start) - start;
> > }
> >
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin
2025-06-09 21:16 ` Barry Song
@ 2025-06-10 12:09 ` Kairui Song
0 siblings, 0 replies; 7+ messages in thread
From: Kairui Song @ 2025-06-10 12:09 UTC (permalink / raw)
To: Barry Song
Cc: linux-mm, stable, Andrew Morton, Hugh Dickins, Baolin Wang,
Kemeng Shi, Chris Li, Nhat Pham, Baoquan He, Usama Arif,
linux-kernel
On Tue, Jun 10, 2025 at 5:17 AM Barry Song <21cnbao@gmail.com> wrote:
>
> On Tue, Jun 10, 2025 at 5:24 AM Kairui Song <ryncsn@gmail.com> wrote:
> >
> > On Tue, Jun 10, 2025 at 1:18 AM Kairui Song <ryncsn@gmail.com> wrote:
> > >
> > > From: Kairui Song <kasong@tencent.com>
> > >
> > > Following softlockup can be easily reproduced on my test machine with:
> > >
> > > echo always > /sys/kernel/mm/transparent_hugepage/hugepages-64kB/enabled
> > > swapon /dev/zram0 # zram0 is a 48G swap device
> > > mkdir -p /sys/fs/cgroup/memory/test
> > > echo 1G > /sys/fs/cgroup/test/memory.max
> > > echo $BASHPID > /sys/fs/cgroup/test/cgroup.procs
> > > while true; do
> > > dd if=/dev/zero of=/tmp/test.img bs=1M count=5120
> > > cat /tmp/test.img > /dev/null
> > > rm /tmp/test.img
> > > done
> > >
> > > Then after a while:
> > > watchdog: BUG: soft lockup - CPU#0 stuck for 763s! [cat:5787]
> > > Modules linked in: zram virtiofs
> > > CPU: 0 UID: 0 PID: 5787 Comm: cat Kdump: loaded Tainted: G L 6.15.0.orig-gf3021d9246bc-dirty #118 PREEMPT(voluntary)·
> > > Tainted: [L]=SOFTLOCKUP
> > > Hardware name: Red Hat KVM/RHEL-AV, BIOS 0.0.0 02/06/2015
> > > RIP: 0010:mpol_shared_policy_lookup+0xd/0x70
> > > Code: e9 b8 b4 ff ff 31 c0 c3 cc cc cc cc 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 41 54 55 53 <48> 8b 1f 48 85 db 74 41 4c 8d 67 08 48 89 fb 48 89 f5 4c 89 e7 e8
> > > RSP: 0018:ffffc90002b1fc28 EFLAGS: 00000202
> > > RAX: 00000000001c20ca RBX: 0000000000724e1e RCX: 0000000000000001
> > > RDX: ffff888118e214c8 RSI: 0000000000057d42 RDI: ffff888118e21518
> > > RBP: 000000000002bec8 R08: 0000000000000001 R09: 0000000000000000
> > > R10: 0000000000000bf4 R11: 0000000000000000 R12: 0000000000000001
> > > R13: 00000000001c20ca R14: 00000000001c20ca R15: 0000000000000000
> > > FS: 00007f03f995c740(0000) GS:ffff88a07ad9a000(0000) knlGS:0000000000000000
> > > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > > CR2: 00007f03f98f1000 CR3: 0000000144626004 CR4: 0000000000770eb0
> > > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > > PKRU: 55555554
> > > Call Trace:
> > > <TASK>
> > > shmem_alloc_folio+0x31/0xc0
> > > shmem_swapin_folio+0x309/0xcf0
> > > ? filemap_get_entry+0x117/0x1e0
> > > ? xas_load+0xd/0xb0
> > > ? filemap_get_entry+0x101/0x1e0
> > > shmem_get_folio_gfp+0x2ed/0x5b0
> > > shmem_file_read_iter+0x7f/0x2e0
> > > vfs_read+0x252/0x330
> > > ksys_read+0x68/0xf0
> > > do_syscall_64+0x4c/0x1c0
> > > entry_SYSCALL_64_after_hwframe+0x76/0x7e
> > > RIP: 0033:0x7f03f9a46991
> > > Code: 00 48 8b 15 81 14 10 00 f7 d8 64 89 02 b8 ff ff ff ff eb bd e8 20 ad 01 00 f3 0f 1e fa 80 3d 35 97 10 00 00 74 13 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 4f c3 66 0f 1f 44 00 00 55 48 89 e5 48 83 ec
> > > RSP: 002b:00007fff3c52bd28 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
> > > RAX: ffffffffffffffda RBX: 0000000000040000 RCX: 00007f03f9a46991
> > > RDX: 0000000000040000 RSI: 00007f03f98ba000 RDI: 0000000000000003
> > > RBP: 00007fff3c52bd50 R08: 0000000000000000 R09: 00007f03f9b9a380
> > > R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000040000
> > > R13: 00007f03f98ba000 R14: 0000000000000003 R15: 0000000000000000
> > > </TASK>
> > >
> > > The reason is simple, readahead brought some order 0 folio in swap
> > > cache, and the swapin mTHP folio being allocated is in confict with it,
> > > so swapcache_prepare fails and causes shmem_swap_alloc_folio to return
> > > -EEXIST, and shmem simply retries again and again causing this loop.
> > >
> > > Fix it by applying a similar fix for anon mTHP swapin.
> > >
> > > The performance change is very slight, time of swapin 10g zero folios
> > > with shmem (test for 12 times):
> > > Before: 2.47s
> > > After: 2.48s
> > >
> > > Fixes: 1dd44c0af4fa1 ("mm: shmem: skip swapcache for swapin of synchronous swap device")
> > > Signed-off-by: Kairui Song <kasong@tencent.com>
>
> Reviewed-by: Barry Song <baohua@kernel.org>
>
> > >
> > > ---
> > >
> > > V1: https://lore.kernel.org/linux-mm/20250608192713.95875-1-ryncsn@gmail.com/
> > > Updates:
> > > - Move non_swapcache_batch check before swapcache_prepare, I was
> > > expecting this could improve the performance, turns out it barely
> > > helps and may even cause more overhead in some cases. [ Barry Song ]
> > > - Remove zero map check, no need to do that for shmem [ Barry Song,
> > > Baolin Wang ]
> > > - Fix build bot error.
> > >
> > > mm/memory.c | 20 --------------------
> > > mm/shmem.c | 4 +++-
> > > mm/swap.h | 23 +++++++++++++++++++++++
> > > 3 files changed, 26 insertions(+), 21 deletions(-)
> > >
> > > diff --git a/mm/memory.c b/mm/memory.c
> > > index 9ead7ab07e8e..3845ed068d74 100644
> > > --- a/mm/memory.c
> > > +++ b/mm/memory.c
> > > @@ -4313,26 +4313,6 @@ static struct folio *__alloc_swap_folio(struct vm_fault *vmf)
> > > }
> > >
> > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> > > -static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> > > -{
> > > - struct swap_info_struct *si = swp_swap_info(entry);
> > > - pgoff_t offset = swp_offset(entry);
> > > - int i;
> > > -
> > > - /*
> > > - * While allocating a large folio and doing swap_read_folio, which is
> > > - * the case the being faulted pte doesn't have swapcache. We need to
> > > - * ensure all PTEs have no cache as well, otherwise, we might go to
> > > - * swap devices while the content is in swapcache.
> > > - */
> > > - for (i = 0; i < max_nr; i++) {
> > > - if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> > > - return i;
> > > - }
> > > -
> > > - return i;
> > > -}
> > > -
> > > /*
> > > * Check if the PTEs within a range are contiguous swap entries
> > > * and have consistent swapcache, zeromap.
> > > diff --git a/mm/shmem.c b/mm/shmem.c
> > > index 73182e904f9c..a4fdfbd086f1 100644
> > > --- a/mm/shmem.c
> > > +++ b/mm/shmem.c
> > > @@ -2256,6 +2256,7 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> > > folio = swap_cache_get_folio(swap, NULL, 0);
> > > order = xa_get_order(&mapping->i_pages, index);
> > > if (!folio) {
> > > + int nr_pages = 1 << order;
> > > bool fallback_order0 = false;
> > >
> > > /* Or update major stats only when swapin succeeds?? */
> > > @@ -2271,7 +2272,8 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index,
> > > * to swapin order-0 folio, as well as for zswap case.
> > > */
> > > if (order > 0 && ((vma && unlikely(userfaultfd_armed(vma))) ||
> > > - !zswap_never_enabled()))
> > > + !zswap_never_enabled() ||
> > > + non_swapcache_batch(swap, nr_pages) != nr_pages))
> > > fallback_order0 = true;
> > >
> > > /* Skip swapcache for synchronous device. */
> > > diff --git a/mm/swap.h b/mm/swap.h
> > > index e87a0f19a0ee..911ad5ff0f89 100644
> > > --- a/mm/swap.h
> > > +++ b/mm/swap.h
> > > @@ -108,6 +108,25 @@ static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr,
> > > return find_next_bit(sis->zeromap, end, start) - start;
> > > }
> > >
> > > +static inline int non_swapcache_batch(swp_entry_t entry, int max_nr)
> > > +{
> > > + struct swap_info_struct *si = swp_swap_info(entry);
> > > + pgoff_t offset = swp_offset(entry);
> > > + int i;
> > > +
> > > + /*
> > > + * While allocating a large folio and doing mTHP swapin, we need to
> > > + * ensure all entries are not cached, otherwise, the mTHP folio will
> > > + * be in conflict with the folio in swap cache.
> > > + */
> > > + for (i = 0; i < max_nr; i++) {
> > > + if ((si->swap_map[offset + i] & SWAP_HAS_CACHE))
> > > + return i;
> > > + }
> > > +
> > > + return i;
> > > +}
>
> Nit:
> This was previously under CONFIG_TRANSPARENT_HUGEPAGE, but is now
> guarded by CONFIG_SWAP. I assume there's no performance impact, since
> if (order > 0) would never be true on platforms where THP is
> disabled, meaning non_swapcache_batch() would never be called?
Yes, there is no performance overhead, order is always 0. But some
clean up can be done later.
The whole `order = xa_get_order` and related checks are not needed at
all for !CONFIG_TRANSPARENT_HUGEPAGE. It's hard to optimize it out
without shuffle too many code here, simply wrap non_swapcache_batch
with CONFIG_TRANSPARENT_HUGEPAGE may cause more trouble for
understanding the code and doesn't help much here, it's really the
caller that should avoid the whole thing. That's not in the scope of a
bugfix.
I have a shmem xarray lookup optimization patch in the swap table
series, now looking at it I found it might be useful for the clean up
here. I'll check if I can send a clean up & optimize series first.
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2025-06-10 12:10 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-06-09 17:17 [PATCH v2] mm/shmem, swap: fix softlockup with mTHP swapin Kairui Song
2025-06-09 17:24 ` Kairui Song
2025-06-09 21:16 ` Barry Song
2025-06-10 12:09 ` Kairui Song
2025-06-09 18:16 ` Nhat Pham
2025-06-09 21:19 ` Barry Song
2025-06-10 12:07 ` Kairui Song
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox