From: Damien Le Moal <dlemoal@kernel.org>
To: Mira Limbeck <m.limbeck@proxmox.com>
Cc: axboe@kernel.dk, hch@lst.de, linux-block@vger.kernel.org,
martin.petersen@oracle.com, Friedrich Weber <f.weber@proxmox.com>
Subject: Re: [PATCH v2] block: Increase BLK_DEF_MAX_SECTORS_CAP
Date: Wed, 1 Apr 2026 04:48:53 +0900 [thread overview]
Message-ID: <ff5e2877-840b-4eb6-b449-bb64fb2e4097@kernel.org> (raw)
In-Reply-To: <291f78bf-4b4a-40dd-867d-053b36c564b3@proxmox.com>
On 3/31/26 21:02, Mira Limbeck wrote:
> Hi,
>
>
> Some of our Proxmox VE users started seeing `unable to handle page
> fault` after switching to our downstream kernel 6.17, and after
> bisecting with the mainline kernel we've identified this patch as the
> first commit (9b8b84879d4adc506b0d3944e20b28d9f3f6994b) where we see
> those errors.
Please test with the current mainline kernel. We do not deal with non-standard
kernels.
> It requires a certain combination of hardware though, so far we've seen
> this with a combination of:
> Broadcom/LSI HBAs with NVMe support (9400, 9500)
> KIOXIA KCD8 NVMes
>
> The Hardware of our test machine consists of:
>
> 81:00.0 Serial Attached SCSI controller [0107]: Broadcom / LSI
> Fusion-MPT 12GSAS/PCIe Secure SAS38xx [1000:00e6]
> Subsystem: Broadcom / LSI 9500-16i Tri-Mode HBA [1000:4050]
> Kernel driver in use: mpt3sas
> Kernel modules: mpt3sas
>
> FW Package Ver(37.00.00.00)
> SAS3816: FWVersion(37.00.00.00), ChipRevision(0x00)
>
> and
>
> Mar 31 10:52:44 pve-test-hba kernel: scsi 16:2:0:0: Direct-Access
> NVMe KIOXIA KCD8XRUG7 0105 PQ: 0 ANSI: 6
>
> The 4 NVMes are exposed as SCSI devices via the Broadcom controller.
>
> ```
> Mar 27 15:03:11 pve-test-hba kernel: sd 3:2:2:0: [sdc] tag#2463 page boundary curr_buff: 0x000000002ce04d26
> Mar 27 15:03:11 pve-test-hba kernel: BUG: unable to handle page fault for address: ff76b049c31fe000
> Mar 27 15:03:11 pve-test-hba kernel: #PF: supervisor write access in kernel mode
> Mar 27 15:03:11 pve-test-hba kernel: #PF: error_code(0x0002) - not-present page
> Mar 27 15:03:11 pve-test-hba kernel: PGD 100010067 P4D 1008c7067 PUD 1008c8067 PMD 11cd9f067 PTE 0
> Mar 27 15:03:11 pve-test-hba kernel: Oops: Oops: 0002 [#1] SMP NOPTI
> Mar 27 15:03:11 pve-test-hba kernel: CPU: 7 UID: 0 PID: 4385 Comm: dmcrypt_write/2 Tainted: G E 6.16.0-rc4-step14-00001-g9b8b84879d4a #16 PREEMPT(voluntary)
> Mar 27 15:03:11 pve-test-hba kernel: Tainted: [E]=UNSIGNED_MODULE
> Mar 27 15:03:11 pve-test-hba kernel: Hardware name: <snip>
> Mar 27 15:03:11 pve-test-hba kernel: RIP: 0010:_base_build_sg_scmd_ieee+0x478/0x590 [mpt3sas]
> Mar 27 15:03:11 pve-test-hba kernel: Code: 20 48 83 c3 20 48 89 d1 48 83 e1 fc 83 e2 01 48 0f 45 d9 4c 8b 73 10 44 8b 63 18 4c 89 e9 4c 8d 69 08 44 85 e8 74 31 45 29 d7 <4c> 89 31 49 83 c1 08 41 83 c0 01 45 29 d4 45 85 ff 7f af 4c 8b 75
> Mar 27 15:03:11 pve-test-hba kernel: RSP: 0018:ff76b049c170b8a8 EFLAGS: 00010206
> Mar 27 15:03:11 pve-test-hba kernel: RAX: 0000000000000fff RBX: ff220bd05e8a0270 RCX: ff76b049c31fe000
> Mar 27 15:03:11 pve-test-hba kernel: RDX: ff76b049c31fe008 RSI: 0000000000000000 RDI: 0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: RBP: ff76b049c170b908 R08: 0000000000000200 R09: 00000000ff161000
> Mar 27 15:03:11 pve-test-hba kernel: R10: 0000000000001000 R11: 0000000000001000 R12: 00000000001a0000
> Mar 27 15:03:11 pve-test-hba kernel: R13: ff76b049c31fe008 R14: 00000000f9600000 R15: 000000000019f000
> Mar 27 15:03:11 pve-test-hba kernel: FS: 0000000000000000(0000) GS:ff220bd3d1ee7000(0000) knlGS:0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> Mar 27 15:03:11 pve-test-hba kernel: CR2: ff76b049c31fe000 CR3: 0000000109f8c007 CR4: 0000000000f71ef0
> Mar 27 15:03:11 pve-test-hba kernel: PKRU: 55555554
> Mar 27 15:03:11 pve-test-hba kernel: Call Trace:
> Mar 27 15:03:11 pve-test-hba kernel: <TASK>
> Mar 27 15:03:11 pve-test-hba kernel: scsih_qcmd+0x37c/0x620 [mpt3sas]
> Mar 27 15:03:11 pve-test-hba kernel: scsi_queue_rq+0x3ec/0xd30
> Mar 27 15:03:11 pve-test-hba kernel: blk_mq_dispatch_rq_list+0x118/0x740
> Mar 27 15:03:11 pve-test-hba kernel: ? sbitmap_get+0x73/0x180
> Mar 27 15:03:11 pve-test-hba kernel: ? sbitmap_get+0x73/0x180
> Mar 27 15:03:11 pve-test-hba kernel: __blk_mq_sched_dispatch_requests+0x3fc/0x5b0
> Mar 27 15:03:11 pve-test-hba kernel: ? elv_attempt_insert_merge+0xa6/0x100
> Mar 27 15:03:11 pve-test-hba kernel: blk_mq_sched_dispatch_requests+0x2d/0x70
> Mar 27 15:03:11 pve-test-hba kernel: blk_mq_run_hw_queue+0x250/0x340
> Mar 27 15:03:11 pve-test-hba kernel: blk_mq_dispatch_list+0x16c/0x450
> Mar 27 15:03:11 pve-test-hba kernel: blk_mq_flush_plug_list+0x62/0x1e0
> Mar 27 15:03:11 pve-test-hba kernel: blk_add_rq_to_plug+0xff/0x1f0
> Mar 27 15:03:11 pve-test-hba kernel: blk_mq_submit_bio+0x616/0x7e0
> Mar 27 15:03:11 pve-test-hba kernel: __submit_bio+0x74/0x290
> Mar 27 15:03:11 pve-test-hba kernel: submit_bio_noacct_nocheck+0x1a2/0x3b0
> Mar 27 15:03:11 pve-test-hba kernel: submit_bio_noacct+0x1a0/0x5b0
> Mar 27 15:03:11 pve-test-hba kernel: dm_submit_bio_remap+0x49/0xb0
> Mar 27 15:03:11 pve-test-hba kernel: dmcrypt_write+0x120/0x150 [dm_crypt]
> Mar 27 15:03:11 pve-test-hba kernel: ? __pfx_dmcrypt_write+0x10/0x10 [dm_crypt]
> Mar 27 15:03:11 pve-test-hba kernel: kthread+0x10a/0x230
> Mar 27 15:03:11 pve-test-hba kernel: ? __pfx_kthread+0x10/0x10
> Mar 27 15:03:11 pve-test-hba kernel: ret_from_fork+0x1d1/0x200
> Mar 27 15:03:11 pve-test-hba kernel: ? __pfx_kthread+0x10/0x10
> Mar 27 15:03:11 pve-test-hba kernel: ret_from_fork_asm+0x1a/0x30
> Mar 27 15:03:11 pve-test-hba kernel: </TASK>
> Mar 27 15:03:11 pve-test-hba kernel: Modules linked in: dm_crypt(E) ebtable_filter(E) ebtables(E) ip_set(E) ip6table_raw(E) ip6table_filter(E) ip6_tables(E) iptable_filter(E) nf_tables(E) sunrpc(E) iptable_raw(E) xt_CT(E) iptable_nat(E) xt>
> Mar 27 15:03:11 pve-test-hba kernel: blake2b_generic(E) xor(E) raid6_pq(E) dm_thin_pool(E) dm_persistent_data(E) dm_bio_prison(E) dm_bufio(E) hid_generic(E) usbmouse(E) rndis_host(E) usbhid(E) cdc_ether(E) hid(E) usbnet(E) mii(E) xhci_pci>
> Mar 27 15:03:11 pve-test-hba kernel: CR2: ff76b049c31fe000
> Mar 27 15:03:11 pve-test-hba kernel: ---[ end trace 0000000000000000 ]---
> Mar 27 15:03:11 pve-test-hba kernel: RIP: 0010:_base_build_sg_scmd_ieee+0x478/0x590 [mpt3sas]
> Mar 27 15:03:11 pve-test-hba kernel: Code: 20 48 83 c3 20 48 89 d1 48 83 e1 fc 83 e2 01 48 0f 45 d9 4c 8b 73 10 44 8b 63 18 4c 89 e9 4c 8d 69 08 44 85 e8 74 31 45 29 d7 <4c> 89 31 49 83 c1 08 41 83 c0 01 45 29 d4 45 85 ff 7f af 4c 8b 75
> Mar 27 15:03:11 pve-test-hba kernel: RSP: 0018:ff76b049c170b8a8 EFLAGS: 00010206
> Mar 27 15:03:11 pve-test-hba kernel: RAX: 0000000000000fff RBX: ff220bd05e8a0270 RCX: ff76b049c31fe000
> Mar 27 15:03:11 pve-test-hba kernel: RDX: ff76b049c31fe008 RSI: 0000000000000000 RDI: 0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: RBP: ff76b049c170b908 R08: 0000000000000200 R09: 00000000ff161000
> Mar 27 15:03:11 pve-test-hba kernel: R10: 0000000000001000 R11: 0000000000001000 R12: 00000000001a0000
> Mar 27 15:03:11 pve-test-hba kernel: R13: ff76b049c31fe008 R14: 00000000f9600000 R15: 000000000019f000
> Mar 27 15:03:11 pve-test-hba kernel: FS: 0000000000000000(0000) GS:ff220bd3d1ee7000(0000) knlGS:0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> Mar 27 15:03:11 pve-test-hba kernel: CR2: ff76b049c31fe000 CR3: 0000000109f8c007 CR4: 0000000000f71ef0
> Mar 27 15:03:11 pve-test-hba kernel: PKRU: 55555554
> Mar 27 15:03:11 pve-test-hba kernel: note: dmcrypt_write/2[4385] exited with irqs disabled
> Mar 27 15:03:11 pve-test-hba kernel: ------------[ cut here ]------------
> Mar 27 15:03:11 pve-test-hba kernel: WARNING: CPU: 7 PID: 4385 at kernel/exit.c:902 do_exit+0x7d3/0xa30
> Mar 27 15:03:11 pve-test-hba kernel: Modules linked in: dm_crypt(E) ebtable_filter(E) ebtables(E) ip_set(E) ip6table_raw(E) ip6table_filter(E) ip6_tables(E) iptable_filter(E) nf_tables(E) sunrpc(E) iptable_raw(E) xt_CT(E) iptable_nat(E) xt>
> Mar 27 15:03:11 pve-test-hba kernel: blake2b_generic(E) xor(E) raid6_pq(E) dm_thin_pool(E) dm_persistent_data(E) dm_bio_prison(E) dm_bufio(E) hid_generic(E) usbmouse(E) rndis_host(E) usbhid(E) cdc_ether(E) hid(E) usbnet(E) mii(E) xhci_pci>
> Mar 27 15:03:11 pve-test-hba kernel: CPU: 7 UID: 0 PID: 4385 Comm: dmcrypt_write/2 Tainted: G D E 6.16.0-rc4-step14-00001-g9b8b84879d4a #16 PREEMPT(voluntary)
> Mar 27 15:03:11 pve-test-hba kernel: Tainted: [D]=DIE, [E]=UNSIGNED_MODULE
> Mar 27 15:03:11 pve-test-hba kernel: Hardware name: <snip>
> Mar 27 15:03:11 pve-test-hba kernel: RIP: 0010:do_exit+0x7d3/0xa30
> Mar 27 15:03:11 pve-test-hba kernel: Code: 48 89 45 c0 48 8b 83 50 0d 00 00 e9 44 fe ff ff 48 8b bb 10 0b 00 00 31 f6 e8 69 e2 ff ff e9 f7 fd ff ff 0f 0b e9 6b f8 ff ff <0f> 0b e9 72 f8 ff ff 4c 89 e6 bf 05 06 00 00 e8 49 46 01 00 e9 ab
> Mar 27 15:03:11 pve-test-hba kernel: RSP: 0018:ff76b049c170bec0 EFLAGS: 00010282
> Mar 27 15:03:11 pve-test-hba kernel: RAX: 0000000000000246 RBX: ff220bd04a4e0000 RCX: 0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: RDX: 000000000000270f RSI: 0000000000002710 RDI: 0000000000000009
> Mar 27 15:03:11 pve-test-hba kernel: RBP: ff76b049c170bf10 R08: 0000000000000000 R09: 0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: R10: 0000000000001000 R11: 0000000000001000 R12: 00000000001a0000
> Mar 27 15:03:11 pve-test-hba kernel: R13: ff76b049c31fe008 R14: 00000000f9600000 R15: 000000000019f000
> Mar 27 15:03:11 pve-test-hba kernel: FS: 0000000000000000(0000) GS:ff220bd3d1ee7000(0000) knlGS:0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> Mar 27 15:03:11 pve-test-hba kernel: CR2: ff76b049c31fe000 CR3: 0000000109f8c007 CR4: 0000000000f71ef0
> Mar 27 15:03:11 pve-test-hba kernel: PKRU: 55555554
> Mar 27 15:03:11 pve-test-hba kernel: note: dmcrypt_write/2[4385] exited with irqs disabled
> Mar 27 15:03:11 pve-test-hba kernel: ------------[ cut here ]------------
> Mar 27 15:03:11 pve-test-hba kernel: WARNING: CPU: 7 PID: 4385 at kernel/exit.c:902 do_exit+0x7d3/0xa30
> Mar 27 15:03:11 pve-test-hba kernel: Modules linked in: dm_crypt(E) ebtable_filter(E) ebtables(E) ip_set(E) ip6table_raw(E) ip6table_filter(E) ip6_tables(E) iptable_filter(E) nf_tables(E) sunrpc(E) iptable_raw(E) xt_CT(E) iptable_nat(E) xt>
> Mar 27 15:03:11 pve-test-hba kernel: blake2b_generic(E) xor(E) raid6_pq(E) dm_thin_pool(E) dm_persistent_data(E) dm_bio_prison(E) dm_bufio(E) hid_generic(E) usbmouse(E) rndis_host(E) usbhid(E) cdc_ether(E) hid(E) usbnet(E) mii(E) xhci_pci>
> Mar 27 15:03:11 pve-test-hba kernel: CPU: 7 UID: 0 PID: 4385 Comm: dmcrypt_write/2 Tainted: G D E 6.16.0-rc4-step14-00001-g9b8b84879d4a #16 PREEMPT(voluntary)
> Mar 27 15:03:11 pve-test-hba kernel: Tainted: [D]=DIE, [E]=UNSIGNED_MODULE
> Mar 27 15:03:11 pve-test-hba kernel: Hardware name: <snip>
> Mar 27 15:03:11 pve-test-hba kernel: RIP: 0010:do_exit+0x7d3/0xa30
> Mar 27 15:03:11 pve-test-hba kernel: Code: 48 89 45 c0 48 8b 83 50 0d 00 00 e9 44 fe ff ff 48 8b bb 10 0b 00 00 31 f6 e8 69 e2 ff ff e9 f7 fd ff ff 0f 0b e9 6b f8 ff ff <0f> 0b e9 72 f8 ff ff 4c 89 e6 bf 05 06 00 00 e8 49 46 01 00 e9 ab
> Mar 27 15:03:11 pve-test-hba kernel: RSP: 0018:ff76b049c170bec0 EFLAGS: 00010282
> Mar 27 15:03:11 pve-test-hba kernel: RAX: 0000000000000246 RBX: ff220bd04a4e0000 RCX: 0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: RDX: 000000000000270f RSI: 0000000000002710 RDI: 0000000000000009
> Mar 27 15:03:11 pve-test-hba kernel: RBP: ff76b049c170bf10 R08: 0000000000000000 R09: 0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000009
> Mar 27 15:03:11 pve-test-hba kernel: R13: ff220bd04a4e0000 R14: ff220bd04a4e0000 R15: 0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: FS: 0000000000000000(0000) GS:ff220bd3d1ee7000(0000) knlGS:0000000000000000
> Mar 27 15:03:11 pve-test-hba kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> Mar 27 15:03:11 pve-test-hba kernel: CR2: ff76b049c31fe000 CR3: 0000000109f8c007 CR4: 0000000000f71ef0
> Mar 27 15:03:11 pve-test-hba kernel: PKRU: 55555554
> Mar 27 15:03:11 pve-test-hba kernel: Call Trace:
> Mar 27 15:03:11 pve-test-hba kernel: <TASK>
> Mar 27 15:03:11 pve-test-hba kernel: make_task_dead+0x81/0x160
> Mar 27 15:03:11 pve-test-hba kernel: rewind_stack_and_make_dead+0x16/0x20
> Mar 27 15:03:11 pve-test-hba kernel: </TASK>
> Mar 27 15:03:11 pve-test-hba kernel: ---[ end trace 0000000000000000 ]---
> ```
>
> Please note that the kernel used here was the last build during
> bisecting, having this patch as the last commit.
> The stack trace looks similar in all tested (bad) versions.
> We've also tested 7.0-rc5, which also triggered the issue.
>
> The easiest way we found to trigger this was to create a Ceph OSD on the
> disks. When they were started on boot, the error was triggered.
>
> So far we are not sure if it's the Broadcom controller, or the disk that
> is causing it in the end.
>
> Since we saw the quirks added for certain devices [0][1], we also tried
> changing the sector size on an unaffected kernel to 8191, 8192 and
> 16384, but could not trigger the issue.
>
> Any ideas what could be the cause for this, or how to troubleshoot this
> further?
>
> Happy to provide any further information if needed.
>
>
> Thanks,
> Mira
>
> [0]
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=2e983271363108b3813b38754eb96d9b1cb252bb
> [1]
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=5f64ae1ef639a2bab7e39497c55f76cc0682f108
>
--
Damien Le Moal
Western Digital Research
next prev parent reply other threads:[~2026-03-31 19:48 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-18 6:00 [PATCH v2] block: Increase BLK_DEF_MAX_SECTORS_CAP Damien Le Moal
2025-06-18 6:17 ` Hannes Reinecke
2025-06-18 8:51 ` Johannes Thumshirn
2025-06-18 9:06 ` John Garry
2025-06-18 9:47 ` Damien Le Moal
2025-06-18 10:19 ` Martin K. Petersen
2025-06-23 13:40 ` Christoph Hellwig
2025-06-24 16:49 ` Jens Axboe
2025-08-27 7:07 ` Sebastian Andrzej Siewior
2025-08-27 7:38 ` Christoph Hellwig
2025-08-27 7:52 ` Sebastian Andrzej Siewior
2025-08-27 8:00 ` Christoph Hellwig
2025-08-27 8:03 ` Damien Le Moal
2025-08-27 8:01 ` Damien Le Moal
2025-08-27 8:42 ` Sebastian Andrzej Siewior
2025-08-27 9:01 ` Damien Le Moal
2025-08-27 10:16 ` Sebastian Andrzej Siewior
2026-03-31 12:02 ` Mira Limbeck
2026-03-31 12:30 ` Mira Limbeck
2026-03-31 19:48 ` Damien Le Moal [this message]
2026-04-01 10:32 ` Mira Limbeck
2026-04-01 20:02 ` Damien Le Moal
2026-04-01 20:55 ` Keith Busch
2026-04-01 23:31 ` Damien Le Moal
2026-04-02 14:33 ` Friedrich Weber
2026-04-02 15:03 ` Keith Busch
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ff5e2877-840b-4eb6-b449-bb64fb2e4097@kernel.org \
--to=dlemoal@kernel.org \
--cc=axboe@kernel.dk \
--cc=f.weber@proxmox.com \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
--cc=m.limbeck@proxmox.com \
--cc=martin.petersen@oracle.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox