October 2024 - Linux-stable-mirror

[PATCH] sched: psi: fix bogus pressure spikes from aggregation race

by Johannes Weiner

Brandon reports sporadic, non-sensical spikes in cumulative pressure time (total=) when reading cpu.pressure at a high rate. This is due to a race condition between reader aggregation and tasks changing states. While it affects all states and all resources captured by PSI, in practice it most likely triggers with CPU pressure, since scheduling events are so frequent compared to other resource events. The race context is the live snooping of ongoing stalls during a pressure read. The read aggregates per-cpu records for stalls that have concluded, but will also incorporate ad-hoc the duration of any active state that hasn't been recorded yet. This is important to get timely measurements of ongoing stalls. Those ad-hoc samples are calculated on-the-fly up to the current time on that CPU; since the stall hasn't concluded, it's expected that this is the minimum amount of stall time that will enter the per-cpu records once it does. The problem is that the path that concludes the state uses a CPU clock read that is not synchronized against aggregators; the clock is read outside of the seqlock protection. This allows aggregators to race and snoop a stall with a longer duration than will actually be recorded. With the recorded stall time being less than the last snapshot remembered by the aggregator, a subsequent sample will underflow and observe a bogus delta value, resulting in an erratic jump in pressure. Fix this by moving the clock read of the state change into the seqlock protection. This ensures no aggregation can snoop live stalls past the time that's recorded when the state concludes. Reported-by: Brandon Duffany <brandon(a)buildbuddy.io> Link: https://bugzilla.kernel.org/show_bug.cgi?id=219194 Link: https://lore.kernel.org/lkml/20240827121851.GB438928@cmpxchg.org/ Fixes: df77430639c9 ("psi: Reduce calls to sched_clock() in psi") Cc: stable(a)vger.kernel.org Signed-off-by: Johannes Weiner <hannes(a)cmpxchg.org> Reviewed-by: Chengming Zhou <chengming.zhou(a)linux.dev> --- kernel/sched/psi.c | 26 ++++++++++++-------------- 1 file changed, 12 insertions(+), 14 deletions(-) diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index 020d58967d4e..84dad1511d1e 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -769,12 +769,13 @@ static void record_times(struct psi_group_cpu *groupc, u64 now) } static void psi_group_change(struct psi_group *group, int cpu, - unsigned int clear, unsigned int set, u64 now, + unsigned int clear, unsigned int set, bool wake_clock) { struct psi_group_cpu *groupc; unsigned int t, m; u32 state_mask; + u64 now; lockdep_assert_rq_held(cpu_rq(cpu)); groupc = per_cpu_ptr(group->pcpu, cpu); @@ -789,6 +790,7 @@ static void psi_group_change(struct psi_group *group, int cpu, * SOME and FULL time these may have resulted in. */ write_seqcount_begin(&groupc->seq); + now = cpu_clock(cpu); /* * Start with TSK_ONCPU, which doesn't have a corresponding @@ -899,18 +901,15 @@ void psi_task_change(struct task_struct *task, int clear, int set) { int cpu = task_cpu(task); struct psi_group *group; - u64 now; if (!task->pid) return; psi_flags_change(task, clear, set); - now = cpu_clock(cpu); - group = task_psi_group(task); do { - psi_group_change(group, cpu, clear, set, now, true); + psi_group_change(group, cpu, clear, set, true); } while ((group = group->parent)); } @@ -919,7 +918,6 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, { struct psi_group *group, *common = NULL; int cpu = task_cpu(prev); - u64 now = cpu_clock(cpu); if (next->pid) { psi_flags_change(next, 0, TSK_ONCPU); @@ -936,7 +934,7 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, break; } - psi_group_change(group, cpu, 0, TSK_ONCPU, now, true); + psi_group_change(group, cpu, 0, TSK_ONCPU, true); } while ((group = group->parent)); } @@ -974,7 +972,7 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, do { if (group == common) break; - psi_group_change(group, cpu, clear, set, now, wake_clock); + psi_group_change(group, cpu, clear, set, wake_clock); } while ((group = group->parent)); /* @@ -986,7 +984,7 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, if ((prev->psi_flags ^ next->psi_flags) & ~TSK_ONCPU) { clear &= ~TSK_ONCPU; for (; group; group = group->parent) - psi_group_change(group, cpu, clear, set, now, wake_clock); + psi_group_change(group, cpu, clear, set, wake_clock); } } } @@ -997,8 +995,8 @@ void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_st int cpu = task_cpu(curr); struct psi_group *group; struct psi_group_cpu *groupc; - u64 now, irq; s64 delta; + u64 irq; if (static_branch_likely(&psi_disabled)) return; @@ -1011,7 +1009,6 @@ void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_st if (prev && task_psi_group(prev) == group) return; - now = cpu_clock(cpu); irq = irq_time_read(cpu); delta = (s64)(irq - rq->psi_irq_time); if (delta < 0) @@ -1019,12 +1016,15 @@ void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_st rq->psi_irq_time = irq; do { + u64 now; + if (!group->enabled) continue; groupc = per_cpu_ptr(group->pcpu, cpu); write_seqcount_begin(&groupc->seq); + now = cpu_clock(cpu); record_times(groupc, now); groupc->times[PSI_IRQ_FULL] += delta; @@ -1223,11 +1223,9 @@ void psi_cgroup_restart(struct psi_group *group) for_each_possible_cpu(cpu) { struct rq *rq = cpu_rq(cpu); struct rq_flags rf; - u64 now; rq_lock_irq(rq, &rf); - now = cpu_clock(cpu); - psi_group_change(group, cpu, 0, 0, now, true); + psi_group_change(group, cpu, 0, 0, true); rq_unlock_irq(rq, &rf); } } -- 2.46.2

9 months, 1 week

1
0
0 0

[PATCH] ALSA: core: add isascii() check to card ID generator

by Jaroslav Kysela

The card identifier should contain only safe ASCII characters. The isalnum() returns true also for characters for non-ASCII characters. Buglink: https://gitlab.freedesktop.org/pipewire/pipewire/-/issues/4135 Link: https://lore.kernel.org/linux-sound/yk3WTvKkwheOon_LzZlJ43PPInz6byYfBzpKkba… Cc: stable(a)vger.kernel.org Reported-by: Barnabás Pőcze <pobrn(a)protonmail.com> Signed-off-by: Jaroslav Kysela <perex(a)perex.cz> --- sound/core/init.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/sound/core/init.c b/sound/core/init.c index b92aa7103589..114fb87de990 100644 --- a/sound/core/init.c +++ b/sound/core/init.c @@ -654,13 +654,19 @@ void snd_card_free(struct snd_card *card) } EXPORT_SYMBOL(snd_card_free); +/* check, if the character is in the valid ASCII range */ +static inline bool safe_ascii_char(char c) +{ + return isascii(c) && isalnum(c); +} + /* retrieve the last word of shortname or longname */ static const char *retrieve_id_from_card_name(const char *name) { const char *spos = name; while (*name) { - if (isspace(*name) && isalnum(name[1])) + if (isspace(*name) && safe_ascii_char(name[1])) spos = name + 1; name++; } @@ -687,12 +693,12 @@ static void copy_valid_id_string(struct snd_card *card, const char *src, { char *id = card->id; - while (*nid && !isalnum(*nid)) + while (*nid && !safe_ascii_char(*nid)) nid++; if (isdigit(*nid)) *id++ = isalpha(*src) ? *src : 'D'; while (*nid && (size_t)(id - card->id) < sizeof(card->id) - 1) { - if (isalnum(*nid)) + if (safe_ascii_char(*nid)) *id++ = *nid; nid++; } @@ -787,7 +793,7 @@ static ssize_t id_store(struct device *dev, struct device_attribute *attr, for (idx = 0; idx < copy; idx++) { c = buf[idx]; - if (!isalnum(c) && c != '_' && c != '-') + if (!safe_ascii_char(c) && c != '_' && c != '-') return -EINVAL; } memcpy(buf1, buf, copy); -- 2.46.0

9 months, 1 week

2
1
0 0

[PATCH net v2] gso: fix udp gso fraglist segmentation after pull from frag_list

by Willem de Bruijn

From: Willem de Bruijn <willemb(a)google.com> Detect gso fraglist skbs with corrupted geometry (see below) and pass these to skb_segment instead of skb_segment_list, as the first can segment them correctly. Valid SKB_GSO_FRAGLIST skbs - consist of two or more segments - the head_skb holds the protocol headers plus first gso_size - one or more frag_list skbs hold exactly one segment - all but the last must be gso_size Optional datapath hooks such as NAT and BPF (bpf_skb_pull_data) can modify these skbs, breaking these invariants. In extreme cases they pull all data into skb linear. For UDP, this causes a NULL ptr deref in __udpv4_gso_segment_list_csum at udp_hdr(seg->next)->dest. Detect invalid geometry due to pull, by checking head_skb size. Don't just drop, as this may blackhole a destination. Convert to be able to pass to regular skb_segment. Link: https://lore.kernel.org/netdev/20240428142913.18666-1-shiming.cheng@mediate… Fixes: 9fd1ff5d2ac7 ("udp: Support UDP fraglist GRO/GSO.") Signed-off-by: Willem de Bruijn <willemb(a)google.com> Cc: stable(a)vger.kernel.org --- v1->v2 - update Fixes tag to point to udp specific patch - reinit uh->check to pseudo csum as required by CHECKSUM_PARTIAL --- net/ipv4/udp_offload.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index d842303587af..a5be6e4ed326 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -296,8 +296,26 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb, return NULL; } - if (skb_shinfo(gso_skb)->gso_type & SKB_GSO_FRAGLIST) - return __udp_gso_segment_list(gso_skb, features, is_ipv6); + if (skb_shinfo(gso_skb)->gso_type & SKB_GSO_FRAGLIST) { + /* Detect modified geometry and pass those to skb_segment. */ + if (skb_pagelen(gso_skb) - sizeof(*uh) == skb_shinfo(gso_skb)->gso_size) + return __udp_gso_segment_list(gso_skb, features, is_ipv6); + + /* Setup csum, as fraglist skips this in udp4_gro_receive. */ + gso_skb->csum_start = skb_transport_header(gso_skb) - gso_skb->head; + gso_skb->csum_offset = offsetof(struct udphdr, check); + gso_skb->ip_summed = CHECKSUM_PARTIAL; + + uh = udp_hdr(gso_skb); + if (is_ipv6) + uh->check = ~udp_v6_check(gso_skb->len, + &ipv6_hdr(gso_skb)->saddr, + &ipv6_hdr(gso_skb)->daddr, 0); + else + uh->check = ~udp_v4_check(gso_skb->len, + ip_hdr(gso_skb)->saddr, + ip_hdr(gso_skb)->daddr, 0); + } skb_pull(gso_skb, sizeof(*uh)); -- 2.46.1.824.gd892dcdcdd-goog

9 months, 1 week

2
1
0 0

[PATCH net] vrf: revert "vrf: Remove unnecessary RCU-bh critical section"

by Willem de Bruijn

From: Willem de Bruijn <willemb(a)google.com> This reverts commit 504fc6f4f7f681d2a03aa5f68aad549d90eab853. dev_queue_xmit_nit is expected to be called with BH disabled. __dev_queue_xmit has the following: /* Disable soft irqs for various locks below. Also * stops preemption for RCU. */ rcu_read_lock_bh(); VRF must follow this invariant. The referenced commit removed this protection. Which triggered a lockdep warning: ================================ WARNING: inconsistent lock state 6.11.0 #1 Tainted: G W -------------------------------- inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage. btserver/134819 [HC0[0]:SC0[0]:HE1:SE1] takes: ffff8882da30c118 (rlock-AF_PACKET){+.?.}-{2:2}, at: tpacket_rcv+0x863/0x3b30 {IN-SOFTIRQ-W} state was registered at: lock_acquire+0x19a/0x4f0 _raw_spin_lock+0x27/0x40 packet_rcv+0xa33/0x1320 __netif_receive_skb_core.constprop.0+0xcb0/0x3a90 __netif_receive_skb_list_core+0x2c9/0x890 netif_receive_skb_list_internal+0x610/0xcc0 [...] other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(rlock-AF_PACKET); <Interrupt> lock(rlock-AF_PACKET); *** DEADLOCK *** Call Trace: <TASK> dump_stack_lvl+0x73/0xa0 mark_lock+0x102e/0x16b0 __lock_acquire+0x9ae/0x6170 lock_acquire+0x19a/0x4f0 _raw_spin_lock+0x27/0x40 tpacket_rcv+0x863/0x3b30 dev_queue_xmit_nit+0x709/0xa40 vrf_finish_direct+0x26e/0x340 [vrf] vrf_l3_out+0x5f4/0xe80 [vrf] __ip_local_out+0x51e/0x7a0 [...] Fixes: 504fc6f4f7f6 ("vrf: Remove unnecessary RCU-bh critical section") Link: https://lore.kernel.org/netdev/20240925185216.1990381-1-greearb@candelatech… Reported-by: Ben Greear <greearb(a)candelatech.com> Signed-off-by: Willem de Bruijn <willemb(a)google.com> Cc: stable(a)vger.kernel.org --- drivers/net/vrf.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 4d8ccaf9a2b4..4087f72f0d2b 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -608,7 +608,9 @@ static void vrf_finish_direct(struct sk_buff *skb) eth_zero_addr(eth->h_dest); eth->h_proto = skb->protocol; + rcu_read_lock_bh(); dev_queue_xmit_nit(skb, vrf_dev); + rcu_read_unlock_bh(); skb_pull(skb, ETH_HLEN); } -- 2.46.1.824.gd892dcdcdd-goog

9 months, 1 week

4
4
0 0

[PATCH] Revert "mmc: mvsdio: Use sg_miter for PIO"

by Linus Walleij

This reverts commit 2761822c00e8c271f10a10affdbd4917d900d7ea. When testing on real hardware the patch does not work. Revert, try to acquire real hardware, and retry. These systems typically don't have highmem anyway so the impact is likely zero. Cc: stable(a)vger.kernel.org Reported-by: Charlie <g4sra(a)protonmail.com> Signed-off-by: Linus Walleij <linus.walleij(a)linaro.org> --- drivers/mmc/host/mvsdio.c | 71 ++++++++++++----------------------------------- 1 file changed, 18 insertions(+), 53 deletions(-) diff --git a/drivers/mmc/host/mvsdio.c b/drivers/mmc/host/mvsdio.c index af7f21888e27..ca01b7d204ba 100644 --- a/drivers/mmc/host/mvsdio.c +++ b/drivers/mmc/host/mvsdio.c @@ -38,9 +38,8 @@ struct mvsd_host { unsigned int xfer_mode; unsigned int intr_en; unsigned int ctrl; - bool use_pio; - struct sg_mapping_iter sg_miter; unsigned int pio_size; + void *pio_ptr; unsigned int sg_frags; unsigned int ns_per_clk; unsigned int clock; @@ -115,18 +114,11 @@ static int mvsd_setup_data(struct mvsd_host *host, struct mmc_data *data) * data when the buffer is not aligned on a 64 byte * boundary. */ - unsigned int miter_flags = SG_MITER_ATOMIC; /* Used from IRQ */ - - if (data->flags & MMC_DATA_READ) - miter_flags |= SG_MITER_TO_SG; - else - miter_flags |= SG_MITER_FROM_SG; - host->pio_size = data->blocks * data->blksz; - sg_miter_start(&host->sg_miter, data->sg, data->sg_len, miter_flags); + host->pio_ptr = sg_virt(data->sg); if (!nodma) - dev_dbg(host->dev, "fallback to PIO for data\n"); - host->use_pio = true; + dev_dbg(host->dev, "fallback to PIO for data at 0x%p size %d\n", + host->pio_ptr, host->pio_size); return 1; } else { dma_addr_t phys_addr; @@ -137,7 +129,6 @@ static int mvsd_setup_data(struct mvsd_host *host, struct mmc_data *data) phys_addr = sg_dma_address(data->sg); mvsd_write(MVSD_SYS_ADDR_LOW, (u32)phys_addr & 0xffff); mvsd_write(MVSD_SYS_ADDR_HI, (u32)phys_addr >> 16); - host->use_pio = false; return 0; } } @@ -297,8 +288,8 @@ static u32 mvsd_finish_data(struct mvsd_host *host, struct mmc_data *data, { void __iomem *iobase = host->base; - if (host->use_pio) { - sg_miter_stop(&host->sg_miter); + if (host->pio_ptr) { + host->pio_ptr = NULL; host->pio_size = 0; } else { dma_unmap_sg(mmc_dev(host->mmc), data->sg, host->sg_frags, @@ -353,12 +344,9 @@ static u32 mvsd_finish_data(struct mvsd_host *host, struct mmc_data *data, static irqreturn_t mvsd_irq(int irq, void *dev) { struct mvsd_host *host = dev; - struct sg_mapping_iter *sgm = &host->sg_miter; void __iomem *iobase = host->base; u32 intr_status, intr_done_mask; int irq_handled = 0; - u16 *p; - int s; intr_status = mvsd_read(MVSD_NOR_INTR_STATUS); dev_dbg(host->dev, "intr 0x%04x intr_en 0x%04x hw_state 0x%04x\n", @@ -382,36 +370,15 @@ static irqreturn_t mvsd_irq(int irq, void *dev) spin_lock(&host->lock); /* PIO handling, if needed. Messy business... */ - if (host->use_pio) { - /* - * As we set sgm->consumed this always gives a valid buffer - * position. - */ - if (!sg_miter_next(sgm)) { - /* This should not happen */ - dev_err(host->dev, "ran out of scatter segments\n"); - spin_unlock(&host->lock); - host->intr_en &= - ~(MVSD_NOR_RX_READY | MVSD_NOR_RX_FIFO_8W | - MVSD_NOR_TX_AVAIL | MVSD_NOR_TX_FIFO_8W); - mvsd_write(MVSD_NOR_INTR_EN, host->intr_en); - return IRQ_HANDLED; - } - p = sgm->addr; - s = sgm->length; - if (s > host->pio_size) - s = host->pio_size; - } - - if (host->use_pio && + if (host->pio_size && (intr_status & host->intr_en & (MVSD_NOR_RX_READY | MVSD_NOR_RX_FIFO_8W))) { - + u16 *p = host->pio_ptr; + int s = host->pio_size; while (s >= 32 && (intr_status & MVSD_NOR_RX_FIFO_8W)) { readsw(iobase + MVSD_FIFO, p, 16); p += 16; s -= 32; - sgm->consumed += 32; intr_status = mvsd_read(MVSD_NOR_INTR_STATUS); } /* @@ -424,7 +391,6 @@ static irqreturn_t mvsd_irq(int irq, void *dev) put_unaligned(mvsd_read(MVSD_FIFO), p++); put_unaligned(mvsd_read(MVSD_FIFO), p++); s -= 4; - sgm->consumed += 4; intr_status = mvsd_read(MVSD_NOR_INTR_STATUS); } if (s && s < 4 && (intr_status & MVSD_NOR_RX_READY)) { @@ -432,13 +398,10 @@ static irqreturn_t mvsd_irq(int irq, void *dev) val[0] = mvsd_read(MVSD_FIFO); val[1] = mvsd_read(MVSD_FIFO); memcpy(p, ((void *)&val) + 4 - s, s); - sgm->consumed += s; s = 0; intr_status = mvsd_read(MVSD_NOR_INTR_STATUS); } - /* PIO transfer done */ - host->pio_size -= sgm->consumed; - if (host->pio_size == 0) { + if (s == 0) { host->intr_en &= ~(MVSD_NOR_RX_READY | MVSD_NOR_RX_FIFO_8W); mvsd_write(MVSD_NOR_INTR_EN, host->intr_en); @@ -450,10 +413,14 @@ static irqreturn_t mvsd_irq(int irq, void *dev) } dev_dbg(host->dev, "pio %d intr 0x%04x hw_state 0x%04x\n", s, intr_status, mvsd_read(MVSD_HW_STATE)); + host->pio_ptr = p; + host->pio_size = s; irq_handled = 1; - } else if (host->use_pio && + } else if (host->pio_size && (intr_status & host->intr_en & (MVSD_NOR_TX_AVAIL | MVSD_NOR_TX_FIFO_8W))) { + u16 *p = host->pio_ptr; + int s = host->pio_size; /* * The TX_FIFO_8W bit is unreliable. When set, bursting * 16 halfwords all at once in the FIFO drops data. Actually @@ -464,7 +431,6 @@ static irqreturn_t mvsd_irq(int irq, void *dev) mvsd_write(MVSD_FIFO, get_unaligned(p++)); mvsd_write(MVSD_FIFO, get_unaligned(p++)); s -= 4; - sgm->consumed += 4; intr_status = mvsd_read(MVSD_NOR_INTR_STATUS); } if (s < 4) { @@ -473,13 +439,10 @@ static irqreturn_t mvsd_irq(int irq, void *dev) memcpy(((void *)&val) + 4 - s, p, s); mvsd_write(MVSD_FIFO, val[0]); mvsd_write(MVSD_FIFO, val[1]); - sgm->consumed += s; s = 0; intr_status = mvsd_read(MVSD_NOR_INTR_STATUS); } - /* PIO transfer done */ - host->pio_size -= sgm->consumed; - if (host->pio_size == 0) { + if (s == 0) { host->intr_en &= ~(MVSD_NOR_TX_AVAIL | MVSD_NOR_TX_FIFO_8W); mvsd_write(MVSD_NOR_INTR_EN, host->intr_en); @@ -487,6 +450,8 @@ static irqreturn_t mvsd_irq(int irq, void *dev) } dev_dbg(host->dev, "pio %d intr 0x%04x hw_state 0x%04x\n", s, intr_status, mvsd_read(MVSD_HW_STATE)); + host->pio_ptr = p; + host->pio_size = s; irq_handled = 1; } --- base-commit: 075dbe9f6e3c21596c5245826a4ee1f1c1676eb8 change-id: 20240927-kirkwood-mmc-regression-986c598d4c9e Best regards, -- Linus Walleij <linus.walleij(a)linaro.org>

9 months, 1 week

2
1
0 0

[PATCH] ceph: fix cap ref leak via netfs init_request

by Patrick Donnelly

From: Patrick Donnelly <pdonnell(a)redhat.com> Log recovered from a user's cluster: <7>[ 5413.970692] ceph: get_cap_refs 00000000958c114b ret 1 got Fr <7>[ 5413.970695] ceph: start_read 00000000958c114b, no cache cap ... <7>[ 5473.934609] ceph: my wanted = Fr, used = Fr, dirty - <7>[ 5473.934616] ceph: revocation: pAsLsXsFr -> pAsLsXs (revoking Fr) <7>[ 5473.934632] ceph: __ceph_caps_issued 00000000958c114b cap 00000000f7784259 issued pAsLsXs <7>[ 5473.934638] ceph: check_caps 10000000e68.fffffffffffffffe file_want - used Fr dirty - flushing - issued pAsLsXs revoking Fr retain pAsLsXsFsr AUTHONLY NOINVAL FLUSH_FORCE The MDS subsequently complains that the kernel client is late releasing caps. Closes: https://tracker.ceph.com/issues/67008 Fixes: 2504470854f8 ("ceph: Make ceph_init_request() check caps on readahead") Signed-off-by: Patrick Donnelly <pdonnell(a)redhat.com> Cc: stable(a)vger.kernel.org --- fs/ceph/addr.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 53fef258c2bc..702c6a730b70 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -489,8 +489,11 @@ static int ceph_init_request(struct netfs_io_request *rreq, struct file *file) rreq->io_streams[0].sreq_max_len = fsc->mount_options->rsize; out: - if (ret < 0) + if (ret < 0) { + if (got) + ceph_put_cap_refs(ceph_inode(inode), got); kfree(priv); + } return ret; } base-commit: e32cde8d2bd7d251a8f9b434143977ddf13dcec6 -- Patrick Donnelly, Ph.D. He / Him / His Red Hat Partner Engineer IBM, Inc. GPG: 19F28A586F808C2402351B93C3301A3E258DD79D

9 months, 1 week

1
0
0 0

[PATCH v2 net] net: Fix an unsafe loop on the list

by Anastasia Kovaleva

The kernel may crash when deleting a genetlink family if there are still listeners for that family: Oops: Kernel access of bad area, sig: 11 [#1] ... NIP [c000000000c080bc] netlink_update_socket_mc+0x3c/0xc0 LR [c000000000c0f764] __netlink_clear_multicast_users+0x74/0xc0 Call Trace: __netlink_clear_multicast_users+0x74/0xc0 genl_unregister_family+0xd4/0x2d0 Change the unsafe loop on the list to a safe one, because inside the loop there is an element removal from this list. Fixes: b8273570f802 ("genetlink: fix netns vs. netlink table locking (2)")\ Cc: stable(a)vger.kernel.org Signed-off-by: Anastasia Kovaleva <a.kovaleva(a)yadro.com> Reviewed-by: Dmitry Bogdanov <d.bogdanov(a)yadro.com> --- include/net/sock.h | 2 ++ net/netlink/af_netlink.c | 3 ++- 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/include/net/sock.h b/include/net/sock.h index c58ca8dd561b..eec77a18602a 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -894,6 +894,8 @@ static inline void sk_add_bind_node(struct sock *sk, hlist_for_each_entry_safe(__sk, tmp, list, sk_node) #define sk_for_each_bound(__sk, list) \ hlist_for_each_entry(__sk, list, sk_bind_node) +#define sk_for_each_bound_safe(__sk, tmp, list) \ + hlist_for_each_entry_safe(__sk, tmp, list, sk_bind_node) /** * sk_for_each_entry_offset_rcu - iterate over a list at a given struct offset diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 0b7a89db3ab7..0a9287fadb47 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -2136,8 +2136,9 @@ void __netlink_clear_multicast_users(struct sock *ksk, unsigned int group) { struct sock *sk; struct netlink_table *tbl = &nl_table[ksk->sk_protocol]; + struct hlist_node *tmp; - sk_for_each_bound(sk, &tbl->mc_list) + sk_for_each_bound_safe(sk, tmp, &tbl->mc_list) netlink_update_socket_mc(nlk_sk(sk), group, 0); } -- 2.40.1

9 months, 1 week

2
3
0 0

[PATCH v2] serial: imx: Update mctrl old_status on RTSD interrupt

by Marek Vasut

When sending data using DMA at high baudrate (4 Mbdps in local test case) to a device with small RX buffer which keeps asserting RTS after every received byte, it is possible that the iMX UART driver would not recognize the falling edge of RTS input signal and get stuck, unable to transmit any more data. This condition happens when the following sequence of events occur: - imx_uart_mctrl_check() is called at some point and takes a snapshot of UART control signal status into sport->old_status using imx_uart_get_hwmctrl(). The RTSS/TIOCM_CTS bit is of interest here (*). - DMA transfer occurs, the remote device asserts RTS signal after each byte. The i.MX UART driver recognizes each such RTS signal change, raises an interrupt with USR1 register RTSD bit set, which leads to invocation of __imx_uart_rtsint(), which calls uart_handle_cts_change(). - If the RTS signal is deasserted, uart_handle_cts_change() clears port->hw_stopped and unblocks the port for further data transfers. - If the RTS is asserted, uart_handle_cts_change() sets port->hw_stopped and blocks the port for further data transfers. This may occur as the last interrupt of a transfer, which means port->hw_stopped remains set and the port remains blocked (**). - Any further data transfer attempts will trigger imx_uart_mctrl_check(), which will read current status of UART control signals by calling imx_uart_get_hwmctrl() (***) and compare it with sport->old_status . - If current status differs from sport->old_status for RTS signal, uart_handle_cts_change() is called and possibly unblocks the port by clearing port->hw_stopped . - If current status does not differ from sport->old_status for RTS signal, no action occurs. This may occur in case prior snapshot (*) was taken before any transfer so the RTS is deasserted, current snapshot (***) was taken after a transfer and therefore RTS is deasserted again, which means current status and sport->old_status are identical. In case (**) triggered when RTS got asserted, and made port->hw_stopped set, the port->hw_stopped will remain set because no change on RTS line is recognized by this driver and uart_handle_cts_change() is not called from here to unblock the port->hw_stopped. Update sport->old_status in __imx_uart_rtsint() accordingly to make imx_uart_mctrl_check() detect such RTS change. Note that TIOCM_CAR and TIOCM_RI bits in sport->old_status do not suffer from this problem. Fixes: ceca629e0b48 ("[ARM] 2971/1: i.MX uart handle rts irq") Reviewed-by: Esben Haabendal <esben(a)geanix.com> Signed-off-by: Marek Vasut <marex(a)denx.de> --- Cc: Christoph Niedermaier <cniedermaier(a)dh-electronics.com> Cc: Christophe JAILLET <christophe.jaillet(a)wanadoo.fr> Cc: Esben Haabendal <esben(a)geanix.com> Cc: Fabio Estevam <festevam(a)gmail.com> Cc: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Cc: Jiri Slaby <jirislaby(a)kernel.org> Cc: Lino Sanfilippo <l.sanfilippo(a)kunbus.com> Cc: Pengutronix Kernel Team <kernel(a)pengutronix.de> Cc: Rasmus Villemoes <linux(a)rasmusvillemoes.dk> Cc: Rickard x Andersson <rickaran(a)axis.com> Cc: Sascha Hauer <s.hauer(a)pengutronix.de> Cc: Shawn Guo <shawnguo(a)kernel.org> Cc: Stefan Eichenberger <stefan.eichenberger(a)toradex.com> Cc: Uwe Kleine-König <u.kleine-koenig(a)baylibre.com> Cc: imx(a)lists.linux.dev Cc: linux-arm-kernel(a)lists.infradead.org Cc: linux-serial(a)vger.kernel.org Cc: stable(a)vger.kernel.org --- V2: - Add code comment - Add RB from Esben --- drivers/tty/serial/imx.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/drivers/tty/serial/imx.c b/drivers/tty/serial/imx.c index 67d4a72eda770..90974d338f3c0 100644 --- a/drivers/tty/serial/imx.c +++ b/drivers/tty/serial/imx.c @@ -762,6 +762,21 @@ static irqreturn_t __imx_uart_rtsint(int irq, void *dev_id) imx_uart_writel(sport, USR1_RTSD, USR1); usr1 = imx_uart_readl(sport, USR1) & USR1_RTSS; + /* + * Update sport->old_status here, so any follow-up calls to + * imx_uart_mctrl_check() will be able to recognize that RTS + * state changed since last imx_uart_mctrl_check() call. + * + * In case RTS has been detected as asserted here and later on + * deasserted by the time imx_uart_mctrl_check() was called, + * imx_uart_mctrl_check() can detect the RTS state change and + * trigger uart_handle_cts_change() to unblock the port for + * further TX transfers. + */ + if (usr1 & USR1_RTSS) + sport->old_status |= TIOCM_CTS; + else + sport->old_status &= ~TIOCM_CTS; uart_handle_cts_change(&sport->port, usr1); wake_up_interruptible(&sport->port.state->port.delta_msr_wait); -- 2.45.2

9 months, 1 week

1
0
0 0

[regression] Regular "cracks" in HDMI sound during playback since backport to 6.1.y for 92afcc310038 ("ALSA: hda: Conditionally use snooping for AMD HDMI")

by Salvatore Bonaccorso

Hi In downstream Debian we got a report from Eric Degenetais, in https://bugs.debian.org/1081833 that after the update to the 6.1.106 based version, there were regular cracks in HDMI sound during playback. Eric was able to bisec the issue down to 92afcc310038ebe5d66c689bb0bf418f5451201c in the v6.1.y series which got applied in 6.1.104. Cf. https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1081833#47 #regzbot introduced: 92afcc310038ebe5d66c689bb0bf418f5451201c #regzbot link: https://bugs.debian.org/1081833 It should be noted that Eric as well tried more recent stable series as well, in particular did test as well 6.10.6 based version back on 20th september, and the issue was reproducible there as well. Is there anything else we can try to provide? Regards, Salvatore

9 months, 1 week

3
3
0 0

Missing 6.11-stable patch

by Jens Axboe

Hi, Arguably the most important block stable patch I don't see in the most recent review series sent out, which is odd because it's certainly marked with fixes and a stable tag. It's this one: commit e3accac1a976e65491a9b9fba82ce8ddbd3d2389 Author: Damien Le Moal <dlemoal(a)kernel.org> Date: Tue Sep 17 22:32:31 2024 +0900 block: Fix elv_iosched_local_module handling of "none" scheduler and it really must go into -stable asap as it's fixing a real issue that I've had multiple users email me about. Can we get this added to the current 6.11-stable series so we don't miss another release? It's also quite possible that I'm blind and it is indeed in the queue or already there, but for the life of me I can't see it. -- Jens Axboe

9 months, 1 week

2
2
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror October 2024