- Kernel - mailweb.openeuler.org

[PATCH OLK-6.6] irqchip/loongson-pch-pic: Update interrupt registration policy
by Hongchen Zhang 07 May '24

07 May '24

From: Baoqi Zhang <zhangbaoqi(a)loongson.cn> The current code is using a fixed mapping between the LS7A interrupt source and the HT interrupt vector. This prevents the utilization of the full interrupt vector space and therefore limits the number of interrupt source in a system. Replace the fixed mapping with a dynamic mapping which allocates a vector when an interrupt source is set up. This avoids that unused sources prevent vectors from being used for other devices. Introduce a mapping table in struct pch_pic, where each interrupt source will allocate an index as a 'hwirq' number from the table in the order of application and set table value as interrupt source number. This hwirq number will be configured as vector in the HT interrupt controller. For an interrupt source, the validity period of the obtained hwirq will last until the system reset. Co-developed-by: Biao Dong <dongbiao(a)loongson.cn> Signed-off-by: Biao Dong <dongbiao(a)loongson.cn> Co-developed-by: Tianyang Zhang <zhangtianyang(a)loongson.cn> Signed-off-by: Tianyang Zhang <zhangtianyang(a)loongson.cn> Signed-off-by: Baoqi Zhang <zhangbaoqi(a)loongson.cn> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Link: https://lore.kernel.org/r/20240422093830.27212-1-zhangtianyang@loongson.cn --- drivers/irqchip/irq-loongson-pch-pic.c | 70 ++++++++++++++++++++------ 1 file changed, 56 insertions(+), 14 deletions(-) diff --git a/drivers/irqchip/irq-loongson-pch-pic.c b/drivers/irqchip/irq-loongson-pch-pic.c index 3b150b6121fc..6b650020fff8 100644 --- a/drivers/irqchip/irq-loongson-pch-pic.c +++ b/drivers/irqchip/irq-loongson-pch-pic.c @@ -33,6 +33,7 @@ #define PIC_COUNT (PIC_COUNT_PER_REG * PIC_REG_COUNT) #define PIC_REG_IDX(irq_id) ((irq_id) / PIC_COUNT_PER_REG) #define PIC_REG_BIT(irq_id) ((irq_id) % PIC_COUNT_PER_REG) +#define PIC_UNDEF_VECTOR 255 #define PIC_COUNT_PER_REG64 64 #define PIC_REG64_COUNT 1 #define PIC_REG64_IDX(irq_id) ((irq_id) / PIC_COUNT_PER_REG64) @@ -50,6 +51,8 @@ struct pch_pic { u32 saved_vec_en[PIC_REG_COUNT]; u32 saved_vec_pol[PIC_REG_COUNT]; u32 saved_vec_edge[PIC_REG_COUNT]; + u8 table[PIC_COUNT]; + int inuse; }; static struct pch_pic *pch_pic_priv[MAX_IO_PICS]; @@ -61,6 +64,11 @@ struct irq_domain *get_pchpic_irq_domain(void) return pch_pic_priv[0]->pic_domain; } +static inline u8 hwirq_to_bit(struct pch_pic *priv, int hirq) +{ + return priv->table[hirq]; +} + static void pch_pic_bitset(struct pch_pic *priv, int offset, int bit) { u32 reg; @@ -89,45 +97,47 @@ static void pch_pic_mask_irq(struct irq_data *d) { struct pch_pic *priv = irq_data_get_irq_chip_data(d); - pch_pic_bitset(priv, PCH_PIC_MASK, d->hwirq); + pch_pic_bitset(priv, PCH_PIC_MASK, hwirq_to_bit(priv, d->hwirq)); irq_chip_mask_parent(d); } static void pch_pic_unmask_irq(struct irq_data *d) { struct pch_pic *priv = irq_data_get_irq_chip_data(d); + int bit = hwirq_to_bit(priv, d->hwirq); writeq(BIT(PIC_REG64_BIT(d->hwirq)), - priv->base + PCH_PIC_CLR + PIC_REG64_IDX(d->hwirq) * 8); + priv->base + PCH_PIC_CLR + PIC_REG64_IDX(bit) * 8); irq_chip_unmask_parent(d); - pch_pic_bitclr(priv, PCH_PIC_MASK, d->hwirq); + pch_pic_bitclr(priv, PCH_PIC_MASK, bit); } static int pch_pic_set_type(struct irq_data *d, unsigned int type) { struct pch_pic *priv = irq_data_get_irq_chip_data(d); + int bit = hwirq_to_bit(priv, d->hwirq); int ret = 0; switch (type) { case IRQ_TYPE_EDGE_RISING: - pch_pic_bitset(priv, PCH_PIC_EDGE, d->hwirq); - pch_pic_bitclr(priv, PCH_PIC_POL, d->hwirq); + pch_pic_bitset(priv, PCH_PIC_EDGE, bit); + pch_pic_bitclr(priv, PCH_PIC_POL, bit); irq_set_handler_locked(d, handle_edge_irq); break; case IRQ_TYPE_EDGE_FALLING: - pch_pic_bitset(priv, PCH_PIC_EDGE, d->hwirq); - pch_pic_bitset(priv, PCH_PIC_POL, d->hwirq); + pch_pic_bitset(priv, PCH_PIC_EDGE, bit); + pch_pic_bitset(priv, PCH_PIC_POL, bit); irq_set_handler_locked(d, handle_edge_irq); break; case IRQ_TYPE_LEVEL_HIGH: - pch_pic_bitclr(priv, PCH_PIC_EDGE, d->hwirq); - pch_pic_bitclr(priv, PCH_PIC_POL, d->hwirq); + pch_pic_bitclr(priv, PCH_PIC_EDGE, bit); + pch_pic_bitclr(priv, PCH_PIC_POL, bit); irq_set_handler_locked(d, handle_level_irq); break; case IRQ_TYPE_LEVEL_LOW: - pch_pic_bitclr(priv, PCH_PIC_EDGE, d->hwirq); - pch_pic_bitset(priv, PCH_PIC_POL, d->hwirq); + pch_pic_bitclr(priv, PCH_PIC_EDGE, bit); + pch_pic_bitset(priv, PCH_PIC_POL, bit); irq_set_handler_locked(d, handle_level_irq); break; default: @@ -142,11 +152,12 @@ static void pch_pic_ack_irq(struct irq_data *d) { unsigned int reg; struct pch_pic *priv = irq_data_get_irq_chip_data(d); + int bit = hwirq_to_bit(priv, d->hwirq); - reg = readl(priv->base + PCH_PIC_EDGE + PIC_REG_IDX(d->hwirq) * 4); + reg = readl(priv->base + PCH_PIC_EDGE + PIC_REG_IDX(bit) * 4); if (reg & BIT(PIC_REG_BIT(d->hwirq))) { writeq(BIT(PIC_REG64_BIT(d->hwirq)), - priv->base + PCH_PIC_CLR + PIC_REG64_IDX(d->hwirq) * 8); + priv->base + PCH_PIC_CLR + PIC_REG64_IDX(bit) * 8); } irq_chip_ack_parent(d); } @@ -168,6 +179,8 @@ static int pch_pic_domain_translate(struct irq_domain *d, { struct pch_pic *priv = d->host_data; struct device_node *of_node = to_of_node(fwspec->fwnode); + unsigned long flags; + int i; if (of_node) { if (fwspec->param_count < 2) @@ -180,12 +193,33 @@ static int pch_pic_domain_translate(struct irq_domain *d, return -EINVAL; *hwirq = fwspec->param[0] - priv->gsi_base; + if (fwspec->param_count > 1) *type = fwspec->param[1] & IRQ_TYPE_SENSE_MASK; else *type = IRQ_TYPE_NONE; } + raw_spin_lock_irqsave(&priv->pic_lock, flags); + /* Check pic-table to confirm if the hwirq has been assigned */ + for (i = 0; i < priv->inuse; i++) { + if (priv->table[i] == *hwirq) { + *hwirq = i; + break; + } + } + if (i == priv->inuse) { + /* Assign a new hwirq in pic-table */ + if (priv->inuse >= PIC_COUNT) { + pr_err("pch-pic domain has no free vectors\n"); + raw_spin_unlock_irqrestore(&priv->pic_lock, flags); + return -EINVAL; + } + priv->table[priv->inuse] = *hwirq; + *hwirq = priv->inuse++; + } + raw_spin_unlock_irqrestore(&priv->pic_lock, flags); + return 0; } @@ -203,6 +237,9 @@ static int pch_pic_alloc(struct irq_domain *domain, unsigned int virq, if (err) return err; + /* Write vector ID */ + writeb(priv->ht_vec_base + hwirq, priv->base + PCH_INT_HTVEC(hwirq_to_bit(priv, hwirq))); + parent_fwspec.fwnode = domain->parent->fwnode; parent_fwspec.param_count = 1; parent_fwspec.param[0] = hwirq + priv->ht_vec_base; @@ -231,7 +268,7 @@ static void pch_pic_reset(struct pch_pic *priv) for (i = 0; i < PIC_COUNT; i++) { /* Write vector ID */ - writeb(priv->ht_vec_base + i, priv->base + PCH_INT_HTVEC(i)); + writeb(priv->ht_vec_base + i, priv->base + PCH_INT_HTVEC(hwirq_to_bit(priv, i))); /* Hardcode route to HT0 Lo */ writeb(1, priv->base + PCH_INT_ROUTE(i)); } @@ -295,6 +332,7 @@ static int pch_pic_init(phys_addr_t addr, unsigned long size, int vec_base, u32 gsi_base) { struct pch_pic *priv; + int i; priv = kzalloc(sizeof(*priv), GFP_KERNEL); if (!priv) @@ -305,6 +343,10 @@ static int pch_pic_init(phys_addr_t addr, unsigned long size, int vec_base, if (!priv->base) goto free_priv; + priv->inuse = 0; + for (i = 0; i < PIC_COUNT; i++) + priv->table[i] = PIC_UNDEF_VECTOR; + priv->ht_vec_base = vec_base; priv->vec_count = ((readq(priv->base) >> 48) & 0xff) + 1; priv->gsi_base = gsi_base; -- 2.33.0

2 1

[PATCH openEuler-1.0-LTS] media: dvb-frontends: avoid stack overflow warnings with clang
by Cai Xinchen 07 May '24

07 May '24

From: Arnd Bergmann <arnd(a)arndb.de> stable inclusion from stable-v4.19.311 commit c073c8cede5abd3836e83d70d72606d11d0759d4 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L9J6 CVE: CVE-2024-27075 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 7a4cf27d1f0538f779bf31b8c99eda394e277119 ] A previous patch worked around a KASAN issue in stv0367, now a similar problem showed up with clang: drivers/media/dvb-frontends/stv0367.c:1222:12: error: stack frame size (3624) exceeds limit (2048) in 'stv0367ter_set_frontend' [-Werror,-Wframe-larger-than] 1214 | static int stv0367ter_set_frontend(struct dvb_frontend *fe) Rework the stv0367_writereg() function to be simpler and mark both register access functions as noinline_for_stack so the temporary i2c_msg structures do not get duplicated on the stack when KASAN_STACK is enabled. Fixes: 3cd890dbe2a4 ("media: dvb-frontends: fix i2c access helpers for KASAN") Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> Reviewed-by: Justin Stitt <justinstitt(a)google.com> Signed-off-by: Hans Verkuil <hverkuil-cisco(a)xs4all.nl> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Cai Xinchen <caixinchen1(a)huawei.com> --- drivers/media/dvb-frontends/stv0367.c | 34 +++++++-------------------- 1 file changed, 8 insertions(+), 26 deletions(-) diff --git a/drivers/media/dvb-frontends/stv0367.c b/drivers/media/dvb-frontends/stv0367.c index 5b91e740e135..a6f2924c137f 100644 --- a/drivers/media/dvb-frontends/stv0367.c +++ b/drivers/media/dvb-frontends/stv0367.c @@ -128,50 +128,32 @@ static const s32 stv0367cab_RF_LookUp2[RF_LOOKUP_TABLE2_SIZE][RF_LOOKUP_TABLE2_S } }; -static -int stv0367_writeregs(struct stv0367_state *state, u16 reg, u8 *data, int len) +static noinline_for_stack +int stv0367_writereg(struct stv0367_state *state, u16 reg, u8 data) { - u8 buf[MAX_XFER_SIZE]; + u8 buf[3] = { MSB(reg), LSB(reg), data }; struct i2c_msg msg = { .addr = state->config->demod_address, .flags = 0, .buf = buf, - .len = len + 2 + .len = 3, }; int ret; - if (2 + len > sizeof(buf)) { - printk(KERN_WARNING - "%s: i2c wr reg=%04x: len=%d is too big!\n", - KBUILD_MODNAME, reg, len); - return -EINVAL; - } - - - buf[0] = MSB(reg); - buf[1] = LSB(reg); - memcpy(buf + 2, data, len); - if (i2cdebug) printk(KERN_DEBUG "%s: [%02x] %02x: %02x\n", __func__, - state->config->demod_address, reg, buf[2]); + state->config->demod_address, reg, data); ret = i2c_transfer(state->i2c, &msg, 1); if (ret != 1) printk(KERN_ERR "%s: i2c write error! ([%02x] %02x: %02x)\n", - __func__, state->config->demod_address, reg, buf[2]); + __func__, state->config->demod_address, reg, data); return (ret != 1) ? -EREMOTEIO : 0; } -static int stv0367_writereg(struct stv0367_state *state, u16 reg, u8 data) -{ - u8 tmp = data; /* see gcc.gnu.org/bugzilla/show_bug.cgi?id=81715 */ - - return stv0367_writeregs(state, reg, &tmp, 1); -} - -static u8 stv0367_readreg(struct stv0367_state *state, u16 reg) +static noinline_for_stack +u8 stv0367_readreg(struct stv0367_state *state, u16 reg) { u8 b0[] = { 0, 0 }; u8 b1[] = { 0 }; -- 2.34.1

2 1

[PATCH OLK-5.10] KVM: Always flush async #PF workqueue when vCPU is being destroyed
by Wei Li 07 May '24

07 May '24

From: Sean Christopherson <seanjc(a)google.com> mainline inclusion from mainline-v6.9-rc1 commit 3d75b8aa5c29058a512db29da7cbee8052724157 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L4ZX CVE: CVE-2024-26976 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit?i… -------------------------------- Always flush the per-vCPU async #PF workqueue when a vCPU is clearing its completion queue, e.g. when a VM and all its vCPUs is being destroyed. KVM must ensure that none of its workqueue callbacks is running when the last reference to the KVM _module_ is put. Gifting a reference to the associated VM prevents the workqueue callback from dereferencing freed vCPU/VM memory, but does not prevent the KVM module from being unloaded before the callback completes. Drop the misguided VM refcount gifting, as calling kvm_put_kvm() from async_pf_execute() if kvm_put_kvm() flushes the async #PF workqueue will result in deadlock. async_pf_execute() can't return until kvm_put_kvm() finishes, and kvm_put_kvm() can't return until async_pf_execute() finishes: WARNING: CPU: 8 PID: 251 at virt/kvm/kvm_main.c:1435 kvm_put_kvm+0x2d/0x320 [kvm] Modules linked in: vhost_net vhost vhost_iotlb tap kvm_intel kvm irqbypass CPU: 8 PID: 251 Comm: kworker/8:1 Tainted: G W 6.6.0-rc1-e7af8d17224a-x86/gmem-vm #119 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015 Workqueue: events async_pf_execute [kvm] RIP: 0010:kvm_put_kvm+0x2d/0x320 [kvm] Call Trace: <TASK> async_pf_execute+0x198/0x260 [kvm] process_one_work+0x145/0x2d0 worker_thread+0x27e/0x3a0 kthread+0xba/0xe0 ret_from_fork+0x2d/0x50 ret_from_fork_asm+0x11/0x20 </TASK> ---[ end trace 0000000000000000 ]--- INFO: task kworker/8:1:251 blocked for more than 120 seconds. Tainted: G W 6.6.0-rc1-e7af8d17224a-x86/gmem-vm #119 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:kworker/8:1 state:D stack:0 pid:251 ppid:2 flags:0x00004000 Workqueue: events async_pf_execute [kvm] Call Trace: <TASK> __schedule+0x33f/0xa40 schedule+0x53/0xc0 schedule_timeout+0x12a/0x140 __wait_for_common+0x8d/0x1d0 __flush_work.isra.0+0x19f/0x2c0 kvm_clear_async_pf_completion_queue+0x129/0x190 [kvm] kvm_arch_destroy_vm+0x78/0x1b0 [kvm] kvm_put_kvm+0x1c1/0x320 [kvm] async_pf_execute+0x198/0x260 [kvm] process_one_work+0x145/0x2d0 worker_thread+0x27e/0x3a0 kthread+0xba/0xe0 ret_from_fork+0x2d/0x50 ret_from_fork_asm+0x11/0x20 </TASK> If kvm_clear_async_pf_completion_queue() actually flushes the workqueue, then there's no need to gift async_pf_execute() a reference because all invocations of async_pf_execute() will be forced to complete before the vCPU and its VM are destroyed/freed. And that in turn fixes the module unloading bug as __fput() won't do module_put() on the last vCPU reference until the vCPU has been freed, e.g. if closing the vCPU file also puts the last reference to the KVM module. Note that kvm_check_async_pf_completion() may also take the work item off the completion queue and so also needs to flush the work queue, as the work will not be seen by kvm_clear_async_pf_completion_queue(). Waiting on the workqueue could theoretically delay a vCPU due to waiting for the work to complete, but that's a very, very small chance, and likely a very small delay. kvm_arch_async_page_present_queued() unconditionally makes a new request, i.e. will effectively delay entering the guest, so the remaining work is really just: trace_kvm_async_pf_completed(addr, cr2_or_gpa); __kvm_vcpu_wake_up(vcpu); mmput(mm); and mmput() can't drop the last reference to the page tables if the vCPU is still alive, i.e. the vCPU won't get stuck tearing down page tables. Add a helper to do the flushing, specifically to deal with "wakeup all" work items, as they aren't actually work items, i.e. are never placed in a workqueue. Trying to flush a bogus workqueue entry rightly makes __flush_work() complain (kudos to whoever added that sanity check). Note, commit 5f6de5cbebee ("KVM: Prevent module exit until all VMs are freed") *tried* to fix the module refcounting issue by having VMs grab a reference to the module, but that only made the bug slightly harder to hit as it gave async_pf_execute() a bit more time to complete before the KVM module could be unloaded. Fixes: af585b921e5d ("KVM: Halt vcpu if page it tries to access is swapped out") Cc: stable(a)vger.kernel.org Cc: David Matlack <dmatlack(a)google.com> Reviewed-by: Xu Yilun <yilun.xu(a)intel.com> Reviewed-by: Vitaly Kuznetsov <vkuznets(a)redhat.com> Link: https://lore.kernel.org/r/20240110011533.503302-2-seanjc@google.com Signed-off-by: Sean Christopherson <seanjc(a)google.com> Conflicts: virt/kvm/async_pf.c [lw: fix context conflict only] Signed-off-by: Wei Li <liwei391(a)huawei.com> --- virt/kvm/async_pf.c | 31 ++++++++++++++++++++++++++----- 1 file changed, 26 insertions(+), 5 deletions(-) diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c index dd777688d14a..952afb1bc83b 100644 --- a/virt/kvm/async_pf.c +++ b/virt/kvm/async_pf.c @@ -88,7 +88,27 @@ static void async_pf_execute(struct work_struct *work) rcuwait_wake_up(&vcpu->wait); mmput(mm); - kvm_put_kvm(vcpu->kvm); +} + +static void kvm_flush_and_free_async_pf_work(struct kvm_async_pf *work) +{ + /* + * The async #PF is "done", but KVM must wait for the work item itself, + * i.e. async_pf_execute(), to run to completion. If KVM is a module, + * KVM must ensure *no* code owned by the KVM (the module) can be run + * after the last call to module_put(). Note, flushing the work item + * is always required when the item is taken off the completion queue. + * E.g. even if the vCPU handles the item in the "normal" path, the VM + * could be terminated before async_pf_execute() completes. + * + * Wake all events skip the queue and go straight done, i.e. don't + * need to be flushed (but sanity check that the work wasn't queued). + */ + if (work->wakeup_all) + WARN_ON_ONCE(work->work.func); + else + flush_work(&work->work); + kmem_cache_free(async_pf_cache, work); } void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu) @@ -115,7 +135,6 @@ void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu) #else if (cancel_work_sync(&work->work)) { mmput(work->mm); - kvm_put_kvm(vcpu->kvm); /* == work->vcpu->kvm */ kmem_cache_free(async_pf_cache, work); } #endif @@ -127,7 +146,10 @@ void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu) list_first_entry(&vcpu->async_pf.done, typeof(*work), link); list_del(&work->link); - kmem_cache_free(async_pf_cache, work); + + spin_unlock(&vcpu->async_pf.lock); + kvm_flush_and_free_async_pf_work(work); + spin_lock(&vcpu->async_pf.lock); } spin_unlock(&vcpu->async_pf.lock); @@ -152,7 +174,7 @@ void kvm_check_async_pf_completion(struct kvm_vcpu *vcpu) list_del(&work->queue); vcpu->async_pf.queued--; - kmem_cache_free(async_pf_cache, work); + kvm_flush_and_free_async_pf_work(work); } } @@ -187,7 +209,6 @@ bool kvm_setup_async_pf(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, work->arch = *arch; work->mm = current->mm; mmget(work->mm); - kvm_get_kvm(work->vcpu->kvm); INIT_WORK(&work->work, async_pf_execute); -- 2.25.1

2 1

[PATCH OLK-6.6] scsi: qla2xxx: Fix double free of the ha->vp_map pointer
by Li Lingfeng 07 May '24

07 May '24

From: Saurav Kashyap <skashyap(a)marvell.com> mainline inclusion from mainline-v6.9-rc2 commit e288285d47784fdcf7c81be56df7d65c6f10c58b category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L4OZ CVE: CVE-2024-26930 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Coverity scan reported potential risk of double free of the pointer ha->vp_map. ha->vp_map was freed in qla2x00_mem_alloc(), and again freed in function qla2x00_mem_free(ha). Assign NULL to vp_map and kfree take care of NULL. Cc: stable(a)vger.kernel.org Signed-off-by: Saurav Kashyap <skashyap(a)marvell.com> Signed-off-by: Nilesh Javali <njavali(a)marvell.com> Link: https://lore.kernel.org/r/20240227164127.36465-8-njavali@marvell.com Reviewed-by: Himanshu Madhani <himanshu.madhani(a)oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Li Lingfeng <lilingfeng3(a)huawei.com> --- drivers/scsi/qla2xxx/qla_os.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index c45eef743c45..53b18f8c96ca 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -4601,6 +4601,7 @@ qla2x00_mem_alloc(struct qla_hw_data *ha, uint16_t req_len, uint16_t rsp_len, ha->init_cb_dma = 0; fail_free_vp_map: kfree(ha->vp_map); + ha->vp_map = NULL; fail: ql_log(ql_log_fatal, NULL, 0x0030, "Memory allocation failure.\n"); -- 2.31.1

2 1

[PATCH openEuler-1.0-LTS 0/2] Fix CVE-2022-48693
by Li Huafei 07 May '24

07 May '24

Liang He (1): soc: brcmstb: pm-arm: Fix refcount leak and __iomem leak bugs Miaoqian Lin (1): soc: bcm: brcmstb: pm: pm-arm: Fix refcount leak in brcmstb_pm_probe drivers/soc/bcm/brcmstb/pm/pm-arm.c | 51 ++++++++++++++++++++++------- 1 file changed, 40 insertions(+), 11 deletions(-) -- 2.25.1

2 3

[PATCH openEuler-1.0-LTS] serial/pmac_zilog: Remove flawed mitigation for rx irq flood
by Yi Yang 07 May '24

07 May '24

From: Finn Thain <fthain(a)linux-m68k.org> stable inclusion from stable-v4.19.313 commit 69a02273e288011b521ee7c1f3ab2c23fda633ce category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5KR CVE: CVE-2024-26999 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 1be3226445362bfbf461c92a5bcdb1723f2e4907 upstream. The mitigation was intended to stop the irq completely. That may be better than a hard lock-up but it turns out that you get a crash anyway if you're using pmac_zilog as a serial console: ttyPZ0: pmz: rx irq flood ! BUG: spinlock recursion on CPU#0, swapper/0 That's because the pr_err() call in pmz_receive_chars() results in pmz_console_write() attempting to lock a spinlock already locked in pmz_interrupt(). With CONFIG_DEBUG_SPINLOCK=y, this produces a fatal BUG splat. The spinlock in question is the one in struct uart_port. Even when it's not fatal, the serial port rx function ceases to work. Also, the iteration limit doesn't play nicely with QEMU, as can be seen in the bug report linked below. A web search for other reports of the error message "pmz: rx irq flood" didn't produce anything. So I don't think this code is needed any more. Remove it. Cc: Benjamin Herrenschmidt <benh(a)kernel.crashing.org> Cc: Michael Ellerman <mpe(a)ellerman.id.au> Cc: Nicholas Piggin <npiggin(a)gmail.com> Cc: Christophe Leroy <christophe.leroy(a)csgroup.eu> Cc: Aneesh Kumar K.V <aneesh.kumar(a)kernel.org> Cc: Naveen N. Rao <naveen.n.rao(a)linux.ibm.com> Cc: Andy Shevchenko <andy.shevchenko(a)gmail.com> Cc: stable(a)kernel.org Cc: linux-m68k(a)lists.linux-m68k.org Link: https://github.com/vivier/qemu-m68k/issues/44 Link: https://lore.kernel.org/all/1078874617.9746.36.camel@gaston/ Acked-by: Michael Ellerman <mpe(a)ellerman.id.au> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable <stable(a)kernel.org> Signed-off-by: Finn Thain <fthain(a)linux-m68k.org> Link: https://lore.kernel.org/r/e853cf2c762f23101cd2ddec0cc0c2be0e72685f.17125682… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/tty/serial/pmac_zilog.c | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/drivers/tty/serial/pmac_zilog.c b/drivers/tty/serial/pmac_zilog.c index 3d21790d961e..2cddcf74f702 100644 --- a/drivers/tty/serial/pmac_zilog.c +++ b/drivers/tty/serial/pmac_zilog.c @@ -220,7 +220,6 @@ static bool pmz_receive_chars(struct uart_pmac_port *uap) { struct tty_port *port; unsigned char ch, r1, drop, error, flag; - int loops = 0; /* Sanity check, make sure the old bug is no longer happening */ if (uap->port.state == NULL) { @@ -303,24 +302,11 @@ static bool pmz_receive_chars(struct uart_pmac_port *uap) if (r1 & Rx_OVR) tty_insert_flip_char(port, 0, TTY_OVERRUN); next_char: - /* We can get stuck in an infinite loop getting char 0 when the - * line is in a wrong HW state, we break that here. - * When that happens, I disable the receive side of the driver. - * Note that what I've been experiencing is a real irq loop where - * I'm getting flooded regardless of the actual port speed. - * Something strange is going on with the HW - */ - if ((++loops) > 1000) - goto flood; ch = read_zsreg(uap, R0); if (!(ch & Rx_CH_AV)) break; } - return true; - flood: - pmz_interrupt_control(uap, 0); - pmz_error("pmz: rx irq flood !\n"); return true; } -- 2.25.1

2 1

[PATCH OLK-5.10] serial/pmac_zilog: Remove flawed mitigation for rx irq flood
by Yi Yang 07 May '24

07 May '24

From: Finn Thain <fthain(a)linux-m68k.org> stable inclusion from stable-v5.10.216 commit ab86cf6f8d24e63e9aca23da5108af1aa5483928 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5KR CVE: CVE-2024-26999 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 1be3226445362bfbf461c92a5bcdb1723f2e4907 upstream. The mitigation was intended to stop the irq completely. That may be better than a hard lock-up but it turns out that you get a crash anyway if you're using pmac_zilog as a serial console: ttyPZ0: pmz: rx irq flood ! BUG: spinlock recursion on CPU#0, swapper/0 That's because the pr_err() call in pmz_receive_chars() results in pmz_console_write() attempting to lock a spinlock already locked in pmz_interrupt(). With CONFIG_DEBUG_SPINLOCK=y, this produces a fatal BUG splat. The spinlock in question is the one in struct uart_port. Even when it's not fatal, the serial port rx function ceases to work. Also, the iteration limit doesn't play nicely with QEMU, as can be seen in the bug report linked below. A web search for other reports of the error message "pmz: rx irq flood" didn't produce anything. So I don't think this code is needed any more. Remove it. Cc: Benjamin Herrenschmidt <benh(a)kernel.crashing.org> Cc: Michael Ellerman <mpe(a)ellerman.id.au> Cc: Nicholas Piggin <npiggin(a)gmail.com> Cc: Christophe Leroy <christophe.leroy(a)csgroup.eu> Cc: Aneesh Kumar K.V <aneesh.kumar(a)kernel.org> Cc: Naveen N. Rao <naveen.n.rao(a)linux.ibm.com> Cc: Andy Shevchenko <andy.shevchenko(a)gmail.com> Cc: stable(a)kernel.org Cc: linux-m68k(a)lists.linux-m68k.org Link: https://github.com/vivier/qemu-m68k/issues/44 Link: https://lore.kernel.org/all/1078874617.9746.36.camel@gaston/ Acked-by: Michael Ellerman <mpe(a)ellerman.id.au> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable <stable(a)kernel.org> Signed-off-by: Finn Thain <fthain(a)linux-m68k.org> Link: https://lore.kernel.org/r/e853cf2c762f23101cd2ddec0cc0c2be0e72685f.17125682… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/tty/serial/pmac_zilog.c | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/drivers/tty/serial/pmac_zilog.c b/drivers/tty/serial/pmac_zilog.c index d6aef8a1f0a4..1d0717fc3729 100644 --- a/drivers/tty/serial/pmac_zilog.c +++ b/drivers/tty/serial/pmac_zilog.c @@ -217,7 +217,6 @@ static bool pmz_receive_chars(struct uart_pmac_port *uap) { struct tty_port *port; unsigned char ch, r1, drop, flag; - int loops = 0; /* Sanity check, make sure the old bug is no longer happening */ if (uap->port.state == NULL) { @@ -298,24 +297,11 @@ static bool pmz_receive_chars(struct uart_pmac_port *uap) if (r1 & Rx_OVR) tty_insert_flip_char(port, 0, TTY_OVERRUN); next_char: - /* We can get stuck in an infinite loop getting char 0 when the - * line is in a wrong HW state, we break that here. - * When that happens, I disable the receive side of the driver. - * Note that what I've been experiencing is a real irq loop where - * I'm getting flooded regardless of the actual port speed. - * Something strange is going on with the HW - */ - if ((++loops) > 1000) - goto flood; ch = read_zsreg(uap, R0); if (!(ch & Rx_CH_AV)) break; } - return true; - flood: - pmz_interrupt_control(uap, 0); - pmz_error("pmz: rx irq flood !\n"); return true; } -- 2.25.1

2 1

[PATCH OLK-6.6] f2fs: compress: fix to guarantee persisting compressed blocks by CP
by Zizhi Wo 07 May '24

07 May '24

From: Chao Yu <chao(a)kernel.org> stable inclusion from stable-v6.6.23 commit 82704e598d7b33c7e45526e34d3c585426319bed category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L9OE CVE: CVE-2024-27035 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 8a430dd49e9cb021372b0ad91e60aeef9c6ced00 ] If data block in compressed cluster is not persisted with metadata during checkpoint, after SPOR, the data may be corrupted, let's guarantee to write compressed page by checkpoint. Fixes: 4c8ff7095bef ("f2fs: support data compression") Reviewed-by: Daeho Jeong <daehojeong(a)google.com> Signed-off-by: Chao Yu <chao(a)kernel.org> Signed-off-by: Jaegeuk Kim <jaegeuk(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- fs/f2fs/compress.c | 4 +++- fs/f2fs/data.c | 17 +++++++++-------- fs/f2fs/f2fs.h | 4 +++- 3 files changed, 15 insertions(+), 10 deletions(-) diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index 372616ca8fb5..dfc59a658836 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -1413,6 +1413,8 @@ void f2fs_compress_write_end_io(struct bio *bio, struct page *page) struct f2fs_sb_info *sbi = bio->bi_private; struct compress_io_ctx *cic = (struct compress_io_ctx *)page_private(page); + enum count_type type = WB_DATA_TYPE(page, + f2fs_is_compressed_page(page)); int i; if (unlikely(bio->bi_status)) @@ -1420,7 +1422,7 @@ void f2fs_compress_write_end_io(struct bio *bio, struct page *page) f2fs_compress_free_page(page); - dec_page_count(sbi, F2FS_WB_DATA); + dec_page_count(sbi, type); if (atomic_dec_return(&cic->pending_pages)) return; diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 8dc8e3027a65..2f67922b43f5 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -48,7 +48,7 @@ void f2fs_destroy_bioset(void) bioset_exit(&f2fs_bioset); } -static bool __is_cp_guaranteed(struct page *page) +bool f2fs_is_cp_guaranteed(struct page *page) { struct address_space *mapping = page->mapping; struct inode *inode; @@ -65,8 +65,6 @@ static bool __is_cp_guaranteed(struct page *page) S_ISDIR(inode->i_mode)) return true; - if (f2fs_is_compressed_page(page)) - return false; if ((S_ISREG(inode->i_mode) && IS_NOQUOTA(inode)) || page_private_gcing(page)) return true; @@ -338,7 +336,7 @@ static void f2fs_write_end_io(struct bio *bio) bio_for_each_segment_all(bvec, bio, iter_all) { struct page *page = bvec->bv_page; - enum count_type type = WB_DATA_TYPE(page); + enum count_type type = WB_DATA_TYPE(page, false); if (page_private_dummy(page)) { clear_page_private_dummy(page); @@ -762,7 +760,7 @@ int f2fs_submit_page_bio(struct f2fs_io_info *fio) wbc_account_cgroup_owner(fio->io_wbc, fio->page, PAGE_SIZE); inc_page_count(fio->sbi, is_read_io(fio->op) ? - __read_io_type(page) : WB_DATA_TYPE(fio->page)); + __read_io_type(page) : WB_DATA_TYPE(fio->page, false)); if (is_read_io(bio_op(bio))) f2fs_submit_read_bio(fio->sbi, bio, fio->type); @@ -973,7 +971,7 @@ int f2fs_merge_page_bio(struct f2fs_io_info *fio) if (fio->io_wbc) wbc_account_cgroup_owner(fio->io_wbc, fio->page, PAGE_SIZE); - inc_page_count(fio->sbi, WB_DATA_TYPE(page)); + inc_page_count(fio->sbi, WB_DATA_TYPE(page, false)); *fio->last_block = fio->new_blkaddr; *fio->bio = bio; @@ -1007,6 +1005,7 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio) enum page_type btype = PAGE_TYPE_OF_BIO(fio->type); struct f2fs_bio_info *io = sbi->write_io[btype] + fio->temp; struct page *bio_page; + enum count_type type; f2fs_bug_on(sbi, is_read_io(fio->op)); @@ -1046,7 +1045,8 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio) /* set submitted = true as a return value */ fio->submitted = 1; - inc_page_count(sbi, WB_DATA_TYPE(bio_page)); + type = WB_DATA_TYPE(bio_page, fio->compressed_page); + inc_page_count(sbi, type); if (io->bio && (!io_is_mergeable(sbi, io->bio, io, fio, io->last_block_in_bio, @@ -1059,7 +1059,8 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio) if (F2FS_IO_ALIGNED(sbi) && (fio->type == DATA || fio->type == NODE) && fio->new_blkaddr & F2FS_IO_SIZE_MASK(sbi)) { - dec_page_count(sbi, WB_DATA_TYPE(bio_page)); + dec_page_count(sbi, WB_DATA_TYPE(bio_page, + fio->compressed_page)); fio->retry = 1; goto skip; } diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index dc79bc8a9845..efd8cdfb38a1 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1075,7 +1075,8 @@ struct f2fs_sm_info { * f2fs monitors the number of several block types such as on-writeback, * dirty dentry blocks, dirty node blocks, and dirty meta blocks. */ -#define WB_DATA_TYPE(p) (__is_cp_guaranteed(p) ? F2FS_WB_CP_DATA : F2FS_WB_DATA) +#define WB_DATA_TYPE(p, f) \ + (f || f2fs_is_cp_guaranteed(p) ? F2FS_WB_CP_DATA : F2FS_WB_DATA) enum count_type { F2FS_DIRTY_DENTS, F2FS_DIRTY_DATA, @@ -3795,6 +3796,7 @@ void f2fs_init_ckpt_req_control(struct f2fs_sb_info *sbi); */ int __init f2fs_init_bioset(void); void f2fs_destroy_bioset(void); +bool f2fs_is_cp_guaranteed(struct page *page); int f2fs_init_bio_entry_cache(void); void f2fs_destroy_bio_entry_cache(void); void f2fs_submit_read_bio(struct f2fs_sb_info *sbi, struct bio *bio, -- 2.39.2

2 1

[PATCH] serial/pmac_zilog: Remove flawed mitigation for rx irq flood
by Yi Yang 07 May '24

07 May '24

From: Finn Thain <fthain(a)linux-m68k.org> stable inclusion from stable-v5.10.216 commit ab86cf6f8d24e63e9aca23da5108af1aa5483928 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5KR CVE: CVE-2024-26999 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 1be3226445362bfbf461c92a5bcdb1723f2e4907 upstream. The mitigation was intended to stop the irq completely. That may be better than a hard lock-up but it turns out that you get a crash anyway if you're using pmac_zilog as a serial console: ttyPZ0: pmz: rx irq flood ! BUG: spinlock recursion on CPU#0, swapper/0 That's because the pr_err() call in pmz_receive_chars() results in pmz_console_write() attempting to lock a spinlock already locked in pmz_interrupt(). With CONFIG_DEBUG_SPINLOCK=y, this produces a fatal BUG splat. The spinlock in question is the one in struct uart_port. Even when it's not fatal, the serial port rx function ceases to work. Also, the iteration limit doesn't play nicely with QEMU, as can be seen in the bug report linked below. A web search for other reports of the error message "pmz: rx irq flood" didn't produce anything. So I don't think this code is needed any more. Remove it. Cc: Benjamin Herrenschmidt <benh(a)kernel.crashing.org> Cc: Michael Ellerman <mpe(a)ellerman.id.au> Cc: Nicholas Piggin <npiggin(a)gmail.com> Cc: Christophe Leroy <christophe.leroy(a)csgroup.eu> Cc: Aneesh Kumar K.V <aneesh.kumar(a)kernel.org> Cc: Naveen N. Rao <naveen.n.rao(a)linux.ibm.com> Cc: Andy Shevchenko <andy.shevchenko(a)gmail.com> Cc: stable(a)kernel.org Cc: linux-m68k(a)lists.linux-m68k.org Link: https://github.com/vivier/qemu-m68k/issues/44 Link: https://lore.kernel.org/all/1078874617.9746.36.camel@gaston/ Acked-by: Michael Ellerman <mpe(a)ellerman.id.au> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable <stable(a)kernel.org> Signed-off-by: Finn Thain <fthain(a)linux-m68k.org> Link: https://lore.kernel.org/r/e853cf2c762f23101cd2ddec0cc0c2be0e72685f.17125682… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/tty/serial/pmac_zilog.c | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/drivers/tty/serial/pmac_zilog.c b/drivers/tty/serial/pmac_zilog.c index d6aef8a1f0a4..1d0717fc3729 100644 --- a/drivers/tty/serial/pmac_zilog.c +++ b/drivers/tty/serial/pmac_zilog.c @@ -217,7 +217,6 @@ static bool pmz_receive_chars(struct uart_pmac_port *uap) { struct tty_port *port; unsigned char ch, r1, drop, flag; - int loops = 0; /* Sanity check, make sure the old bug is no longer happening */ if (uap->port.state == NULL) { @@ -298,24 +297,11 @@ static bool pmz_receive_chars(struct uart_pmac_port *uap) if (r1 & Rx_OVR) tty_insert_flip_char(port, 0, TTY_OVERRUN); next_char: - /* We can get stuck in an infinite loop getting char 0 when the - * line is in a wrong HW state, we break that here. - * When that happens, I disable the receive side of the driver. - * Note that what I've been experiencing is a real irq loop where - * I'm getting flooded regardless of the actual port speed. - * Something strange is going on with the HW - */ - if ((++loops) > 1000) - goto flood; ch = read_zsreg(uap, R0); if (!(ch & Rx_CH_AV)) break; } - return true; - flood: - pmz_interrupt_control(uap, 0); - pmz_error("pmz: rx irq flood !\n"); return true; } -- 2.25.1

1 0

[PATCH] serial/pmac_zilog: Remove flawed mitigation for rx irq flood
by Yi Yang 07 May '24

07 May '24

From: Finn Thain <fthain(a)linux-m68k.org> stable inclusion from stable-v4.19.313 commit 69a02273e288011b521ee7c1f3ab2c23fda633ce category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5KR CVE: CVE-2024-26999 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 1be3226445362bfbf461c92a5bcdb1723f2e4907 upstream. The mitigation was intended to stop the irq completely. That may be better than a hard lock-up but it turns out that you get a crash anyway if you're using pmac_zilog as a serial console: ttyPZ0: pmz: rx irq flood ! BUG: spinlock recursion on CPU#0, swapper/0 That's because the pr_err() call in pmz_receive_chars() results in pmz_console_write() attempting to lock a spinlock already locked in pmz_interrupt(). With CONFIG_DEBUG_SPINLOCK=y, this produces a fatal BUG splat. The spinlock in question is the one in struct uart_port. Even when it's not fatal, the serial port rx function ceases to work. Also, the iteration limit doesn't play nicely with QEMU, as can be seen in the bug report linked below. A web search for other reports of the error message "pmz: rx irq flood" didn't produce anything. So I don't think this code is needed any more. Remove it. Cc: Benjamin Herrenschmidt <benh(a)kernel.crashing.org> Cc: Michael Ellerman <mpe(a)ellerman.id.au> Cc: Nicholas Piggin <npiggin(a)gmail.com> Cc: Christophe Leroy <christophe.leroy(a)csgroup.eu> Cc: Aneesh Kumar K.V <aneesh.kumar(a)kernel.org> Cc: Naveen N. Rao <naveen.n.rao(a)linux.ibm.com> Cc: Andy Shevchenko <andy.shevchenko(a)gmail.com> Cc: stable(a)kernel.org Cc: linux-m68k(a)lists.linux-m68k.org Link: https://github.com/vivier/qemu-m68k/issues/44 Link: https://lore.kernel.org/all/1078874617.9746.36.camel@gaston/ Acked-by: Michael Ellerman <mpe(a)ellerman.id.au> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable <stable(a)kernel.org> Signed-off-by: Finn Thain <fthain(a)linux-m68k.org> Link: https://lore.kernel.org/r/e853cf2c762f23101cd2ddec0cc0c2be0e72685f.17125682… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/tty/serial/pmac_zilog.c | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/drivers/tty/serial/pmac_zilog.c b/drivers/tty/serial/pmac_zilog.c index 3d21790d961e..2cddcf74f702 100644 --- a/drivers/tty/serial/pmac_zilog.c +++ b/drivers/tty/serial/pmac_zilog.c @@ -220,7 +220,6 @@ static bool pmz_receive_chars(struct uart_pmac_port *uap) { struct tty_port *port; unsigned char ch, r1, drop, error, flag; - int loops = 0; /* Sanity check, make sure the old bug is no longer happening */ if (uap->port.state == NULL) { @@ -303,24 +302,11 @@ static bool pmz_receive_chars(struct uart_pmac_port *uap) if (r1 & Rx_OVR) tty_insert_flip_char(port, 0, TTY_OVERRUN); next_char: - /* We can get stuck in an infinite loop getting char 0 when the - * line is in a wrong HW state, we break that here. - * When that happens, I disable the receive side of the driver. - * Note that what I've been experiencing is a real irq loop where - * I'm getting flooded regardless of the actual port speed. - * Something strange is going on with the HW - */ - if ((++loops) > 1000) - goto flood; ch = read_zsreg(uap, R0); if (!(ch & Rx_CH_AV)) break; } - return true; - flood: - pmz_interrupt_control(uap, 0); - pmz_error("pmz: rx irq flood !\n"); return true; } -- 2.25.1

1 0

[PATCH OLK-5.10] erofs: fix pcluster use-after-free on UP platforms
by Zizhi Wo 07 May '24

07 May '24

From: Gao Xiang <hsiangkao(a)linux.alibaba.com> stable inclusion from stable-v5.15.68 commit 8ddd001cef5e82d19192e6861068463ecca5f556 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9LK3Y CVE: CVE-2022-48674 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 2f44013e39984c127c6efedf70e6b5f4e9dcf315 ] During stress testing with CONFIG_SMP disabled, KASAN reports as below: ================================================================== BUG: KASAN: use-after-free in __mutex_lock+0xe5/0xc30 Read of size 8 at addr ffff8881094223f8 by task stress/7789 CPU: 0 PID: 7789 Comm: stress Not tainted 6.0.0-rc1-00002-g0d53d2e882f9 #3 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 Call Trace: <TASK> .. __mutex_lock+0xe5/0xc30 .. z_erofs_do_read_page+0x8ce/0x1560 .. z_erofs_readahead+0x31c/0x580 .. Freed by task 7787 kasan_save_stack+0x1e/0x40 kasan_set_track+0x20/0x30 kasan_set_free_info+0x20/0x40 __kasan_slab_free+0x10c/0x190 kmem_cache_free+0xed/0x380 rcu_core+0x3d5/0xc90 __do_softirq+0x12d/0x389 Last potentially related work creation: kasan_save_stack+0x1e/0x40 __kasan_record_aux_stack+0x97/0xb0 call_rcu+0x3d/0x3f0 erofs_shrink_workstation+0x11f/0x210 erofs_shrink_scan+0xdc/0x170 shrink_slab.constprop.0+0x296/0x530 drop_slab+0x1c/0x70 drop_caches_sysctl_handler+0x70/0x80 proc_sys_call_handler+0x20a/0x2f0 vfs_write+0x555/0x6c0 ksys_write+0xbe/0x160 do_syscall_64+0x3b/0x90 The root cause is that erofs_workgroup_unfreeze() doesn't reset to orig_val thus it causes a race that the pcluster reuses unexpectedly before freeing. Since UP platforms are quite rare now, such path becomes unnecessary. Let's drop such specific-designed path directly instead. Fixes: 73f5c66df3e2 ("staging: erofs: fix `erofs_workgroup_{try_to_freeze, unfreeze}'") Reviewed-by: Yue Hu <huyue2(a)coolpad.com> Reviewed-by: Chao Yu <chao(a)kernel.org> Link: https://lore.kernel.org/r/20220902045710.109530-1-hsiangkao@linux.alibaba.c… Signed-off-by: Gao Xiang <hsiangkao(a)linux.alibaba.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: fs/erofs/internal.h Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- fs/erofs/internal.h | 29 ----------------------------- 1 file changed, 29 deletions(-) diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index ce52f708a403..5b3a9f5c282d 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -125,7 +125,6 @@ struct erofs_workgroup { atomic_t refcount; }; -#if defined(CONFIG_SMP) static inline bool erofs_workgroup_try_to_freeze(struct erofs_workgroup *grp, int val) { @@ -154,34 +153,6 @@ static inline int erofs_wait_on_workgroup_freezed(struct erofs_workgroup *grp) return atomic_cond_read_relaxed(&grp->refcount, VAL != EROFS_LOCKED_MAGIC); } -#else -static inline bool erofs_workgroup_try_to_freeze(struct erofs_workgroup *grp, - int val) -{ - preempt_disable(); - /* no need to spin on UP platforms, let's just disable preemption. */ - if (val != atomic_read(&grp->refcount)) { - preempt_enable(); - return false; - } - return true; -} - -static inline void erofs_workgroup_unfreeze(struct erofs_workgroup *grp, - int orig_val) -{ - preempt_enable(); -} - -static inline int erofs_wait_on_workgroup_freezed(struct erofs_workgroup *grp) -{ - int v = atomic_read(&grp->refcount); - - /* workgroup is never freezed on uniprocessor systems */ - DBG_BUGON(v == EROFS_LOCKED_MAGIC); - return v; -} -#endif /* !CONFIG_SMP */ /* hard limit of pages per compressed cluster */ #define Z_EROFS_CLUSTER_MAX_PAGES (CONFIG_EROFS_FS_CLUSTER_PAGE_LIMIT) -- 2.39.2

2 1

[PATCH OLK-6.6] scsi: core: Fix unremoved procfs host directory regression
by Li Lingfeng 07 May '24

07 May '24

From: "Guilherme G. Piccoli" <gpiccoli(a)igalia.com> mainline inclusion from mainline-v6.9-rc2 commit f23a4d6e07570826fe95023ca1aa96a011fa9f84 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9HJP4 CVE: CVE-2024-26935 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Commit fc663711b944 ("scsi: core: Remove the /proc/scsi/${proc_name} directory earlier") fixed a bug related to modules loading/unloading, by adding a call to scsi_proc_hostdir_rm() on scsi_remove_host(). But that led to a potential duplicate call to the hostdir_rm() routine, since it's also called from scsi_host_dev_release(). That triggered a regression report, which was then fixed by commit be03df3d4bfe ("scsi: core: Fix a procfs host directory removal regression"). The fix just dropped the hostdir_rm() call from dev_release(). But it happens that this proc directory is created on scsi_host_alloc(), and that function "pairs" with scsi_host_dev_release(), while scsi_remove_host() pairs with scsi_add_host(). In other words, it seems the reason for removing the proc directory on dev_release() was meant to cover cases in which a SCSI host structure was allocated, but the call to scsi_add_host() didn't happen. And that pattern happens to exist in some error paths, for example. Syzkaller causes that by using USB raw gadget device, error'ing on usb-storage driver, at usb_stor_probe2(). By checking that path, we can see that the BadDevice label leads to a scsi_host_put() after a SCSI host allocation, but there's no call to scsi_add_host() in such path. That leads to messages like this in dmesg (and a leak of the SCSI host proc structure): usb-storage 4-1:87.51: USB Mass Storage device detected proc_dir_entry 'scsi/usb-storage' already registered WARNING: CPU: 1 PID: 3519 at fs/proc/generic.c:377 proc_register+0x347/0x4e0 fs/proc/generic.c:376 The proper fix seems to still call scsi_proc_hostdir_rm() on dev_release(), but guard that with the state check for SHOST_CREATED; there is even a comment in scsi_host_dev_release() detailing that: such conditional is meant for cases where the SCSI host was allocated but there was no calls to {add,remove}_host(), like the usb-storage case. This is what we propose here and with that, the error path of usb-storage does not trigger the warning anymore. Reported-by: syzbot+c645abf505ed21f931b5(a)syzkaller.appspotmail.com Fixes: be03df3d4bfe ("scsi: core: Fix a procfs host directory removal regression") Cc: stable(a)vger.kernel.org Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: John Garry <john.g.garry(a)oracle.com> Cc: Shin'ichiro Kawasaki <shinichiro.kawasaki(a)wdc.com> Signed-off-by: Guilherme G. Piccoli <gpiccoli(a)igalia.com> Link: https://lore.kernel.org/r/20240313113006.2834799-1-gpiccoli@igalia.com Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Li Lingfeng <lilingfeng3(a)huawei.com> --- drivers/scsi/hosts.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index d7f51b84f3c7..445f4a220df3 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -353,12 +353,13 @@ static void scsi_host_dev_release(struct device *dev) if (shost->shost_state == SHOST_CREATED) { /* - * Free the shost_dev device name here if scsi_host_alloc() - * and scsi_host_put() have been called but neither + * Free the shost_dev device name and remove the proc host dir + * here if scsi_host_{alloc,put}() have been called but neither * scsi_host_add() nor scsi_remove_host() has been called. * This avoids that the memory allocated for the shost_dev - * name is leaked. + * name as well as the proc dir structure are leaked. */ + scsi_proc_hostdir_rm(shost->hostt); kfree(dev_name(&shost->shost_dev)); } -- 2.31.1

2 1

[PATCH openEuler-1.0-LTS] ALSA: emu10k1: Fix out of bounds access in snd_emu10k1_pcm_channel_alloc()
by Zhang Xiaoxu 07 May '24

07 May '24

From: Tasos Sahanidis <tasos(a)tasossah.com> mainline inclusion from mainline-v6.0-rc5 commit d29f59051d3a07b81281b2df2b8c9dfe4716067f category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9LKE1 CVE: CVE-2022-48702 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- The voice allocator sometimes begins allocating from near the end of the array and then wraps around, however snd_emu10k1_pcm_channel_alloc() accesses the newly allocated voices as if it never wrapped around. This results in out of bounds access if the first voice has a high enough index so that first_voice + requested_voice_count > NUM_G (64). The more voices are requested, the more likely it is for this to occur. This was initially discovered using PipeWire, however it can be reproduced by calling aplay multiple times with 16 channels: aplay -r 48000 -D plughw:CARD=Live,DEV=3 -c 16 /dev/zero UBSAN: array-index-out-of-bounds in sound/pci/emu10k1/emupcm.c:127:40 index 65 is out of range for type 'snd_emu10k1_voice [64]' CPU: 1 PID: 31977 Comm: aplay Tainted: G W IOE 6.0.0-rc2-emu10k1+ #7 Hardware name: ASUSTEK COMPUTER INC P5W DH Deluxe/P5W DH Deluxe, BIOS 3002 07/22/2010 Call Trace: <TASK> dump_stack_lvl+0x49/0x63 dump_stack+0x10/0x16 ubsan_epilogue+0x9/0x3f __ubsan_handle_out_of_bounds.cold+0x44/0x49 snd_emu10k1_playback_hw_params+0x3bc/0x420 [snd_emu10k1] snd_pcm_hw_params+0x29f/0x600 [snd_pcm] snd_pcm_common_ioctl+0x188/0x1410 [snd_pcm] ? exit_to_user_mode_prepare+0x35/0x170 ? do_syscall_64+0x69/0x90 ? syscall_exit_to_user_mode+0x26/0x50 ? do_syscall_64+0x69/0x90 ? exit_to_user_mode_prepare+0x35/0x170 snd_pcm_ioctl+0x27/0x40 [snd_pcm] __x64_sys_ioctl+0x95/0xd0 do_syscall_64+0x5c/0x90 ? do_syscall_64+0x69/0x90 ? do_syscall_64+0x69/0x90 entry_SYSCALL_64_after_hwframe+0x63/0xcd Signed-off-by: Tasos Sahanidis <tasos(a)tasossah.com> Cc: <stable(a)vger.kernel.org> Link: https://lore.kernel.org/r/3707dcab-320a-62ff-63c0-73fc201ef756@tasossah.com Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5(a)huawei.com> --- sound/pci/emu10k1/emupcm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/emu10k1/emupcm.c b/sound/pci/emu10k1/emupcm.c index 9f2b6097f486..623776b13f8d 100644 --- a/sound/pci/emu10k1/emupcm.c +++ b/sound/pci/emu10k1/emupcm.c @@ -137,7 +137,7 @@ static int snd_emu10k1_pcm_channel_alloc(struct snd_emu10k1_pcm * epcm, int voic epcm->voices[0]->epcm = epcm; if (voices > 1) { for (i = 1; i < voices; i++) { - epcm->voices[i] = &epcm->emu->voices[epcm->voices[0]->number + i]; + epcm->voices[i] = &epcm->emu->voices[(epcm->voices[0]->number + i) % NUM_G]; epcm->voices[i]->epcm = epcm; } } -- 2.34.1

2 1

[PATCH OLK-5.10] perf/x86/amd: Fix crash due to race between amd_pmu_enable_all, perf NMI and throttling
by Luo Gengkun 07 May '24

07 May '24

From: Ravi Bangoria <ravi.bangoria(a)amd.com> mainline inclusion from mainline-v6.1-rc6 commit baa014b9543c8e5e94f5d15b66abfe60750b8284 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9MMNG Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- amd_pmu_enable_all() does: if (!test_bit(idx, cpuc->active_mask)) continue; amd_pmu_enable_event(cpuc->events[idx]); A perf NMI of another event can come between these two steps. Perf NMI handler internally disables and enables _all_ events, including the one which nmi-intercepted amd_pmu_enable_all() was in process of enabling. If that unintentionally enabled event has very low sampling period and causes immediate successive NMI, causing the event to be throttled, cpuc->events[idx] and cpuc->active_mask gets cleared by x86_pmu_stop(). This will result in amd_pmu_enable_event() getting called with event=NULL when amd_pmu_enable_all() resumes after handling the NMIs. This causes a kernel crash: BUG: kernel NULL pointer dereference, address: 0000000000000198 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page [...] Call Trace: <TASK> amd_pmu_enable_all+0x68/0xb0 ctx_resched+0xd9/0x150 event_function+0xb8/0x130 ? hrtimer_start_range_ns+0x141/0x4a0 ? perf_duration_warn+0x30/0x30 remote_function+0x4d/0x60 __flush_smp_call_function_queue+0xc4/0x500 flush_smp_call_function_queue+0x11d/0x1b0 do_idle+0x18f/0x2d0 cpu_startup_entry+0x19/0x20 start_secondary+0x121/0x160 secondary_startup_64_no_verify+0xe5/0xeb </TASK> amd_pmu_disable_all()/amd_pmu_enable_all() calls inside perf NMI handler were recently added as part of BRS enablement but I'm not sure whether we really need them. We can just disable BRS in the beginning and enable it back while returning from NMI. This will solve the issue by not enabling those events whose active_masks are set but are not yet enabled in hw pmu. Fixes: ada543459cab ("perf/x86/amd: Add AMD Fam19h Branch Sampling support") Reported-by: Linux Kernel Functional Testing <lkft(a)linaro.org> Signed-off-by: Ravi Bangoria <ravi.bangoria(a)amd.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Link: https://lkml.kernel.org/r/20221114044029.373-1-ravi.bangoria@amd.com Signed-off-by: Luo Gengkun <luogengkun2(a)huawei.com> --- arch/x86/events/amd/core.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/arch/x86/events/amd/core.c b/arch/x86/events/amd/core.c index 2c082abb930b..3127519281f5 100644 --- a/arch/x86/events/amd/core.c +++ b/arch/x86/events/amd/core.c @@ -900,8 +900,7 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) pmu_enabled = cpuc->enabled; cpuc->enabled = 0; - /* stop everything (includes BRS) */ - amd_pmu_disable_all(); + amd_brs_disable_all(); /* Drain BRS is in use (could be inactive) */ if (cpuc->lbr_users) @@ -912,7 +911,7 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) cpuc->enabled = pmu_enabled; if (pmu_enabled) - amd_pmu_enable_all(0); + amd_brs_enable_all(); return amd_pmu_adjust_nmi_window(handled); } -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] erofs: fix pcluster use-after-free on UP platforms
by Zizhi Wo 07 May '24

07 May '24

From: Gao Xiang <hsiangkao(a)linux.alibaba.com> stable inclusion from stable-v5.15.68 commit 8ddd001cef5e82d19192e6861068463ecca5f556 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9LK3Y CVE: CVE-2022-48674 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 2f44013e39984c127c6efedf70e6b5f4e9dcf315 ] During stress testing with CONFIG_SMP disabled, KASAN reports as below: ================================================================== BUG: KASAN: use-after-free in __mutex_lock+0xe5/0xc30 Read of size 8 at addr ffff8881094223f8 by task stress/7789 CPU: 0 PID: 7789 Comm: stress Not tainted 6.0.0-rc1-00002-g0d53d2e882f9 #3 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 Call Trace: <TASK> .. __mutex_lock+0xe5/0xc30 .. z_erofs_do_read_page+0x8ce/0x1560 .. z_erofs_readahead+0x31c/0x580 .. Freed by task 7787 kasan_save_stack+0x1e/0x40 kasan_set_track+0x20/0x30 kasan_set_free_info+0x20/0x40 __kasan_slab_free+0x10c/0x190 kmem_cache_free+0xed/0x380 rcu_core+0x3d5/0xc90 __do_softirq+0x12d/0x389 Last potentially related work creation: kasan_save_stack+0x1e/0x40 __kasan_record_aux_stack+0x97/0xb0 call_rcu+0x3d/0x3f0 erofs_shrink_workstation+0x11f/0x210 erofs_shrink_scan+0xdc/0x170 shrink_slab.constprop.0+0x296/0x530 drop_slab+0x1c/0x70 drop_caches_sysctl_handler+0x70/0x80 proc_sys_call_handler+0x20a/0x2f0 vfs_write+0x555/0x6c0 ksys_write+0xbe/0x160 do_syscall_64+0x3b/0x90 The root cause is that erofs_workgroup_unfreeze() doesn't reset to orig_val thus it causes a race that the pcluster reuses unexpectedly before freeing. Since UP platforms are quite rare now, such path becomes unnecessary. Let's drop such specific-designed path directly instead. Fixes: 73f5c66df3e2 ("staging: erofs: fix `erofs_workgroup_{try_to_freeze, unfreeze}'") Reviewed-by: Yue Hu <huyue2(a)coolpad.com> Reviewed-by: Chao Yu <chao(a)kernel.org> Link: https://lore.kernel.org/r/20220902045710.109530-1-hsiangkao@linux.alibaba.c… Signed-off-by: Gao Xiang <hsiangkao(a)linux.alibaba.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: fs/erofs/internal.h drivers/staging/erofs/internal.h Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- drivers/staging/erofs/internal.h | 29 ----------------------------- 1 file changed, 29 deletions(-) diff --git a/drivers/staging/erofs/internal.h b/drivers/staging/erofs/internal.h index 8ce37091db20..541326c457f0 100644 --- a/drivers/staging/erofs/internal.h +++ b/drivers/staging/erofs/internal.h @@ -186,7 +186,6 @@ struct erofs_workgroup { #define EROFS_LOCKED_MAGIC (INT_MIN | 0xE0F510CCL) -#if defined(CONFIG_SMP) static inline bool erofs_workgroup_try_to_freeze(struct erofs_workgroup *grp, int val) { @@ -215,34 +214,6 @@ static inline int erofs_wait_on_workgroup_freezed(struct erofs_workgroup *grp) return atomic_cond_read_relaxed(&grp->refcount, VAL != EROFS_LOCKED_MAGIC); } -#else -static inline bool erofs_workgroup_try_to_freeze(struct erofs_workgroup *grp, - int val) -{ - preempt_disable(); - /* no need to spin on UP platforms, let's just disable preemption. */ - if (val != atomic_read(&grp->refcount)) { - preempt_enable(); - return false; - } - return true; -} - -static inline void erofs_workgroup_unfreeze(struct erofs_workgroup *grp, - int orig_val) -{ - preempt_enable(); -} - -static inline int erofs_wait_on_workgroup_freezed(struct erofs_workgroup *grp) -{ - int v = atomic_read(&grp->refcount); - - /* workgroup is never freezed on uniprocessor systems */ - DBG_BUGON(v == EROFS_LOCKED_MAGIC); - return v; -} -#endif static inline bool erofs_workgroup_get(struct erofs_workgroup *grp, int *ocnt) { -- 2.39.2

2 1

[PATCH OLK-6.6] block: fix deadlock between bd_link_disk_holder and partition scan
by Li Nan 07 May '24

07 May '24

mainline inclusion from mainline-v6.9-rc1 commit 03f12122b20b6e6028e9ed69030a49f9cffcbb75 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9IVH5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- 'open_mutex' of gendisk is used to protect open/close block devices. But in bd_link_disk_holder(), it is used to protect the creation of symlink between holding disk and slave bdev, which introduces some issues. When bd_link_disk_holder() is called, the driver is usually in the process of initialization/modification and may suspend submitting io. At this time, any io hold 'open_mutex', such as scanning partitions, can cause deadlocks. For example, in raid: T1 T2 bdev_open_by_dev lock open_mutex [1] ... efi_partition ... md_submit_bio md_ioctl mddev_syspend -> suspend all io md_add_new_disk bind_rdev_to_array bd_link_disk_holder try lock open_mutex [2] md_handle_request -> wait mddev_resume T1 scan partition, T2 add a new device to raid. T1 waits for T2 to resume mddev, but T2 waits for open_mutex held by T1. Deadlock occurs. Fix it by introducing a local mutex 'blk_holder_mutex' to replace 'open_mutex'. Fixes: 1b0a2d950ee2 ("md: use new apis to suspend array for ioctls involed array reconfiguration") Reported-by: mgperkow(a)gmail.com Closes: https://bugzilla.kernel.org/show_bug.cgi?id=218459 Signed-off-by: Li Nan <linan122(a)huawei.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Yu Kuai <yukuai3(a)huawei.com> Link: https://lore.kernel.org/r/20240221090122.1281868-1-linan666@huaweicloud.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> --- block/holder.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/block/holder.c b/block/holder.c index 37d18c13d958..791091a7eac2 100644 --- a/block/holder.c +++ b/block/holder.c @@ -8,6 +8,8 @@ struct bd_holder_disk { int refcnt; }; +static DEFINE_MUTEX(blk_holder_mutex); + static struct bd_holder_disk *bd_find_holder_disk(struct block_device *bdev, struct gendisk *disk) { @@ -80,7 +82,7 @@ int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk) kobject_get(bdev->bd_holder_dir); mutex_unlock(&bdev->bd_disk->open_mutex); - mutex_lock(&disk->open_mutex); + mutex_lock(&blk_holder_mutex); WARN_ON_ONCE(!bdev->bd_holder); holder = bd_find_holder_disk(bdev, disk); @@ -108,7 +110,7 @@ int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk) goto out_del_symlink; list_add(&holder->list, &disk->slave_bdevs); - mutex_unlock(&disk->open_mutex); + mutex_unlock(&blk_holder_mutex); return 0; out_del_symlink: @@ -116,7 +118,7 @@ int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk) out_free_holder: kfree(holder); out_unlock: - mutex_unlock(&disk->open_mutex); + mutex_unlock(&blk_holder_mutex); if (ret) kobject_put(bdev->bd_holder_dir); return ret; @@ -140,7 +142,7 @@ void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk) if (WARN_ON_ONCE(!disk->slave_dir)) return; - mutex_lock(&disk->open_mutex); + mutex_lock(&blk_holder_mutex); holder = bd_find_holder_disk(bdev, disk); if (!WARN_ON_ONCE(holder == NULL) && !--holder->refcnt) { del_symlink(disk->slave_dir, bdev_kobj(bdev)); @@ -149,6 +151,6 @@ void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk) list_del_init(&holder->list); kfree(holder); } - mutex_unlock(&disk->open_mutex); + mutex_unlock(&blk_holder_mutex); } EXPORT_SYMBOL_GPL(bd_unlink_disk_holder); -- 2.39.2

2 1

[PATCH OLK-5.10 v2] scsi: core: Fix unremoved procfs host directory regression
by Li Lingfeng 07 May '24

07 May '24

From: "Guilherme G. Piccoli" <gpiccoli(a)igalia.com> stable inclusion from stable-v5.10.215 commit 5c2386ba80e779a92ec3bb64ccadbedd88f779b1 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9HJP4 CVE: CVE-2024-26935 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit f23a4d6e07570826fe95023ca1aa96a011fa9f84 upstream. Commit fc663711b944 ("scsi: core: Remove the /proc/scsi/${proc_name} directory earlier") fixed a bug related to modules loading/unloading, by adding a call to scsi_proc_hostdir_rm() on scsi_remove_host(). But that led to a potential duplicate call to the hostdir_rm() routine, since it's also called from scsi_host_dev_release(). That triggered a regression report, which was then fixed by commit be03df3d4bfe ("scsi: core: Fix a procfs host directory removal regression"). The fix just dropped the hostdir_rm() call from dev_release(). But it happens that this proc directory is created on scsi_host_alloc(), and that function "pairs" with scsi_host_dev_release(), while scsi_remove_host() pairs with scsi_add_host(). In other words, it seems the reason for removing the proc directory on dev_release() was meant to cover cases in which a SCSI host structure was allocated, but the call to scsi_add_host() didn't happen. And that pattern happens to exist in some error paths, for example. Syzkaller causes that by using USB raw gadget device, error'ing on usb-storage driver, at usb_stor_probe2(). By checking that path, we can see that the BadDevice label leads to a scsi_host_put() after a SCSI host allocation, but there's no call to scsi_add_host() in such path. That leads to messages like this in dmesg (and a leak of the SCSI host proc structure): usb-storage 4-1:87.51: USB Mass Storage device detected proc_dir_entry 'scsi/usb-storage' already registered WARNING: CPU: 1 PID: 3519 at fs/proc/generic.c:377 proc_register+0x347/0x4e0 fs/proc/generic.c:376 The proper fix seems to still call scsi_proc_hostdir_rm() on dev_release(), but guard that with the state check for SHOST_CREATED; there is even a comment in scsi_host_dev_release() detailing that: such conditional is meant for cases where the SCSI host was allocated but there was no calls to {add,remove}_host(), like the usb-storage case. This is what we propose here and with that, the error path of usb-storage does not trigger the warning anymore. Reported-by: syzbot+c645abf505ed21f931b5(a)syzkaller.appspotmail.com Fixes: be03df3d4bfe ("scsi: core: Fix a procfs host directory removal regression") Cc: stable(a)vger.kernel.org Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: John Garry <john.g.garry(a)oracle.com> Cc: Shin'ichiro Kawasaki <shinichiro.kawasaki(a)wdc.com> Signed-off-by: Guilherme G. Piccoli <gpiccoli(a)igalia.com> Link: https://lore.kernel.org/r/20240313113006.2834799-1-gpiccoli@igalia.com Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Li Lingfeng <lilingfeng3(a)huawei.com> --- v1->v2: Add CVE num in commit message. drivers/scsi/hosts.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index 5bac4936c647..3feaa61f948f 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -334,12 +334,13 @@ static void scsi_host_dev_release(struct device *dev) if (shost->shost_state == SHOST_CREATED) { /* - * Free the shost_dev device name here if scsi_host_alloc() - * and scsi_host_put() have been called but neither + * Free the shost_dev device name and remove the proc host dir + * here if scsi_host_{alloc,put}() have been called but neither * scsi_host_add() nor scsi_host_remove() has been called. * This avoids that the memory allocated for the shost_dev - * name is leaked. + * name as well as the proc dir structure are leaked. */ + scsi_proc_hostdir_rm(shost->hostt); kfree(dev_name(&shost->shost_dev)); } -- 2.31.1

2 1

[PATCH openEuler-1.0-LTS] ALSA: emu10k1: Fix out of bounds access in snd_emu10k1_pcm_channel_alloc()
by Zhang Xiaoxu 07 May '24

07 May '24

From: Tasos Sahanidis <tasos(a)tasossah.com> mainline inclusion from mainline-v6.0-rc5 commit d29f59051d3a07b81281b2df2b8c9dfe4716067f category: bugfix bugzilla: NA CVE: CVE-2022-48702 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- The voice allocator sometimes begins allocating from near the end of the array and then wraps around, however snd_emu10k1_pcm_channel_alloc() accesses the newly allocated voices as if it never wrapped around. This results in out of bounds access if the first voice has a high enough index so that first_voice + requested_voice_count > NUM_G (64). The more voices are requested, the more likely it is for this to occur. This was initially discovered using PipeWire, however it can be reproduced by calling aplay multiple times with 16 channels: aplay -r 48000 -D plughw:CARD=Live,DEV=3 -c 16 /dev/zero UBSAN: array-index-out-of-bounds in sound/pci/emu10k1/emupcm.c:127:40 index 65 is out of range for type 'snd_emu10k1_voice [64]' CPU: 1 PID: 31977 Comm: aplay Tainted: G W IOE 6.0.0-rc2-emu10k1+ #7 Hardware name: ASUSTEK COMPUTER INC P5W DH Deluxe/P5W DH Deluxe, BIOS 3002 07/22/2010 Call Trace: <TASK> dump_stack_lvl+0x49/0x63 dump_stack+0x10/0x16 ubsan_epilogue+0x9/0x3f __ubsan_handle_out_of_bounds.cold+0x44/0x49 snd_emu10k1_playback_hw_params+0x3bc/0x420 [snd_emu10k1] snd_pcm_hw_params+0x29f/0x600 [snd_pcm] snd_pcm_common_ioctl+0x188/0x1410 [snd_pcm] ? exit_to_user_mode_prepare+0x35/0x170 ? do_syscall_64+0x69/0x90 ? syscall_exit_to_user_mode+0x26/0x50 ? do_syscall_64+0x69/0x90 ? exit_to_user_mode_prepare+0x35/0x170 snd_pcm_ioctl+0x27/0x40 [snd_pcm] __x64_sys_ioctl+0x95/0xd0 do_syscall_64+0x5c/0x90 ? do_syscall_64+0x69/0x90 ? do_syscall_64+0x69/0x90 entry_SYSCALL_64_after_hwframe+0x63/0xcd Signed-off-by: Tasos Sahanidis <tasos(a)tasossah.com> Cc: <stable(a)vger.kernel.org> Link: https://lore.kernel.org/r/3707dcab-320a-62ff-63c0-73fc201ef756@tasossah.com Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5(a)huawei.com> --- sound/pci/emu10k1/emupcm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/emu10k1/emupcm.c b/sound/pci/emu10k1/emupcm.c index 9f2b6097f486..623776b13f8d 100644 --- a/sound/pci/emu10k1/emupcm.c +++ b/sound/pci/emu10k1/emupcm.c @@ -137,7 +137,7 @@ static int snd_emu10k1_pcm_channel_alloc(struct snd_emu10k1_pcm * epcm, int voic epcm->voices[0]->epcm = epcm; if (voices > 1) { for (i = 1; i < voices; i++) { - epcm->voices[i] = &epcm->emu->voices[epcm->voices[0]->number + i]; + epcm->voices[i] = &epcm->emu->voices[(epcm->voices[0]->number + i) % NUM_G]; epcm->voices[i]->epcm = epcm; } } -- 2.34.1

2 1

[PATCH OLK-5.10] scsi: core: Fix unremoved procfs host directory regression
by Li Lingfeng 07 May '24

07 May '24

From: "Guilherme G. Piccoli" <gpiccoli(a)igalia.com> stable inclusion from stable-v5.10.215 commit 5c2386ba80e779a92ec3bb64ccadbedd88f779b1 category: bugfix bugzilla: 189877 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit f23a4d6e07570826fe95023ca1aa96a011fa9f84 upstream. Commit fc663711b944 ("scsi: core: Remove the /proc/scsi/${proc_name} directory earlier") fixed a bug related to modules loading/unloading, by adding a call to scsi_proc_hostdir_rm() on scsi_remove_host(). But that led to a potential duplicate call to the hostdir_rm() routine, since it's also called from scsi_host_dev_release(). That triggered a regression report, which was then fixed by commit be03df3d4bfe ("scsi: core: Fix a procfs host directory removal regression"). The fix just dropped the hostdir_rm() call from dev_release(). But it happens that this proc directory is created on scsi_host_alloc(), and that function "pairs" with scsi_host_dev_release(), while scsi_remove_host() pairs with scsi_add_host(). In other words, it seems the reason for removing the proc directory on dev_release() was meant to cover cases in which a SCSI host structure was allocated, but the call to scsi_add_host() didn't happen. And that pattern happens to exist in some error paths, for example. Syzkaller causes that by using USB raw gadget device, error'ing on usb-storage driver, at usb_stor_probe2(). By checking that path, we can see that the BadDevice label leads to a scsi_host_put() after a SCSI host allocation, but there's no call to scsi_add_host() in such path. That leads to messages like this in dmesg (and a leak of the SCSI host proc structure): usb-storage 4-1:87.51: USB Mass Storage device detected proc_dir_entry 'scsi/usb-storage' already registered WARNING: CPU: 1 PID: 3519 at fs/proc/generic.c:377 proc_register+0x347/0x4e0 fs/proc/generic.c:376 The proper fix seems to still call scsi_proc_hostdir_rm() on dev_release(), but guard that with the state check for SHOST_CREATED; there is even a comment in scsi_host_dev_release() detailing that: such conditional is meant for cases where the SCSI host was allocated but there was no calls to {add,remove}_host(), like the usb-storage case. This is what we propose here and with that, the error path of usb-storage does not trigger the warning anymore. Reported-by: syzbot+c645abf505ed21f931b5(a)syzkaller.appspotmail.com Fixes: be03df3d4bfe ("scsi: core: Fix a procfs host directory removal regression") Cc: stable(a)vger.kernel.org Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: John Garry <john.g.garry(a)oracle.com> Cc: Shin'ichiro Kawasaki <shinichiro.kawasaki(a)wdc.com> Signed-off-by: Guilherme G. Piccoli <gpiccoli(a)igalia.com> Link: https://lore.kernel.org/r/20240313113006.2834799-1-gpiccoli@igalia.com Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Li Lingfeng <lilingfeng3(a)huawei.com> --- drivers/scsi/hosts.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index 5bac4936c647..3feaa61f948f 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -334,12 +334,13 @@ static void scsi_host_dev_release(struct device *dev) if (shost->shost_state == SHOST_CREATED) { /* - * Free the shost_dev device name here if scsi_host_alloc() - * and scsi_host_put() have been called but neither + * Free the shost_dev device name and remove the proc host dir + * here if scsi_host_{alloc,put}() have been called but neither * scsi_host_add() nor scsi_host_remove() has been called. * This avoids that the memory allocated for the shost_dev - * name is leaked. + * name as well as the proc dir structure are leaked. */ + scsi_proc_hostdir_rm(shost->hostt); kfree(dev_name(&shost->shost_dev)); } -- 2.31.1

2 1

[PATCH OLK-5.10 0/1] CVE-2023-52650
by Yuntao Liu 07 May '24

07 May '24

CVE-2023-52650 Chen Ni (1): [Backport] drm/tegra: dsi: Add missing check for of_find_device_by_node drivers/gpu/drm/tegra/dsi.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) -- 2.34.1

2 2

[PATCH openEuler-1.0-LTS 0/1] CVE-2023-52650
by Yuntao Liu 07 May '24

07 May '24

CVE-2023-52650 Chen Ni (1): [Backport] drm/tegra: dsi: Add missing check for of_find_device_by_node drivers/gpu/drm/tegra/dsi.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) -- 2.34.1

2 2

[PATCH OLK-5.10] fs: sysfs: Fix reference leak in sysfs_break_active_protection()
by Yifan Qiao 07 May '24

07 May '24

From: Alan Stern <stern(a)rowland.harvard.edu> stable inclusion from stable-v6.8.8 commit a4c99b57d43bab45225ba92d574a8683f9edc8e4 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5IQ CVE: CVE-2024-26993 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit a90bca2228c0646fc29a72689d308e5fe03e6d78 upstream. The sysfs_break_active_protection() routine has an obvious reference leak in its error path. If the call to kernfs_find_and_get() fails then kn will be NULL, so the companion sysfs_unbreak_active_protection() routine won't get called (and would only cause an access violation by trying to dereference kn->parent if it was called). As a result, the reference to kobj acquired at the start of the function will never be released. Fix the leak by adding an explicit kobject_put() call when kn is NULL. Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu> Fixes: 2afc9166f79b ("scsi: sysfs: Introduce sysfs_{un,}break_active_protection()") Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: stable(a)vger.kernel.org Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Acked-by: Tejun Heo <tj(a)kernel.org> Link: https://lore.kernel.org/r/8a4d3f0f-c5e3-4b70-a188-0ca433f9e6f9@rowland.harv… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/sysfs/file.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index a12ac0356c69..f21e73d10724 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -450,6 +450,8 @@ struct kernfs_node *sysfs_break_active_protection(struct kobject *kobj, kn = kernfs_find_and_get(kobj->sd, attr->name); if (kn) kernfs_break_active_protection(kn); + else + kobject_put(kobj); return kn; } EXPORT_SYMBOL_GPL(sysfs_break_active_protection); -- 2.39.2

2 1

[PATCH OLK-6.6] fs: sysfs: Fix reference leak in sysfs_break_active_protection()
by Yifan Qiao 07 May '24

07 May '24

From: Alan Stern <stern(a)rowland.harvard.edu> stable inclusion from stable-v6.8.8 commit a4c99b57d43bab45225ba92d574a8683f9edc8e4 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5IQ CVE: CVE-2024-26993 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit a90bca2228c0646fc29a72689d308e5fe03e6d78 upstream. The sysfs_break_active_protection() routine has an obvious reference leak in its error path. If the call to kernfs_find_and_get() fails then kn will be NULL, so the companion sysfs_unbreak_active_protection() routine won't get called (and would only cause an access violation by trying to dereference kn->parent if it was called). As a result, the reference to kobj acquired at the start of the function will never be released. Fix the leak by adding an explicit kobject_put() call when kn is NULL. Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu> Fixes: 2afc9166f79b ("scsi: sysfs: Introduce sysfs_{un,}break_active_protection()") Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: stable(a)vger.kernel.org Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Acked-by: Tejun Heo <tj(a)kernel.org> Link: https://lore.kernel.org/r/8a4d3f0f-c5e3-4b70-a188-0ca433f9e6f9@rowland.harv… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/sysfs/file.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index a12ac0356c69..f21e73d10724 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -450,6 +450,8 @@ struct kernfs_node *sysfs_break_active_protection(struct kobject *kobj, kn = kernfs_find_and_get(kobj->sd, attr->name); if (kn) kernfs_break_active_protection(kn); + else + kobject_put(kobj); return kn; } EXPORT_SYMBOL_GPL(sysfs_break_active_protection); -- 2.39.2

2 1

[PATCH openEuler-1.0-LTS] fs: sysfs: Fix reference leak in sysfs_break_active_protection()
by Yifan Qiao 07 May '24

07 May '24

From: Alan Stern <stern(a)rowland.harvard.edu> stable inclusion from stable-v6.8.8 commit a4c99b57d43bab45225ba92d574a8683f9edc8e4 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5IQ CVE: CVE-2024-26993 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit a90bca2228c0646fc29a72689d308e5fe03e6d78 upstream. The sysfs_break_active_protection() routine has an obvious reference leak in its error path. If the call to kernfs_find_and_get() fails then kn will be NULL, so the companion sysfs_unbreak_active_protection() routine won't get called (and would only cause an access violation by trying to dereference kn->parent if it was called). As a result, the reference to kobj acquired at the start of the function will never be released. Fix the leak by adding an explicit kobject_put() call when kn is NULL. Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu> Fixes: 2afc9166f79b ("scsi: sysfs: Introduce sysfs_{un,}break_active_protection()") Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: stable(a)vger.kernel.org Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Acked-by: Tejun Heo <tj(a)kernel.org> Link: https://lore.kernel.org/r/8a4d3f0f-c5e3-4b70-a188-0ca433f9e6f9@rowland.harv… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/sysfs/file.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index a12ac0356c69..f21e73d10724 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -450,6 +450,8 @@ struct kernfs_node *sysfs_break_active_protection(struct kobject *kobj, kn = kernfs_find_and_get(kobj->sd, attr->name); if (kn) kernfs_break_active_protection(kn); + else + kobject_put(kobj); return kn; } EXPORT_SYMBOL_GPL(sysfs_break_active_protection); -- 2.39.2

2 1

[openeuler:OLK-5.10] BUILD SUCCESS 93f42c5daddd692f4cc1f56387774d46e083147c
by kernel test robot 07 May '24

07 May '24

tree/branch: https://gitee.com/openeuler/kernel.git OLK-5.10 branch HEAD: 93f42c5daddd692f4cc1f56387774d46e083147c !6798 media: ir_toy: fix a memleak in irtoy_tx Warning ids grouped by kconfigs: clang_recent_errors |-- arm64-allyesconfig | |-- Documentation-devicetree-bindings-arm-cpu.yaml:properties:capacity-dmips-mhz:ref-should-not-be-valid-under-const:ref | |-- Documentation-devicetree-bindings-arm-cpu.yaml:title:ARM-CPUs-bindings-should-not-be-valid-under-pattern:(-Bb-inding-Ss-chema) | |-- Documentation-devicetree-bindings-arm-cpus.yaml:examples:cpus-arm-pbha-performance-only-bits-arm-pbha-no-aliases-bits-ncpu-device_type-cpu-compatible-arm-cortex-a57-...-n-is-not-of-type-array | |-- Documentation-devicetree-bindings-arm-cpus.yaml:maintainers-is-a-required-property | |-- description:Display-controller-reference-clock-source-is-not-of-type-object-boolean | |-- description:Display-controller-reference-clock-source-is-too-short | |-- description:Offset-and-length-of-the-memory-mapped-registers-is-too-short | |-- items-is-not-one-of-type-description-dependencies-dependentRequired-dependentSchemas-properties-patternProperties-additionalProperties-unevaluatedProperties-deprecated-required-not-allOf-anyOf-oneOf-r | `-- minItems-is-not-one-of-type-description-dependencies-dependentRequired-dependentSchemas-properties-patternProperties-additionalProperties-unevaluatedProperties-deprecated-required-not-allOf-anyOf-oneO `-- x86_64-allnoconfig |-- Warning:openEuler-MAINTAINERS-references-a-file-that-doesn-t-exist:Documentation-networking-hinic3.rst `-- drivers-net-ethernet-mucse-rnpm-rnpm_common.h:linux-version.h-not-needed. elapsed time: 729m configs tested: 35 configs skipped: 149 The following configs have been built successfully. More configs may be tested in the coming days. tested configs: arm64 allmodconfig clang arm64 allnoconfig gcc arm64 defconfig gcc arm64 randconfig-001-20240507 clang arm64 randconfig-002-20240507 clang arm64 randconfig-003-20240507 clang arm64 randconfig-004-20240507 clang x86_64 allnoconfig clang x86_64 allyesconfig clang x86_64 buildonly-randconfig-001-20240507 clang x86_64 buildonly-randconfig-002-20240507 clang x86_64 buildonly-randconfig-003-20240507 clang x86_64 buildonly-randconfig-004-20240507 clang x86_64 buildonly-randconfig-005-20240507 clang x86_64 buildonly-randconfig-006-20240507 gcc x86_64 defconfig gcc x86_64 randconfig-001-20240507 gcc x86_64 randconfig-002-20240507 clang x86_64 randconfig-003-20240507 clang x86_64 randconfig-004-20240507 gcc x86_64 randconfig-005-20240507 gcc x86_64 randconfig-006-20240507 gcc x86_64 randconfig-011-20240507 gcc x86_64 randconfig-012-20240507 clang x86_64 randconfig-013-20240507 clang x86_64 randconfig-014-20240507 clang x86_64 randconfig-015-20240507 clang x86_64 randconfig-016-20240507 clang x86_64 randconfig-071-20240507 gcc x86_64 randconfig-072-20240507 gcc x86_64 randconfig-073-20240507 clang x86_64 randconfig-074-20240507 clang x86_64 randconfig-075-20240507 clang x86_64 randconfig-076-20240507 clang x86_64 rhel-8.3-rust clang -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openEuler-1.0-LTS] nouveau: fix instmem race condition around ptr stores
by Yipeng Zou 07 May '24

07 May '24

From: Dave Airlie <airlied(a)redhat.com> stable inclusion from stable-v5.10 commit 13d76b2f443dc371842916dd8768009ff1594716 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9LK8M CVE: CVE-2024-26984 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit fff1386cc889d8fb4089d285f883f8cba62d82ce upstream. Running a lot of VK CTS in parallel against nouveau, once every few hours you might see something like this crash. BUG: kernel NULL pointer dereference, address: 0000000000000008 PGD 8000000114e6e067 P4D 8000000114e6e067 PUD 109046067 PMD 0 Oops: 0000 [#1] PREEMPT SMP PTI CPU: 7 PID: 53891 Comm: deqp-vk Not tainted 6.8.0-rc6+ #27 Hardware name: Gigabyte Technology Co., Ltd. Z390 I AORUS PRO WIFI/Z390 I AORUS PRO WIFI-CF, BIOS F8 11/05/2021 RIP: 0010:gp100_vmm_pgt_mem+0xe3/0x180 [nouveau] Code: c7 48 01 c8 49 89 45 58 85 d2 0f 84 95 00 00 00 41 0f b7 46 12 49 8b 7e 08 89 da 42 8d 2c f8 48 8b 47 08 41 83 c7 01 48 89 ee <48> 8b 40 08 ff d0 0f 1f 00 49 8b 7e 08 48 89 d9 48 8d 75 04 48 c1 RSP: 0000:ffffac20c5857838 EFLAGS: 00010202 RAX: 0000000000000000 RBX: 00000000004d8001 RCX: 0000000000000001 RDX: 00000000004d8001 RSI: 00000000000006d8 RDI: ffffa07afe332180 RBP: 00000000000006d8 R08: ffffac20c5857ad0 R09: 0000000000ffff10 R10: 0000000000000001 R11: ffffa07af27e2de0 R12: 000000000000001c R13: ffffac20c5857ad0 R14: ffffa07a96fe9040 R15: 000000000000001c FS: 00007fe395eed7c0(0000) GS:ffffa07e2c980000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000008 CR3: 000000011febe001 CR4: 00000000003706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ... ? gp100_vmm_pgt_mem+0xe3/0x180 [nouveau] ? gp100_vmm_pgt_mem+0x37/0x180 [nouveau] nvkm_vmm_iter+0x351/0xa20 [nouveau] ? __pfx_nvkm_vmm_ref_ptes+0x10/0x10 [nouveau] ? __pfx_gp100_vmm_pgt_mem+0x10/0x10 [nouveau] ? __pfx_gp100_vmm_pgt_mem+0x10/0x10 [nouveau] ? __lock_acquire+0x3ed/0x2170 ? __pfx_gp100_vmm_pgt_mem+0x10/0x10 [nouveau] nvkm_vmm_ptes_get_map+0xc2/0x100 [nouveau] ? __pfx_nvkm_vmm_ref_ptes+0x10/0x10 [nouveau] ? __pfx_gp100_vmm_pgt_mem+0x10/0x10 [nouveau] nvkm_vmm_map_locked+0x224/0x3a0 [nouveau] Adding any sort of useful debug usually makes it go away, so I hand wrote the function in a line, and debugged the asm. Every so often pt->memory->ptrs is NULL. This ptrs ptr is set in the nv50_instobj_acquire called from nvkm_kmap. If Thread A and Thread B both get to nv50_instobj_acquire around the same time, and Thread A hits the refcount_set line, and in lockstep thread B succeeds at refcount_inc_not_zero, there is a chance the ptrs value won't have been stored since refcount_set is unordered. Force a memory barrier here, I picked smp_mb, since we want it on all CPUs and it's write followed by a read. v2: use paired smp_rmb/smp_wmb. Cc: <stable(a)vger.kernel.org> Fixes: be55287aa5ba ("drm/nouveau/imem/nv50: embed nvkm_instobj directly into nv04_instobj") Signed-off-by: Dave Airlie <airlied(a)redhat.com> Signed-off-by: Danilo Krummrich <dakr(a)redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240411011510.2546857-1-airl… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yipeng Zou <zouyipeng(a)huawei.com> --- drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c index db48a1daca0c..f8ca79eaa7f7 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c @@ -221,8 +221,11 @@ nv50_instobj_acquire(struct nvkm_memory *memory) void __iomem *map = NULL; /* Already mapped? */ - if (refcount_inc_not_zero(&iobj->maps)) + if (refcount_inc_not_zero(&iobj->maps)) { + /* read barrier match the wmb on refcount set */ + smp_rmb(); return iobj->map; + } /* Take the lock, and re-check that another thread hasn't * already mapped the object in the meantime. @@ -249,6 +252,8 @@ nv50_instobj_acquire(struct nvkm_memory *memory) iobj->base.memory.ptrs = &nv50_instobj_fast; else iobj->base.memory.ptrs = &nv50_instobj_slow; + /* barrier to ensure the ptrs are written before refcount is set */ + smp_wmb(); refcount_set(&iobj->maps, 1); } -- 2.34.1

1 0

[PATCH OLK-5.10] kprobes: Fix check for probe enabled in kill_kprobe()
by Pu Lehui 07 May '24

07 May '24

From: Li Huafei <lihuafei1(a)huawei.com> mainline inclusion from mainline-v6.2-rc1 commit 0c76ef3f26d5ef2ac2c21b47e7620cff35809fbb category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9MXH3 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- In kill_kprobe(), the check whether disarm_kprobe_ftrace() needs to be called always fails. This is because before that we set the KPROBE_FLAG_GONE flag for kprobe so that "!kprobe_disabled(p)" is always false. The disarm_kprobe_ftrace() call introduced by commit: 0cb2f1372baa ("kprobes: Fix NULL pointer dereference at kprobe_ftrace_handler") to fix the NULL pointer reference problem. When the probe is enabled, if we do not disarm it, this problem still exists. Fix it by putting the probe enabled check before setting the KPROBE_FLAG_GONE flag. Link: https://lore.kernel.org/all/20221126114316.201857-1-lihuafei1@huawei.com/ Fixes: 3031313eb3d54 ("kprobes: Fix to check probe enabled before disarm_kprobe_ftrace()") Signed-off-by: Li Huafei <lihuafei1(a)huawei.com> Acked-by: Masami Hiramatsu (Google) <mhiramat(a)kernel.org> Reviewed-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> Signed-off-by: Masami Hiramatsu (Google) <mhiramat(a)kernel.org> Signed-off-by: Pu Lehui <pulehui(a)huawei.com> --- kernel/kprobes.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 6fd66a47b476..325a81f9c38f 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -2281,6 +2281,14 @@ static void kill_kprobe(struct kprobe *p) if (WARN_ON_ONCE(kprobe_gone(p))) return; + /* + * The module is going away. We should disarm the kprobe which + * is using ftrace, because ftrace framework is still available at + * MODULE_STATE_GOING notification. + */ + if (kprobe_ftrace(p) && !kprobe_disabled(p) && !kprobes_all_disarmed) + disarm_kprobe_ftrace(p); + p->flags |= KPROBE_FLAG_GONE; if (kprobe_aggrprobe(p)) { /* @@ -2297,14 +2305,6 @@ static void kill_kprobe(struct kprobe *p) * the original probed function (which will be freed soon) any more. */ arch_remove_kprobe(p); - - /* - * The module is going away. We should disarm the kprobe which - * is using ftrace, because ftrace framework is still available at - * MODULE_STATE_GOING notification. - */ - if (kprobe_ftrace(p) && !kprobe_disabled(p) && !kprobes_all_disarmed) - disarm_kprobe_ftrace(p); } /* Disable one kprobe */ -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] fs: sysfs: Fix reference leak in sysfs_break_active_protection()
by Yifan Qiao 07 May '24

07 May '24

From: Alan Stern <stern(a)rowland.harvard.edu> stable inclusion from stable-v6.8.8 commit a4c99b57d43bab45225ba92d574a8683f9edc8e4 category: bugfix bugzilla: 189875 CVE: CVE-2024-26993 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit a90bca2228c0646fc29a72689d308e5fe03e6d78 upstream. The sysfs_break_active_protection() routine has an obvious reference leak in its error path. If the call to kernfs_find_and_get() fails then kn will be NULL, so the companion sysfs_unbreak_active_protection() routine won't get called (and would only cause an access violation by trying to dereference kn->parent if it was called). As a result, the reference to kobj acquired at the start of the function will never be released. Fix the leak by adding an explicit kobject_put() call when kn is NULL. Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu> Fixes: 2afc9166f79b ("scsi: sysfs: Introduce sysfs_{un,}break_active_protection()") Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: stable(a)vger.kernel.org Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Acked-by: Tejun Heo <tj(a)kernel.org> Link: https://lore.kernel.org/r/8a4d3f0f-c5e3-4b70-a188-0ca433f9e6f9@rowland.harv… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/sysfs/file.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index a12ac0356c69..f21e73d10724 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -450,6 +450,8 @@ struct kernfs_node *sysfs_break_active_protection(struct kobject *kobj, kn = kernfs_find_and_get(kobj->sd, attr->name); if (kn) kernfs_break_active_protection(kn); + else + kobject_put(kobj); return kn; } EXPORT_SYMBOL_GPL(sysfs_break_active_protection); -- 2.39.2

2 1

[PATCH OLK-5.10] fs: sysfs: Fix reference leak in sysfs_break_active_protection()
by Yifan Qiao 07 May '24

07 May '24

From: Alan Stern <stern(a)rowland.harvard.edu> stable inclusion from stable-v6.8.8 commit a4c99b57d43bab45225ba92d574a8683f9edc8e4 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5IQ CVE: CVE-2024-26993 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit a90bca2228c0646fc29a72689d308e5fe03e6d78 upstream. The sysfs_break_active_protection() routine has an obvious reference leak in its error path. If the call to kernfs_find_and_get() fails then kn will be NULL, so the companion sysfs_unbreak_active_protection() routine won't get called (and would only cause an access violation by trying to dereference kn->parent if it was called). As a result, the reference to kobj acquired at the start of the function will never be released. Fix the leak by adding an explicit kobject_put() call when kn is NULL. Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu> Fixes: 2afc9166f79b ("scsi: sysfs: Introduce sysfs_{un,}break_active_protection()") Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: stable(a)vger.kernel.org Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Acked-by: Tejun Heo <tj(a)kernel.org> Link: https://lore.kernel.org/r/8a4d3f0f-c5e3-4b70-a188-0ca433f9e6f9@rowland.harv… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/sysfs/file.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index a12ac0356c69..f21e73d10724 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -450,6 +450,8 @@ struct kernfs_node *sysfs_break_active_protection(struct kobject *kobj, kn = kernfs_find_and_get(kobj->sd, attr->name); if (kn) kernfs_break_active_protection(kn); + else + kobject_put(kobj); return kn; } EXPORT_SYMBOL_GPL(sysfs_break_active_protection); -- 2.39.2

2 1

[PATCH OLK-6.6] fs: sysfs: Fix reference leak in sysfs_break_active_protection()
by Yifan Qiao 07 May '24

07 May '24

From: Alan Stern <stern(a)rowland.harvard.edu> stable inclusion from stable-v6.8.8 commit a4c99b57d43bab45225ba92d574a8683f9edc8e4 category: bugfix bugzilla: 189875 CVE: CVE-2024-26993 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit a90bca2228c0646fc29a72689d308e5fe03e6d78 upstream. The sysfs_break_active_protection() routine has an obvious reference leak in its error path. If the call to kernfs_find_and_get() fails then kn will be NULL, so the companion sysfs_unbreak_active_protection() routine won't get called (and would only cause an access violation by trying to dereference kn->parent if it was called). As a result, the reference to kobj acquired at the start of the function will never be released. Fix the leak by adding an explicit kobject_put() call when kn is NULL. Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu> Fixes: 2afc9166f79b ("scsi: sysfs: Introduce sysfs_{un,}break_active_protection()") Cc: Bart Van Assche <bvanassche(a)acm.org> Cc: stable(a)vger.kernel.org Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Acked-by: Tejun Heo <tj(a)kernel.org> Link: https://lore.kernel.org/r/8a4d3f0f-c5e3-4b70-a188-0ca433f9e6f9@rowland.harv… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/sysfs/file.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index a12ac0356c69..f21e73d10724 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -450,6 +450,8 @@ struct kernfs_node *sysfs_break_active_protection(struct kobject *kobj, kn = kernfs_find_and_get(kobj->sd, attr->name); if (kn) kernfs_break_active_protection(kn); + else + kobject_put(kobj); return kn; } EXPORT_SYMBOL_GPL(sysfs_break_active_protection); -- 2.39.2

2 1

[PATCH OLK-5.10] perf/x86/amd: Fix crash due to race between amd_pmu_enable_all, perf NMI and throttling
by Luo Gengkun 07 May '24

07 May '24

From: Ravi Bangoria <ravi.bangoria(a)amd.com> mainline inclusion from mainline-v6.1-rc6v6.1 commit baa014b9543c8e5e94f5d15b66abfe60750b8284 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9MMNG Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- amd_pmu_enable_all() does: if (!test_bit(idx, cpuc->active_mask)) continue; amd_pmu_enable_event(cpuc->events[idx]); A perf NMI of another event can come between these two steps. Perf NMI handler internally disables and enables _all_ events, including the one which nmi-intercepted amd_pmu_enable_all() was in process of enabling. If that unintentionally enabled event has very low sampling period and causes immediate successive NMI, causing the event to be throttled, cpuc->events[idx] and cpuc->active_mask gets cleared by x86_pmu_stop(). This will result in amd_pmu_enable_event() getting called with event=NULL when amd_pmu_enable_all() resumes after handling the NMIs. This causes a kernel crash: BUG: kernel NULL pointer dereference, address: 0000000000000198 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page [...] Call Trace: <TASK> amd_pmu_enable_all+0x68/0xb0 ctx_resched+0xd9/0x150 event_function+0xb8/0x130 ? hrtimer_start_range_ns+0x141/0x4a0 ? perf_duration_warn+0x30/0x30 remote_function+0x4d/0x60 __flush_smp_call_function_queue+0xc4/0x500 flush_smp_call_function_queue+0x11d/0x1b0 do_idle+0x18f/0x2d0 cpu_startup_entry+0x19/0x20 start_secondary+0x121/0x160 secondary_startup_64_no_verify+0xe5/0xeb </TASK> amd_pmu_disable_all()/amd_pmu_enable_all() calls inside perf NMI handler were recently added as part of BRS enablement but I'm not sure whether we really need them. We can just disable BRS in the beginning and enable it back while returning from NMI. This will solve the issue by not enabling those events whose active_masks are set but are not yet enabled in hw pmu. Fixes: ada543459cab ("perf/x86/amd: Add AMD Fam19h Branch Sampling support") Reported-by: Linux Kernel Functional Testing <lkft(a)linaro.org> Signed-off-by: Ravi Bangoria <ravi.bangoria(a)amd.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Link: https://lkml.kernel.org/r/20221114044029.373-1-ravi.bangoria@amd.com Signed-off-by: Luo Gengkun <luogengkun2(a)huawei.com> --- arch/x86/events/amd/core.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/arch/x86/events/amd/core.c b/arch/x86/events/amd/core.c index 2c082abb930b..3127519281f5 100644 --- a/arch/x86/events/amd/core.c +++ b/arch/x86/events/amd/core.c @@ -900,8 +900,7 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) pmu_enabled = cpuc->enabled; cpuc->enabled = 0; - /* stop everything (includes BRS) */ - amd_pmu_disable_all(); + amd_brs_disable_all(); /* Drain BRS is in use (could be inactive) */ if (cpuc->lbr_users) @@ -912,7 +911,7 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) cpuc->enabled = pmu_enabled; if (pmu_enabled) - amd_pmu_enable_all(0); + amd_brs_enable_all(); return amd_pmu_adjust_nmi_window(handled); } -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] fat: fix uninitialized field in nostale filehandles
by Yifan Qiao 07 May '24

07 May '24

From: Jan Kara <jack(a)suse.cz> stable inclusion from stable-v6.8.3 commit cdd33d54e789d229d6d5007cbf3f53965ca1a5c6 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5QJ CVE: CVE-2024-26973 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit fde2497d2bc3a063d8af88b258dbadc86bd7b57c ] When fat_encode_fh_nostale() encodes file handle without a parent it stores only first 10 bytes of the file handle. However the length of the file handle must be a multiple of 4 so the file handle is actually 12 bytes long and the last two bytes remain uninitialized. This is not great at we potentially leak uninitialized information with the handle to userspace. Properly initialize the full handle length. Link: https://lkml.kernel.org/r/20240205122626.13701-1-jack@suse.cz Reported-by: syzbot+3ce5dea5b1539ff36769(a)syzkaller.appspotmail.com Fixes: ea3983ace6b7 ("fat: restructure export_operations") Signed-off-by: Jan Kara <jack(a)suse.cz> Acked-by: OGAWA Hirofumi <hirofumi(a)mail.parknet.co.jp> Cc: Amir Goldstein <amir73il(a)gmail.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/fat/nfs.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/fs/fat/nfs.c b/fs/fat/nfs.c index af191371c352..bab63eeaf9cb 100644 --- a/fs/fat/nfs.c +++ b/fs/fat/nfs.c @@ -130,6 +130,12 @@ fat_encode_fh_nostale(struct inode *inode, __u32 *fh, int *lenp, fid->parent_i_gen = parent->i_generation; type = FILEID_FAT_WITH_PARENT; *lenp = FAT_FID_SIZE_WITH_PARENT; + } else { + /* + * We need to initialize this field because the fh is actually + * 12 bytes long + */ + fid->parent_i_pos_hi = 0; } return type; -- 2.39.2

2 1

[PATCH OLK-5.10] fat: fix uninitialized field in nostale filehandles
by Yifan Qiao 07 May '24

07 May '24

From: Jan Kara <jack(a)suse.cz> stable inclusion from stable-v6.8.3 commit cdd33d54e789d229d6d5007cbf3f53965ca1a5c6 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5QJ CVE: CVE-2024-26973 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit fde2497d2bc3a063d8af88b258dbadc86bd7b57c ] When fat_encode_fh_nostale() encodes file handle without a parent it stores only first 10 bytes of the file handle. However the length of the file handle must be a multiple of 4 so the file handle is actually 12 bytes long and the last two bytes remain uninitialized. This is not great at we potentially leak uninitialized information with the handle to userspace. Properly initialize the full handle length. Link: https://lkml.kernel.org/r/20240205122626.13701-1-jack@suse.cz Reported-by: syzbot+3ce5dea5b1539ff36769(a)syzkaller.appspotmail.com Fixes: ea3983ace6b7 ("fat: restructure export_operations") Signed-off-by: Jan Kara <jack(a)suse.cz> Acked-by: OGAWA Hirofumi <hirofumi(a)mail.parknet.co.jp> Cc: Amir Goldstein <amir73il(a)gmail.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/fat/nfs.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/fs/fat/nfs.c b/fs/fat/nfs.c index af191371c352..bab63eeaf9cb 100644 --- a/fs/fat/nfs.c +++ b/fs/fat/nfs.c @@ -130,6 +130,12 @@ fat_encode_fh_nostale(struct inode *inode, __u32 *fh, int *lenp, fid->parent_i_gen = parent->i_generation; type = FILEID_FAT_WITH_PARENT; *lenp = FAT_FID_SIZE_WITH_PARENT; + } else { + /* + * We need to initialize this field because the fh is actually + * 12 bytes long + */ + fid->parent_i_pos_hi = 0; } return type; -- 2.39.2

2 1

[PATCH OLK-6.6] fat: fix uninitialized field in nostale filehandles
by Yifan Qiao 07 May '24

07 May '24

From: Jan Kara <jack(a)suse.cz> stable inclusion from stable-v6.8.3 commit cdd33d54e789d229d6d5007cbf3f53965ca1a5c6 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L5QJ CVE: CVE-2024-26973 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit fde2497d2bc3a063d8af88b258dbadc86bd7b57c ] When fat_encode_fh_nostale() encodes file handle without a parent it stores only first 10 bytes of the file handle. However the length of the file handle must be a multiple of 4 so the file handle is actually 12 bytes long and the last two bytes remain uninitialized. This is not great at we potentially leak uninitialized information with the handle to userspace. Properly initialize the full handle length. Link: https://lkml.kernel.org/r/20240205122626.13701-1-jack@suse.cz Reported-by: syzbot+3ce5dea5b1539ff36769(a)syzkaller.appspotmail.com Fixes: ea3983ace6b7 ("fat: restructure export_operations") Signed-off-by: Jan Kara <jack(a)suse.cz> Acked-by: OGAWA Hirofumi <hirofumi(a)mail.parknet.co.jp> Cc: Amir Goldstein <amir73il(a)gmail.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yifan Qiao <qiaoyifan4(a)huawei.com> --- fs/fat/nfs.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/fs/fat/nfs.c b/fs/fat/nfs.c index af191371c352..bab63eeaf9cb 100644 --- a/fs/fat/nfs.c +++ b/fs/fat/nfs.c @@ -130,6 +130,12 @@ fat_encode_fh_nostale(struct inode *inode, __u32 *fh, int *lenp, fid->parent_i_gen = parent->i_generation; type = FILEID_FAT_WITH_PARENT; *lenp = FAT_FID_SIZE_WITH_PARENT; + } else { + /* + * We need to initialize this field because the fh is actually + * 12 bytes long + */ + fid->parent_i_pos_hi = 0; } return type; -- 2.39.2

2 1

[openEuler-1.0-LTS] media: v4l2-mem2mem: fix a memleak in v4l2_m2m_register_entity
by Yipeng Zou 07 May '24

07 May '24

From: Zhipeng Lu <alexious(a)zju.edu.cn> stable inclusion from stable-v5.10.214 commit afd2a82fe300032f63f8be5d6cd6981e75f8bbf2 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9L9JK CVE: CVE-2024-27077 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- The entity->name (i.e. name) is allocated in v4l2_m2m_register_entity but isn't freed in its following error-handling paths. This patch adds such deallocation to prevent memleak of entity->name. Fixes: be2fff656322 ("media: add helpers for memory-to-memory media controller") Signed-off-by: Zhipeng Lu <alexious(a)zju.edu.cn> Signed-off-by: Hans Verkuil <hverkuil-cisco(a)xs4all.nl> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yipeng Zou <zouyipeng(a)huawei.com> --- drivers/media/v4l2-core/v4l2-mem2mem.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/media/v4l2-core/v4l2-mem2mem.c b/drivers/media/v4l2-core/v4l2-mem2mem.c index 692b0597a35c..0a9be4c5f194 100644 --- a/drivers/media/v4l2-core/v4l2-mem2mem.c +++ b/drivers/media/v4l2-core/v4l2-mem2mem.c @@ -773,11 +773,17 @@ static int v4l2_m2m_register_entity(struct media_device *mdev, entity->function = function; ret = media_entity_pads_init(entity, num_pads, pads); - if (ret) + if (ret) { + kfree(entity->name); + entity->name = NULL; return ret; + } ret = media_device_register_entity(mdev, entity); - if (ret) + if (ret) { + kfree(entity->name); + entity->name = NULL; return ret; + } return 0; } -- 2.34.1

1 0

[PATCH OLK-6.6 0/2] Introduce CONFIG_ARCH_CUSTOM_NUMA_DISTANCE
by Hui Tang 07 May '24

07 May '24

This patchset allow arch adjust node_relaim_distance, as follow: Hui Tang (2): arm64/numa: Support node_reclaim_distance adjust for arch config: enable COBFIG_ARCH_CUSTOM_NUMA_DISTANCE for arm64 arch/arm64/Kconfig | 13 ++++ arch/arm64/configs/openeuler_defconfig | 1 + drivers/base/arch_numa.c | 85 ++++++++++++++++++++++++++ 3 files changed, 99 insertions(+) -- 2.34.1

2 3

[openeuler:openEuler-1.0-LTS] BUILD SUCCESS 48d720404c6ef5c0c29a6019fa34590b3632cf4c
by kernel test robot 07 May '24

07 May '24

tree/branch: https://gitee.com/openeuler/kernel.git openEuler-1.0-LTS branch HEAD: 48d720404c6ef5c0c29a6019fa34590b3632cf4c !6705 quota: fix CVE-2024-26878 Warning ids grouped by kconfigs: gcc_recent_errors |-- x86_64-buildonly-randconfig-002-20240506 | |-- fs-proc-array.c:warning:gtime-may-be-used-uninitialized-in-this-function | |-- fs-proc-array.c:warning:maj_flt-may-be-used-uninitialized-in-this-function | `-- fs-proc-array.c:warning:min_flt-may-be-used-uninitialized-in-this-function |-- x86_64-buildonly-randconfig-006-20240506 | |-- fs-proc-array.c:warning:gtime-may-be-used-uninitialized | |-- fs-proc-array.c:warning:maj_flt-may-be-used-uninitialized | `-- fs-proc-array.c:warning:min_flt-may-be-used-uninitialized |-- x86_64-defconfig | |-- include-linux-list.h:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node | |-- include-linux-plist.h:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node | `-- mm-swapfile.c:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node |-- x86_64-randconfig-014-20240506 | |-- include-linux-compiler.h:warning:array-subscript-index-is-outside-array-bounds-of-u32-aka-unsigned-int | `-- include-linux-compiler.h:warning:array-subscript-unknown-is-outside-array-bounds-of-const-u32-aka-const-unsigned-int |-- x86_64-randconfig-071-20240506 | |-- include-linux-compiler.h:warning:array-subscript-index-is-outside-array-bounds-of-u32-aka-unsigned-int | `-- include-linux-compiler.h:warning:array-subscript-unknown-is-outside-array-bounds-of-const-u32-aka-const-unsigned-int |-- x86_64-randconfig-073-20240506 | |-- include-linux-compiler.h:warning:array-subscript-index-is-outside-array-bounds-of-u32-aka-unsigned-int | |-- include-linux-compiler.h:warning:array-subscript-unknown-is-outside-array-bounds-of-const-u32-aka-const-unsigned-int | |-- include-linux-list.h:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node | |-- include-linux-plist.h:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node | `-- mm-swapfile.c:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node |-- x86_64-randconfig-104-20240506 | |-- include-linux-compiler.h:warning:array-subscript-index-is-outside-array-bounds-of-u32-aka-unsigned-int | `-- include-linux-compiler.h:warning:array-subscript-unknown-is-outside-array-bounds-of-const-u32-aka-const-unsigned-int |-- x86_64-randconfig-121-20240507 | `-- net-ipv4-arp.c:sparse:sparse:incompatible-types-in-comparison-expression-(different-type-sizes): `-- x86_64-randconfig-123-20240507 |-- include-linux-compiler.h:warning:array-subscript-index-is-outside-array-bounds-of-u32-aka-unsigned-int |-- include-linux-compiler.h:warning:array-subscript-unknown-is-outside-array-bounds-of-const-u32-aka-const-unsigned-int |-- include-linux-list.h:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node |-- include-linux-plist.h:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node |-- mm-swapfile.c:warning:array-subscript-pfo_ret__-is-outside-array-bounds-of-struct-plist_node `-- net-ipv4-arp.c:sparse:sparse:incompatible-types-in-comparison-expression-(different-type-sizes): clang_recent_errors |-- x86_64-allyesconfig | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-001-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-002-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-003-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-004-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-005-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-006-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-072-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-074-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-076-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-101-20240506 | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) |-- x86_64-randconfig-122-20240507 | |-- net-ipv4-arp.c:sparse:sparse:incompatible-types-in-comparison-expression-(different-type-sizes): | `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) `-- x86_64-rhel-8.3-rust `-- net-ipv4-arp.c:warning:comparison-of-distinct-pointer-types-(-typeof-(dev-addr_len)-(aka-unsigned-char-)-and-typeof-(sizeof-(r-arp_ha.sa_data))-(aka-unsigned-long-)) elapsed time: 721m configs tested: 35 configs skipped: 148 The following configs have been built successfully. More configs may be tested in the coming days. tested configs: arm64 allmodconfig gcc arm64 allnoconfig gcc arm64 defconfig gcc arm64 randconfig-001-20240506 gcc arm64 randconfig-002-20240506 gcc arm64 randconfig-003-20240506 gcc arm64 randconfig-004-20240506 gcc x86_64 allnoconfig clang x86_64 allyesconfig clang x86_64 buildonly-randconfig-001-20240506 gcc x86_64 buildonly-randconfig-002-20240506 gcc x86_64 buildonly-randconfig-003-20240506 gcc x86_64 buildonly-randconfig-004-20240506 clang x86_64 buildonly-randconfig-005-20240506 clang x86_64 buildonly-randconfig-006-20240506 gcc x86_64 defconfig gcc x86_64 randconfig-001-20240506 clang x86_64 randconfig-002-20240506 clang x86_64 randconfig-003-20240506 clang x86_64 randconfig-004-20240506 clang x86_64 randconfig-005-20240506 clang x86_64 randconfig-006-20240506 clang x86_64 randconfig-011-20240506 gcc x86_64 randconfig-012-20240506 gcc x86_64 randconfig-013-20240506 gcc x86_64 randconfig-014-20240506 gcc x86_64 randconfig-015-20240506 gcc x86_64 randconfig-016-20240506 gcc x86_64 randconfig-071-20240506 gcc x86_64 randconfig-072-20240506 clang x86_64 randconfig-073-20240506 gcc x86_64 randconfig-074-20240506 clang x86_64 randconfig-075-20240506 gcc x86_64 randconfig-076-20240506 clang x86_64 rhel-8.3-rust clang -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH OLK-6.6] sched/fair: Optimize performance by simplifying cpu_util_without()
by Li Zetao 06 May '24

06 May '24

hulk inclusion category: performance bugzilla: https://gitee.com/openeuler/kernel/issues/I9MSNE?from=project-issue CVE: NA -------------------------------- Considering that the high-frequency function cpu_util without is only called when waking up or creating for the first time, in this scenario, the performance can be optimized by simplifying the function. Signed-off-by: Zhang Qiao <zhangqiao22(a)huawei.com> Signed-off-by: Li Zetao <lizetao1(a)huawei.com> --- kernel/sched/fair.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7b0cb2f090da..010dbf2047e5 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -8530,13 +8530,23 @@ unsigned long cpu_util_cfs_boost(int cpu) * utilization of the specified task, whenever the task is currently * contributing to the CPU utilization. */ -static unsigned long cpu_util_without(int cpu, struct task_struct *p) +static inline unsigned long cpu_util_without(int cpu, struct task_struct *p) { - /* Task has no contribution or is new */ - if (cpu != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time)) - p = NULL; + struct cfs_rq *cfs_rq = &cpu_rq(cpu)->cfs; + unsigned long util = READ_ONCE(cfs_rq->avg.util_avg); + /* + * If @dst_cpu is -1 or @p migrates from @cpu to @dst_cpu remove its + * contribution. If @p migrates from another CPU to @cpu add its + * contribution. In all the other cases @cpu is not impacted by the + * migration so its util_avg is already correct. + */ + if (sched_feat(UTIL_EST)) { + unsigned long util_est; + util_est = READ_ONCE(cfs_rq->avg.util_est.enqueued); + util = max(util, util_est); + } - return cpu_util(cpu, p, -1, 0); + return min(util, capacity_orig_of(cpu)); } /* -- 2.34.1

2 1

[PATCH OLK-6.6 v3 0/2] block: fix discard ioctl
by Li Nan 06 May '24

06 May '24

Fix patch format. Li Nan (2): block: fix overflow in blk_ioctl_discard() block: check io size before submit discard block/blk-lib.c | 8 ++++++++ block/ioctl.c | 5 +++-- 2 files changed, 11 insertions(+), 2 deletions(-) -- 2.39.2

2 3

[PATCH openEuler-1.0-LTS] s390/dasd: fix Oops in dasd_alias_get_start_dev due to missing pavgroup
by Li Zetao 06 May '24

06 May '24

From: Stefan Haberland <sth(a)linux.ibm.com> stable inclusion from stable-v4.19.260 commit aaba5ff2742043705bc4c02fd0b2b246e2e16da1 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9KHGT CVE: CVE-2022-48636 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit db7ba07108a48c0f95b74fabbfd5d63e924f992d upstream. Fix Oops in dasd_alias_get_start_dev() function caused by the pavgroup pointer being NULL. The pavgroup pointer is checked on the entrance of the function but without the lcu->lock being held. Therefore there is a race window between dasd_alias_get_start_dev() and _lcu_update() which sets pavgroup to NULL with the lcu->lock held. Fix by checking the pavgroup pointer with lcu->lock held. Cc: <stable(a)vger.kernel.org> # 2.6.25+ Fixes: 8e09f21574ea ("[S390] dasd: add hyper PAV support to DASD device driver, part 1") Signed-off-by: Stefan Haberland <sth(a)linux.ibm.com> Reviewed-by: Jan Hoeppner <hoeppner(a)linux.ibm.com> Link: https://lore.kernel.org/r/20220919154931.4123002-2-sth@linux.ibm.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Li Zetao <lizetao1(a)huawei.com> --- drivers/s390/block/dasd_alias.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/s390/block/dasd_alias.c b/drivers/s390/block/dasd_alias.c index 99f86612f775..88530a4952a9 100644 --- a/drivers/s390/block/dasd_alias.c +++ b/drivers/s390/block/dasd_alias.c @@ -657,12 +657,12 @@ int dasd_alias_remove_device(struct dasd_device *device) struct dasd_device *dasd_alias_get_start_dev(struct dasd_device *base_device) { struct dasd_eckd_private *alias_priv, *private = base_device->private; - struct alias_pav_group *group = private->pavgroup; struct alias_lcu *lcu = private->lcu; struct dasd_device *alias_device; + struct alias_pav_group *group; unsigned long flags; - if (!group || !lcu) + if (!lcu) return NULL; if (lcu->pav == NO_PAV || lcu->flags & (NEED_UAC_UPDATE | UPDATE_PENDING)) @@ -679,6 +679,11 @@ struct dasd_device *dasd_alias_get_start_dev(struct dasd_device *base_device) } spin_lock_irqsave(&lcu->lock, flags); + group = private->pavgroup; + if (!group) { + spin_unlock_irqrestore(&lcu->lock, flags); + return NULL; + } alias_device = group->next; if (!alias_device) { if (list_empty(&group->aliaslist)) { -- 2.34.1

2 3

[PATCH OLK-6.6 v2 0/2] block: fix discard ioctl
by Li Nan 06 May '24

06 May '24

Li Nan (2): block: fix overflow in blk_ioctl_discard() block: check io size before submit discard block/blk-lib.c | 8 ++++++++ block/ioctl.c | 5 +++-- 2 files changed, 11 insertions(+), 2 deletions(-) -- 2.39.2

2 3

[PATCH OLK-5.10] mm: memcg: fix stale protection of reclaim target memcg
by Cai Xinchen 06 May '24

06 May '24

From: Yosry Ahmed <yosryahmed(a)google.com> mainline inclusion from mainline-v6.2-rc1 commit adb8213014b25c7f1d75d5b219becaadcd695efb category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9MD18 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Patch series "mm: memcg: fix protection of reclaim target memcg", v3. This series fixes a bug in calculating the protection of the reclaim target memcg where we end up using stale effective protection values from the last reclaim operation, instead of completely ignoring the protection of the reclaim target as intended. More detailed explanation and examples in patch 1, which includes the fix. Patches 2 & 3 introduce a selftest case that catches the bug. This patch (of 3): When we are doing memcg reclaim, the intended behavior is that we ignore any protection (memory.min, memory.low) of the target memcg (but not its children). Ever since the patch pointed to by the "Fixes" tag, we actually read a stale value for the target memcg protection when deciding whether to skip the memcg or not because it is protected. If the stale value happens to be high enough, we don't reclaim from the target memcg. Essentially, in some cases we may falsely skip reclaiming from the target memcg of reclaim because we read a stale protection value from last time we reclaimed from it. During reclaim, mem_cgroup_calculate_protection() is used to determine the effective protection (emin and elow) values of a memcg. The protection of the reclaim target is ignored, but we cannot set their effective protection to 0 due to a limitation of the current implementation (see comment in mem_cgroup_protection()). Instead, we leave their effective protection values unchaged, and later ignore it in mem_cgroup_protection(). However, mem_cgroup_protection() is called later in shrink_lruvec()->get_scan_count(), which is after the mem_cgroup_below_{min/low}() checks in shrink_node_memcgs(). As a result, the stale effective protection values of the target memcg may lead us to skip reclaiming from the target memcg entirely, before calling shrink_lruvec(). This can be even worse with recursive protection, where the stale target memcg protection can be higher than its standalone protection. See two examples below (a similar version of example (a) is added to test_memcontrol in a later patch). (a) A simple example with proactive reclaim is as follows. Consider the following hierarchy: ROOT | A | B (memory.min = 10M) Consider the following scenario: - B has memory.current = 10M. - The system undergoes global reclaim (or memcg reclaim in A). - In shrink_node_memcgs(): - mem_cgroup_calculate_protection() calculates the effective min (emin) of B as 10M. - mem_cgroup_below_min() returns true for B, we do not reclaim from B. - Now if we want to reclaim 5M from B using proactive reclaim (memory.reclaim), we should be able to, as the protection of the target memcg should be ignored. - In shrink_node_memcgs(): - mem_cgroup_calculate_protection() immediately returns for B without doing anything, as B is the target memcg, relying on mem_cgroup_protection() to ignore B's stale effective min (still 10M). - mem_cgroup_below_min() reads the stale effective min for B and we skip it instead of ignoring its protection as intended, as we never reach mem_cgroup_protection(). (b) An more complex example with recursive protection is as follows. Consider the following hierarchy with memory_recursiveprot: ROOT | A (memory.min = 50M) | B (memory.min = 10M, memory.high = 40M) Consider the following scenario: - B has memory.current = 35M. - The system undergoes global reclaim (target memcg is NULL). - B will have an effective min of 50M (all of A's unclaimed protection). - B will not be reclaimed from. - Now allocate 10M more memory in B, pushing it above it's high limit. - The system undergoes memcg reclaim from B (target memcg is B). - Like example (a), we do nothing in mem_cgroup_calculate_protection(), then call mem_cgroup_below_min(), which will read the stale effective min for B (50M) and skip it. In this case, it's even worse because we are not just considering B's standalone protection (10M), but we are reading a much higher stale protection (50M) which will cause us to not reclaim from B at all. This is an artifact of commit 45c7f7e1ef17 ("mm, memcg: decouple e{low,min} state mutations from protection checks") which made mem_cgroup_calculate_protection() only change the state without returning any value. Before that commit, we used to return MEMCG_PROT_NONE for the target memcg, which would cause us to skip the mem_cgroup_below_{min/low}() checks. After that commit we do not return anything and we end up checking the min & low effective protections for the target memcg, which are stale. Update mem_cgroup_supports_protection() to also check if we are reclaiming from the target, and rename it to mem_cgroup_unprotected() (now returns true if we should not protect the memcg, much simpler logic). Link: https://lkml.kernel.org/r/20221202031512.1365483-1-yosryahmed@google.com Link: https://lkml.kernel.org/r/20221202031512.1365483-2-yosryahmed@google.com Fixes: 45c7f7e1ef17 ("mm, memcg: decouple e{low,min} state mutations from protection checks") Signed-off-by: Yosry Ahmed <yosryahmed(a)google.com> Reviewed-by: Roman Gushchin <roman.gushchin(a)linux.dev> Cc: Chris Down <chris(a)chrisdown.name> Cc: David Rientjes <rientjes(a)google.com> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Matthew Wilcox <willy(a)infradead.org> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Muchun Song <songmuchun(a)bytedance.com> Cc: Shakeel Butt <shakeelb(a)google.com> Cc: Tejun Heo <tj(a)kernel.org> Cc: Vasily Averin <vasily.averin(a)linux.dev> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: Yu Zhao <yuzhao(a)google.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Conflicts: mm/vmscan.c Signed-off-by: Cai Xinchen <caixinchen1(a)huawei.com> --- include/linux/memcontrol.h | 31 +++++++++++++++++++++---------- mm/vmscan.c | 4 ++-- 2 files changed, 23 insertions(+), 12 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 24794b56eaaf..287c54141a90 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -769,28 +769,32 @@ static inline void mem_cgroup_protection(struct mem_cgroup *root, void mem_cgroup_calculate_protection(struct mem_cgroup *root, struct mem_cgroup *memcg); -static inline bool mem_cgroup_supports_protection(struct mem_cgroup *memcg) +static inline bool mem_cgroup_unprotected(struct mem_cgroup *target, + struct mem_cgroup *memcg) { /* * The root memcg doesn't account charges, and doesn't support - * protection. + * protection. The target memcg's protection is ignored, see + * mem_cgroup_calculate_protection() and mem_cgroup_protection() */ - return !mem_cgroup_disabled() && !mem_cgroup_is_root(memcg); - + return mem_cgroup_disabled() || mem_cgroup_is_root(memcg) || + memcg == target; } -static inline bool mem_cgroup_below_low(struct mem_cgroup *memcg) +static inline bool mem_cgroup_below_low(struct mem_cgroup *target, + struct mem_cgroup *memcg) { - if (!mem_cgroup_supports_protection(memcg)) + if (mem_cgroup_unprotected(target, memcg)) return false; return READ_ONCE(memcg->memory.elow) >= page_counter_read(&memcg->memory); } -static inline bool mem_cgroup_below_min(struct mem_cgroup *memcg) +static inline bool mem_cgroup_below_min(struct mem_cgroup *target, + struct mem_cgroup *memcg) { - if (!mem_cgroup_supports_protection(memcg)) + if (mem_cgroup_unprotected(target, memcg)) return false; return READ_ONCE(memcg->memory.emin) >= @@ -1378,12 +1382,19 @@ static inline void mem_cgroup_calculate_protection(struct mem_cgroup *root, { } -static inline bool mem_cgroup_below_low(struct mem_cgroup *memcg) +static inline bool mem_cgroup_unprotected(struct mem_cgroup *target, + struct mem_cgroup *memcg) +{ + return true; +} +static inline bool mem_cgroup_below_low(struct mem_cgroup *target, + struct mem_cgroup *memcg) { return false; } -static inline bool mem_cgroup_below_min(struct mem_cgroup *memcg) +static inline bool mem_cgroup_below_min(struct mem_cgroup *target, + struct mem_cgroup *memcg) { return false; } diff --git a/mm/vmscan.c b/mm/vmscan.c index 3d383c7126e3..044bf496885b 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2860,13 +2860,13 @@ static void shrink_node_memcgs(pg_data_t *pgdat, struct scan_control *sc) mem_cgroup_calculate_protection(target_memcg, memcg); - if (mem_cgroup_below_min(memcg)) { + if (mem_cgroup_below_min(target_memcg, memcg)) { /* * Hard protection. * If there is no reclaimable memory, OOM. */ continue; - } else if (mem_cgroup_below_low(memcg)) { + } else if (mem_cgroup_below_low(target_memcg, memcg)) { /* * Soft protection. * Respect the protection only as long as -- 2.34.1

2 1

[PATCH OLK-5.10] drivers: base: transport_class: fix possible memory leak
by dinglongwei 06 May '24

06 May '24

From: Yang Yingliang <yangyingliang(a)huawei.com> mainline inclusion from mainline-v6.3-rc1 commit a86367803838b369fe5486ac18771d14723c258c category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9MMNG Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Current some drivers(like iscsi) call transport_register_device() failed, they don't call transport_destroy_device() to release the memory allocated in transport_setup_device(), because they don't know what was done, it should be internal thing to release the resource in register function. So fix this leak by calling destroy function inside register function. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Link: https://lore.kernel.org/r/20221110102307.3492557-1-yangyingliang@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: dinglongwei <dinglongwei1(a)huawei.com> --- include/linux/transport_class.h | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/include/linux/transport_class.h b/include/linux/transport_class.h index 63076fb835e3..2efc271a96fa 100644 --- a/include/linux/transport_class.h +++ b/include/linux/transport_class.h @@ -70,8 +70,14 @@ void transport_destroy_device(struct device *); static inline int transport_register_device(struct device *dev) { + int ret; + transport_setup_device(dev); - return transport_add_device(dev); + ret = transport_add_device(dev); + if (ret) + transport_destroy_device(dev); + + return ret; } static inline void -- 2.17.1

2 1

[PATCH] mm: memcg: fix stale protection of reclaim target memcg
by Cai Xinchen 06 May '24

06 May '24

From: Yosry Ahmed <yosryahmed(a)google.com> mainline inclusion from mainline-v6.2-rc1 commit adb8213014b25c7f1d75d5b219becaadcd695efb category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9MD18 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Patch series "mm: memcg: fix protection of reclaim target memcg", v3. This series fixes a bug in calculating the protection of the reclaim target memcg where we end up using stale effective protection values from the last reclaim operation, instead of completely ignoring the protection of the reclaim target as intended. More detailed explanation and examples in patch 1, which includes the fix. Patches 2 & 3 introduce a selftest case that catches the bug. This patch (of 3): When we are doing memcg reclaim, the intended behavior is that we ignore any protection (memory.min, memory.low) of the target memcg (but not its children). Ever since the patch pointed to by the "Fixes" tag, we actually read a stale value for the target memcg protection when deciding whether to skip the memcg or not because it is protected. If the stale value happens to be high enough, we don't reclaim from the target memcg. Essentially, in some cases we may falsely skip reclaiming from the target memcg of reclaim because we read a stale protection value from last time we reclaimed from it. During reclaim, mem_cgroup_calculate_protection() is used to determine the effective protection (emin and elow) values of a memcg. The protection of the reclaim target is ignored, but we cannot set their effective protection to 0 due to a limitation of the current implementation (see comment in mem_cgroup_protection()). Instead, we leave their effective protection values unchaged, and later ignore it in mem_cgroup_protection(). However, mem_cgroup_protection() is called later in shrink_lruvec()->get_scan_count(), which is after the mem_cgroup_below_{min/low}() checks in shrink_node_memcgs(). As a result, the stale effective protection values of the target memcg may lead us to skip reclaiming from the target memcg entirely, before calling shrink_lruvec(). This can be even worse with recursive protection, where the stale target memcg protection can be higher than its standalone protection. See two examples below (a similar version of example (a) is added to test_memcontrol in a later patch). (a) A simple example with proactive reclaim is as follows. Consider the following hierarchy: ROOT | A | B (memory.min = 10M) Consider the following scenario: - B has memory.current = 10M. - The system undergoes global reclaim (or memcg reclaim in A). - In shrink_node_memcgs(): - mem_cgroup_calculate_protection() calculates the effective min (emin) of B as 10M. - mem_cgroup_below_min() returns true for B, we do not reclaim from B. - Now if we want to reclaim 5M from B using proactive reclaim (memory.reclaim), we should be able to, as the protection of the target memcg should be ignored. - In shrink_node_memcgs(): - mem_cgroup_calculate_protection() immediately returns for B without doing anything, as B is the target memcg, relying on mem_cgroup_protection() to ignore B's stale effective min (still 10M). - mem_cgroup_below_min() reads the stale effective min for B and we skip it instead of ignoring its protection as intended, as we never reach mem_cgroup_protection(). (b) An more complex example with recursive protection is as follows. Consider the following hierarchy with memory_recursiveprot: ROOT | A (memory.min = 50M) | B (memory.min = 10M, memory.high = 40M) Consider the following scenario: - B has memory.current = 35M. - The system undergoes global reclaim (target memcg is NULL). - B will have an effective min of 50M (all of A's unclaimed protection). - B will not be reclaimed from. - Now allocate 10M more memory in B, pushing it above it's high limit. - The system undergoes memcg reclaim from B (target memcg is B). - Like example (a), we do nothing in mem_cgroup_calculate_protection(), then call mem_cgroup_below_min(), which will read the stale effective min for B (50M) and skip it. In this case, it's even worse because we are not just considering B's standalone protection (10M), but we are reading a much higher stale protection (50M) which will cause us to not reclaim from B at all. This is an artifact of commit 45c7f7e1ef17 ("mm, memcg: decouple e{low,min} state mutations from protection checks") which made mem_cgroup_calculate_protection() only change the state without returning any value. Before that commit, we used to return MEMCG_PROT_NONE for the target memcg, which would cause us to skip the mem_cgroup_below_{min/low}() checks. After that commit we do not return anything and we end up checking the min & low effective protections for the target memcg, which are stale. Update mem_cgroup_supports_protection() to also check if we are reclaiming from the target, and rename it to mem_cgroup_unprotected() (now returns true if we should not protect the memcg, much simpler logic). Link: https://lkml.kernel.org/r/20221202031512.1365483-1-yosryahmed@google.com Link: https://lkml.kernel.org/r/20221202031512.1365483-2-yosryahmed@google.com Fixes: 45c7f7e1ef17 ("mm, memcg: decouple e{low,min} state mutations from protection checks") Signed-off-by: Yosry Ahmed <yosryahmed(a)google.com> Reviewed-by: Roman Gushchin <roman.gushchin(a)linux.dev> Cc: Chris Down <chris(a)chrisdown.name> Cc: David Rientjes <rientjes(a)google.com> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Matthew Wilcox <willy(a)infradead.org> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Muchun Song <songmuchun(a)bytedance.com> Cc: Shakeel Butt <shakeelb(a)google.com> Cc: Tejun Heo <tj(a)kernel.org> Cc: Vasily Averin <vasily.averin(a)linux.dev> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: Yu Zhao <yuzhao(a)google.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Conflicts: mm/vmscan.c Signed-off-by: Cai Xinchen <caixinchen1(a)huawei.com> --- include/linux/memcontrol.h | 31 +++++++++++++++++++++---------- mm/vmscan.c | 4 ++-- 2 files changed, 23 insertions(+), 12 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 24794b56eaaf..287c54141a90 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -769,28 +769,32 @@ static inline void mem_cgroup_protection(struct mem_cgroup *root, void mem_cgroup_calculate_protection(struct mem_cgroup *root, struct mem_cgroup *memcg); -static inline bool mem_cgroup_supports_protection(struct mem_cgroup *memcg) +static inline bool mem_cgroup_unprotected(struct mem_cgroup *target, + struct mem_cgroup *memcg) { /* * The root memcg doesn't account charges, and doesn't support - * protection. + * protection. The target memcg's protection is ignored, see + * mem_cgroup_calculate_protection() and mem_cgroup_protection() */ - return !mem_cgroup_disabled() && !mem_cgroup_is_root(memcg); - + return mem_cgroup_disabled() || mem_cgroup_is_root(memcg) || + memcg == target; } -static inline bool mem_cgroup_below_low(struct mem_cgroup *memcg) +static inline bool mem_cgroup_below_low(struct mem_cgroup *target, + struct mem_cgroup *memcg) { - if (!mem_cgroup_supports_protection(memcg)) + if (mem_cgroup_unprotected(target, memcg)) return false; return READ_ONCE(memcg->memory.elow) >= page_counter_read(&memcg->memory); } -static inline bool mem_cgroup_below_min(struct mem_cgroup *memcg) +static inline bool mem_cgroup_below_min(struct mem_cgroup *target, + struct mem_cgroup *memcg) { - if (!mem_cgroup_supports_protection(memcg)) + if (mem_cgroup_unprotected(target, memcg)) return false; return READ_ONCE(memcg->memory.emin) >= @@ -1378,12 +1382,19 @@ static inline void mem_cgroup_calculate_protection(struct mem_cgroup *root, { } -static inline bool mem_cgroup_below_low(struct mem_cgroup *memcg) +static inline bool mem_cgroup_unprotected(struct mem_cgroup *target, + struct mem_cgroup *memcg) +{ + return true; +} +static inline bool mem_cgroup_below_low(struct mem_cgroup *target, + struct mem_cgroup *memcg) { return false; } -static inline bool mem_cgroup_below_min(struct mem_cgroup *memcg) +static inline bool mem_cgroup_below_min(struct mem_cgroup *target, + struct mem_cgroup *memcg) { return false; } diff --git a/mm/vmscan.c b/mm/vmscan.c index 3d383c7126e3..044bf496885b 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2860,13 +2860,13 @@ static void shrink_node_memcgs(pg_data_t *pgdat, struct scan_control *sc) mem_cgroup_calculate_protection(target_memcg, memcg); - if (mem_cgroup_below_min(memcg)) { + if (mem_cgroup_below_min(target_memcg, memcg)) { /* * Hard protection. * If there is no reclaimable memory, OOM. */ continue; - } else if (mem_cgroup_below_low(memcg)) { + } else if (mem_cgroup_below_low(target_memcg, memcg)) { /* * Soft protection. * Respect the protection only as long as -- 2.34.1

1 0

[PATCH OLK-5.10 0/5] bugfix patches from mainline
by Xiongfeng Wang 06 May '24

06 May '24

Frederic Weisbecker (2): posix-cpu-timers: Force next expiration recalc after itimer reset posix-cpu-timers: Prevent spuriously armed 0-value itimer Jiri Slaby (1): x86/vdso: Use proper modifier for len's format specifier in extract() Rafael J. Wysocki (2): PM: runtime: Simplify locking in pm_runtime_put_suppliers() PM: runtime: Fix supplier device management during consumer probe arch/x86/entry/vdso/vdso2c.h | 2 +- drivers/base/core.c | 10 ++++++++++ drivers/base/power/runtime.c | 10 +--------- kernel/time/posix-cpu-timers.c | 5 ++--- 4 files changed, 14 insertions(+), 13 deletions(-) -- 2.20.1

2 6

[PATCH openEuler-1.0-LTS] drivers: base: transport_class: fix resource leak when transport_add_device() fails
by dinglongwei 06 May '24

06 May '24

From: Yang Yingliang <yangyingliang(a)huawei.com> mainline inclusion from mainline-v6.3-rc1 commit e5da06b27ff5a148e42265c8e306670a9d913969 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9MMNG Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The normal call sequence of using transport class is: Add path: transport_setup_device() transport_setup_classdev() // call sas_host_setup() here transport_add_device() // if fails, need call transport_destroy_device() transport_configure_device() Remove path: transport_remove_device() transport_remove_classdev // call sas_host_remove() here transport_destroy_device() If transport_add_device() fails, need call transport_destroy_device() to free memory, but in this case, ->remove() is not called, and the resources allocated in ->setup() are leaked. So fix these leaks by calling ->remove() in transport_add_class_device() if it returns error. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Link: https://lore.kernel.org/r/20221115031638.3816551-1-yangyingliang@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: dinglongwei <dinglongwei1(a)huawei.com> --- drivers/base/transport_class.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/base/transport_class.c b/drivers/base/transport_class.c index 5ed86ded6e6b..4c75dce60551 100644 --- a/drivers/base/transport_class.c +++ b/drivers/base/transport_class.c @@ -151,12 +151,27 @@ static int transport_add_class_device(struct attribute_container *cont, struct device *dev, struct device *classdev) { + struct transport_class *tclass = class_to_transport_class(cont->class); int error = attribute_container_add_class_device(classdev); struct transport_container *tcont = attribute_container_to_transport_container(cont); - if (!error && tcont->statistics) + if (error) + goto err_remove; + + if (tcont->statistics) { error = sysfs_create_group(&classdev->kobj, tcont->statistics); + if (error) + goto err_del; + } + + return 0; + +err_del: + attribute_container_class_device_del(classdev); +err_remove: + if (tclass->remove) + tclass->remove(tcont, dev, classdev); return error; } -- 2.17.1

2 1

[PATCH OLK-5.10] drivers: base: transport_class: fix resource leak when transport_add_device() fails
by dinglongwei 06 May '24

06 May '24

From: Yang Yingliang <yangyingliang(a)huawei.com> mainline inclusion from mainline-v6.3-rc1 commit e5da06b27ff5a148e42265c8e306670a9d913969 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9MMNG Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The normal call sequence of using transport class is: Add path: transport_setup_device() transport_setup_classdev() // call sas_host_setup() here transport_add_device() // if fails, need call transport_destroy_device() transport_configure_device() Remove path: transport_remove_device() transport_remove_classdev // call sas_host_remove() here transport_destroy_device() If transport_add_device() fails, need call transport_destroy_device() to free memory, but in this case, ->remove() is not called, and the resources allocated in ->setup() are leaked. So fix these leaks by calling ->remove() in transport_add_class_device() if it returns error. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Link: https://lore.kernel.org/r/20221115031638.3816551-1-yangyingliang@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: dinglongwei <dinglongwei1(a)huawei.com> --- drivers/base/transport_class.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/base/transport_class.c b/drivers/base/transport_class.c index ccc86206e508..09ee2a1e35bb 100644 --- a/drivers/base/transport_class.c +++ b/drivers/base/transport_class.c @@ -155,12 +155,27 @@ static int transport_add_class_device(struct attribute_container *cont, struct device *dev, struct device *classdev) { + struct transport_class *tclass = class_to_transport_class(cont->class); int error = attribute_container_add_class_device(classdev); struct transport_container *tcont = attribute_container_to_transport_container(cont); - if (!error && tcont->statistics) + if (error) + goto err_remove; + + if (tcont->statistics) { error = sysfs_create_group(&classdev->kobj, tcont->statistics); + if (error) + goto err_del; + } + + return 0; + +err_del: + attribute_container_class_device_del(classdev); +err_remove: + if (tclass->remove) + tclass->remove(tcont, dev, classdev); return error; } -- 2.17.1

2 1

[PATCH OLK-6.6 0/2] Introduce CONFIG_ARCH_CUSTOM_NUMA_DISTANCE
by Hui Tang 06 May '24

06 May '24

This patchset allow arch adjust node_relaim_distance, as follow: Hui Tang (2): arm64/numa: Support node_reclaim_distance adjust for arch config: enable COBFIG_ARCH_CUSTOM_NUMA_DISTANCE for arm64 arch/arm64/Kconfig | 13 +++++ arch/arm64/configs/openeuler_defconfig | 1 + drivers/base/arch_numa.c | 79 ++++++++++++++++++++++++++ 3 files changed, 93 insertions(+) -- 2.34.1

4 7

[PATCH OLK-5.10] memcg: fix possible use-after-free in memcg_write_event_control()
by Cai Xinchen 06 May '24

06 May '24

From: Tejun Heo <tj(a)kernel.org> stable inclusion from stable-v5.10.159 commit f1f7f36cf682fa59db15e2089039a2eeb58ff2ad category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7NTXH Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 4a7ba45b1a435e7097ca0f79a847d0949d0eb088 upstream. memcg_write_event_control() accesses the dentry->d_name of the specified control fd to route the write call. As a cgroup interface file can't be renamed, it's safe to access d_name as long as the specified file is a regular cgroup file. Also, as these cgroup interface files can't be removed before the directory, it's safe to access the parent too. Prior to 347c4a874710 ("memcg: remove cgroup_event->cft"), there was a call to __file_cft() which verified that the specified file is a regular cgroupfs file before further accesses. The cftype pointer returned from __file_cft() was no longer necessary and the commit inadvertently dropped the file type check with it allowing any file to slip through. With the invarients broken, the d_name and parent accesses can now race against renames and removals of arbitrary files and cause use-after-free's. Fix the bug by resurrecting the file type check in __file_cft(). Now that cgroupfs is implemented through kernfs, checking the file operations needs to go through a layer of indirection. Instead, let's check the superblock and dentry type. Link: https://lkml.kernel.org/r/Y5FRm/cfcKPGzWwl@slm.duckdns.org Fixes: 347c4a874710 ("memcg: remove cgroup_event->cft") Signed-off-by: Tejun Heo <tj(a)kernel.org> Reported-by: Jann Horn <jannh(a)google.com> Acked-by: Roman Gushchin <roman.gushchin(a)linux.dev> Acked-by: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Linus Torvalds <torvalds(a)linux-foundation.org> Cc: Michal Hocko <mhocko(a)kernel.org> Cc: Muchun Song <songmuchun(a)bytedance.com> Cc: Shakeel Butt <shakeelb(a)google.com> Cc: <stable(a)vger.kernel.org> [3.14+] Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Cai Xinchen <caixinchen1(a)huawei.com> --- include/linux/cgroup.h | 1 + kernel/cgroup/cgroup-internal.h | 1 - mm/memcontrol.c | 15 +++++++++++++-- 3 files changed, 14 insertions(+), 3 deletions(-) diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 1456b23d3ba1..b7ffef090ea7 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -68,6 +68,7 @@ struct css_task_iter { struct list_head iters_node; /* css_set->task_iters */ }; +extern struct file_system_type cgroup_fs_type; extern struct cgroup_root cgrp_dfl_root; extern struct css_set init_css_set; diff --git a/kernel/cgroup/cgroup-internal.h b/kernel/cgroup/cgroup-internal.h index 6fc71b98209e..470f196b7f74 100644 --- a/kernel/cgroup/cgroup-internal.h +++ b/kernel/cgroup/cgroup-internal.h @@ -169,7 +169,6 @@ extern struct mutex cgroup_mutex; extern spinlock_t css_set_lock; extern struct cgroup_subsys *cgroup_subsys[]; extern struct list_head cgroup_roots; -extern struct file_system_type cgroup_fs_type; /* iterate across the hierarchies */ #define for_each_root(root) \ diff --git a/mm/memcontrol.c b/mm/memcontrol.c index f5add1247f43..2aa0167863b7 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -5391,6 +5391,7 @@ static ssize_t memcg_write_event_control(struct kernfs_open_file *of, unsigned int efd, cfd; struct fd efile; struct fd cfile; + struct dentry *cdentry; const char *name; char *endp; int ret; @@ -5441,6 +5442,16 @@ static ssize_t memcg_write_event_control(struct kernfs_open_file *of, if (ret < 0) goto out_put_cfile; + /* + * The control file must be a regular cgroup1 file. As a regular cgroup + * file can't be renamed, it's safe to access its name afterwards. + */ + cdentry = cfile.file->f_path.dentry; + if (cdentry->d_sb->s_type != &cgroup_fs_type || !d_is_reg(cdentry)) { + ret = -EINVAL; + goto out_put_cfile; + } + /* * Determine the event callbacks and set them in @event. This used * to be done via struct cftype but cgroup core no longer knows @@ -5449,7 +5460,7 @@ static ssize_t memcg_write_event_control(struct kernfs_open_file *of, * * DO NOT ADD NEW FILES. */ - name = cfile.file->f_path.dentry->d_name.name; + name = cdentry->d_name.name; if (!strcmp(name, "memory.usage_in_bytes")) { event->register_event = mem_cgroup_usage_register_event; @@ -5473,7 +5484,7 @@ static ssize_t memcg_write_event_control(struct kernfs_open_file *of, * automatically removed on cgroup destruction but the removal is * asynchronous, so take an extra ref on @css. */ - cfile_css = css_tryget_online_from_dir(cfile.file->f_path.dentry->d_parent, + cfile_css = css_tryget_online_from_dir(cdentry->d_parent, &memory_cgrp_subsys); ret = -EINVAL; if (IS_ERR(cfile_css)) -- 2.34.1

2 1

[PATCH OLK-5.10 0/2] CVE-2024-26923
by Liu Jian 06 May '24

06 May '24

CVE-2024-26923 Kuniyuki Iwashima (1): af_unix: Suppress false-positive lockdep splat for spin_lock() in __unix_gc(). Michal Luczaj (1): af_unix: Fix garbage collector racing against connect() include/net/af_unix.h | 3 +++ net/unix/garbage.c | 18 +++++++++++++++++- 2 files changed, 20 insertions(+), 1 deletion(-) -- 2.34.1

2 3

2024

2023

2022

2021

2020

2019