- Kernel - mailweb.openeuler.org

[PATCH OLK-6.6] KVM: arm64: vgic-v3: Restrict ICH_AP1Rn_EL2 accesses to 64-bit only with vNMI
by Jinqian Yang 15 Sep '26

15 Sep '26

virt inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9836 -------------------------------------------------------------------- ICH_AP1Rn_EL2 is 64-bit only when vNMI is in use, with the NMI priority in ICH_AP1R0_EL2[63:32]; otherwise the upper bits are RES0. Gate the save/restore access width on vgic_v3_vcpu_has_vnmi() which checks both vgic.has_nmi and ID_AA64PFR1_EL1.NMI, so that the NMI priority is preserved across context switch only when vNMI is actually enabled for the guest. Also drop the early assignment of kvm->arch.pfr1_nmi in kvm_arch_init_vm(). Without it vNMI is no longer enabled by default at VM creation on hosts with FEAT_NMI; instead the VMM must explicitly set ID_AA64PFR1_EL1.NMI via SET_ONE_REG to enable vNMI for the guest. Signed-off-by: Jinqian Yang <yangjinqian1(a)huawei.com> --- arch/arm64/include/asm/kvm_hyp.h | 1 + arch/arm64/kvm/arm.c | 3 -- arch/arm64/kvm/hyp/vgic-v3-sr.c | 55 +++++++++++++++++++++++++++----- 3 files changed, 48 insertions(+), 11 deletions(-) diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h index 127ad280e68b..c4121bf0aadc 100644 --- a/arch/arm64/include/asm/kvm_hyp.h +++ b/arch/arm64/include/asm/kvm_hyp.h @@ -82,6 +82,7 @@ void __vgic_v3_activate_traps(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if); +bool vgic_v3_vcpu_has_vnmi(struct kvm_vcpu *vcpu); int __vgic_v3_perform_cpuif_access(struct kvm_vcpu *vcpu); #ifdef __KVM_NVHE_HYPERVISOR__ diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index 2b560eba3e0b..9a4be08730da 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -381,9 +381,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) /* The maximum number of VCPUs is limited by the host's GIC model */ kvm->max_vcpus = kvm_arm_default_max_vcpus(); - if (cpus_have_const_cap(ARM64_HAS_NMI) && !static_branch_unlikely(&vgic_v3_cpuif_trap)) - kvm->arch.pfr1_nmi = ID_AA64PFR1_EL1_NMI_IMP; - kvm_arm_init_hypercalls(kvm); if (!kvm_is_realm(kvm)) diff --git a/arch/arm64/kvm/hyp/vgic-v3-sr.c b/arch/arm64/kvm/hyp/vgic-v3-sr.c index 7a9e132f68db..ab406332762e 100644 --- a/arch/arm64/kvm/hyp/vgic-v3-sr.c +++ b/arch/arm64/kvm/hyp/vgic-v3-sr.c @@ -335,10 +335,40 @@ void __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if) write_gicreg(0, ICH_HCR_EL2); } +static struct kvm_vcpu *vgic_v3_cpu_if_to_vcpu(struct vgic_v3_cpu_if *cpu_if) +{ + struct vgic_cpu *vgic_cpu = container_of(cpu_if, struct vgic_cpu, vgic_v3); + + return container_of(vgic_cpu, struct kvm_vcpu, arch.vgic_cpu); +} + +/* + * vNMI is only usable if the guest has selected NMI support + * (vgic.has_nmi) and the vcpu has FEAT_NMI (pfr1_nmi). + * + * This drives the width of the ICH_AP1Rn_EL2 accesses: with vNMI the + * registers are 64bit (the NMI priority lives in ICH_AP1R0_EL2[63]), + * without it only the low 32 bits are valid. + */ +bool vgic_v3_vcpu_has_vnmi(struct kvm_vcpu *vcpu) +{ + struct kvm *kvm = kern_hyp_va(vcpu->kvm); + + return kvm->arch.vgic.has_nmi && + kvm->arch.pfr1_nmi == ID_AA64PFR1_EL1_NMI_IMP; +} + void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if) { u64 val; u32 nr_pre_bits; + /* + * ICH_AP1Rn_EL2 are only 64bit when the vcpu has vNMI: the NMI + * priority is encoded in ICH_AP1R0_EL2[63]. Without vNMI, the + * upper bits are RES0 and only the low 32 bits must be + * preserved. + */ + bool vnmi = vgic_v3_vcpu_has_vnmi(vgic_v3_cpu_if_to_vcpu(cpu_if)); val = read_gicreg(ICH_VTR_EL2); nr_pre_bits = vtr_to_nr_pre_bits(val); @@ -357,14 +387,18 @@ void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if) switch (nr_pre_bits) { case 7: - cpu_if->vgic_ap1r[3] = __vgic_v3_read_ap1rn(3); - cpu_if->vgic_ap1r[2] = __vgic_v3_read_ap1rn(2); + cpu_if->vgic_ap1r[3] = vnmi ? __vgic_v3_read_ap1rn(3) + : (u32)__vgic_v3_read_ap1rn(3); + cpu_if->vgic_ap1r[2] = vnmi ? __vgic_v3_read_ap1rn(2) + : (u32)__vgic_v3_read_ap1rn(2); fallthrough; case 6: - cpu_if->vgic_ap1r[1] = __vgic_v3_read_ap1rn(1); + cpu_if->vgic_ap1r[1] = vnmi ? __vgic_v3_read_ap1rn(1) + : (u32)__vgic_v3_read_ap1rn(1); fallthrough; default: - cpu_if->vgic_ap1r[0] = __vgic_v3_read_ap1rn(0); + cpu_if->vgic_ap1r[0] = vnmi ? __vgic_v3_read_ap1rn(0) + : (u32)__vgic_v3_read_ap1rn(0); } } @@ -372,6 +406,7 @@ void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if) { u64 val; u32 nr_pre_bits; + bool vnmi = vgic_v3_vcpu_has_vnmi(vgic_v3_cpu_if_to_vcpu(cpu_if)); val = read_gicreg(ICH_VTR_EL2); nr_pre_bits = vtr_to_nr_pre_bits(val); @@ -390,14 +425,18 @@ void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if) switch (nr_pre_bits) { case 7: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[3], 3); - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[2], 2); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[3] + : (u32)cpu_if->vgic_ap1r[3], 3); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[2] + : (u32)cpu_if->vgic_ap1r[2], 2); fallthrough; case 6: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[1], 1); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[1] + : (u32)cpu_if->vgic_ap1r[1], 1); fallthrough; default: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[0], 0); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[0] + : (u32)cpu_if->vgic_ap1r[0], 0); } } -- 2.33.0

2 1

[PATCH OLK-5.10 0/2] CVE-2026-68396
by Zhou Minqiang 14 Sep '26

14 Sep '26

CVE-2026-68396 Zhou Minqiang (2): scsi: core: wake eh reliably when using scsi_schedule_eh scsi: core: Fix kabi breakage of struct Scsi_Host drivers/scsi/hosts.c | 13 +++++++++++++ drivers/scsi/scsi_error.c | 24 +++++++++++++++++++++++- drivers/scsi/scsi_priv.h | 1 + include/scsi/scsi_host.h | 7 ++++++- 4 files changed, 43 insertions(+), 2 deletions(-) -- 2.52.0

2 3

[PATCH OLK-5.10] scsi: core: pair EH runtime PM get and put
by Zhou Minqiang 14 Sep '26

14 Sep '26

mainline inclusion from mainline-v7.2 commit 872f486259ae0bc6b73ca4735a15d013241f73e9 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18399 CVE: CVE-2026-74754 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- shost->eh_noresume is currently consulted twice in one error handling iteration: once before scsi_autopm_get_host() and once again before scsi_autopm_put_host(). That is racy when a PM-triggered error path flips shost->eh_noresume while the SCSI EH thread is still running. The problem flow looks like this: PM path ufshcd_set_dev_pwr_mode() shost->eh_noresume = 1 ufshcd_execute_start_stop <-- trigger EH ... shost->eh_noresume = 0 EH path scsi_error_handler() if (!shost->eh_noresume) scsi_autopm_get_host() <-- skipped ... if (!shost->eh_noresume) scsi_autopm_put_host() <-- executed later In that case one EH iteration can skip autoresume on entry and still drop a runtime PM reference on exit. That leaves an unmatched runtime PM put and can trigger a runtime PM usage count underflow. Fix this by making eh_noresume a regular bool so it can be accessed with READ_ONCE() and WRITE_ONCE(). Snapshot it once per EH iteration and use that snapshot for both runtime PM get and put decisions. Fixes: ae0751ffc77e ("[SCSI] add flag to skip the runtime PM calls on the host") Signed-off-by: Hongjie Fang <hongjiefang(a)asrmicro.com> Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Reviewed-by: Peter Wang <peter.wang(a)mediatek.com> Link: https://patch.msgid.link/20260729111614.2407559-1-hongjiefang@asrmicro.com Signed-off-by: Martin K. Petersen (Oracle) <mkp(a)kernel.org> Conflicts: drivers/scsi/scsi_error.c drivers/scsi/ufs/ufshcd.c [ctx conflicts] Signed-off-by: Zhou Minqiang <zhouminqiang2(a)huawei.com> --- drivers/scsi/scsi_error.c | 9 +++++++-- drivers/scsi/ufs/ufshcd.c | 4 ++++ 2 files changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 18512b33a190..90f275ac43c6 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -2204,6 +2204,8 @@ static void scsi_unjam_host(struct Scsi_Host *shost) int scsi_error_handler(void *data) { struct Scsi_Host *shost = data; + bool eh_noresume; + unsigned long flags; /* * We use TASK_INTERRUPTIBLE so that the thread is not @@ -2245,7 +2247,10 @@ int scsi_error_handler(void *data) * what we need to do to get it up and online again (if we can). * If we fail, we end up taking the thing offline. */ - if (!shost->eh_noresume && scsi_autopm_get_host(shost) != 0) { + spin_lock_irqsave(shost->host_lock, flags); + eh_noresume = shost->eh_noresume; + spin_unlock_irqrestore(shost->host_lock, flags); + if (!eh_noresume && scsi_autopm_get_host(shost) != 0) { SCSI_LOG_ERROR_RECOVERY(1, shost_printk(KERN_ERR, shost, "scsi_eh_%d: unable to autoresume\n", @@ -2269,7 +2274,7 @@ int scsi_error_handler(void *data) * which are still online. */ scsi_restart_operations(shost); - if (!shost->eh_noresume) + if (!eh_noresume) scsi_autopm_put_host(shost); } __set_current_state(TASK_RUNNING); diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c index ae00892ba44a..b3c6f41a5284 100644 --- a/drivers/scsi/ufs/ufshcd.c +++ b/drivers/scsi/ufs/ufshcd.c @@ -8291,7 +8291,9 @@ static int ufshcd_set_dev_pwr_mode(struct ufs_hba *hba, * we are functional while we are here, skip host resume in error * handling context. */ + spin_lock_irqsave(hba->host->host_lock, flags); hba->host->eh_noresume = 1; + spin_unlock_irqrestore(hba->host->host_lock, flags); if (hba->wlun_dev_clr_ua) { ret = ufshcd_send_request_sense(hba, sdp); if (ret) @@ -8321,7 +8323,9 @@ static int ufshcd_set_dev_pwr_mode(struct ufs_hba *hba, hba->curr_dev_pwr_mode = pwr_mode; out: scsi_device_put(sdp); + spin_lock_irqsave(hba->host->host_lock, flags); hba->host->eh_noresume = 0; + spin_unlock_irqrestore(hba->host->host_lock, flags); return ret; } -- 2.52.0

2 1

[PATCH OLK-6.6 0/8] arm64: Add FEAT_TLBID support
by Zeng Heng 14 Sep '26

14 Sep '26

TLBI broadcasting all PEs introduces performance noise. By combining hardware and software, TLBID (TLBI Domain) limit TLBI to an appropriate scope, avoiding the performance overhead caused by broadcasting. Arm Architecture defines the FEAT_TLBID extension since Armv9.3, when present, allows the OSPM to target TLB invalidations to a specific domain. A TLBI domain can be contained in multiple parent domains. The TLBI table lists all the PE and SMMU components and the TLBI Domains that each component belongs to. The feature is detected at runtime, and will remain disabled if the system does not implement the feature. Jinjiang Tu (1): arm64: tlbflush: Optimize flush_tlb_mm() by using TLBID Liao Chang (1): acpi: arm64: Add basic skeleton for TLBI Domains Marc Zyngier (4): arm64: cpufeature: Add ID_AA64MMFR4_EL1 handling arm64: sysreg: Add layout for ID_AA64MMFR4_EL1 arm64: cpufeatures: Add missing ID_AA64MMFR4_EL1 to __read_sysreg_by_encoding() arm64: sysreg: Update ID_AA64MMFR4_EL1 description Zeng Heng (2): arm64: cpufeature: Add TLBID (Domain-based TLB Invalidation) detection arm64: mm: Track CPUs that a task has run on for TLBID optimization arch/arm64/Kconfig | 20 ++ arch/arm64/include/asm/cpu.h | 2 + arch/arm64/include/asm/cpufeature.h | 13 ++ arch/arm64/include/asm/tlbflush.h | 66 +++++- arch/arm64/include/asm/tlbidomain.h | 34 +++ arch/arm64/kernel/cpufeature.c | 20 ++ arch/arm64/kernel/cpuinfo.c | 4 + arch/arm64/mm/Makefile | 1 + arch/arm64/mm/context.c | 9 +- arch/arm64/mm/tlbidomain.c | 339 ++++++++++++++++++++++++++++ arch/arm64/tools/cpucaps | 2 +- arch/arm64/tools/sysreg | 64 ++++++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/arm64/init.c | 2 + drivers/acpi/arm64/init.h | 1 + drivers/acpi/arm64/tlbid.c | 107 +++++++++ drivers/acpi/tables.c | 9 + include/acpi/actbl2.h | 29 +++ include/linux/acpi.h | 1 + 19 files changed, 718 insertions(+), 6 deletions(-) create mode 100644 arch/arm64/include/asm/tlbidomain.h create mode 100644 arch/arm64/mm/tlbidomain.c create mode 100644 drivers/acpi/arm64/tlbid.c -- 2.43.0

2 9

[PATCH OLK-6.6 0/2] CVE-2026-68396
by Zhou Minqiang 14 Sep '26

14 Sep '26

CVE-2026-68396 Zhou Minqiang (2): scsi: core: wake eh reliably when using scsi_schedule_eh scsi: core: Fix kabi breakage of struct Scsi_Host drivers/scsi/hosts.c | 13 +++++++++++++ drivers/scsi/scsi_error.c | 24 +++++++++++++++++++++++- drivers/scsi/scsi_priv.h | 1 + include/scsi/scsi_host.h | 7 ++++++- 4 files changed, 43 insertions(+), 2 deletions(-) -- 2.52.0

2 3

[PATCH OLK-6.6] scsi: core: pair EH runtime PM get and put
by Zhou Minqiang 14 Sep '26

14 Sep '26

mainline inclusion from mainline-v7.2 commit 872f486259ae0bc6b73ca4735a15d013241f73e9 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18399 CVE: CVE-2026-74754 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- [ Upstream commit 872f486259ae0bc6b73ca4735a15d013241f73e9 ] shost->eh_noresume is currently consulted twice in one error handling iteration: once before scsi_autopm_get_host() and once again before scsi_autopm_put_host(). That is racy when a PM-triggered error path flips shost->eh_noresume while the SCSI EH thread is still running. The problem flow looks like this: PM path ufshcd_set_dev_pwr_mode() shost->eh_noresume = 1 ufshcd_execute_start_stop <-- trigger EH ... shost->eh_noresume = 0 EH path scsi_error_handler() if (!shost->eh_noresume) scsi_autopm_get_host() <-- skipped ... if (!shost->eh_noresume) scsi_autopm_put_host() <-- executed later In that case one EH iteration can skip autoresume on entry and still drop a runtime PM reference on exit. That leaves an unmatched runtime PM put and can trigger a runtime PM usage count underflow. Fix this by making eh_noresume a regular bool so it can be accessed with READ_ONCE() and WRITE_ONCE(). Snapshot it once per EH iteration and use that snapshot for both runtime PM get and put decisions. Fixes: ae0751ffc77e ("[SCSI] add flag to skip the runtime PM calls on the host") Signed-off-by: Hongjie Fang <hongjiefang(a)asrmicro.com> Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Reviewed-by: Peter Wang <peter.wang(a)mediatek.com> Link: https://patch.msgid.link/20260729111614.2407559-1-hongjiefang@asrmicro.com Signed-off-by: Martin K. Petersen (Oracle) <mkp(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: drivers/scsi/scsi_error.c drivers/ufs/core/ufshcd.c [ctx conflicts] Signed-off-by: Zhou Minqiang <zhouminqiang2(a)huawei.com> --- drivers/scsi/scsi_error.c | 9 +++++++-- drivers/ufs/core/ufshcd.c | 4 ++++ 2 files changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 9b0ff9fad104..18474e441b87 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -2293,6 +2293,8 @@ static void scsi_unjam_host(struct Scsi_Host *shost) int scsi_error_handler(void *data) { struct Scsi_Host *shost = data; + bool eh_noresume; + unsigned long flags; /* * We use TASK_INTERRUPTIBLE so that the thread is not @@ -2334,7 +2336,10 @@ int scsi_error_handler(void *data) * what we need to do to get it up and online again (if we can). * If we fail, we end up taking the thing offline. */ - if (!shost->eh_noresume && scsi_autopm_get_host(shost) != 0) { + spin_lock_irqsave(shost->host_lock, flags); + eh_noresume = shost->eh_noresume; + spin_unlock_irqrestore(shost->host_lock, flags); + if (!eh_noresume && scsi_autopm_get_host(shost) != 0) { SCSI_LOG_ERROR_RECOVERY(1, shost_printk(KERN_ERR, shost, "scsi_eh_%d: unable to autoresume\n", @@ -2358,7 +2363,7 @@ int scsi_error_handler(void *data) * which are still online. */ scsi_restart_operations(shost); - if (!shost->eh_noresume) + if (!eh_noresume) scsi_autopm_put_host(shost); } __set_current_state(TASK_RUNNING); diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index 06f2626063a1..724e0225c932 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -9613,7 +9613,9 @@ static int ufshcd_set_dev_pwr_mode(struct ufs_hba *hba, * we are functional while we are here, skip host resume in error * handling context. */ + spin_lock_irqsave(hba->host->host_lock, flags); hba->host->eh_noresume = 1; + spin_unlock_irqrestore(hba->host->host_lock, flags); /* * Current function would be generally called from the power management @@ -9643,7 +9645,9 @@ static int ufshcd_set_dev_pwr_mode(struct ufs_hba *hba, } scsi_device_put(sdp); + spin_lock_irqsave(hba->host->host_lock, flags); hba->host->eh_noresume = 0; + spin_unlock_irqrestore(hba->host->host_lock, flags); return ret; } -- 2.52.0

2 1

[PATCH OLK-6.6 v2] md/raid1,raid10: fix bio accounting for split md cloned bios
by Zhou Minqiang 14 Sep '26

14 Sep '26

mainline inclusion from mainline-v7.2-rc1 commit ba976e3501111d11c550848b3b7341a73035f582 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17965 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Use md_cloned_bio() to control bio accounting instead of relying on r1bio_existed in raid1 or the io_accounting flag in raid10. The previous logic does not reliably reflect whether a bio is an md cloned bio. When a failed bio is split and resubmitted via bio_submit_split_bioset() on the error path, this can lead to either double accounting for md cloned bios, or missing accounting for bios returned from bio_submit_split_bioset() Fix this by using md_cloned_bio() to detect md cloned bios and skip accounting accordingly. Fixes: bb2a9acefaf9 ("md/raid1: switch to use md_account_bio() for io accounting") Fixes: 820455238366 ("md/raid10: switch to use md_account_bio() for io accounting") Signed-off-by: Abd-Alrhman Masalkhi <abd.masalkhi(a)gmail.com> Reviewed-by: Xiao Ni <xiao(a)kernel.org> Link: https://patch.msgid.link/20260501114652.590037-4-abd.masalkhi@gmail.com Signed-off-by: Yu Kuai <yukuai(a)fygo.io> Signed-off-by: Zhou Minqiang <zhouminqiang2(a)huawei.com> --- drivers/md/raid1.c | 2 +- drivers/md/raid10.c | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index a4662d5077ba..80c1802c26e1 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -1412,7 +1412,7 @@ static void raid1_read_request(struct mddev *mddev, struct bio *bio, } r1_bio->read_disk = rdisk; - if (!r1bio_existed) { + if (likely(!md_cloned_bio(mddev, bio))) { md_account_bio(mddev, &bio); r1_bio->master_bio = bio; } diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 316b6c96d413..e8ce8cdd7fce 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -1140,7 +1140,7 @@ static bool regular_request_wait(struct mddev *mddev, struct r10conf *conf, } static void raid10_read_request(struct mddev *mddev, struct bio *bio, - struct r10bio *r10_bio, bool io_accounting) + struct r10bio *r10_bio) { struct r10conf *conf = mddev->private; struct bio *read_bio; @@ -1218,7 +1218,7 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio, } slot = r10_bio->read_slot; - if (io_accounting) { + if (likely(!md_cloned_bio(mddev, bio))) { md_account_bio(mddev, &bio); r10_bio->master_bio = bio; } @@ -1514,7 +1514,7 @@ static bool __make_request(struct mddev *mddev, struct bio *bio, int sectors) ret = true; if (bio_data_dir(bio) == READ) - raid10_read_request(mddev, bio, r10_bio, true); + raid10_read_request(mddev, bio, r10_bio); else ret = raid10_write_request(mddev, bio, r10_bio); @@ -2839,7 +2839,7 @@ static void handle_read_error(struct mddev *mddev, struct r10bio *r10_bio) rdev_dec_pending(rdev, mddev); r10_bio->state = 0; - raid10_read_request(mddev, r10_bio->master_bio, r10_bio, false); + raid10_read_request(mddev, r10_bio->master_bio, r10_bio); /* * allow_barrier after re-submit to ensure no sync io * can be issued while regular io pending. -- 2.52.0

2 1

[PATCH OLK-5.10] ipv4: validate IPV4_DEVCONF attributes properly
by JiangJieHua 14 Sep '26

14 Sep '26

hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9803 -------------------------------- syzbot reported a UBSAN undefined behavior issue in vif_delete(): UBSAN: Undefined behaviour in net/ipv4/ipmr.c:720:43 signed integer overflow: -2147483648 - 1 cannot be represented in type 'int' MC_FORWARDING is a read-only sysctl (0444), but netlink RTM_SETLINK allows userspace to modify it via inet_validate_link_af() without permission checks, bypassing the read-only restriction. A corrupted counter can lead to inconsistent multicast routing state. Fix by rejecting IPV4_DEVCONF_MC_FORWARDING in inet_validate_link_af() with -EINVAL. Fixes: 9f0f7272ac95 ("ipv4: AF_INET link address family") Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- net/ipv4/devinet.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 9ac7d47d27b81..6822408e8f8d2 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1987,6 +1987,9 @@ static int inet_validate_link_af(const struct net_device *dev, if (cfgid <= 0 || cfgid > IPV4_DEVCONF_MAX) return -EINVAL; + + if (cfgid == IPV4_DEVCONF_MC_FORWARDING) + return -EINVAL; } } -- 2.33.8

2 1

[PATCH OLK-6.6] ipv4: validate IPV4_DEVCONF attributes properly
by JiangJieHua 14 Sep '26

14 Sep '26

hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9803 -------------------------------- syzbot reported a UBSAN undefined behavior issue in vif_delete(): UBSAN: Undefined behaviour in net/ipv4/ipmr.c:720:43 signed integer overflow: -2147483648 - 1 cannot be represented in type 'int' MC_FORWARDING is a read-only sysctl (0444), but netlink RTM_SETLINK allows userspace to modify it via inet_validate_link_af() without permission checks, bypassing the read-only restriction. A corrupted counter can lead to inconsistent multicast routing state. Fix by rejecting IPV4_DEVCONF_MC_FORWARDING in inet_validate_link_af() with -EINVAL. Fixes: 9f0f7272ac95 ("ipv4: AF_INET link address family") Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- net/ipv4/devinet.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 798497c8b1923..9b42ef2cc4ae2 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -2028,6 +2028,9 @@ static int inet_validate_link_af(const struct net_device *dev, if (cfgid <= 0 || cfgid > IPV4_DEVCONF_MAX) return -EINVAL; + + if (cfgid == IPV4_DEVCONF_MC_FORWARDING) + return -EINVAL; } } -- 2.33.8

2 1

[PATCH OLK-6.6] tracing/probes: Fix use-after-free on field name/type of events with multiple probes
by Tengda Wu 14 Sep '26

14 Sep '26

From: Henry Martin <bsdhenrymartin(a)gmail.com> mainline inclusion from mainline-v7.3-rc2 commit 86b7a239ec6b14a7544200ede85474c6f5526049 category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9956 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The fields of a probe-based dynamic event (kprobe, uprobe, eprobe and fprobe events) are created in traceprobe_define_arg_fields() by handing the probe_arg name/type strings to trace_define_field(), which only stores the pointers without copying. Those strings are owned by the trace_probe and are freed when that probe is removed. An event can have several probes attached. The field list is defined only once, by the first probe that registers the event, but it is kept alive by any surviving sibling probe. Deleting just that first probe by symbol - # primary A: fields are defined from A's args echo 'p:kprobes/ev vfs_read a1=$arg1' > kprobe_events # append B: shares A's event call echo 'p:kprobes/ev vfs_write a1=$arg1' >> kprobe_events # delete only A (matched by symbol), B survives echo '-:kprobes/ev vfs_read' >> kprobe_events frees A's args (trace_probe_cleanup() -> traceprobe_free_probe_arg()), but trace_probe_unlink() keeps the trace_probe_event because the probe list is not empty. The event call stays registered via B while its fields now reference freed memory. Any field lookup then reads it, e.g. echo 'a1 == 1' > events/kprobes/ev/filter BUG: KASAN: slab-use-after-free in strcmp+0xa7/0xb0 Call Trace: strcmp trace_find_event_field parse_pred process_preds create_filter apply_event_filter event_filter_write field->name references parg->name (kstrdup'd, freed with the probe) and, for array arguments, field->type references parg->fmt (kmalloc'd, freed with the probe) - the scalar type otherwise points at the static fmttype rodata, which is safe. Have traceprobe_define_arg_fields() duplicate the name and type strings and anchor the copies on the trace_probe_event, which embeds the event call and outlives every individual probe; trace_probe_event_free() releases them. The reproducer above triggers reliably; the field lookup and the delete both run under event_mutex, so this is a dangling reference after removal rather than a race. The issue was found by the autokbug dynamic kernel fuzzer at Tencent Yunding Lab. Link: https://lore.kernel.org/all/20260826030009.1855331-1-bsdhenrymartin@gmail.c… Fixes: ca89bc071d5e4 ("tracing/kprobe: Add multi-probe per event support") Signed-off-by: Henry Martin <bsdhenrymartin(a)gmail.com> Signed-off-by: Masami Hiramatsu (Google) <mhiramat(a)kernel.org> Signed-off-by: Tengda Wu <wutengda2(a)huawei.com> --- kernel/trace/trace_probe.c | 48 +++++++++++++++++++++++++++++++++++++- kernel/trace/trace_probe.h | 2 ++ 2 files changed, 49 insertions(+), 1 deletion(-) diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 066e154b91ee..1c05937d2ee0 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -1782,19 +1782,60 @@ int traceprobe_set_print_fmt(struct trace_probe *tp, enum probe_print_type ptype int traceprobe_define_arg_fields(struct trace_event_call *event_call, size_t offset, struct trace_probe *tp) { + struct trace_probe_event *tpe = trace_probe_event_from_call(event_call); int ret, i; + /* + * A field created by trace_define_field() only stores the name and + * type pointers, it does not copy the strings. Here they point into + * the probe_arg of @tp, which is freed when @tp is removed. For an + * event with multiple probes attached, the field list is defined + * once by the first probe but kept alive by the surviving siblings, + * so removing that first probe would leave the fields referencing + * freed memory. Duplicate the strings and anchor the copies on the + * trace_probe_event, which lives as long as the field list itself. + * + * event_define_fields() ignores the return value of this hook, so + * if a previous attempt failed before creating any field, it may + * call here again. Release duplicates left behind by such an + * attempt before starting over. + */ + for (i = 0; i < tpe->nr_field_strings; i++) + kfree(tpe->field_strings[i]); + kfree(tpe->field_strings); + tpe->field_strings = NULL; + tpe->nr_field_strings = 0; + + if (tp->nr_args) { + tpe->field_strings = kcalloc(tp->nr_args * 2, sizeof(char *), + GFP_KERNEL); + if (!tpe->field_strings) + return -ENOMEM; + } + /* Set argument names as fields */ for (i = 0; i < tp->nr_args; i++) { struct probe_arg *parg = &tp->args[i]; const char *fmt = parg->type->fmttype; int size = parg->type->size; + char *name, *type; if (parg->fmt) fmt = parg->fmt; if (parg->count) size *= parg->count; - ret = trace_define_field(event_call, fmt, parg->name, + + name = kstrdup(parg->name, GFP_KERNEL); + type = kstrdup(fmt, GFP_KERNEL); + if (!name || !type) { + kfree(name); + kfree(type); + return -ENOMEM; + } + tpe->field_strings[tpe->nr_field_strings++] = name; + tpe->field_strings[tpe->nr_field_strings++] = type; + + ret = trace_define_field(event_call, type, name, offset + parg->offset, size, parg->type->is_signed, FILTER_OTHER); @@ -1806,6 +1847,11 @@ int traceprobe_define_arg_fields(struct trace_event_call *event_call, static void trace_probe_event_free(struct trace_probe_event *tpe) { + int i; + + for (i = 0; i < tpe->nr_field_strings; i++) + kfree(tpe->field_strings[i]); + kfree(tpe->field_strings); kfree(tpe->class.system); kfree(tpe->call.name); kfree(tpe->call.print_fmt); diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index e8efa55b3a18..47982338bd42 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -246,6 +246,8 @@ struct trace_probe_event { struct trace_event_call call; struct list_head files; struct list_head probes; + char **field_strings; + int nr_field_strings; struct trace_uprobe_filter filter[]; }; -- 2.34.1

2 1