mm/slub.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
When CONFIG_MEM_ALLOC_PROFILING_DEBUG is enabled, the following warning
may be noticed:
[ 3959.023862] ------------[ cut here ]------------
[ 3959.023891] alloc_tag was not cleared (got tag for lib/xarray.c:378)
[ 3959.023947] WARNING: ./include/linux/alloc_tag.h:155 at alloc_tag_add+0x128/0x178, CPU#6: mkfs.ntfs/113998
[ 3959.023978] Modules linked in: dns_resolver tun brd overlay exfat btrfs blake2b libblake2b xor xor_neon raid6_pq loop sctp ip6_udp_tunnel udp_tunnel ext4 crc16 mbcache jbd2 rfkill sunrpc vfat fat sg fuse nfnetlink sr_mod virtio_gpu cdrom drm_client_lib virtio_dma_buf drm_shmem_helper drm_kms_helper ghash_ce drm sm4 backlight virtio_net net_failover virtio_scsi failover virtio_console virtio_blk virtio_mmio dm_mirror dm_region_hash dm_log dm_multipath dm_mod i2c_dev aes_neon_bs aes_ce_blk [last unloaded: hwpoison_inject]
[ 3959.024170] CPU: 6 UID: 0 PID: 113998 Comm: mkfs.ntfs Kdump: loaded Tainted: G W 6.19.0-rc7+ #7 PREEMPT(voluntary)
[ 3959.024182] Tainted: [W]=WARN
[ 3959.024186] Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022
[ 3959.024192] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 3959.024199] pc : alloc_tag_add+0x128/0x178
[ 3959.024207] lr : alloc_tag_add+0x128/0x178
[ 3959.024214] sp : ffff80008b696d60
[ 3959.024219] x29: ffff80008b696d60 x28: 0000000000000000 x27: 0000000000000240
[ 3959.024232] x26: 0000000000000000 x25: 0000000000000240 x24: ffff800085d17860
[ 3959.024245] x23: 0000000000402800 x22: ffff0000c0012dc0 x21: 00000000000002d0
[ 3959.024257] x20: ffff0000e6ef3318 x19: ffff800085ae0410 x18: 0000000000000000
[ 3959.024269] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
[ 3959.024281] x14: 0000000000000000 x13: 0000000000000001 x12: ffff600064101293
[ 3959.024292] x11: 1fffe00064101292 x10: ffff600064101292 x9 : dfff800000000000
[ 3959.024305] x8 : 00009fff9befed6e x7 : ffff000320809493 x6 : 0000000000000001
[ 3959.024316] x5 : ffff000320809490 x4 : ffff600064101293 x3 : ffff800080691838
[ 3959.024328] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000d5bcd640
[ 3959.024340] Call trace:
[ 3959.024346] alloc_tag_add+0x128/0x178 (P)
[ 3959.024355] __alloc_tagging_slab_alloc_hook+0x11c/0x1a8
[ 3959.024362] kmem_cache_alloc_lru_noprof+0x1b8/0x5e8
[ 3959.024369] xas_alloc+0x304/0x4f0
[ 3959.024381] xas_create+0x1e0/0x4a0
[ 3959.024388] xas_store+0x68/0xda8
[ 3959.024395] __filemap_add_folio+0x5b0/0xbd8
[ 3959.024409] filemap_add_folio+0x16c/0x7e0
[ 3959.024416] __filemap_get_folio_mpol+0x2dc/0x9e8
[ 3959.024424] iomap_get_folio+0xfc/0x180
[ 3959.024435] __iomap_get_folio+0x2f8/0x4b8
[ 3959.024441] iomap_write_begin+0x198/0xc18
[ 3959.024448] iomap_write_iter+0x2ec/0x8f8
[ 3959.024454] iomap_file_buffered_write+0x19c/0x290
[ 3959.024461] blkdev_write_iter+0x38c/0x978
[ 3959.024470] vfs_write+0x4d4/0x928
[ 3959.024482] ksys_write+0xfc/0x1f8
[ 3959.024489] __arm64_sys_write+0x74/0xb0
[ 3959.024496] invoke_syscall+0xd4/0x258
[ 3959.024507] el0_svc_common.constprop.0+0xb4/0x240
[ 3959.024514] do_el0_svc+0x48/0x68
[ 3959.024520] el0_svc+0x40/0xf8
[ 3959.024526] el0t_64_sync_handler+0xa0/0xe8
[ 3959.024533] el0t_64_sync+0x1ac/0x1b0
[ 3959.024540] ---[ end trace 0000000000000000 ]---
This is due to a race condition that occurs when two threads concurrently
perform allocation and freeing operations on the same slab object.
When a process is preparing to allocate a slab object, another process
successfully preempts the CPU, and then proceeds to free a slab object.
However, before the freeing process can invoke `alloc_tag_sub()`, it is
preempted again by the original allocating process. At this point, the
allocating process acquires the same slab object, and subsequently triggers
a warning when it invokes `alloc_tag_add()`.
Signed-off-by: Hao Ge <hao.ge@linux.dev>
---
Hi Suren
I'm not sure if my solution still has any issues, so I'd like to
get your suggestions on it. At the very least, my understanding is
that READ_ONCE and WRITE_ONCE should be used in pairs.
I look forward to your suggestions.
Thanks
---
mm/slub.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/mm/slub.c b/mm/slub.c
index f77b7407c51b..0d84fc917a89 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -2261,8 +2261,13 @@ __alloc_tagging_slab_alloc_hook(struct kmem_cache *s, void *object, gfp_t flags)
* If other users appear then mem_alloc_profiling_enabled()
* check should be added before alloc_tag_add().
*/
- if (likely(obj_exts))
+ if (likely(obj_exts)) {
+
+ while (!READ_ONCE(obj_exts->ref.ct))
+ cpu_relax();
+
alloc_tag_add(&obj_exts->ref, current->alloc_tag, s->size);
+ }
else
alloc_tag_set_inaccurate(current->alloc_tag);
}
--
2.25.1On Tue, Feb 03, 2026 at 03:30:06PM +0800, Hao Ge wrote:
> When CONFIG_MEM_ALLOC_PROFILING_DEBUG is enabled, the following warning
> may be noticed:
>
> [ 3959.023862] ------------[ cut here ]------------
> [ 3959.023891] alloc_tag was not cleared (got tag for lib/xarray.c:378)
> [ 3959.023947] WARNING: ./include/linux/alloc_tag.h:155 at alloc_tag_add+0x128/0x178, CPU#6: mkfs.ntfs/113998
> [ 3959.023978] Modules linked in: dns_resolver tun brd overlay exfat btrfs blake2b libblake2b xor xor_neon raid6_pq loop sctp ip6_udp_tunnel udp_tunnel ext4 crc16 mbcache jbd2 rfkill sunrpc vfat fat sg fuse nfnetlink sr_mod virtio_gpu cdrom drm_client_lib virtio_dma_buf drm_shmem_helper drm_kms_helper ghash_ce drm sm4 backlight virtio_net net_failover virtio_scsi failover virtio_console virtio_blk virtio_mmio dm_mirror dm_region_hash dm_log dm_multipath dm_mod i2c_dev aes_neon_bs aes_ce_blk [last unloaded: hwpoison_inject]
> [ 3959.024170] CPU: 6 UID: 0 PID: 113998 Comm: mkfs.ntfs Kdump: loaded Tainted: G W 6.19.0-rc7+ #7 PREEMPT(voluntary)
> [ 3959.024182] Tainted: [W]=WARN
> [ 3959.024186] Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022
> [ 3959.024192] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [ 3959.024199] pc : alloc_tag_add+0x128/0x178
> [ 3959.024207] lr : alloc_tag_add+0x128/0x178
> [ 3959.024214] sp : ffff80008b696d60
> [ 3959.024219] x29: ffff80008b696d60 x28: 0000000000000000 x27: 0000000000000240
> [ 3959.024232] x26: 0000000000000000 x25: 0000000000000240 x24: ffff800085d17860
> [ 3959.024245] x23: 0000000000402800 x22: ffff0000c0012dc0 x21: 00000000000002d0
> [ 3959.024257] x20: ffff0000e6ef3318 x19: ffff800085ae0410 x18: 0000000000000000
> [ 3959.024269] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
> [ 3959.024281] x14: 0000000000000000 x13: 0000000000000001 x12: ffff600064101293
> [ 3959.024292] x11: 1fffe00064101292 x10: ffff600064101292 x9 : dfff800000000000
> [ 3959.024305] x8 : 00009fff9befed6e x7 : ffff000320809493 x6 : 0000000000000001
> [ 3959.024316] x5 : ffff000320809490 x4 : ffff600064101293 x3 : ffff800080691838
> [ 3959.024328] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000d5bcd640
> [ 3959.024340] Call trace:
> [ 3959.024346] alloc_tag_add+0x128/0x178 (P)
> [ 3959.024355] __alloc_tagging_slab_alloc_hook+0x11c/0x1a8
> [ 3959.024362] kmem_cache_alloc_lru_noprof+0x1b8/0x5e8
> [ 3959.024369] xas_alloc+0x304/0x4f0
> [ 3959.024381] xas_create+0x1e0/0x4a0
> [ 3959.024388] xas_store+0x68/0xda8
> [ 3959.024395] __filemap_add_folio+0x5b0/0xbd8
> [ 3959.024409] filemap_add_folio+0x16c/0x7e0
> [ 3959.024416] __filemap_get_folio_mpol+0x2dc/0x9e8
> [ 3959.024424] iomap_get_folio+0xfc/0x180
> [ 3959.024435] __iomap_get_folio+0x2f8/0x4b8
> [ 3959.024441] iomap_write_begin+0x198/0xc18
> [ 3959.024448] iomap_write_iter+0x2ec/0x8f8
> [ 3959.024454] iomap_file_buffered_write+0x19c/0x290
> [ 3959.024461] blkdev_write_iter+0x38c/0x978
> [ 3959.024470] vfs_write+0x4d4/0x928
> [ 3959.024482] ksys_write+0xfc/0x1f8
> [ 3959.024489] __arm64_sys_write+0x74/0xb0
> [ 3959.024496] invoke_syscall+0xd4/0x258
> [ 3959.024507] el0_svc_common.constprop.0+0xb4/0x240
> [ 3959.024514] do_el0_svc+0x48/0x68
> [ 3959.024520] el0_svc+0x40/0xf8
> [ 3959.024526] el0t_64_sync_handler+0xa0/0xe8
> [ 3959.024533] el0t_64_sync+0x1ac/0x1b0
> [ 3959.024540] ---[ end trace 0000000000000000 ]---
Hi Hao, on which commit did you observe this warning?
> This is due to a race condition that occurs when two threads concurrently
> perform allocation and freeing operations on the same slab object.
>
> When a process is preparing to allocate a slab object, another process
> successfully preempts the CPU, and then proceeds to free a slab object.
> However, before the freeing process can invoke `alloc_tag_sub()`, it is
> preempted again by the original allocating process. At this point, the
> allocating process acquires the same slab object, and subsequently triggers
> a warning when it invokes `alloc_tag_add()`.
The explanation doesn't make sense to me, because alloc_tag_sub()
should have been called before it's added back to freelist or sheaf
before other threads can allocate it, or am I missing something?
> Signed-off-by: Hao Ge <hao.ge@linux.dev>
> ---
> mm/slub.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index f77b7407c51b..0d84fc917a89 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -2261,8 +2261,13 @@ __alloc_tagging_slab_alloc_hook(struct kmem_cache *s, void *object, gfp_t flags)
> * If other users appear then mem_alloc_profiling_enabled()
> * check should be added before alloc_tag_add().
> */
> - if (likely(obj_exts))
> + if (likely(obj_exts)) {
> +
> + while (!READ_ONCE(obj_exts->ref.ct))
> + cpu_relax();
I don't think this is acceptable - it shouldn't wait forever even when
there is a real bug that doesn't clear the tag.
> +
> alloc_tag_add(&obj_exts->ref, current->alloc_tag, s->size);
> + }
> else
> alloc_tag_set_inaccurate(current->alloc_tag);
> }
> --
> 2.25.1
>
--
Cheers,
Harry / Hyeonggon
Hi Harry
On 2026/2/3 17:44, Harry Yoo wrote:
> On Tue, Feb 03, 2026 at 03:30:06PM +0800, Hao Ge wrote:
>> When CONFIG_MEM_ALLOC_PROFILING_DEBUG is enabled, the following warning
>> may be noticed:
>>
>> [ 3959.023862] ------------[ cut here ]------------
>> [ 3959.023891] alloc_tag was not cleared (got tag for lib/xarray.c:378)
>> [ 3959.023947] WARNING: ./include/linux/alloc_tag.h:155 at alloc_tag_add+0x128/0x178, CPU#6: mkfs.ntfs/113998
>> [ 3959.023978] Modules linked in: dns_resolver tun brd overlay exfat btrfs blake2b libblake2b xor xor_neon raid6_pq loop sctp ip6_udp_tunnel udp_tunnel ext4 crc16 mbcache jbd2 rfkill sunrpc vfat fat sg fuse nfnetlink sr_mod virtio_gpu cdrom drm_client_lib virtio_dma_buf drm_shmem_helper drm_kms_helper ghash_ce drm sm4 backlight virtio_net net_failover virtio_scsi failover virtio_console virtio_blk virtio_mmio dm_mirror dm_region_hash dm_log dm_multipath dm_mod i2c_dev aes_neon_bs aes_ce_blk [last unloaded: hwpoison_inject]
>> [ 3959.024170] CPU: 6 UID: 0 PID: 113998 Comm: mkfs.ntfs Kdump: loaded Tainted: G W 6.19.0-rc7+ #7 PREEMPT(voluntary)
>> [ 3959.024182] Tainted: [W]=WARN
>> [ 3959.024186] Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022
>> [ 3959.024192] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
>> [ 3959.024199] pc : alloc_tag_add+0x128/0x178
>> [ 3959.024207] lr : alloc_tag_add+0x128/0x178
>> [ 3959.024214] sp : ffff80008b696d60
>> [ 3959.024219] x29: ffff80008b696d60 x28: 0000000000000000 x27: 0000000000000240
>> [ 3959.024232] x26: 0000000000000000 x25: 0000000000000240 x24: ffff800085d17860
>> [ 3959.024245] x23: 0000000000402800 x22: ffff0000c0012dc0 x21: 00000000000002d0
>> [ 3959.024257] x20: ffff0000e6ef3318 x19: ffff800085ae0410 x18: 0000000000000000
>> [ 3959.024269] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
>> [ 3959.024281] x14: 0000000000000000 x13: 0000000000000001 x12: ffff600064101293
>> [ 3959.024292] x11: 1fffe00064101292 x10: ffff600064101292 x9 : dfff800000000000
>> [ 3959.024305] x8 : 00009fff9befed6e x7 : ffff000320809493 x6 : 0000000000000001
>> [ 3959.024316] x5 : ffff000320809490 x4 : ffff600064101293 x3 : ffff800080691838
>> [ 3959.024328] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000d5bcd640
>> [ 3959.024340] Call trace:
>> [ 3959.024346] alloc_tag_add+0x128/0x178 (P)
>> [ 3959.024355] __alloc_tagging_slab_alloc_hook+0x11c/0x1a8
>> [ 3959.024362] kmem_cache_alloc_lru_noprof+0x1b8/0x5e8
>> [ 3959.024369] xas_alloc+0x304/0x4f0
>> [ 3959.024381] xas_create+0x1e0/0x4a0
>> [ 3959.024388] xas_store+0x68/0xda8
>> [ 3959.024395] __filemap_add_folio+0x5b0/0xbd8
>> [ 3959.024409] filemap_add_folio+0x16c/0x7e0
>> [ 3959.024416] __filemap_get_folio_mpol+0x2dc/0x9e8
>> [ 3959.024424] iomap_get_folio+0xfc/0x180
>> [ 3959.024435] __iomap_get_folio+0x2f8/0x4b8
>> [ 3959.024441] iomap_write_begin+0x198/0xc18
>> [ 3959.024448] iomap_write_iter+0x2ec/0x8f8
>> [ 3959.024454] iomap_file_buffered_write+0x19c/0x290
>> [ 3959.024461] blkdev_write_iter+0x38c/0x978
>> [ 3959.024470] vfs_write+0x4d4/0x928
>> [ 3959.024482] ksys_write+0xfc/0x1f8
>> [ 3959.024489] __arm64_sys_write+0x74/0xb0
>> [ 3959.024496] invoke_syscall+0xd4/0x258
>> [ 3959.024507] el0_svc_common.constprop.0+0xb4/0x240
>> [ 3959.024514] do_el0_svc+0x48/0x68
>> [ 3959.024520] el0_svc+0x40/0xf8
>> [ 3959.024526] el0t_64_sync_handler+0xa0/0xe8
>> [ 3959.024533] el0t_64_sync+0x1ac/0x1b0
>> [ 3959.024540] ---[ end trace 0000000000000000 ]---
> Hi Hao, on which commit did you observe this warning?
I've actually encountered this a few times already – it's been present
in previous versions,
in fact – but the occurrence probability is extremely low.
As such, it's not possible to bisect the exact commit that introduced
the issue.
It is worth noting, however, that all the call traces I have observed
are related to xas.
>> This is due to a race condition that occurs when two threads concurrently
>> perform allocation and freeing operations on the same slab object.
>>
>> When a process is preparing to allocate a slab object, another process
>> successfully preempts the CPU, and then proceeds to free a slab object.
>> However, before the freeing process can invoke `alloc_tag_sub()`, it is
>> preempted again by the original allocating process. At this point, the
>> allocating process acquires the same slab object, and subsequently triggers
>> a warning when it invokes `alloc_tag_add()`.
> The explanation doesn't make sense to me, because alloc_tag_sub()
> should have been called before it's added back to freelist or sheaf
> before other threads can allocate it, or am I missing something?
You are correct. Likely mental fatigue on my part – I cleared my head
afterward and found this scenario does not exist.
As you noted, alloc_tag_sub is invoked first, then the object is added
back to the freelist, so the race condition I described is probably
non-existent.
Therefore, we may need to revisit our assumptions and take a closer look
at the code corresponding to XAS.
Thank you for taking the time to review this with me.
Thanks
Best Regards
Hao
>
>> Signed-off-by: Hao Ge <hao.ge@linux.dev>
>> ---
>> mm/slub.c | 7 ++++++-
>> 1 file changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/mm/slub.c b/mm/slub.c
>> index f77b7407c51b..0d84fc917a89 100644
>> --- a/mm/slub.c
>> +++ b/mm/slub.c
>> @@ -2261,8 +2261,13 @@ __alloc_tagging_slab_alloc_hook(struct kmem_cache *s, void *object, gfp_t flags)
>> * If other users appear then mem_alloc_profiling_enabled()
>> * check should be added before alloc_tag_add().
>> */
>> - if (likely(obj_exts))
>> + if (likely(obj_exts)) {
>> +
>> + while (!READ_ONCE(obj_exts->ref.ct))
>> + cpu_relax();
> I don't think this is acceptable - it shouldn't wait forever even when
> there is a real bug that doesn't clear the tag.
>
>> +
>> alloc_tag_add(&obj_exts->ref, current->alloc_tag, s->size);
>> + }
>> else
>> alloc_tag_set_inaccurate(current->alloc_tag);
>> }
>> --
>> 2.25.1
>>
On Tue, Feb 03, 2026 at 06:02:07PM +0800, Hao Ge wrote:
> Hi Harry
>
>
> On 2026/2/3 17:44, Harry Yoo wrote:
> > On Tue, Feb 03, 2026 at 03:30:06PM +0800, Hao Ge wrote:
> > > When CONFIG_MEM_ALLOC_PROFILING_DEBUG is enabled, the following warning
> > > may be noticed:
> > >
> > > [ 3959.023862] ------------[ cut here ]------------
> > > [ 3959.023891] alloc_tag was not cleared (got tag for lib/xarray.c:378)
> > > [ 3959.023947] WARNING: ./include/linux/alloc_tag.h:155 at alloc_tag_add+0x128/0x178, CPU#6: mkfs.ntfs/113998
> > > [ 3959.023978] Modules linked in: dns_resolver tun brd overlay exfat btrfs blake2b libblake2b xor xor_neon raid6_pq loop sctp ip6_udp_tunnel udp_tunnel ext4 crc16 mbcache jbd2 rfkill sunrpc vfat fat sg fuse nfnetlink sr_mod virtio_gpu cdrom drm_client_lib virtio_dma_buf drm_shmem_helper drm_kms_helper ghash_ce drm sm4 backlight virtio_net net_failover virtio_scsi failover virtio_console virtio_blk virtio_mmio dm_mirror dm_region_hash dm_log dm_multipath dm_mod i2c_dev aes_neon_bs aes_ce_blk [last unloaded: hwpoison_inject]
> > > [ 3959.024170] CPU: 6 UID: 0 PID: 113998 Comm: mkfs.ntfs Kdump: loaded Tainted: G W 6.19.0-rc7+ #7 PREEMPT(voluntary)
> > > [ 3959.024182] Tainted: [W]=WARN
> > > [ 3959.024186] Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022
> > > [ 3959.024192] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > > [ 3959.024199] pc : alloc_tag_add+0x128/0x178
> > > [ 3959.024207] lr : alloc_tag_add+0x128/0x178
> > > [ 3959.024214] sp : ffff80008b696d60
> > > [ 3959.024219] x29: ffff80008b696d60 x28: 0000000000000000 x27: 0000000000000240
> > > [ 3959.024232] x26: 0000000000000000 x25: 0000000000000240 x24: ffff800085d17860
> > > [ 3959.024245] x23: 0000000000402800 x22: ffff0000c0012dc0 x21: 00000000000002d0
> > > [ 3959.024257] x20: ffff0000e6ef3318 x19: ffff800085ae0410 x18: 0000000000000000
> > > [ 3959.024269] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
> > > [ 3959.024281] x14: 0000000000000000 x13: 0000000000000001 x12: ffff600064101293
> > > [ 3959.024292] x11: 1fffe00064101292 x10: ffff600064101292 x9 : dfff800000000000
> > > [ 3959.024305] x8 : 00009fff9befed6e x7 : ffff000320809493 x6 : 0000000000000001
> > > [ 3959.024316] x5 : ffff000320809490 x4 : ffff600064101293 x3 : ffff800080691838
> > > [ 3959.024328] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000d5bcd640
> > > [ 3959.024340] Call trace:
> > > [ 3959.024346] alloc_tag_add+0x128/0x178 (P)
> > > [ 3959.024355] __alloc_tagging_slab_alloc_hook+0x11c/0x1a8
> > > [ 3959.024362] kmem_cache_alloc_lru_noprof+0x1b8/0x5e8
> > > [ 3959.024369] xas_alloc+0x304/0x4f0
> > > [ 3959.024381] xas_create+0x1e0/0x4a0
> > > [ 3959.024388] xas_store+0x68/0xda8
> > > [ 3959.024395] __filemap_add_folio+0x5b0/0xbd8
> > > [ 3959.024409] filemap_add_folio+0x16c/0x7e0
> > > [ 3959.024416] __filemap_get_folio_mpol+0x2dc/0x9e8
> > > [ 3959.024424] iomap_get_folio+0xfc/0x180
> > > [ 3959.024435] __iomap_get_folio+0x2f8/0x4b8
> > > [ 3959.024441] iomap_write_begin+0x198/0xc18
> > > [ 3959.024448] iomap_write_iter+0x2ec/0x8f8
> > > [ 3959.024454] iomap_file_buffered_write+0x19c/0x290
> > > [ 3959.024461] blkdev_write_iter+0x38c/0x978
> > > [ 3959.024470] vfs_write+0x4d4/0x928
> > > [ 3959.024482] ksys_write+0xfc/0x1f8
> > > [ 3959.024489] __arm64_sys_write+0x74/0xb0
> > > [ 3959.024496] invoke_syscall+0xd4/0x258
> > > [ 3959.024507] el0_svc_common.constprop.0+0xb4/0x240
> > > [ 3959.024514] do_el0_svc+0x48/0x68
> > > [ 3959.024520] el0_svc+0x40/0xf8
> > > [ 3959.024526] el0t_64_sync_handler+0xa0/0xe8
> > > [ 3959.024533] el0t_64_sync+0x1ac/0x1b0
> > > [ 3959.024540] ---[ end trace 0000000000000000 ]---
> > Hi Hao, on which commit did you observe this warning?
>
>
> I've actually encountered this a few times already – it's been present in
> previous versions,
>
> in fact – but the occurrence probability is extremely low.
>
> As such, it's not possible to bisect the exact commit that introduced the
> issue.
>
> It is worth noting, however, that all the call traces I have observed are
> related to xas.
>
>
> > > This is due to a race condition that occurs when two threads concurrently
> > > perform allocation and freeing operations on the same slab object.
> > >
> > > When a process is preparing to allocate a slab object, another process
> > > successfully preempts the CPU, and then proceeds to free a slab object.
> > > However, before the freeing process can invoke `alloc_tag_sub()`, it is
> > > preempted again by the original allocating process. At this point, the
> > > allocating process acquires the same slab object, and subsequently triggers
> > > a warning when it invokes `alloc_tag_add()`.
> > The explanation doesn't make sense to me, because alloc_tag_sub()
> > should have been called before it's added back to freelist or sheaf
> > before other threads can allocate it, or am I missing something?
>
> You are correct. Likely mental fatigue on my part – I cleared my head
> afterward and found this scenario does not exist.
>
> As you noted, alloc_tag_sub is invoked first, then the object is added back
> to the freelist, so the race condition I described is probably non-existent.
>
> Therefore, we may need to revisit our assumptions and take a closer look at
> the code corresponding to XAS.
Thanks for the reporting.
My suspicion is that this issue is not related to XAS. I suspect it may be
caused by a missing call to alloc_tagging_slab_free_hook() on some free path,
leaving an uncleared tag behind.
And then I reviewed all call sites of alloc_tagging_slab_free_hook() and my
understanding is that it should be invoked on every path that frees slab
objects.
Then after reading some code, I noticed that in memcg_slab_post_alloc_hook(),
when __memcg_slab_post_alloc_hook() fails, there are two different free paths
depending on whether size == 1 or size != 1. In the kmem_cache_free_bulk() path
we do call alloc_tagging_slab_free_hook(). However, in
memcg_alloc_abort_single() we don't, and I think that omission of
alloc_tagging_slab_free_hook() could explain the problem(unless I'm missing some
other details...)
--
Thanks,
Hao Li
>
> Thank you for taking the time to review this with me.
>
> Thanks
>
> Best Regards
>
> Hao
>
> >
> > > Signed-off-by: Hao Ge <hao.ge@linux.dev>
> > > ---
> > > mm/slub.c | 7 ++++++-
> > > 1 file changed, 6 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/mm/slub.c b/mm/slub.c
> > > index f77b7407c51b..0d84fc917a89 100644
> > > --- a/mm/slub.c
> > > +++ b/mm/slub.c
> > > @@ -2261,8 +2261,13 @@ __alloc_tagging_slab_alloc_hook(struct kmem_cache *s, void *object, gfp_t flags)
> > > * If other users appear then mem_alloc_profiling_enabled()
> > > * check should be added before alloc_tag_add().
> > > */
> > > - if (likely(obj_exts))
> > > + if (likely(obj_exts)) {
> > > +
> > > + while (!READ_ONCE(obj_exts->ref.ct))
> > > + cpu_relax();
> > I don't think this is acceptable - it shouldn't wait forever even when
> > there is a real bug that doesn't clear the tag.
> >
> > > +
> > > alloc_tag_add(&obj_exts->ref, current->alloc_tag, s->size);
> > > + }
> > > else
> > > alloc_tag_set_inaccurate(current->alloc_tag);
> > > }
> > > --
> > > 2.25.1
> > >
>
Hi Hao Li On 2026/2/4 14:21, Hao Li wrote: > On Tue, Feb 03, 2026 at 06:02:07PM +0800, Hao Ge wrote: >> Hi Harry >> >> >> On 2026/2/3 17:44, Harry Yoo wrote: >>> On Tue, Feb 03, 2026 at 03:30:06PM +0800, Hao Ge wrote: >>>> When CONFIG_MEM_ALLOC_PROFILING_DEBUG is enabled, the following warning >>>> may be noticed: >>>> >>>> [ 3959.023862] ------------[ cut here ]------------ >>>> [ 3959.023891] alloc_tag was not cleared (got tag for lib/xarray.c:378) >>>> [ 3959.023947] WARNING: ./include/linux/alloc_tag.h:155 at alloc_tag_add+0x128/0x178, CPU#6: mkfs.ntfs/113998 >>>> [ 3959.023978] Modules linked in: dns_resolver tun brd overlay exfat btrfs blake2b libblake2b xor xor_neon raid6_pq loop sctp ip6_udp_tunnel udp_tunnel ext4 crc16 mbcache jbd2 rfkill sunrpc vfat fat sg fuse nfnetlink sr_mod virtio_gpu cdrom drm_client_lib virtio_dma_buf drm_shmem_helper drm_kms_helper ghash_ce drm sm4 backlight virtio_net net_failover virtio_scsi failover virtio_console virtio_blk virtio_mmio dm_mirror dm_region_hash dm_log dm_multipath dm_mod i2c_dev aes_neon_bs aes_ce_blk [last unloaded: hwpoison_inject] >>>> [ 3959.024170] CPU: 6 UID: 0 PID: 113998 Comm: mkfs.ntfs Kdump: loaded Tainted: G W 6.19.0-rc7+ #7 PREEMPT(voluntary) >>>> [ 3959.024182] Tainted: [W]=WARN >>>> [ 3959.024186] Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022 >>>> [ 3959.024192] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--) >>>> [ 3959.024199] pc : alloc_tag_add+0x128/0x178 >>>> [ 3959.024207] lr : alloc_tag_add+0x128/0x178 >>>> [ 3959.024214] sp : ffff80008b696d60 >>>> [ 3959.024219] x29: ffff80008b696d60 x28: 0000000000000000 x27: 0000000000000240 >>>> [ 3959.024232] x26: 0000000000000000 x25: 0000000000000240 x24: ffff800085d17860 >>>> [ 3959.024245] x23: 0000000000402800 x22: ffff0000c0012dc0 x21: 00000000000002d0 >>>> [ 3959.024257] x20: ffff0000e6ef3318 x19: ffff800085ae0410 x18: 0000000000000000 >>>> [ 3959.024269] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000 >>>> [ 3959.024281] x14: 0000000000000000 x13: 0000000000000001 x12: ffff600064101293 >>>> [ 3959.024292] x11: 1fffe00064101292 x10: ffff600064101292 x9 : dfff800000000000 >>>> [ 3959.024305] x8 : 00009fff9befed6e x7 : ffff000320809493 x6 : 0000000000000001 >>>> [ 3959.024316] x5 : ffff000320809490 x4 : ffff600064101293 x3 : ffff800080691838 >>>> [ 3959.024328] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000d5bcd640 >>>> [ 3959.024340] Call trace: >>>> [ 3959.024346] alloc_tag_add+0x128/0x178 (P) >>>> [ 3959.024355] __alloc_tagging_slab_alloc_hook+0x11c/0x1a8 >>>> [ 3959.024362] kmem_cache_alloc_lru_noprof+0x1b8/0x5e8 >>>> [ 3959.024369] xas_alloc+0x304/0x4f0 >>>> [ 3959.024381] xas_create+0x1e0/0x4a0 >>>> [ 3959.024388] xas_store+0x68/0xda8 >>>> [ 3959.024395] __filemap_add_folio+0x5b0/0xbd8 >>>> [ 3959.024409] filemap_add_folio+0x16c/0x7e0 >>>> [ 3959.024416] __filemap_get_folio_mpol+0x2dc/0x9e8 >>>> [ 3959.024424] iomap_get_folio+0xfc/0x180 >>>> [ 3959.024435] __iomap_get_folio+0x2f8/0x4b8 >>>> [ 3959.024441] iomap_write_begin+0x198/0xc18 >>>> [ 3959.024448] iomap_write_iter+0x2ec/0x8f8 >>>> [ 3959.024454] iomap_file_buffered_write+0x19c/0x290 >>>> [ 3959.024461] blkdev_write_iter+0x38c/0x978 >>>> [ 3959.024470] vfs_write+0x4d4/0x928 >>>> [ 3959.024482] ksys_write+0xfc/0x1f8 >>>> [ 3959.024489] __arm64_sys_write+0x74/0xb0 >>>> [ 3959.024496] invoke_syscall+0xd4/0x258 >>>> [ 3959.024507] el0_svc_common.constprop.0+0xb4/0x240 >>>> [ 3959.024514] do_el0_svc+0x48/0x68 >>>> [ 3959.024520] el0_svc+0x40/0xf8 >>>> [ 3959.024526] el0t_64_sync_handler+0xa0/0xe8 >>>> [ 3959.024533] el0t_64_sync+0x1ac/0x1b0 >>>> [ 3959.024540] ---[ end trace 0000000000000000 ]--- >>> Hi Hao, on which commit did you observe this warning? >> >> I've actually encountered this a few times already – it's been present in >> previous versions, >> >> in fact – but the occurrence probability is extremely low. >> >> As such, it's not possible to bisect the exact commit that introduced the >> issue. >> >> It is worth noting, however, that all the call traces I have observed are >> related to xas. >> >> >>>> This is due to a race condition that occurs when two threads concurrently >>>> perform allocation and freeing operations on the same slab object. >>>> >>>> When a process is preparing to allocate a slab object, another process >>>> successfully preempts the CPU, and then proceeds to free a slab object. >>>> However, before the freeing process can invoke `alloc_tag_sub()`, it is >>>> preempted again by the original allocating process. At this point, the >>>> allocating process acquires the same slab object, and subsequently triggers >>>> a warning when it invokes `alloc_tag_add()`. >>> The explanation doesn't make sense to me, because alloc_tag_sub() >>> should have been called before it's added back to freelist or sheaf >>> before other threads can allocate it, or am I missing something? >> You are correct. Likely mental fatigue on my part – I cleared my head >> afterward and found this scenario does not exist. >> >> As you noted, alloc_tag_sub is invoked first, then the object is added back >> to the freelist, so the race condition I described is probably non-existent. >> >> Therefore, we may need to revisit our assumptions and take a closer look at >> the code corresponding to XAS. > Thanks for the reporting. > > My suspicion is that this issue is not related to XAS. I suspect it may be > caused by a missing call to alloc_tagging_slab_free_hook() on some free path, > leaving an uncleared tag behind. > > And then I reviewed all call sites of alloc_tagging_slab_free_hook() and my > understanding is that it should be invoked on every path that frees slab > objects. > > Then after reading some code, I noticed that in memcg_slab_post_alloc_hook(), > when __memcg_slab_post_alloc_hook() fails, there are two different free paths > depending on whether size == 1 or size != 1. In the kmem_cache_free_bulk() path > we do call alloc_tagging_slab_free_hook(). However, in > memcg_alloc_abort_single() we don't, and I think that omission of > alloc_tagging_slab_free_hook() could explain the problem(unless I'm missing some > other details...) Your analysis is correct. Nice catch. After adding a WARN_ON(1) at this location, the following log was observed: https://elixir.bootlin.com/linux/v6.19-rc5/source/mm/slub.c#L2342 [ 3966.117006] ------------[ cut here ]------------ [ 3966.117031] WARNING: mm/slub.c:2342 at kmem_cache_alloc_lru_noprof+0x4cc/0x5e8, CPU#1: madvise06/114564 [ 3966.117053] Modules linked in: dns_resolver tun brd overlay exfat btrfs blake2b libblake2b xor xor_neon raid6_pq loop sctp ip6_udp_tunnel udp_tunnel ext4 crc16 mbcache jbd2 rfkill sunrpc vfat fat sg fuse nfnetlink virtio_gpu drm_client_lib virtio_dma_buf drm_shmem_helper drm_kms_helper sr_mod cdrom ghash_ce drm virtio_net backlight virtio_blk net_failover failover sm4 virtio_console virtio_scsi virtio_mmio dm_mirror dm_region_hash dm_log dm_multipath dm_mod i2c_dev aes_neon_bs aes_ce_blk [ 3966.117218] CPU: 1 UID: 0 PID: 114564 Comm: madvise06 Kdump: loaded Not tainted 6.19.0-rc7+ #10 PREEMPT(voluntary) [ 3966.117226] Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022 [ 3966.117231] pstate: 104000c5 (nzcV daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 3966.117238] pc : kmem_cache_alloc_lru_noprof+0x4cc/0x5e8 [ 3966.117244] lr : kmem_cache_alloc_lru_noprof+0x4c8/0x5e8 [ 3966.117250] sp : ffff8000a99772d0 [ 3966.117254] x29: ffff8000a99772e0 x28: 0000000000000000 x27: 0000000000000240 [ 3966.117267] x26: 0000000000000000 x25: 0000000000000240 x24: 0000000000000000 [ 3966.117279] x23: 0000000000402800 x22: ffff800082caf864 x21: ffff000100c8bb18 [ 3966.117291] x20: 0000000000402800 x19: ffff0000c0012dc0 x18: ffff8000a9977cdd [ 3966.117303] x17: 0000000000000000 x16: 0000000000000000 x15: ffff8000a9977cb0 [ 3966.117315] x14: ffff8000a9977ca8 x13: 0000000041b58ab3 x12: ffff6000185d78c9 [ 3966.117326] x11: 1fffe000185d78c8 x10: ffff6000185d78c8 x9 : ffff800080b6908c [ 3966.117338] x8 : 00009fffe7a28738 x7 : ffff0000c2ebc647 x6 : 0000000000000001 [ 3966.117350] x5 : ffff0000c2ebc640 x4 : 1fffe000185d78d8 x3 : 1fffe00025b598f0 [ 3966.117362] x2 : 0000000000000000 x1 : 0000000000000003 x0 : 0000000000000000 [ 3966.117373] Call trace: [ 3966.117378] kmem_cache_alloc_lru_noprof+0x4cc/0x5e8 (P) [ 3966.117386] xas_alloc+0x304/0x4f0 [ 3966.117395] xas_create+0x1e0/0x4a0 [ 3966.117402] xas_store+0x68/0xda8 [ 3966.117410] shmem_add_to_page_cache+0x4f4/0x6a8 [ 3966.117418] shmem_alloc_and_add_folio+0x360/0xd40 [ 3966.117425] shmem_get_folio_gfp+0x424/0x10d0 [ 3966.117432] shmem_fault+0x1a4/0x6c8 [ 3966.117438] __do_fault+0xd0/0x6c0 [ 3966.117447] do_fault+0x2e8/0xbc0 [ 3966.117453] handle_pte_fault+0x43c/0x7b8 [ 3966.117460] __handle_mm_fault+0x308/0xb88 [ 3966.117467] handle_mm_fault+0x238/0x7b8 [ 3966.117473] do_page_fault+0x1cc/0x1138 [ 3966.117481] do_translation_fault+0x80/0x130 [ 3966.117488] do_mem_abort+0x74/0x1b8 [ 3966.117496] el0_da+0x4c/0xf8 [ 3966.117503] el0t_64_sync_handler+0xd0/0xe8 [ 3966.117509] el0t_64_sync+0x1ac/0x1b0 [ 3966.117516] ---[ end trace 0000000000000000 ]--- The bug was subsequently reproduced, with the following log: [ 4015.055460] ------------[ cut here ]------------ [ 4015.055480] alloc_tag was not cleared (got tag for lib/xarray.c:378) [ 4015.055512] WARNING: ./include/linux/alloc_tag.h:155 at alloc_tag_add+0x128/0x178, CPU#3: kworker/u37:0/110627 [ 4015.055534] Modules linked in: dns_resolver tun brd overlay exfat btrfs blake2b libblake2b xor xor_neon raid6_pq loop sctp ip6_udp_tunnel udp_tunnel ext4 crc16 mbcache jbd2 rfkill sunrpc vfat fat sg fuse nfnetlink virtio_gpu drm_client_lib virtio_dma_buf drm_shmem_helper drm_kms_helper sr_mod cdrom ghash_ce drm virtio_net backlight virtio_blk net_failover failover sm4 virtio_console virtio_scsi virtio_mmio dm_mirror dm_region_hash dm_log dm_multipath dm_mod i2c_dev aes_neon_bs aes_ce_blk [last unloaded: hwpoison_inject] [ 4015.055707] CPU: 3 UID: 0 PID: 110627 Comm: kworker/u37:0 Kdump: loaded Tainted: G W 6.19.0-rc7+ #10 PREEMPT(voluntary) [ 4015.055716] Tainted: [W]=WARN [ 4015.055727] Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022 [ 4015.055734] Workqueue: loop0 loop_rootcg_workfn [loop] [ 4015.055749] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 4015.055756] pc : alloc_tag_add+0x128/0x178 [ 4015.055764] lr : alloc_tag_add+0x128/0x178 [ 4015.055771] sp : ffff80008d7a71c0 [ 4015.055775] x29: ffff80008d7a71c0 x28: 0000000000000000 x27: 0000000000000240 [ 4015.055788] x26: 0000000000000000 x25: 0000000000000240 x24: 0000000000000000 [ 4015.055800] x23: 0000000000402800 x22: ffff0000c0012dc0 x21: 00000000000002d0 [ 4015.055812] x20: ffff0000e4743358 x19: ffff800085ae0410 x18: 0000000000000000 [ 4015.055824] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000 [ 4015.055836] x14: 0000000000000000 x13: 0000000000000001 x12: ffff6000640eda93 [ 4015.055848] x11: 1fffe000640eda92 x10: ffff6000640eda92 x9 : dfff800000000000 [ 4015.055860] x8 : 00009fff9bf1256e x7 : ffff00032076d493 x6 : 0000000000000001 [ 4015.055872] x5 : ffff00032076d490 x4 : ffff6000640eda93 x3 : ffff800080691838 [ 4015.055883] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000d46795c0 [ 4015.055896] Call trace: [ 4015.055900] alloc_tag_add+0x128/0x178 (P) [ 4015.055909] __alloc_tagging_slab_alloc_hook+0x11c/0x1a8 [ 4015.055916] kmem_cache_alloc_lru_noprof+0x1b8/0x5e8 [ 4015.055922] xas_alloc+0x304/0x4f0 [ 4015.055931] xas_create+0x1e0/0x4a0 [ 4015.055938] xas_store+0x68/0xda8 [ 4015.055945] shmem_add_to_page_cache+0x4f4/0x6a8 [ 4015.055953] shmem_alloc_and_add_folio+0x360/0xd40 [ 4015.055960] shmem_get_folio_gfp+0x424/0x10d0 [ 4015.055966] shmem_write_begin+0x14c/0x458 [ 4015.055973] generic_perform_write+0x2d8/0x5d8 [ 4015.055981] shmem_file_write_iter+0xe8/0x118 [ 4015.055987] lo_rw_aio.isra.0+0x91c/0xe20 [loop] [ 4015.055995] loop_process_work+0x328/0xeb0 [loop] [ 4015.056003] loop_rootcg_workfn+0x28/0x40 [loop] [ 4015.056011] process_one_work+0x5ac/0x1110 [ 4015.056019] worker_thread+0x724/0xb30 [ 4015.056027] kthread+0x2fc/0x3a8 [ 4015.056034] ret_from_fork+0x10/0x20 [ 4015.056042] ---[ end trace 0000000000000000 ]--- This v2 patch incorporates your suggested fix. https://lore.kernel.org/all/20260204101401.202762-1-hao.ge@linux.dev/ I really appreciate your help. Best Regards Hao >
On Wed, Feb 04, 2026 at 06:24:42PM +0800, Hao Ge wrote: > Hi Hao Li > ... > > This v2 patch incorporates your suggested fix. > > https://lore.kernel.org/all/20260204101401.202762-1-hao.ge@linux.dev/ > > I really appreciate your help. Happy to help. > > > Best Regards > > Hao > > >
On 2026/2/3 15:30, Hao Ge wrote:
> When CONFIG_MEM_ALLOC_PROFILING_DEBUG is enabled, the following warning
> may be noticed:
>
> [ 3959.023862] ------------[ cut here ]------------
> [ 3959.023891] alloc_tag was not cleared (got tag for lib/xarray.c:378)
> [ 3959.023947] WARNING: ./include/linux/alloc_tag.h:155 at alloc_tag_add+0x128/0x178, CPU#6: mkfs.ntfs/113998
> [ 3959.023978] Modules linked in: dns_resolver tun brd overlay exfat btrfs blake2b libblake2b xor xor_neon raid6_pq loop sctp ip6_udp_tunnel udp_tunnel ext4 crc16 mbcache jbd2 rfkill sunrpc vfat fat sg fuse nfnetlink sr_mod virtio_gpu cdrom drm_client_lib virtio_dma_buf drm_shmem_helper drm_kms_helper ghash_ce drm sm4 backlight virtio_net net_failover virtio_scsi failover virtio_console virtio_blk virtio_mmio dm_mirror dm_region_hash dm_log dm_multipath dm_mod i2c_dev aes_neon_bs aes_ce_blk [last unloaded: hwpoison_inject]
> [ 3959.024170] CPU: 6 UID: 0 PID: 113998 Comm: mkfs.ntfs Kdump: loaded Tainted: G W 6.19.0-rc7+ #7 PREEMPT(voluntary)
> [ 3959.024182] Tainted: [W]=WARN
> [ 3959.024186] Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022
> [ 3959.024192] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [ 3959.024199] pc : alloc_tag_add+0x128/0x178
> [ 3959.024207] lr : alloc_tag_add+0x128/0x178
> [ 3959.024214] sp : ffff80008b696d60
> [ 3959.024219] x29: ffff80008b696d60 x28: 0000000000000000 x27: 0000000000000240
> [ 3959.024232] x26: 0000000000000000 x25: 0000000000000240 x24: ffff800085d17860
> [ 3959.024245] x23: 0000000000402800 x22: ffff0000c0012dc0 x21: 00000000000002d0
> [ 3959.024257] x20: ffff0000e6ef3318 x19: ffff800085ae0410 x18: 0000000000000000
> [ 3959.024269] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
> [ 3959.024281] x14: 0000000000000000 x13: 0000000000000001 x12: ffff600064101293
> [ 3959.024292] x11: 1fffe00064101292 x10: ffff600064101292 x9 : dfff800000000000
> [ 3959.024305] x8 : 00009fff9befed6e x7 : ffff000320809493 x6 : 0000000000000001
> [ 3959.024316] x5 : ffff000320809490 x4 : ffff600064101293 x3 : ffff800080691838
> [ 3959.024328] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000d5bcd640
> [ 3959.024340] Call trace:
> [ 3959.024346] alloc_tag_add+0x128/0x178 (P)
> [ 3959.024355] __alloc_tagging_slab_alloc_hook+0x11c/0x1a8
> [ 3959.024362] kmem_cache_alloc_lru_noprof+0x1b8/0x5e8
> [ 3959.024369] xas_alloc+0x304/0x4f0
> [ 3959.024381] xas_create+0x1e0/0x4a0
> [ 3959.024388] xas_store+0x68/0xda8
> [ 3959.024395] __filemap_add_folio+0x5b0/0xbd8
> [ 3959.024409] filemap_add_folio+0x16c/0x7e0
> [ 3959.024416] __filemap_get_folio_mpol+0x2dc/0x9e8
> [ 3959.024424] iomap_get_folio+0xfc/0x180
> [ 3959.024435] __iomap_get_folio+0x2f8/0x4b8
> [ 3959.024441] iomap_write_begin+0x198/0xc18
> [ 3959.024448] iomap_write_iter+0x2ec/0x8f8
> [ 3959.024454] iomap_file_buffered_write+0x19c/0x290
> [ 3959.024461] blkdev_write_iter+0x38c/0x978
> [ 3959.024470] vfs_write+0x4d4/0x928
> [ 3959.024482] ksys_write+0xfc/0x1f8
> [ 3959.024489] __arm64_sys_write+0x74/0xb0
> [ 3959.024496] invoke_syscall+0xd4/0x258
> [ 3959.024507] el0_svc_common.constprop.0+0xb4/0x240
> [ 3959.024514] do_el0_svc+0x48/0x68
> [ 3959.024520] el0_svc+0x40/0xf8
> [ 3959.024526] el0t_64_sync_handler+0xa0/0xe8
> [ 3959.024533] el0t_64_sync+0x1ac/0x1b0
> [ 3959.024540] ---[ end trace 0000000000000000 ]---
>
> This is due to a race condition that occurs when two threads concurrently
> perform allocation and freeing operations on the same slab object.
>
> When a process is preparing to allocate a slab object, another process
> successfully preempts the CPU, and then proceeds to free a slab object.
> However, before the freeing process can invoke `alloc_tag_sub()`, it is
> preempted again by the original allocating process. At this point, the
> allocating process acquires the same slab object, and subsequently triggers
> a warning when it invokes `alloc_tag_add()`.
>
> Signed-off-by: Hao Ge <hao.ge@linux.dev>
> ---
> Hi Suren
>
> I'm not sure if my solution still has any issues, so I'd like to
> get your suggestions on it. At the very least, my understanding is
> that READ_ONCE and WRITE_ONCE should be used in pairs.
>
> I look forward to your suggestions.
>
> Thanks
> ---
> mm/slub.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index f77b7407c51b..0d84fc917a89 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -2261,8 +2261,13 @@ __alloc_tagging_slab_alloc_hook(struct kmem_cache *s, void *object, gfp_t flags)
> * If other users appear then mem_alloc_profiling_enabled()
> * check should be added before alloc_tag_add().
> */
> - if (likely(obj_exts))
> + if (likely(obj_exts)) {
> +
> + while (!READ_ONCE(obj_exts->ref.ct))
> + cpu_relax();
> +
My apologies for forgetting to update this part when submitting the
patch — I've already made the change in my local environment:
while (READ_ONCE(obj_exts->ref.ct))
cpu_relax();
I'd also welcome any thoughts or feedback from others in the community.
> alloc_tag_add(&obj_exts->ref, current->alloc_tag, s->size);
> + }
> else
> alloc_tag_set_inaccurate(current->alloc_tag);
> }
© 2016 - 2026 Red Hat, Inc.