- Linux-stable-mirror - lists.linaro.org

FAILED: patch "[PATCH] scsi: mpi3mr: Serialize admin queue BAR writes on 32-bit" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.1-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y git checkout FETCH_HEAD git cherry-pick -x c91e140c82eb58724c435f623702e51cc7896646 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082105-penalize-quadrant-9ad9@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From c91e140c82eb58724c435f623702e51cc7896646 Mon Sep 17 00:00:00 2001 From: Ranjan Kumar <ranjan.kumar(a)broadcom.com> Date: Sat, 28 Jun 2025 01:15:38 +0530 Subject: [PATCH] scsi: mpi3mr: Serialize admin queue BAR writes on 32-bit systems On 32-bit systems, 64-bit BAR writes to admin queue registers are performed as two 32-bit writes. Without locking, this can cause partial writes when accessed concurrently. Updated per-queue spinlocks is used to serialize these writes and prevent race conditions. Fixes: 824a156633df ("scsi: mpi3mr: Base driver code") Cc: stable(a)vger.kernel.org Signed-off-by: Ranjan Kumar <ranjan.kumar(a)broadcom.com> Link: https://lore.kernel.org/r/20250627194539.48851-4-ranjan.kumar@broadcom.com Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> diff --git a/drivers/scsi/mpi3mr/mpi3mr.h b/drivers/scsi/mpi3mr/mpi3mr.h index bf272dd69d23..f2201108ea91 100644 --- a/drivers/scsi/mpi3mr/mpi3mr.h +++ b/drivers/scsi/mpi3mr/mpi3mr.h @@ -1137,6 +1137,8 @@ struct scmd_priv { * @logdata_buf: Circular buffer to store log data entries * @logdata_buf_idx: Index of entry in buffer to store * @logdata_entry_sz: log data entry size + * @adm_req_q_bar_writeq_lock: Admin request queue lock + * @adm_reply_q_bar_writeq_lock: Admin reply queue lock * @pend_large_data_sz: Counter to track pending large data * @io_throttle_data_length: I/O size to track in 512b blocks * @io_throttle_high: I/O size to start throttle in 512b blocks @@ -1339,6 +1341,8 @@ struct mpi3mr_ioc { u8 *logdata_buf; u16 logdata_buf_idx; u16 logdata_entry_sz; + spinlock_t adm_req_q_bar_writeq_lock; + spinlock_t adm_reply_q_bar_writeq_lock; atomic_t pend_large_data_sz; u32 io_throttle_data_length; diff --git a/drivers/scsi/mpi3mr/mpi3mr_fw.c b/drivers/scsi/mpi3mr/mpi3mr_fw.c index 8976582946a2..0152d31d430a 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_fw.c +++ b/drivers/scsi/mpi3mr/mpi3mr_fw.c @@ -23,17 +23,22 @@ module_param(poll_queues, int, 0444); MODULE_PARM_DESC(poll_queues, "Number of queues for io_uring poll mode. (Range 1 - 126)"); #if defined(writeq) && defined(CONFIG_64BIT) -static inline void mpi3mr_writeq(__u64 b, void __iomem *addr) +static inline void mpi3mr_writeq(__u64 b, void __iomem *addr, + spinlock_t *write_queue_lock) { writeq(b, addr); } #else -static inline void mpi3mr_writeq(__u64 b, void __iomem *addr) +static inline void mpi3mr_writeq(__u64 b, void __iomem *addr, + spinlock_t *write_queue_lock) { __u64 data_out = b; + unsigned long flags; + spin_lock_irqsave(write_queue_lock, flags); writel((u32)(data_out), addr); writel((u32)(data_out >> 32), (addr + 4)); + spin_unlock_irqrestore(write_queue_lock, flags); } #endif @@ -2954,9 +2959,11 @@ static int mpi3mr_setup_admin_qpair(struct mpi3mr_ioc *mrioc) (mrioc->num_admin_req); writel(num_admin_entries, &mrioc->sysif_regs->admin_queue_num_entries); mpi3mr_writeq(mrioc->admin_req_dma, - &mrioc->sysif_regs->admin_request_queue_address); + &mrioc->sysif_regs->admin_request_queue_address, + &mrioc->adm_req_q_bar_writeq_lock); mpi3mr_writeq(mrioc->admin_reply_dma, - &mrioc->sysif_regs->admin_reply_queue_address); + &mrioc->sysif_regs->admin_reply_queue_address, + &mrioc->adm_reply_q_bar_writeq_lock); writel(mrioc->admin_req_pi, &mrioc->sysif_regs->admin_request_queue_pi); writel(mrioc->admin_reply_ci, &mrioc->sysif_regs->admin_reply_queue_ci); return retval; diff --git a/drivers/scsi/mpi3mr/mpi3mr_os.c b/drivers/scsi/mpi3mr/mpi3mr_os.c index ce444efd859e..0d1c9bbb13b5 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_os.c +++ b/drivers/scsi/mpi3mr/mpi3mr_os.c @@ -5365,6 +5365,8 @@ mpi3mr_probe(struct pci_dev *pdev, const struct pci_device_id *id) spin_lock_init(&mrioc->tgtdev_lock); spin_lock_init(&mrioc->watchdog_lock); spin_lock_init(&mrioc->chain_buf_lock); + spin_lock_init(&mrioc->adm_req_q_bar_writeq_lock); + spin_lock_init(&mrioc->adm_reply_q_bar_writeq_lock); spin_lock_init(&mrioc->sas_node_lock); spin_lock_init(&mrioc->trigger_lock);

3 days, 1 hour

2
2
0 0

FAILED: patch "[PATCH] scsi: mpi3mr: Serialize admin queue BAR writes on 32-bit" failed to apply to 6.6-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.6-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.6.y git checkout FETCH_HEAD git cherry-pick -x c91e140c82eb58724c435f623702e51cc7896646 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082105-fit-porridge-a9dc@gregkh' --subject-prefix 'PATCH 6.6.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From c91e140c82eb58724c435f623702e51cc7896646 Mon Sep 17 00:00:00 2001 From: Ranjan Kumar <ranjan.kumar(a)broadcom.com> Date: Sat, 28 Jun 2025 01:15:38 +0530 Subject: [PATCH] scsi: mpi3mr: Serialize admin queue BAR writes on 32-bit systems On 32-bit systems, 64-bit BAR writes to admin queue registers are performed as two 32-bit writes. Without locking, this can cause partial writes when accessed concurrently. Updated per-queue spinlocks is used to serialize these writes and prevent race conditions. Fixes: 824a156633df ("scsi: mpi3mr: Base driver code") Cc: stable(a)vger.kernel.org Signed-off-by: Ranjan Kumar <ranjan.kumar(a)broadcom.com> Link: https://lore.kernel.org/r/20250627194539.48851-4-ranjan.kumar@broadcom.com Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> diff --git a/drivers/scsi/mpi3mr/mpi3mr.h b/drivers/scsi/mpi3mr/mpi3mr.h index bf272dd69d23..f2201108ea91 100644 --- a/drivers/scsi/mpi3mr/mpi3mr.h +++ b/drivers/scsi/mpi3mr/mpi3mr.h @@ -1137,6 +1137,8 @@ struct scmd_priv { * @logdata_buf: Circular buffer to store log data entries * @logdata_buf_idx: Index of entry in buffer to store * @logdata_entry_sz: log data entry size + * @adm_req_q_bar_writeq_lock: Admin request queue lock + * @adm_reply_q_bar_writeq_lock: Admin reply queue lock * @pend_large_data_sz: Counter to track pending large data * @io_throttle_data_length: I/O size to track in 512b blocks * @io_throttle_high: I/O size to start throttle in 512b blocks @@ -1339,6 +1341,8 @@ struct mpi3mr_ioc { u8 *logdata_buf; u16 logdata_buf_idx; u16 logdata_entry_sz; + spinlock_t adm_req_q_bar_writeq_lock; + spinlock_t adm_reply_q_bar_writeq_lock; atomic_t pend_large_data_sz; u32 io_throttle_data_length; diff --git a/drivers/scsi/mpi3mr/mpi3mr_fw.c b/drivers/scsi/mpi3mr/mpi3mr_fw.c index 8976582946a2..0152d31d430a 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_fw.c +++ b/drivers/scsi/mpi3mr/mpi3mr_fw.c @@ -23,17 +23,22 @@ module_param(poll_queues, int, 0444); MODULE_PARM_DESC(poll_queues, "Number of queues for io_uring poll mode. (Range 1 - 126)"); #if defined(writeq) && defined(CONFIG_64BIT) -static inline void mpi3mr_writeq(__u64 b, void __iomem *addr) +static inline void mpi3mr_writeq(__u64 b, void __iomem *addr, + spinlock_t *write_queue_lock) { writeq(b, addr); } #else -static inline void mpi3mr_writeq(__u64 b, void __iomem *addr) +static inline void mpi3mr_writeq(__u64 b, void __iomem *addr, + spinlock_t *write_queue_lock) { __u64 data_out = b; + unsigned long flags; + spin_lock_irqsave(write_queue_lock, flags); writel((u32)(data_out), addr); writel((u32)(data_out >> 32), (addr + 4)); + spin_unlock_irqrestore(write_queue_lock, flags); } #endif @@ -2954,9 +2959,11 @@ static int mpi3mr_setup_admin_qpair(struct mpi3mr_ioc *mrioc) (mrioc->num_admin_req); writel(num_admin_entries, &mrioc->sysif_regs->admin_queue_num_entries); mpi3mr_writeq(mrioc->admin_req_dma, - &mrioc->sysif_regs->admin_request_queue_address); + &mrioc->sysif_regs->admin_request_queue_address, + &mrioc->adm_req_q_bar_writeq_lock); mpi3mr_writeq(mrioc->admin_reply_dma, - &mrioc->sysif_regs->admin_reply_queue_address); + &mrioc->sysif_regs->admin_reply_queue_address, + &mrioc->adm_reply_q_bar_writeq_lock); writel(mrioc->admin_req_pi, &mrioc->sysif_regs->admin_request_queue_pi); writel(mrioc->admin_reply_ci, &mrioc->sysif_regs->admin_reply_queue_ci); return retval; diff --git a/drivers/scsi/mpi3mr/mpi3mr_os.c b/drivers/scsi/mpi3mr/mpi3mr_os.c index ce444efd859e..0d1c9bbb13b5 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_os.c +++ b/drivers/scsi/mpi3mr/mpi3mr_os.c @@ -5365,6 +5365,8 @@ mpi3mr_probe(struct pci_dev *pdev, const struct pci_device_id *id) spin_lock_init(&mrioc->tgtdev_lock); spin_lock_init(&mrioc->watchdog_lock); spin_lock_init(&mrioc->chain_buf_lock); + spin_lock_init(&mrioc->adm_req_q_bar_writeq_lock); + spin_lock_init(&mrioc->adm_reply_q_bar_writeq_lock); spin_lock_init(&mrioc->sas_node_lock); spin_lock_init(&mrioc->trigger_lock);

3 days, 2 hours

2
2
0 0

FAILED: patch "[PATCH] selftests: mptcp: pm: check flush doesn't reset limits" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y git checkout FETCH_HEAD git cherry-pick -x 452690be7de2f91cc0de68cb9e95252875b33503 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082203-populate-sublease-ef51@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 452690be7de2f91cc0de68cb9e95252875b33503 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" <matttbe(a)kernel.org> Date: Fri, 15 Aug 2025 19:28:21 +0200 Subject: [PATCH] selftests: mptcp: pm: check flush doesn't reset limits This modification is linked to the parent commit where the received ADD_ADDR limit was accidentally reset when the endpoints were flushed. To validate that, the test is now flushing endpoints after having set new limits, and before checking them. The 'Fixes' tag here below is the same as the one from the previous commit: this patch here is not fixing anything wrong in the selftests, but it validates the previous fix for an issue introduced by this commit ID. Fixes: 01cacb00b35c ("mptcp: add netlink-based PM") Cc: stable(a)vger.kernel.org Reviewed-by: Mat Martineau <martineau(a)kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe(a)kernel.org> Link: https://patch.msgid.link/20250815-net-mptcp-misc-fixes-6-17-rc2-v1-3-521fe9… Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> diff --git a/tools/testing/selftests/net/mptcp/pm_netlink.sh b/tools/testing/selftests/net/mptcp/pm_netlink.sh index 2e6648a2b2c0..ac7ec6f94023 100755 --- a/tools/testing/selftests/net/mptcp/pm_netlink.sh +++ b/tools/testing/selftests/net/mptcp/pm_netlink.sh @@ -198,6 +198,7 @@ set_limits 1 9 2>/dev/null check "get_limits" "${default_limits}" "subflows above hard limit" set_limits 8 8 +flush_endpoint ## to make sure it doesn't affect the limits check "get_limits" "$(format_limits 8 8)" "set limits" flush_endpoint

3 days, 2 hours

4
3
0 0

FAILED: patch "[PATCH] mmc: sdhci-pci-gli: GL9763e: Mask the replay timer timeout of" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x 340be332e420ed37d15d4169a1b4174e912ad6cb # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082253-ashes-nutmeg-b7d8@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 340be332e420ed37d15d4169a1b4174e912ad6cb Mon Sep 17 00:00:00 2001 From: Victor Shih <victor.shih(a)genesyslogic.com.tw> Date: Thu, 31 Jul 2025 14:57:52 +0800 Subject: [PATCH] mmc: sdhci-pci-gli: GL9763e: Mask the replay timer timeout of AER Due to a flaw in the hardware design, the GL9763e replay timer frequently times out when ASPM is enabled. As a result, the warning messages will often appear in the system log when the system accesses the GL9763e PCI config. Therefore, the replay timer timeout must be masked. Signed-off-by: Victor Shih <victor.shih(a)genesyslogic.com.tw> Fixes: 1ae1d2d6e555 ("mmc: sdhci-pci-gli: Add Genesys Logic GL9763E support") Cc: stable(a)vger.kernel.org Acked-by: Adrian Hunter <adrian.hunter(a)intel.com> Link: https://lore.kernel.org/r/20250731065752.450231-4-victorshihgli@gmail.com Signed-off-by: Ulf Hansson <ulf.hansson(a)linaro.org> diff --git a/drivers/mmc/host/sdhci-pci-gli.c b/drivers/mmc/host/sdhci-pci-gli.c index 436f0460222f..3a1de477e9af 100644 --- a/drivers/mmc/host/sdhci-pci-gli.c +++ b/drivers/mmc/host/sdhci-pci-gli.c @@ -1782,6 +1782,9 @@ static void gl9763e_hw_setting(struct sdhci_pci_slot *slot) value |= FIELD_PREP(GLI_9763E_HS400_RXDLY, GLI_9763E_HS400_RXDLY_5); pci_write_config_dword(pdev, PCIE_GLI_9763E_CLKRXDLY, value); + /* mask the replay timer timeout of AER */ + sdhci_gli_mask_replay_timer_timeout(pdev); + pci_read_config_dword(pdev, PCIE_GLI_9763E_VHS, &value); value &= ~GLI_9763E_VHS_REV; value |= FIELD_PREP(GLI_9763E_VHS_REV, GLI_9763E_VHS_REV_R);

3 days, 2 hours

1
0
0 0

FAILED: patch "[PATCH] mmc: sdhci-pci-gli: GL9763e: Mask the replay timer timeout of" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y git checkout FETCH_HEAD git cherry-pick -x 340be332e420ed37d15d4169a1b4174e912ad6cb # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082253-from-kindred-7d7e@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 340be332e420ed37d15d4169a1b4174e912ad6cb Mon Sep 17 00:00:00 2001 From: Victor Shih <victor.shih(a)genesyslogic.com.tw> Date: Thu, 31 Jul 2025 14:57:52 +0800 Subject: [PATCH] mmc: sdhci-pci-gli: GL9763e: Mask the replay timer timeout of AER Due to a flaw in the hardware design, the GL9763e replay timer frequently times out when ASPM is enabled. As a result, the warning messages will often appear in the system log when the system accesses the GL9763e PCI config. Therefore, the replay timer timeout must be masked. Signed-off-by: Victor Shih <victor.shih(a)genesyslogic.com.tw> Fixes: 1ae1d2d6e555 ("mmc: sdhci-pci-gli: Add Genesys Logic GL9763E support") Cc: stable(a)vger.kernel.org Acked-by: Adrian Hunter <adrian.hunter(a)intel.com> Link: https://lore.kernel.org/r/20250731065752.450231-4-victorshihgli@gmail.com Signed-off-by: Ulf Hansson <ulf.hansson(a)linaro.org> diff --git a/drivers/mmc/host/sdhci-pci-gli.c b/drivers/mmc/host/sdhci-pci-gli.c index 436f0460222f..3a1de477e9af 100644 --- a/drivers/mmc/host/sdhci-pci-gli.c +++ b/drivers/mmc/host/sdhci-pci-gli.c @@ -1782,6 +1782,9 @@ static void gl9763e_hw_setting(struct sdhci_pci_slot *slot) value |= FIELD_PREP(GLI_9763E_HS400_RXDLY, GLI_9763E_HS400_RXDLY_5); pci_write_config_dword(pdev, PCIE_GLI_9763E_CLKRXDLY, value); + /* mask the replay timer timeout of AER */ + sdhci_gli_mask_replay_timer_timeout(pdev); + pci_read_config_dword(pdev, PCIE_GLI_9763E_VHS, &value); value &= ~GLI_9763E_VHS_REV; value |= FIELD_PREP(GLI_9763E_VHS_REV, GLI_9763E_VHS_REV_R);

3 days, 2 hours

1
0
0 0

FAILED: patch "[PATCH] mm/mremap: fix WARN with uffd that has remap events disabled" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y git checkout FETCH_HEAD git cherry-pick -x 772e5b4a5e8360743645b9a466842d16092c4f94 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082218-untidy-blaspheme-a9c3@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 772e5b4a5e8360743645b9a466842d16092c4f94 Mon Sep 17 00:00:00 2001 From: David Hildenbrand <david(a)redhat.com> Date: Mon, 18 Aug 2025 19:53:58 +0200 Subject: [PATCH] mm/mremap: fix WARN with uffd that has remap events disabled Registering userfaultd on a VMA that spans at least one PMD and then mremap()'ing that VMA can trigger a WARN when recovering from a failed page table move due to a page table allocation error. The code ends up doing the right thing (recurse, avoiding moving actual page tables), but triggering that WARN is unpleasant: WARNING: CPU: 2 PID: 6133 at mm/mremap.c:357 move_normal_pmd mm/mremap.c:357 [inline] WARNING: CPU: 2 PID: 6133 at mm/mremap.c:357 move_pgt_entry mm/mremap.c:595 [inline] WARNING: CPU: 2 PID: 6133 at mm/mremap.c:357 move_page_tables+0x3832/0x44a0 mm/mremap.c:852 Modules linked in: CPU: 2 UID: 0 PID: 6133 Comm: syz.0.19 Not tainted 6.17.0-rc1-syzkaller-00004-g53e760d89498 #0 PREEMPT(full) Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 RIP: 0010:move_normal_pmd mm/mremap.c:357 [inline] RIP: 0010:move_pgt_entry mm/mremap.c:595 [inline] RIP: 0010:move_page_tables+0x3832/0x44a0 mm/mremap.c:852 Code: ... RSP: 0018:ffffc900037a76d8 EFLAGS: 00010293 RAX: 0000000000000000 RBX: 0000000032930007 RCX: ffffffff820c6645 RDX: ffff88802e56a440 RSI: ffffffff820c7201 RDI: 0000000000000007 RBP: ffff888037728fc0 R08: 0000000000000007 R09: 0000000000000000 R10: 0000000032930007 R11: 0000000000000000 R12: 0000000000000000 R13: ffffc900037a79a8 R14: 0000000000000001 R15: dffffc0000000000 FS: 000055556316a500(0000) GS:ffff8880d68bc000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b30863fff CR3: 0000000050171000 CR4: 0000000000352ef0 Call Trace: <TASK> copy_vma_and_data+0x468/0x790 mm/mremap.c:1215 move_vma+0x548/0x1780 mm/mremap.c:1282 mremap_to+0x1b7/0x450 mm/mremap.c:1406 do_mremap+0xfad/0x1f80 mm/mremap.c:1921 __do_sys_mremap+0x119/0x170 mm/mremap.c:1977 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xcd/0x4c0 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f00d0b8ebe9 Code: ... RSP: 002b:00007ffe5ea5ee98 EFLAGS: 00000246 ORIG_RAX: 0000000000000019 RAX: ffffffffffffffda RBX: 00007f00d0db5fa0 RCX: 00007f00d0b8ebe9 RDX: 0000000000400000 RSI: 0000000000c00000 RDI: 0000200000000000 RBP: 00007ffe5ea5eef0 R08: 0000200000c00000 R09: 0000000000000000 R10: 0000000000000003 R11: 0000000000000246 R12: 0000000000000002 R13: 00007f00d0db5fa0 R14: 00007f00d0db5fa0 R15: 0000000000000005 </TASK> The underlying issue is that we recurse during the original page table move, but not during the recovery move. Fix it by checking for both VMAs and performing the check before the pmd_none() sanity check. Add a new helper where we perform+document that check for the PMD and PUD level. Thanks to Harry for bisecting. Link: https://lkml.kernel.org/r/20250818175358.1184757-1-david@redhat.com Fixes: 0cef0bb836e3 ("mm: clear uffd-wp PTE/PMD state on mremap()") Signed-off-by: David Hildenbrand <david(a)redhat.com> Reported-by: syzbot+4d9a13f0797c46a29e42(a)syzkaller.appspotmail.com Closes: https://lkml.kernel.org/r/689bb893.050a0220.7f033.013a.GAE@google.com Tested-by: Harry Yoo <harry.yoo(a)oracle.com> Cc: "Liam R. Howlett" <Liam.Howlett(a)oracle.com> Cc: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: Jann Horn <jannh(a)google.com> Cc: Pedro Falcato <pfalcato(a)suse.de> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> diff --git a/mm/mremap.c b/mm/mremap.c index 33b642076205..e618a706aff5 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -323,6 +323,25 @@ static inline bool arch_supports_page_table_move(void) } #endif +static inline bool uffd_supports_page_table_move(struct pagetable_move_control *pmc) +{ + /* + * If we are moving a VMA that has uffd-wp registered but with + * remap events disabled (new VMA will not be registered with uffd), we + * need to ensure that the uffd-wp state is cleared from all pgtables. + * This means recursing into lower page tables in move_page_tables(). + * + * We might get called with VMAs reversed when recovering from a + * failed page table move. In that case, the + * "old"-but-actually-"originally new" VMA during recovery will not have + * a uffd context. Recursing into lower page tables during the original + * move but not during the recovery move will cause trouble, because we + * run into already-existing page tables. So check both VMAs. + */ + return !vma_has_uffd_without_event_remap(pmc->old) && + !vma_has_uffd_without_event_remap(pmc->new); +} + #ifdef CONFIG_HAVE_MOVE_PMD static bool move_normal_pmd(struct pagetable_move_control *pmc, pmd_t *old_pmd, pmd_t *new_pmd) @@ -335,6 +354,8 @@ static bool move_normal_pmd(struct pagetable_move_control *pmc, if (!arch_supports_page_table_move()) return false; + if (!uffd_supports_page_table_move(pmc)) + return false; /* * The destination pmd shouldn't be established, free_pgtables() * should have released it. @@ -361,15 +382,6 @@ static bool move_normal_pmd(struct pagetable_move_control *pmc, if (WARN_ON_ONCE(!pmd_none(*new_pmd))) return false; - /* If this pmd belongs to a uffd vma with remap events disabled, we need - * to ensure that the uffd-wp state is cleared from all pgtables. This - * means recursing into lower page tables in move_page_tables(), and we - * can reuse the existing code if we simply treat the entry as "not - * moved". - */ - if (vma_has_uffd_without_event_remap(vma)) - return false; - /* * We don't have to worry about the ordering of src and dst * ptlocks because exclusive mmap_lock prevents deadlock. @@ -418,6 +430,8 @@ static bool move_normal_pud(struct pagetable_move_control *pmc, if (!arch_supports_page_table_move()) return false; + if (!uffd_supports_page_table_move(pmc)) + return false; /* * The destination pud shouldn't be established, free_pgtables() * should have released it. @@ -425,15 +439,6 @@ static bool move_normal_pud(struct pagetable_move_control *pmc, if (WARN_ON_ONCE(!pud_none(*new_pud))) return false; - /* If this pud belongs to a uffd vma with remap events disabled, we need - * to ensure that the uffd-wp state is cleared from all pgtables. This - * means recursing into lower page tables in move_page_tables(), and we - * can reuse the existing code if we simply treat the entry as "not - * moved". - */ - if (vma_has_uffd_without_event_remap(vma)) - return false; - /* * We don't have to worry about the ordering of src and dst * ptlocks because exclusive mmap_lock prevents deadlock.

3 days, 2 hours

1
0
0 0

FAILED: patch "[PATCH] scsi: mpi3mr: Serialize admin queue BAR writes on 32-bit" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y git checkout FETCH_HEAD git cherry-pick -x c91e140c82eb58724c435f623702e51cc7896646 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082104-negative-wildcat-59ac@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From c91e140c82eb58724c435f623702e51cc7896646 Mon Sep 17 00:00:00 2001 From: Ranjan Kumar <ranjan.kumar(a)broadcom.com> Date: Sat, 28 Jun 2025 01:15:38 +0530 Subject: [PATCH] scsi: mpi3mr: Serialize admin queue BAR writes on 32-bit systems On 32-bit systems, 64-bit BAR writes to admin queue registers are performed as two 32-bit writes. Without locking, this can cause partial writes when accessed concurrently. Updated per-queue spinlocks is used to serialize these writes and prevent race conditions. Fixes: 824a156633df ("scsi: mpi3mr: Base driver code") Cc: stable(a)vger.kernel.org Signed-off-by: Ranjan Kumar <ranjan.kumar(a)broadcom.com> Link: https://lore.kernel.org/r/20250627194539.48851-4-ranjan.kumar@broadcom.com Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> diff --git a/drivers/scsi/mpi3mr/mpi3mr.h b/drivers/scsi/mpi3mr/mpi3mr.h index bf272dd69d23..f2201108ea91 100644 --- a/drivers/scsi/mpi3mr/mpi3mr.h +++ b/drivers/scsi/mpi3mr/mpi3mr.h @@ -1137,6 +1137,8 @@ struct scmd_priv { * @logdata_buf: Circular buffer to store log data entries * @logdata_buf_idx: Index of entry in buffer to store * @logdata_entry_sz: log data entry size + * @adm_req_q_bar_writeq_lock: Admin request queue lock + * @adm_reply_q_bar_writeq_lock: Admin reply queue lock * @pend_large_data_sz: Counter to track pending large data * @io_throttle_data_length: I/O size to track in 512b blocks * @io_throttle_high: I/O size to start throttle in 512b blocks @@ -1339,6 +1341,8 @@ struct mpi3mr_ioc { u8 *logdata_buf; u16 logdata_buf_idx; u16 logdata_entry_sz; + spinlock_t adm_req_q_bar_writeq_lock; + spinlock_t adm_reply_q_bar_writeq_lock; atomic_t pend_large_data_sz; u32 io_throttle_data_length; diff --git a/drivers/scsi/mpi3mr/mpi3mr_fw.c b/drivers/scsi/mpi3mr/mpi3mr_fw.c index 8976582946a2..0152d31d430a 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_fw.c +++ b/drivers/scsi/mpi3mr/mpi3mr_fw.c @@ -23,17 +23,22 @@ module_param(poll_queues, int, 0444); MODULE_PARM_DESC(poll_queues, "Number of queues for io_uring poll mode. (Range 1 - 126)"); #if defined(writeq) && defined(CONFIG_64BIT) -static inline void mpi3mr_writeq(__u64 b, void __iomem *addr) +static inline void mpi3mr_writeq(__u64 b, void __iomem *addr, + spinlock_t *write_queue_lock) { writeq(b, addr); } #else -static inline void mpi3mr_writeq(__u64 b, void __iomem *addr) +static inline void mpi3mr_writeq(__u64 b, void __iomem *addr, + spinlock_t *write_queue_lock) { __u64 data_out = b; + unsigned long flags; + spin_lock_irqsave(write_queue_lock, flags); writel((u32)(data_out), addr); writel((u32)(data_out >> 32), (addr + 4)); + spin_unlock_irqrestore(write_queue_lock, flags); } #endif @@ -2954,9 +2959,11 @@ static int mpi3mr_setup_admin_qpair(struct mpi3mr_ioc *mrioc) (mrioc->num_admin_req); writel(num_admin_entries, &mrioc->sysif_regs->admin_queue_num_entries); mpi3mr_writeq(mrioc->admin_req_dma, - &mrioc->sysif_regs->admin_request_queue_address); + &mrioc->sysif_regs->admin_request_queue_address, + &mrioc->adm_req_q_bar_writeq_lock); mpi3mr_writeq(mrioc->admin_reply_dma, - &mrioc->sysif_regs->admin_reply_queue_address); + &mrioc->sysif_regs->admin_reply_queue_address, + &mrioc->adm_reply_q_bar_writeq_lock); writel(mrioc->admin_req_pi, &mrioc->sysif_regs->admin_request_queue_pi); writel(mrioc->admin_reply_ci, &mrioc->sysif_regs->admin_reply_queue_ci); return retval; diff --git a/drivers/scsi/mpi3mr/mpi3mr_os.c b/drivers/scsi/mpi3mr/mpi3mr_os.c index ce444efd859e..0d1c9bbb13b5 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_os.c +++ b/drivers/scsi/mpi3mr/mpi3mr_os.c @@ -5365,6 +5365,8 @@ mpi3mr_probe(struct pci_dev *pdev, const struct pci_device_id *id) spin_lock_init(&mrioc->tgtdev_lock); spin_lock_init(&mrioc->watchdog_lock); spin_lock_init(&mrioc->chain_buf_lock); + spin_lock_init(&mrioc->adm_req_q_bar_writeq_lock); + spin_lock_init(&mrioc->adm_reply_q_bar_writeq_lock); spin_lock_init(&mrioc->sas_node_lock); spin_lock_init(&mrioc->trigger_lock);

3 days, 2 hours

2
2
0 0

FAILED: patch "[PATCH] NFS: Fix a race when updating an existing write" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.4.y git checkout FETCH_HEAD git cherry-pick -x 76d2e3890fb169168c73f2e4f8375c7cc24a765e # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082226-unnerve-ripening-2dfa@gregkh' --subject-prefix 'PATCH 5.4.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 76d2e3890fb169168c73f2e4f8375c7cc24a765e Mon Sep 17 00:00:00 2001 From: Trond Myklebust <trond.myklebust(a)hammerspace.com> Date: Sat, 16 Aug 2025 07:25:20 -0700 Subject: [PATCH] NFS: Fix a race when updating an existing write After nfs_lock_and_join_requests() tests for whether the request is still attached to the mapping, nothing prevents a call to nfs_inode_remove_request() from succeeding until we actually lock the page group. The reason is that whoever called nfs_inode_remove_request() doesn't necessarily have a lock on the page group head. So in order to avoid races, let's take the page group lock earlier in nfs_lock_and_join_requests(), and hold it across the removal of the request in nfs_inode_remove_request(). Reported-by: Jeff Layton <jlayton(a)kernel.org> Tested-by: Joe Quanaim <jdq(a)meta.com> Tested-by: Andrew Steffen <aksteffen(a)meta.com> Reviewed-by: Jeff Layton <jlayton(a)kernel.org> Fixes: bd37d6fce184 ("NFSv4: Convert nfs_lock_and_join_requests() to use nfs_page_find_head_request()") Cc: stable(a)vger.kernel.org Signed-off-by: Trond Myklebust <trond.myklebust(a)hammerspace.com> diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 11968dcb7243..6e69ce43a13f 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -253,13 +253,14 @@ nfs_page_group_unlock(struct nfs_page *req) nfs_page_clear_headlock(req); } -/* - * nfs_page_group_sync_on_bit_locked +/** + * nfs_page_group_sync_on_bit_locked - Test if all requests have @bit set + * @req: request in page group + * @bit: PG_* bit that is used to sync page group * * must be called with page group lock held */ -static bool -nfs_page_group_sync_on_bit_locked(struct nfs_page *req, unsigned int bit) +bool nfs_page_group_sync_on_bit_locked(struct nfs_page *req, unsigned int bit) { struct nfs_page *head = req->wb_head; struct nfs_page *tmp; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index fa5c41d0989a..8b7c04737967 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -153,20 +153,10 @@ nfs_page_set_inode_ref(struct nfs_page *req, struct inode *inode) } } -static int -nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode) +static void nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode) { - int ret; - - if (!test_bit(PG_REMOVE, &req->wb_flags)) - return 0; - ret = nfs_page_group_lock(req); - if (ret) - return ret; if (test_and_clear_bit(PG_REMOVE, &req->wb_flags)) nfs_page_set_inode_ref(req, inode); - nfs_page_group_unlock(req); - return 0; } /** @@ -585,19 +575,18 @@ retry: } } + ret = nfs_page_group_lock(head); + if (ret < 0) + goto out_unlock; + /* Ensure that nobody removed the request before we locked it */ if (head != folio->private) { + nfs_page_group_unlock(head); nfs_unlock_and_release_request(head); goto retry; } - ret = nfs_cancel_remove_inode(head, inode); - if (ret < 0) - goto out_unlock; - - ret = nfs_page_group_lock(head); - if (ret < 0) - goto out_unlock; + nfs_cancel_remove_inode(head, inode); /* lock each request in the page group */ for (subreq = head->wb_this_page; @@ -786,7 +775,8 @@ static void nfs_inode_remove_request(struct nfs_page *req) { struct nfs_inode *nfsi = NFS_I(nfs_page_to_inode(req)); - if (nfs_page_group_sync_on_bit(req, PG_REMOVE)) { + nfs_page_group_lock(req); + if (nfs_page_group_sync_on_bit_locked(req, PG_REMOVE)) { struct folio *folio = nfs_page_to_folio(req->wb_head); struct address_space *mapping = folio->mapping; @@ -798,6 +788,7 @@ static void nfs_inode_remove_request(struct nfs_page *req) } spin_unlock(&mapping->i_private_lock); } + nfs_page_group_unlock(req); if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) { atomic_long_dec(&nfsi->nrequests); diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 169b4ae30ff4..9aed39abc94b 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -160,6 +160,7 @@ extern void nfs_join_page_group(struct nfs_page *head, extern int nfs_page_group_lock(struct nfs_page *); extern void nfs_page_group_unlock(struct nfs_page *); extern bool nfs_page_group_sync_on_bit(struct nfs_page *, unsigned int); +extern bool nfs_page_group_sync_on_bit_locked(struct nfs_page *, unsigned int); extern int nfs_page_set_headlock(struct nfs_page *req); extern void nfs_page_clear_headlock(struct nfs_page *req); extern bool nfs_async_iocounter_wait(struct rpc_task *, struct nfs_lock_context *);

3 days, 2 hours

1
0
0 0

FAILED: patch "[PATCH] NFS: Fix a race when updating an existing write" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y git checkout FETCH_HEAD git cherry-pick -x 76d2e3890fb169168c73f2e4f8375c7cc24a765e # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082225-attribute-embark-823b@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 76d2e3890fb169168c73f2e4f8375c7cc24a765e Mon Sep 17 00:00:00 2001 From: Trond Myklebust <trond.myklebust(a)hammerspace.com> Date: Sat, 16 Aug 2025 07:25:20 -0700 Subject: [PATCH] NFS: Fix a race when updating an existing write After nfs_lock_and_join_requests() tests for whether the request is still attached to the mapping, nothing prevents a call to nfs_inode_remove_request() from succeeding until we actually lock the page group. The reason is that whoever called nfs_inode_remove_request() doesn't necessarily have a lock on the page group head. So in order to avoid races, let's take the page group lock earlier in nfs_lock_and_join_requests(), and hold it across the removal of the request in nfs_inode_remove_request(). Reported-by: Jeff Layton <jlayton(a)kernel.org> Tested-by: Joe Quanaim <jdq(a)meta.com> Tested-by: Andrew Steffen <aksteffen(a)meta.com> Reviewed-by: Jeff Layton <jlayton(a)kernel.org> Fixes: bd37d6fce184 ("NFSv4: Convert nfs_lock_and_join_requests() to use nfs_page_find_head_request()") Cc: stable(a)vger.kernel.org Signed-off-by: Trond Myklebust <trond.myklebust(a)hammerspace.com> diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 11968dcb7243..6e69ce43a13f 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -253,13 +253,14 @@ nfs_page_group_unlock(struct nfs_page *req) nfs_page_clear_headlock(req); } -/* - * nfs_page_group_sync_on_bit_locked +/** + * nfs_page_group_sync_on_bit_locked - Test if all requests have @bit set + * @req: request in page group + * @bit: PG_* bit that is used to sync page group * * must be called with page group lock held */ -static bool -nfs_page_group_sync_on_bit_locked(struct nfs_page *req, unsigned int bit) +bool nfs_page_group_sync_on_bit_locked(struct nfs_page *req, unsigned int bit) { struct nfs_page *head = req->wb_head; struct nfs_page *tmp; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index fa5c41d0989a..8b7c04737967 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -153,20 +153,10 @@ nfs_page_set_inode_ref(struct nfs_page *req, struct inode *inode) } } -static int -nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode) +static void nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode) { - int ret; - - if (!test_bit(PG_REMOVE, &req->wb_flags)) - return 0; - ret = nfs_page_group_lock(req); - if (ret) - return ret; if (test_and_clear_bit(PG_REMOVE, &req->wb_flags)) nfs_page_set_inode_ref(req, inode); - nfs_page_group_unlock(req); - return 0; } /** @@ -585,19 +575,18 @@ retry: } } + ret = nfs_page_group_lock(head); + if (ret < 0) + goto out_unlock; + /* Ensure that nobody removed the request before we locked it */ if (head != folio->private) { + nfs_page_group_unlock(head); nfs_unlock_and_release_request(head); goto retry; } - ret = nfs_cancel_remove_inode(head, inode); - if (ret < 0) - goto out_unlock; - - ret = nfs_page_group_lock(head); - if (ret < 0) - goto out_unlock; + nfs_cancel_remove_inode(head, inode); /* lock each request in the page group */ for (subreq = head->wb_this_page; @@ -786,7 +775,8 @@ static void nfs_inode_remove_request(struct nfs_page *req) { struct nfs_inode *nfsi = NFS_I(nfs_page_to_inode(req)); - if (nfs_page_group_sync_on_bit(req, PG_REMOVE)) { + nfs_page_group_lock(req); + if (nfs_page_group_sync_on_bit_locked(req, PG_REMOVE)) { struct folio *folio = nfs_page_to_folio(req->wb_head); struct address_space *mapping = folio->mapping; @@ -798,6 +788,7 @@ static void nfs_inode_remove_request(struct nfs_page *req) } spin_unlock(&mapping->i_private_lock); } + nfs_page_group_unlock(req); if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) { atomic_long_dec(&nfsi->nrequests); diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 169b4ae30ff4..9aed39abc94b 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -160,6 +160,7 @@ extern void nfs_join_page_group(struct nfs_page *head, extern int nfs_page_group_lock(struct nfs_page *); extern void nfs_page_group_unlock(struct nfs_page *); extern bool nfs_page_group_sync_on_bit(struct nfs_page *, unsigned int); +extern bool nfs_page_group_sync_on_bit_locked(struct nfs_page *, unsigned int); extern int nfs_page_set_headlock(struct nfs_page *req); extern void nfs_page_clear_headlock(struct nfs_page *req); extern bool nfs_async_iocounter_wait(struct rpc_task *, struct nfs_lock_context *);

3 days, 2 hours

1
0
0 0

FAILED: patch "[PATCH] NFS: Fix a race when updating an existing write" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x 76d2e3890fb169168c73f2e4f8375c7cc24a765e # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025082225-cling-drainer-d884@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 76d2e3890fb169168c73f2e4f8375c7cc24a765e Mon Sep 17 00:00:00 2001 From: Trond Myklebust <trond.myklebust(a)hammerspace.com> Date: Sat, 16 Aug 2025 07:25:20 -0700 Subject: [PATCH] NFS: Fix a race when updating an existing write After nfs_lock_and_join_requests() tests for whether the request is still attached to the mapping, nothing prevents a call to nfs_inode_remove_request() from succeeding until we actually lock the page group. The reason is that whoever called nfs_inode_remove_request() doesn't necessarily have a lock on the page group head. So in order to avoid races, let's take the page group lock earlier in nfs_lock_and_join_requests(), and hold it across the removal of the request in nfs_inode_remove_request(). Reported-by: Jeff Layton <jlayton(a)kernel.org> Tested-by: Joe Quanaim <jdq(a)meta.com> Tested-by: Andrew Steffen <aksteffen(a)meta.com> Reviewed-by: Jeff Layton <jlayton(a)kernel.org> Fixes: bd37d6fce184 ("NFSv4: Convert nfs_lock_and_join_requests() to use nfs_page_find_head_request()") Cc: stable(a)vger.kernel.org Signed-off-by: Trond Myklebust <trond.myklebust(a)hammerspace.com> diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 11968dcb7243..6e69ce43a13f 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -253,13 +253,14 @@ nfs_page_group_unlock(struct nfs_page *req) nfs_page_clear_headlock(req); } -/* - * nfs_page_group_sync_on_bit_locked +/** + * nfs_page_group_sync_on_bit_locked - Test if all requests have @bit set + * @req: request in page group + * @bit: PG_* bit that is used to sync page group * * must be called with page group lock held */ -static bool -nfs_page_group_sync_on_bit_locked(struct nfs_page *req, unsigned int bit) +bool nfs_page_group_sync_on_bit_locked(struct nfs_page *req, unsigned int bit) { struct nfs_page *head = req->wb_head; struct nfs_page *tmp; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index fa5c41d0989a..8b7c04737967 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -153,20 +153,10 @@ nfs_page_set_inode_ref(struct nfs_page *req, struct inode *inode) } } -static int -nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode) +static void nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode) { - int ret; - - if (!test_bit(PG_REMOVE, &req->wb_flags)) - return 0; - ret = nfs_page_group_lock(req); - if (ret) - return ret; if (test_and_clear_bit(PG_REMOVE, &req->wb_flags)) nfs_page_set_inode_ref(req, inode); - nfs_page_group_unlock(req); - return 0; } /** @@ -585,19 +575,18 @@ retry: } } + ret = nfs_page_group_lock(head); + if (ret < 0) + goto out_unlock; + /* Ensure that nobody removed the request before we locked it */ if (head != folio->private) { + nfs_page_group_unlock(head); nfs_unlock_and_release_request(head); goto retry; } - ret = nfs_cancel_remove_inode(head, inode); - if (ret < 0) - goto out_unlock; - - ret = nfs_page_group_lock(head); - if (ret < 0) - goto out_unlock; + nfs_cancel_remove_inode(head, inode); /* lock each request in the page group */ for (subreq = head->wb_this_page; @@ -786,7 +775,8 @@ static void nfs_inode_remove_request(struct nfs_page *req) { struct nfs_inode *nfsi = NFS_I(nfs_page_to_inode(req)); - if (nfs_page_group_sync_on_bit(req, PG_REMOVE)) { + nfs_page_group_lock(req); + if (nfs_page_group_sync_on_bit_locked(req, PG_REMOVE)) { struct folio *folio = nfs_page_to_folio(req->wb_head); struct address_space *mapping = folio->mapping; @@ -798,6 +788,7 @@ static void nfs_inode_remove_request(struct nfs_page *req) } spin_unlock(&mapping->i_private_lock); } + nfs_page_group_unlock(req); if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) { atomic_long_dec(&nfsi->nrequests); diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 169b4ae30ff4..9aed39abc94b 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -160,6 +160,7 @@ extern void nfs_join_page_group(struct nfs_page *head, extern int nfs_page_group_lock(struct nfs_page *); extern void nfs_page_group_unlock(struct nfs_page *); extern bool nfs_page_group_sync_on_bit(struct nfs_page *, unsigned int); +extern bool nfs_page_group_sync_on_bit_locked(struct nfs_page *, unsigned int); extern int nfs_page_set_headlock(struct nfs_page *req); extern void nfs_page_clear_headlock(struct nfs_page *req); extern bool nfs_async_iocounter_wait(struct rpc_task *, struct nfs_lock_context *);

3 days, 2 hours

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror