June 2024 - Linux-stable-mirror

[PATCH v5] af_packet: Handle outgoing VLAN packets without hardware offloading

by Chengen Du

The issue initially stems from libpcap. The ethertype will be overwritten as the VLAN TPID if the network interface lacks hardware VLAN offloading. In the outbound packet path, if hardware VLAN offloading is unavailable, the VLAN tag is inserted into the payload but then cleared from the sk_buff struct. Consequently, this can lead to a false negative when checking for the presence of a VLAN tag, causing the packet sniffing outcome to lack VLAN tag information (i.e., TCI-TPID). As a result, the packet capturing tool may be unable to parse packets as expected. The TCI-TPID is missing because the prb_fill_vlan_info() function does not modify the tp_vlan_tci/tp_vlan_tpid values, as the information is in the payload and not in the sk_buff struct. The skb_vlan_tag_present() function only checks vlan_all in the sk_buff struct. In cooked mode, the L2 header is stripped, preventing the packet capturing tool from determining the correct TCI-TPID value. Additionally, the protocol in SLL is incorrect, which means the packet capturing tool cannot parse the L3 header correctly. Link: https://github.com/the-tcpdump-group/libpcap/issues/1105 Link: https://lore.kernel.org/netdev/20240520070348.26725-1-chengen.du@canonical.… Fixes: 393e52e33c6c ("packet: deliver VLAN TCI to userspace") Cc: stable(a)vger.kernel.org Signed-off-by: Chengen Du <chengen.du(a)canonical.com> --- net/packet/af_packet.c | 64 ++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 62 insertions(+), 2 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index ea3ebc160e25..53d51ac87ac6 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -538,6 +538,52 @@ static void *packet_current_frame(struct packet_sock *po, return packet_lookup_frame(po, rb, rb->head, status); } +static u16 vlan_get_tci(struct sk_buff *skb) +{ + unsigned int vlan_depth = skb->mac_len; + struct vlan_hdr vhdr, *vh; + u8 *skb_head = skb->data; + int skb_len = skb->len; + + if (vlan_depth) { + if (WARN_ON(vlan_depth < VLAN_HLEN)) + return 0; + vlan_depth -= VLAN_HLEN; + } else { + vlan_depth = ETH_HLEN; + } + + skb_push(skb, skb->data - skb_mac_header(skb)); + vh = skb_header_pointer(skb, vlan_depth, sizeof(vhdr), &vhdr); + if (skb_head != skb->data) { + skb->data = skb_head; + skb->len = skb_len; + } + if (unlikely(!vh)) + return 0; + + return ntohs(vh->h_vlan_TCI); +} + +static __be16 vlan_get_protocol_dgram(struct sk_buff *skb) +{ + __be16 proto = skb->protocol; + + if (unlikely(eth_type_vlan(proto))) { + u8 *skb_head = skb->data; + int skb_len = skb->len; + + skb_push(skb, skb->data - skb_mac_header(skb)); + proto = __vlan_get_protocol(skb, proto, NULL); + if (skb_head != skb->data) { + skb->data = skb_head; + skb->len = skb_len; + } + } + + return proto; +} + static void prb_del_retire_blk_timer(struct tpacket_kbdq_core *pkc) { del_timer_sync(&pkc->retire_blk_timer); @@ -1011,6 +1057,10 @@ static void prb_fill_vlan_info(struct tpacket_kbdq_core *pkc, ppd->hv1.tp_vlan_tci = skb_vlan_tag_get(pkc->skb); ppd->hv1.tp_vlan_tpid = ntohs(pkc->skb->vlan_proto); ppd->tp_status = TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID; + } else if (unlikely(eth_type_vlan(pkc->skb->protocol))) { + ppd->hv1.tp_vlan_tci = vlan_get_tci(pkc->skb); + ppd->hv1.tp_vlan_tpid = ntohs(pkc->skb->protocol); + ppd->tp_status = TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID; } else { ppd->hv1.tp_vlan_tci = 0; ppd->hv1.tp_vlan_tpid = 0; @@ -2428,6 +2478,10 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, h.h2->tp_vlan_tci = skb_vlan_tag_get(skb); h.h2->tp_vlan_tpid = ntohs(skb->vlan_proto); status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID; + } else if (unlikely(eth_type_vlan(skb->protocol))) { + h.h2->tp_vlan_tci = vlan_get_tci(skb); + h.h2->tp_vlan_tpid = ntohs(skb->protocol); + status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID; } else { h.h2->tp_vlan_tci = 0; h.h2->tp_vlan_tpid = 0; @@ -2457,7 +2511,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, sll->sll_halen = dev_parse_header(skb, sll->sll_addr); sll->sll_family = AF_PACKET; sll->sll_hatype = dev->type; - sll->sll_protocol = skb->protocol; + sll->sll_protocol = (sk->sk_type == SOCK_DGRAM) ? + vlan_get_protocol_dgram(skb) : skb->protocol; sll->sll_pkttype = skb->pkt_type; if (unlikely(packet_sock_flag(po, PACKET_SOCK_ORIGDEV))) sll->sll_ifindex = orig_dev->ifindex; @@ -3482,7 +3537,8 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, /* Original length was stored in sockaddr_ll fields */ origlen = PACKET_SKB_CB(skb)->sa.origlen; sll->sll_family = AF_PACKET; - sll->sll_protocol = skb->protocol; + sll->sll_protocol = (sock->type == SOCK_DGRAM) ? + vlan_get_protocol_dgram(skb) : skb->protocol; } sock_recv_cmsgs(msg, sk, skb); @@ -3539,6 +3595,10 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, aux.tp_vlan_tci = skb_vlan_tag_get(skb); aux.tp_vlan_tpid = ntohs(skb->vlan_proto); aux.tp_status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID; + } else if (unlikely(eth_type_vlan(skb->protocol))) { + aux.tp_vlan_tci = vlan_get_tci(skb); + aux.tp_vlan_tpid = ntohs(skb->protocol); + aux.tp_status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID; } else { aux.tp_vlan_tci = 0; aux.tp_vlan_tpid = 0; -- 2.43.0

1 year

3
7
0 0

[PATCH] scsi: sd: Use READ(16) when reading block zero on large capacity disks

by Martin K. Petersen

Commit 321da3dc1f3c ("scsi: sd: usb_storage: uas: Access media prior to querying device properties") triggered a read to LBA 0 before attempting to inquire about device characteristics. This was done because some protocol bridge devices will return generic values until an attached storage device's media has been accessed. Pierre Tomon reported that this change caused problems on a large capacity external drive connected via a bridge device. The bridge in question does not appear to implement the READ(10) command. Issue a READ(16) instead of READ(10) when a device has been identified as preferring 16-byte commands (use_16_for_rw heuristic). Cc: stable(a)vger.kernel.org Fixes: 321da3dc1f3c ("scsi: sd: usb_storage: uas: Access media prior to querying device properties") Link: https://bugzilla.kernel.org/show_bug.cgi?id=218890 Reported-by: Pierre Tomon <pierretom+12(a)ik.me> Suggested-by: Alan Stern <stern(a)rowland.harvard.edu> Tested-by: Pierre Tomon <pierretom+12(a)ik.me> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> --- drivers/scsi/sd.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 65cdc8b77e35..6759bd5af58a 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -3572,16 +3572,23 @@ static bool sd_validate_opt_xfer_size(struct scsi_disk *sdkp, static void sd_read_block_zero(struct scsi_disk *sdkp) { - unsigned int buf_len = sdkp->device->sector_size; - char *buffer, cmd[10] = { }; + struct scsi_device *sdev = sdkp->device; + unsigned int buf_len = sdev->sector_size; + char *buffer, cmd[16] = { }; buffer = kmalloc(buf_len, GFP_KERNEL); if (!buffer) return; - cmd[0] = READ_10; - put_unaligned_be32(0, &cmd[2]); /* Logical block address 0 */ - put_unaligned_be16(1, &cmd[7]); /* Transfer 1 logical block */ + if (sdev->use_16_for_rw) { + cmd[0] = READ_16; + put_unaligned_be64(0, &cmd[2]); /* Logical block address 0 */ + put_unaligned_be32(1, &cmd[10]);/* Transfer 1 logical block */ + } else { + cmd[0] = READ_10; + put_unaligned_be32(0, &cmd[2]); /* Logical block address 0 */ + put_unaligned_be16(1, &cmd[7]); /* Transfer 1 logical block */ + } scsi_execute_cmd(sdkp->device, cmd, REQ_OP_DRV_IN, buffer, buf_len, SD_TIMEOUT, sdkp->max_retries, NULL); -- 2.45.1

1 year

2
4
0 0

[PATCH 1/2] mm: page_ref: remove folio_try_get_rcu()

by Yang Shi

The below bug was reported on a non-SMP kernel: [ 275.267158][ T4335] ------------[ cut here ]------------ [ 275.267949][ T4335] kernel BUG at include/linux/page_ref.h:275! [ 275.268526][ T4335] invalid opcode: 0000 [#1] KASAN PTI [ 275.269001][ T4335] CPU: 0 PID: 4335 Comm: trinity-c3 Not tainted 6.7.0-rc4-00061-gefa7df3e3bb5 #1 [ 275.269787][ T4335] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-debian-1.16.2-1 04/01/2014 [ 275.270679][ T4335] RIP: 0010:try_get_folio (include/linux/page_ref.h:275 (discriminator 3) mm/gup.c:79 (discriminator 3)) [ 275.272813][ T4335] RSP: 0018:ffffc90005dcf650 EFLAGS: 00010202 [ 275.273346][ T4335] RAX: 0000000000000246 RBX: ffffea00066e0000 RCX: 0000000000000000 [ 275.274032][ T4335] RDX: fffff94000cdc007 RSI: 0000000000000004 RDI: ffffea00066e0034 [ 275.274719][ T4335] RBP: ffffea00066e0000 R08: 0000000000000000 R09: fffff94000cdc006 [ 275.275404][ T4335] R10: ffffea00066e0037 R11: 0000000000000000 R12: 0000000000000136 [ 275.276106][ T4335] R13: ffffea00066e0034 R14: dffffc0000000000 R15: ffffea00066e0008 [ 275.276790][ T4335] FS: 00007fa2f9b61740(0000) GS:ffffffff89d0d000(0000) knlGS:0000000000000000 [ 275.277570][ T4335] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 275.278143][ T4335] CR2: 00007fa2f6c00000 CR3: 0000000134b04000 CR4: 00000000000406f0 [ 275.278833][ T4335] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 275.279521][ T4335] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 275.280201][ T4335] Call Trace: [ 275.280499][ T4335] <TASK> [ 275.280751][ T4335] ? die (arch/x86/kernel/dumpstack.c:421 arch/x86/kernel/dumpstack.c:434 arch/x86/kernel/dumpstack.c:447) [ 275.281087][ T4335] ? do_trap (arch/x86/kernel/traps.c:112 arch/x86/kernel/traps.c:153) [ 275.281463][ T4335] ? try_get_folio (include/linux/page_ref.h:275 (discriminator 3) mm/gup.c:79 (discriminator 3)) [ 275.281884][ T4335] ? try_get_folio (include/linux/page_ref.h:275 (discriminator 3) mm/gup.c:79 (discriminator 3)) [ 275.282300][ T4335] ? do_error_trap (arch/x86/kernel/traps.c:174) [ 275.282711][ T4335] ? try_get_folio (include/linux/page_ref.h:275 (discriminator 3) mm/gup.c:79 (discriminator 3)) [ 275.283129][ T4335] ? handle_invalid_op (arch/x86/kernel/traps.c:212) [ 275.283561][ T4335] ? try_get_folio (include/linux/page_ref.h:275 (discriminator 3) mm/gup.c:79 (discriminator 3)) [ 275.283990][ T4335] ? exc_invalid_op (arch/x86/kernel/traps.c:264) [ 275.284415][ T4335] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:568) [ 275.284859][ T4335] ? try_get_folio (include/linux/page_ref.h:275 (discriminator 3) mm/gup.c:79 (discriminator 3)) [ 275.285278][ T4335] try_grab_folio (mm/gup.c:148) [ 275.285684][ T4335] __get_user_pages (mm/gup.c:1297 (discriminator 1)) [ 275.286111][ T4335] ? __pfx___get_user_pages (mm/gup.c:1188) [ 275.286579][ T4335] ? __pfx_validate_chain (kernel/locking/lockdep.c:3825) [ 275.287034][ T4335] ? mark_lock (kernel/locking/lockdep.c:4656 (discriminator 1)) [ 275.287416][ T4335] __gup_longterm_locked (mm/gup.c:1509 mm/gup.c:2209) [ 275.288192][ T4335] ? __pfx___gup_longterm_locked (mm/gup.c:2204) [ 275.288697][ T4335] ? __pfx_lock_acquire (kernel/locking/lockdep.c:5722) [ 275.289135][ T4335] ? __pfx___might_resched (kernel/sched/core.c:10106) [ 275.289595][ T4335] pin_user_pages_remote (mm/gup.c:3350) [ 275.290041][ T4335] ? __pfx_pin_user_pages_remote (mm/gup.c:3350) [ 275.290545][ T4335] ? find_held_lock (kernel/locking/lockdep.c:5244 (discriminator 1)) [ 275.290961][ T4335] ? mm_access (kernel/fork.c:1573) [ 275.291353][ T4335] process_vm_rw_single_vec+0x142/0x360 [ 275.291900][ T4335] ? __pfx_process_vm_rw_single_vec+0x10/0x10 [ 275.292471][ T4335] ? mm_access (kernel/fork.c:1573) [ 275.292859][ T4335] process_vm_rw_core+0x272/0x4e0 [ 275.293384][ T4335] ? hlock_class (arch/x86/include/asm/bitops.h:227 arch/x86/include/asm/bitops.h:239 include/asm-generic/bitops/instrumented-non-atomic.h:142 kernel/locking/lockdep.c:228) [ 275.293780][ T4335] ? __pfx_process_vm_rw_core+0x10/0x10 [ 275.294350][ T4335] process_vm_rw (mm/process_vm_access.c:284) [ 275.294748][ T4335] ? __pfx_process_vm_rw (mm/process_vm_access.c:259) [ 275.295197][ T4335] ? __task_pid_nr_ns (include/linux/rcupdate.h:306 (discriminator 1) include/linux/rcupdate.h:780 (discriminator 1) kernel/pid.c:504 (discriminator 1)) [ 275.295634][ T4335] __x64_sys_process_vm_readv (mm/process_vm_access.c:291) [ 275.296139][ T4335] ? syscall_enter_from_user_mode (kernel/entry/common.c:94 kernel/entry/common.c:112) [ 275.296642][ T4335] do_syscall_64 (arch/x86/entry/common.c:51 (discriminator 1) arch/x86/entry/common.c:82 (discriminator 1)) [ 275.297032][ T4335] ? __task_pid_nr_ns (include/linux/rcupdate.h:306 (discriminator 1) include/linux/rcupdate.h:780 (discriminator 1) kernel/pid.c:504 (discriminator 1)) [ 275.297470][ T4335] ? lockdep_hardirqs_on_prepare (kernel/locking/lockdep.c:4300 kernel/locking/lockdep.c:4359) [ 275.297988][ T4335] ? do_syscall_64 (arch/x86/include/asm/cpufeature.h:171 arch/x86/entry/common.c:97) [ 275.298389][ T4335] ? lockdep_hardirqs_on_prepare (kernel/locking/lockdep.c:4300 kernel/locking/lockdep.c:4359) [ 275.298906][ T4335] ? do_syscall_64 (arch/x86/include/asm/cpufeature.h:171 arch/x86/entry/common.c:97) [ 275.299304][ T4335] ? do_syscall_64 (arch/x86/include/asm/cpufeature.h:171 arch/x86/entry/common.c:97) [ 275.299703][ T4335] ? do_syscall_64 (arch/x86/include/asm/cpufeature.h:171 arch/x86/entry/common.c:97) [ 275.300115][ T4335] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:129) This BUG is the VM_BUG_ON(!in_atomic() && !irqs_disabled()) assertion in folio_ref_try_add_rcu() for non-SMP kernel. The process_vm_readv() calls GUP to pin the THP. An optimization for pinning THP instroduced by commit 57edfcfd3419 ("mm/gup: accelerate thp gup even for "pages != NULL"") calls try_grab_folio() to pin the THP, but try_grab_folio() is supposed to be called in atomic context for non-SMP kernel, for example, irq disabled or preemption disabled, due to the optimization introduced by commit e286781d5f2e ("mm: speculative page references"). The commit efa7df3e3bb5 ("mm: align larger anonymous mappings on THP boundaries") is not actually the root cause although it was bisected to. It just makes the problem exposed more likely. The follow up discussion suggested the optimization for non-SMP kernel may be out-dated and not worth it anymore [1]. So removing the optimization to silence the BUG. However calling try_grab_folio() in GUP slow path actually is unnecessary, so the following patch will clean this up. [1] https://lore.kernel.org/linux-mm/821cf1d6-92b9-4ac4-bacc-d8f2364ac14f@paulm… Fixes: 57edfcfd3419 ("mm/gup: accelerate thp gup even for "pages != NULL"") Reported-by: kernel test robot <oliver.sang(a)intel.com> Cc: linux-stable <stable(a)vger.kernel.org> v6.6+ Signed-off-by: Yang Shi <yang(a)os.amperecomputing.com> --- include/linux/page_ref.h | 49 ++-------------------------------------- mm/filemap.c | 10 ++++---- mm/gup.c | 2 +- 3 files changed, 8 insertions(+), 53 deletions(-) diff --git a/include/linux/page_ref.h b/include/linux/page_ref.h index 1acf5bac7f50..490d0ad6e56d 100644 --- a/include/linux/page_ref.h +++ b/include/linux/page_ref.h @@ -258,54 +258,9 @@ static inline bool folio_try_get(struct folio *folio) return folio_ref_add_unless(folio, 1, 0); } -static inline bool folio_ref_try_add_rcu(struct folio *folio, int count) -{ -#ifdef CONFIG_TINY_RCU - /* - * The caller guarantees the folio will not be freed from interrupt - * context, so (on !SMP) we only need preemption to be disabled - * and TINY_RCU does that for us. - */ -# ifdef CONFIG_PREEMPT_COUNT - VM_BUG_ON(!in_atomic() && !irqs_disabled()); -# endif - VM_BUG_ON_FOLIO(folio_ref_count(folio) == 0, folio); - folio_ref_add(folio, count); -#else - if (unlikely(!folio_ref_add_unless(folio, count, 0))) { - /* Either the folio has been freed, or will be freed. */ - return false; - } -#endif - return true; -} - -/** - * folio_try_get_rcu - Attempt to increase the refcount on a folio. - * @folio: The folio. - * - * This is a version of folio_try_get() optimised for non-SMP kernels. - * If you are still holding the rcu_read_lock() after looking up the - * page and know that the page cannot have its refcount decreased to - * zero in interrupt context, you can use this instead of folio_try_get(). - * - * Example users include get_user_pages_fast() (as pages are not unmapped - * from interrupt context) and the page cache lookups (as pages are not - * truncated from interrupt context). We also know that pages are not - * frozen in interrupt context for the purposes of splitting or migration. - * - * You can also use this function if you're holding a lock that prevents - * pages being frozen & removed; eg the i_pages lock for the page cache - * or the mmap_lock or page table lock for page tables. In this case, - * it will always succeed, and you could have used a plain folio_get(), - * but it's sometimes more convenient to have a common function called - * from both locked and RCU-protected contexts. - * - * Return: True if the reference count was successfully incremented. - */ -static inline bool folio_try_get_rcu(struct folio *folio) +static inline bool folio_ref_try_add(struct folio *folio, int count) { - return folio_ref_try_add_rcu(folio, 1); + return folio_ref_add_unless(folio, count, 0); } static inline int page_ref_freeze(struct page *page, int count) diff --git a/mm/filemap.c b/mm/filemap.c index 9fe5c02ae92e..0fb5f3097094 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1847,7 +1847,7 @@ void *filemap_get_entry(struct address_space *mapping, pgoff_t index) if (!folio || xa_is_value(folio)) goto out; - if (!folio_try_get_rcu(folio)) + if (!folio_try_get(folio)) goto repeat; if (unlikely(folio != xas_reload(&xas))) { @@ -2001,7 +2001,7 @@ static inline struct folio *find_get_entry(struct xa_state *xas, pgoff_t max, if (!folio || xa_is_value(folio)) return folio; - if (!folio_try_get_rcu(folio)) + if (!folio_try_get(folio)) goto reset; if (unlikely(folio != xas_reload(xas))) { @@ -2181,7 +2181,7 @@ unsigned filemap_get_folios_contig(struct address_space *mapping, if (xa_is_value(folio)) goto update_start; - if (!folio_try_get_rcu(folio)) + if (!folio_try_get(folio)) goto retry; if (unlikely(folio != xas_reload(&xas))) @@ -2313,7 +2313,7 @@ static void filemap_get_read_batch(struct address_space *mapping, break; if (xa_is_sibling(folio)) break; - if (!folio_try_get_rcu(folio)) + if (!folio_try_get(folio)) goto retry; if (unlikely(folio != xas_reload(&xas))) @@ -3473,7 +3473,7 @@ static struct folio *next_uptodate_folio(struct xa_state *xas, continue; if (folio_test_locked(folio)) continue; - if (!folio_try_get_rcu(folio)) + if (!folio_try_get(folio)) continue; /* Has the page moved or been split? */ if (unlikely(folio != xas_reload(xas))) diff --git a/mm/gup.c b/mm/gup.c index e17466fd62bb..17f89e8d31f1 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -78,7 +78,7 @@ static inline struct folio *try_get_folio(struct page *page, int refs) folio = page_folio(page); if (WARN_ON_ONCE(folio_ref_count(folio) < 0)) return NULL; - if (unlikely(!folio_ref_try_add_rcu(folio, refs))) + if (unlikely(!folio_ref_try_add(folio, refs))) return NULL; /* -- 2.41.0

1 year

4
7
0 0

[PATCH v4] perf/core: Fix missing wakeup when waiting for context reference

by Haifeng Xu

In our production environment, we found many hung tasks which are blocked for more than 18 hours. Their call traces are like this: [346278.191038] __schedule+0x2d8/0x890 [346278.191046] schedule+0x4e/0xb0 [346278.191049] perf_event_free_task+0x220/0x270 [346278.191056] ? init_wait_var_entry+0x50/0x50 [346278.191060] copy_process+0x663/0x18d0 [346278.191068] kernel_clone+0x9d/0x3d0 [346278.191072] __do_sys_clone+0x5d/0x80 [346278.191076] __x64_sys_clone+0x25/0x30 [346278.191079] do_syscall_64+0x5c/0xc0 [346278.191083] ? syscall_exit_to_user_mode+0x27/0x50 [346278.191086] ? do_syscall_64+0x69/0xc0 [346278.191088] ? irqentry_exit_to_user_mode+0x9/0x20 [346278.191092] ? irqentry_exit+0x19/0x30 [346278.191095] ? exc_page_fault+0x89/0x160 [346278.191097] ? asm_exc_page_fault+0x8/0x30 [346278.191102] entry_SYSCALL_64_after_hwframe+0x44/0xae The task was waiting for the refcount become to 1, but from the vmcore, we found the refcount has already been 1. It seems that the task didn't get woken up by perf_event_release_kernel() and got stuck forever. The below scenario may cause the problem. Thread A Thread B ... ... perf_event_free_task perf_event_release_kernel ... acquire event->child_mutex ... get_ctx ... release event->child_mutex acquire ctx->mutex ... perf_free_event (acquire/release event->child_mutex) ... release ctx->mutex wait_var_event acquire ctx->mutex acquire event->child_mutex # move existing events to free_list release event->child_mutex release ctx->mutex put_ctx ... ... In this case, all events of the ctx have been freed, so we couldn't find the ctx in free_list and Thread A will miss the wakeup. It's thus necessary to add a wakeup after dropping the reference. Fixes: 1cf8dfe8a661 ("perf/core: Fix race between close() and fork()") Cc: stable(a)vger.kernel.org Signed-off-by: Haifeng Xu <haifeng.xu(a)shopee.com> Reviewed-by: Frederic Weisbecker <frederic(a)kernel.org> Acked-by: Mark Rutland <mark.rutland(a)arm.com> --- Changes since v1: - Add the fixed tag. - Simplify v1's patch. (Frederic) Changes since v2: - Use Reviewed-by tag instead of Signed-off-by tag. Changes since v3: - Add Acked-by tag. - Cc stable(a)vger.kernel.org. (Mark) --- kernel/events/core.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/kernel/events/core.c b/kernel/events/core.c index 4f0c45ab8d7d..15c35070db6a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5340,6 +5340,7 @@ int perf_event_release_kernel(struct perf_event *event) again: mutex_lock(&event->child_mutex); list_for_each_entry(child, &event->child_list, child_list) { + void *var = NULL; /* * Cannot change, child events are not migrated, see the @@ -5380,11 +5381,23 @@ int perf_event_release_kernel(struct perf_event *event) * this can't be the last reference. */ put_event(event); + } else { + var = &ctx->refcount; } mutex_unlock(&event->child_mutex); mutex_unlock(&ctx->mutex); put_ctx(ctx); + + if (var) { + /* + * If perf_event_free_task() has deleted all events from the + * ctx while the child_mutex got released above, make sure to + * notify about the preceding put_ctx(). + */ + smp_mb(); /* pairs with wait_var_event() */ + wake_up_var(var); + } goto again; } mutex_unlock(&event->child_mutex); -- 2.25.1

1 year

4
3
0 0

[PATCH] locking/atomic: fix trivial typo in comment

by Carlos Llamas

For atomic_sub_and_test() the @i parameter is the value to subtract, not add. Fix the kerneldoc comment accordingly. Fixes: ad8110706f38 ("locking/atomic: scripts: generate kerneldoc comments") Cc: Mark Rutland <mark.rutland(a)arm.com> Cc: stable(a)vger.kernel.org Signed-off-by: Carlos Llamas <cmllamas(a)google.com> --- include/linux/atomic/atomic-instrumented.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/atomic/atomic-instrumented.h b/include/linux/atomic/atomic-instrumented.h index debd487fe971..12b558c05384 100644 --- a/include/linux/atomic/atomic-instrumented.h +++ b/include/linux/atomic/atomic-instrumented.h @@ -1349,7 +1349,7 @@ atomic_try_cmpxchg_relaxed(atomic_t *v, int *old, int new) /** * atomic_sub_and_test() - atomic subtract and test if zero with full ordering - * @i: int value to add + * @i: int value to subtract * @v: pointer to atomic_t * * Atomically updates @v to (@v - @i) with full ordering. -- 2.45.0.rc1.225.g2a3ae87e7f-goog

1 year

6
10
0 0

[PATCH stable 5.15] scripts/gdb: fix SB_* constants parsing

by Florian Fainelli

commit 6a59cb5158bff13b80f116305155fbe4967a5010 upstream --0000000000009a0c9905fd9173ad Content-Transfer-Encoding: 8bit After f15afbd34d8f ("fs: fix undefined behavior in bit shift for SB_NOUSER") the constants were changed from plain integers which LX_VALUE() can parse to constants using the BIT() macro which causes the following: Reading symbols from build/linux-custom/vmlinux...done. Traceback (most recent call last): File "/home/fainelli/work/buildroot/output/arm64/build/linux-custom/vmlinux-gdb.py", line 25, in <module> import linux.constants File "/home/fainelli/work/buildroot/output/arm64/build/linux-custom/scripts/gdb/linux/constants.py", line 5 LX_SB_RDONLY = ((((1UL))) << (0)) Use LX_GDBPARSED() which does not suffer from that issue. Fixes: f15afbd34d8f ("fs: fix undefined behavior in bit shift for SB_NOUSER") Link: https://lkml.kernel.org/r/20230607221337.2781730-1-florian.fainelli@broadco… Signed-off-by: Florian Fainelli <florian.fainelli(a)broadcom.com> Acked-by: Christian Brauner <brauner(a)kernel.org> Cc: Hao Ge <gehao(a)kylinos.cn> Cc: Jan Kiszka <jan.kiszka(a)siemens.com> Cc: Kieran Bingham <kbingham(a)kernel.org> Cc: Luis Chamberlain <mcgrof(a)kernel.org> Cc: Pankaj Raghav <p.raghav(a)samsung.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Florian Fainelli <florian.fainelli(a)broadcom.com> --- scripts/gdb/linux/constants.py.in | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/scripts/gdb/linux/constants.py.in b/scripts/gdb/linux/constants.py.in index 08f0587d15ea..0ff707bc1896 100644 --- a/scripts/gdb/linux/constants.py.in +++ b/scripts/gdb/linux/constants.py.in @@ -46,12 +46,12 @@ if IS_BUILTIN(CONFIG_COMMON_CLK): LX_GDBPARSED(CLK_GET_RATE_NOCACHE) /* linux/fs.h */ -LX_VALUE(SB_RDONLY) -LX_VALUE(SB_SYNCHRONOUS) -LX_VALUE(SB_MANDLOCK) -LX_VALUE(SB_DIRSYNC) -LX_VALUE(SB_NOATIME) -LX_VALUE(SB_NODIRATIME) +LX_GDBPARSED(SB_RDONLY) +LX_GDBPARSED(SB_SYNCHRONOUS) +LX_GDBPARSED(SB_MANDLOCK) +LX_GDBPARSED(SB_DIRSYNC) +LX_GDBPARSED(SB_NOATIME) +LX_GDBPARSED(SB_NODIRATIME) /* linux/htimer.h */ LX_GDBPARSED(hrtimer_resolution) -- 2.34.1

1 year

1
1
0 0

[PATCH AUTOSEL 5.4 1/2] nvme-multipath: find NUMA path only for online numa-node

by Sasha Levin

From: Nilay Shroff <nilay(a)linux.ibm.com> [ Upstream commit d3a043733f25d743f3aa617c7f82dbcb5ee2211a ] In current native multipath design when a shared namespace is created, we loop through each possible numa-node, calculate the NUMA distance of that node from each nvme controller and then cache the optimal IO path for future reference while sending IO. The issue with this design is that we may refer to the NUMA distance table for an offline node which may not be populated at the time and so we may inadvertently end up finding and caching a non-optimal path for IO. Then latter when the corresponding numa-node becomes online and hence the NUMA distance table entry for that node is created, ideally we should re-calculate the multipath node distance for the newly added node however that doesn't happen unless we rescan/reset the controller. So essentially, we may keep using non-optimal IO path for a node which is made online after namespace is created. This patch helps fix this issue ensuring that when a shared namespace is created, we calculate the multipath node distance for each online numa-node instead of each possible numa-node. Then latter when a node becomes online and we receive any IO on that newly added node, we would calculate the multipath node distance for newly added node but this time NUMA distance table would have been already populated for newly added node. Hence we would be able to correctly calculate the multipath node distance and choose the optimal path for the IO. Signed-off-by: Nilay Shroff <nilay(a)linux.ibm.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Keith Busch <kbusch(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/nvme/host/multipath.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index 811f7b96b5517..c993548403f5c 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -436,7 +436,7 @@ static void nvme_mpath_set_live(struct nvme_ns *ns) int node, srcu_idx; srcu_idx = srcu_read_lock(&head->srcu); - for_each_node(node) + for_each_online_node(node) __nvme_find_path(head, node); srcu_read_unlock(&head->srcu, srcu_idx); } -- 2.43.0

1 year

1
1
0 0

[PATCH AUTOSEL 5.15 01/12] nvme-multipath: find NUMA path only for online numa-node

by Sasha Levin

From: Nilay Shroff <nilay(a)linux.ibm.com> [ Upstream commit d3a043733f25d743f3aa617c7f82dbcb5ee2211a ] In current native multipath design when a shared namespace is created, we loop through each possible numa-node, calculate the NUMA distance of that node from each nvme controller and then cache the optimal IO path for future reference while sending IO. The issue with this design is that we may refer to the NUMA distance table for an offline node which may not be populated at the time and so we may inadvertently end up finding and caching a non-optimal path for IO. Then latter when the corresponding numa-node becomes online and hence the NUMA distance table entry for that node is created, ideally we should re-calculate the multipath node distance for the newly added node however that doesn't happen unless we rescan/reset the controller. So essentially, we may keep using non-optimal IO path for a node which is made online after namespace is created. This patch helps fix this issue ensuring that when a shared namespace is created, we calculate the multipath node distance for each online numa-node instead of each possible numa-node. Then latter when a node becomes online and we receive any IO on that newly added node, we would calculate the multipath node distance for newly added node but this time NUMA distance table would have been already populated for newly added node. Hence we would be able to correctly calculate the multipath node distance and choose the optimal path for the IO. Signed-off-by: Nilay Shroff <nilay(a)linux.ibm.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Keith Busch <kbusch(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/nvme/host/multipath.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index 73eddb67f0d24..9ddaa599a9cd0 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -516,7 +516,7 @@ static void nvme_mpath_set_live(struct nvme_ns *ns) int node, srcu_idx; srcu_idx = srcu_read_lock(&head->srcu); - for_each_node(node) + for_each_online_node(node) __nvme_find_path(head, node); srcu_read_unlock(&head->srcu, srcu_idx); } -- 2.43.0

1 year

1
11
0 0

[PATCH AUTOSEL 6.8 01/18] nvme-multipath: find NUMA path only for online numa-node

by Sasha Levin

From: Nilay Shroff <nilay(a)linux.ibm.com> [ Upstream commit d3a043733f25d743f3aa617c7f82dbcb5ee2211a ] In current native multipath design when a shared namespace is created, we loop through each possible numa-node, calculate the NUMA distance of that node from each nvme controller and then cache the optimal IO path for future reference while sending IO. The issue with this design is that we may refer to the NUMA distance table for an offline node which may not be populated at the time and so we may inadvertently end up finding and caching a non-optimal path for IO. Then latter when the corresponding numa-node becomes online and hence the NUMA distance table entry for that node is created, ideally we should re-calculate the multipath node distance for the newly added node however that doesn't happen unless we rescan/reset the controller. So essentially, we may keep using non-optimal IO path for a node which is made online after namespace is created. This patch helps fix this issue ensuring that when a shared namespace is created, we calculate the multipath node distance for each online numa-node instead of each possible numa-node. Then latter when a node becomes online and we receive any IO on that newly added node, we would calculate the multipath node distance for newly added node but this time NUMA distance table would have been already populated for newly added node. Hence we would be able to correctly calculate the multipath node distance and choose the optimal path for the IO. Signed-off-by: Nilay Shroff <nilay(a)linux.ibm.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Keith Busch <kbusch(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/nvme/host/multipath.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index 75386d3e0f981..615fbdc09d1cc 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -594,7 +594,7 @@ static void nvme_mpath_set_live(struct nvme_ns *ns) int node, srcu_idx; srcu_idx = srcu_read_lock(&head->srcu); - for_each_node(node) + for_each_online_node(node) __nvme_find_path(head, node); srcu_read_unlock(&head->srcu, srcu_idx); } -- 2.43.0

1 year

1
17
0 0

[PATCH AUTOSEL 6.9 01/23] nvme-multipath: find NUMA path only for online numa-node

by Sasha Levin

From: Nilay Shroff <nilay(a)linux.ibm.com> [ Upstream commit d3a043733f25d743f3aa617c7f82dbcb5ee2211a ] In current native multipath design when a shared namespace is created, we loop through each possible numa-node, calculate the NUMA distance of that node from each nvme controller and then cache the optimal IO path for future reference while sending IO. The issue with this design is that we may refer to the NUMA distance table for an offline node which may not be populated at the time and so we may inadvertently end up finding and caching a non-optimal path for IO. Then latter when the corresponding numa-node becomes online and hence the NUMA distance table entry for that node is created, ideally we should re-calculate the multipath node distance for the newly added node however that doesn't happen unless we rescan/reset the controller. So essentially, we may keep using non-optimal IO path for a node which is made online after namespace is created. This patch helps fix this issue ensuring that when a shared namespace is created, we calculate the multipath node distance for each online numa-node instead of each possible numa-node. Then latter when a node becomes online and we receive any IO on that newly added node, we would calculate the multipath node distance for newly added node but this time NUMA distance table would have been already populated for newly added node. Hence we would be able to correctly calculate the multipath node distance and choose the optimal path for the IO. Signed-off-by: Nilay Shroff <nilay(a)linux.ibm.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Keith Busch <kbusch(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/nvme/host/multipath.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index d16e976ae1a47..9c1e135b8df3b 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -595,7 +595,7 @@ static void nvme_mpath_set_live(struct nvme_ns *ns) int node, srcu_idx; srcu_idx = srcu_read_lock(&head->srcu); - for_each_node(node) + for_each_online_node(node) __nvme_find_path(head, node); srcu_read_unlock(&head->srcu, srcu_idx); } -- 2.43.0

1 year

1
22
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror June 2024