- 17 Dec, 2017 40 commits
-
-
Leon Romanovsky authored
[ Upstream commit 7d7d065a ] Chelsio cxgb4 HW is big-endian, hence there is need to properly annotate r2 and stag fields as __be32 and not __u32 to fix the following sparse warnings. drivers/infiniband/hw/cxgb4/qp.c:614:16: warning: incorrect type in assignment (different base types) expected unsigned int [unsigned] [usertype] r2 got restricted __be32 [usertype] <noident> drivers/infiniband/hw/cxgb4/qp.c:615:18: warning: incorrect type in assignment (different base types) expected unsigned int [unsigned] [usertype] stag got restricted __be32 [usertype] <noident> Cc: Steve Wise <swise@opengridcomputing.com> Signed-off-by:
Leon Romanovsky <leon@kernel.org> Reviewed-by:
Steve Wise <swise@opengridcomputing.com> Signed-off-by:
Doug Ledford <dledford@redhat.com> Signed-off-by:
Sasha Levin <alexander.levin@verizon.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Zdenek Kabelac authored
[ Upstream commit 0868b99c ] When bitmap is resized, the old kalloced chunks just are not released once the resized bitmap starts to use new space. This fixes in particular kmemleak reports like this one: unreferenced object 0xffff8f4311e9c000 (size 4096): comm "lvm", pid 19333, jiffies 4295263268 (age 528.265s) hex dump (first 32 bytes): 02 80 02 80 02 80 02 80 02 80 02 80 02 80 02 80 ................ 02 80 02 80 02 80 02 80 02 80 02 80 02 80 02 80 ................ backtrace: [<ffffffffa69471ca>] kmemleak_alloc+0x4a/0xa0 [<ffffffffa628c10e>] kmem_cache_alloc_trace+0x14e/0x2e0 [<ffffffffa676cfec>] bitmap_checkpage+0x7c/0x110 [<ffffffffa676d0c5>] bitmap_get_counter+0x45/0xd0 [<ffffffffa676d6b3>] bitmap_set_memory_bits+0x43/0xe0 [<ffffffffa676e41c>] bitmap_init_from_disk+0x23c/0x530 [<ffffffffa676f1ae>] bitmap_load+0xbe/0x160 [<ffffffffc04c47d3>] raid_preresume+0x203/0x2f0 [dm_raid] [<ffffffffa677762f>] dm_table_resume_targets+0x4f/0xe0 [<ffffffffa6774b52>] dm_resume+0x122/0x140 [<ffffffffa6779b9f>] dev_suspend+0x18f/0x290 [<ffffffffa677a3a7>] ctl_ioctl+0x287/0x560 [<ffffffffa677a693>] dm_ctl_ioctl+0x13/0x20 [<ffffffffa62d6b46>] do_vfs_ioctl+0xa6/0x750 [<ffffffffa62d7269>] SyS_ioctl+0x79/0x90 [<ffffffffa6956d41>] entry_SYSCALL_64_fastpath+0x1f/0xc2 Signed-off-by:
Zdenek Kabelac <zkabelac@redhat.com> Signed-off-by:
Shaohua Li <shli@fb.com> Signed-off-by:
-
Heinz Mauelshagen authored
[ Upstream commit 23397844 ] Requesting a sync on an active raid device via a table reload (see 'sync' parameter in Documentation/device-mapper/dm-raid.txt) skips the super_load() call that defines the superblock size (rdev->sb_size) -- resulting in an oops if/when super_sync()->memset() is called. Fix by moving the initialization of the superblock start and size out of super_load() to the caller (analyse_superblocks). Signed-off-by:
Heinz Mauelshagen <heinzm@redhat.com> Signed-off-by:
Mike Snitzer <snitzer@redhat.com> Signed-off-by:
-
Paul Moore authored
[ Upstream commit 173743dd ] Prior to this patch we enabled audit in audit_init(), which is too late for PID 1 as the standard initcalls are run after the PID 1 task is forked. This means that we never allocate an audit_context (see audit_alloc()) for PID 1 and therefore miss a lot of audit events generated by PID 1. This patch enables audit as early as possible to help ensure that when PID 1 is forked it can allocate an audit_context if required. Reviewed-by:
Richard Guy Briggs <rgb@redhat.com> Signed-off-by:
Paul Moore <paul@paul-moore.com> Signed-off-by:
-
Steve Grubb authored
[ Upstream commit 33e8a907 ] The API to end auditing has historically been for auditd to set the pid to 0. This patch restores that functionality. See: https://github.com/linux-audit/audit-kernel/issues/69Reviewed-by:
Steve Grubb <sgrubb@redhat.com> Signed-off-by:
-
Israel Rukshin authored
[ Upstream commit 43b92fd2 ] A NULL deref happens when nvmet_rdma_remove_one() is called more than once (e.g. while connected via 2 ports). The first call frees the queues related to the first ib_device but doesn't remove them from the queue list. While calling nvmet_rdma_remove_one() for the second ib_device it goes over the full queue list again and we get the NULL deref. Fixes: f1d4ef7d ("nvmet-rdma: register ib_client to not deadlock in device removal") Signed-off-by:
Israel Rukshin <israelr@mellanox.com> Reviewed-by:
Max Gurtovoy <maxg@mellanox.com> Reviewed-by:
Sagi Grimberg <sagi@grmberg.me> Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
Jens Axboe <axboe@kernel.dk> Signed-off-by:
-
Bart Van Assche authored
[ Upstream commit aba7afc5 ] Avoid that removal of a request queue sporadically triggers the following warning: list_del corruption. next->prev should be ffff8807d649b970, but was 6b6b6b6b6b6b6b6b WARNING: CPU: 3 PID: 342 at lib/list_debug.c:56 __list_del_entry_valid+0x92/0xa0 Call Trace: process_one_work+0x11b/0x660 worker_thread+0x3d/0x3b0 kthread+0x129/0x140 ret_from_fork+0x27/0x40 Signed-off-by:
Bart Van Assche <bart.vanassche@wdc.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.com> Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by:
-
Hongxu Jia authored
[ Upstream commit 8dc7a31f ] Compile ide-atapi failed with defining macro "DEBUG" ... |drivers/ide/ide-atapi.c:285:52: error: 'struct request' has no member named 'cmd'; did you mean 'csd'? | debug_log("%s: rq->cmd[0]: 0x%x\n", __func__, rq->cmd[0]); ... Since we split the scsi_request out of struct request, it missed do the same thing on debug_log Fixes: 82ed4db4 ("block: split scsi_request out of struct request") Signed-off-by:
Hongxu Jia <hongxu.jia@windriver.com> Signed-off-by:
-
Keefe Liu authored
[ Upstream commit ca29fd7c ] When process the outbound packet of ipv6, we should assign the master device to output device other than input device. Signed-off-by:
Keefe Liu <liuqifa@huawei.com> Acked-by:
Mahesh Bandewar <maheshb@google.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Vaidyanathan Srinivasan authored
[ Upstream commit 8d4e10e9 ] On PowerNV platforms, firmware provides exit latency and target residency for each of the idle states in nano seconds. Cpuidle framework expects the values in micro seconds. Round up to nearest micro seconds to avoid errors in cases where the values are defined as fractional micro seconds. Default idle state of 'snooze' has exit latency of zero. If other states have fractional micro second exit latency, they would get rounded down to zero micro second and make cpuidle framework choose deeper idle state when snooze loop is the right choice. Reported-by:
Anton Blanchard <anton@samba.org> Signed-off-by:
Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com> Reviewed-by:
Gautham R. Shenoy <ego@linux.vnet.ibm.com> Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au> Signed-off-by:
-
Masahiro Yamada authored
[ Upstream commit 433dc2eb ] Some $(call cc-option,...) are invoked very early, even before KBUILD_CFLAGS, etc. are initialized. The returned string from $(call cc-option,...) depends on KBUILD_CPPFLAGS, KBUILD_CFLAGS, and GCC_PLUGINS_CFLAGS. Since they are exported, they are not empty when the top Makefile is recursively invoked. The recursion occurs in several places. For example, the top Makefile invokes itself for silentoldconfig. "make tinyconfig", "make rpm-pkg" are the cases, too. In those cases, the second call of cc-option from the same line runs a different shell command due to non-pristine KBUILD_CFLAGS. To get the same result all the time, KBUILD_* and GCC_PLUGINS_CFLAGS must be initialized before any call of cc-option. This avoids garbage data in the .cache.mk file. Move all calls of cc-option below the config targets because target compiler flags are unnecessary for Kconfig. Signed-off-by:
Masahiro Yamada <yamada.masahiro@socionext.com> Reviewed-by:
Douglas Anderson <dianders@chromium.org> Signed-off-by:
-
Marc Zyngier authored
commit 64afe6e9 upstream. The current pending table parsing code assumes that we keep the previous read of the pending bits, but keep that variable in the current block, making sure it is discarded on each loop. We end-up using whatever is on the stack. Who knows, it might just be the right thing... Fixes: 33d3bc95 ("KVM: arm64: vgic-its: Read initial LPI pending table") Reported-by:
AKASHI Takahiro <takahiro.akashi@linaro.org> Reviewed-by:
Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by:
Marc Zyngier <marc.zyngier@arm.com> Signed-off-by:
-
Al Viro authored
commit a5739435 upstream. 1) it's fput() or sock_release(), not both 2) don't do fd_install() until the last failure exit. 3) not a bug per se, but... don't attach socket to struct file until it's set up. Take reserving descriptor into the caller, move fd_install() to the caller, sanitize failure exits and calling conventions. Acked-by:
Tom Herbert <tom@herbertland.com> Signed-off-by:
Al Viro <viro@zeniv.linux.org.uk> Signed-off-by:
-
Jeff Layton authored
commit 4d2dc2cc upstream. Currently, we're capping the values too low in the F_GETLK64 case. The fields in that structure are 64-bit values, so we shouldn't need to do any sort of fixup there. Make sure we check that assumption at build time in the future however by ensuring that the sizes we're copying will fit. With this, we no longer need COMPAT_LOFF_T_MAX either, so remove it. Fixes: 94073ad7 (fs/locks: don't mess with the address limit in compat_fcntl64) Reported-by:
Vitaly Lipatov <lav@etersoft.ru> Signed-off-by:
Jeff Layton <jlayton@redhat.com> Reviewed-by:
David Howells <dhowells@redhat.com> Signed-off-by:
-
Vincent Pelletier authored
commit 30bf90cc upstream. Found using DEBUG_ATOMIC_SLEEP while submitting an AIO read operation: [ 100.853642] BUG: sleeping function called from invalid context at mm/slab.h:421 [ 100.861148] in_atomic(): 1, irqs_disabled(): 1, pid: 1880, name: python [ 100.867954] 2 locks held by python/1880: [ 100.867961] #0: (&epfile->mutex){....}, at: [<f8188627>] ffs_mutex_lock+0x27/0x30 [usb_f_fs] [ 100.868020] #1: (&(&ffs->eps_lock)->rlock){....}, at: [<f818ad4b>] ffs_epfile_io.isra.17+0x24b/0x590 [usb_f_fs] [ 100.868076] CPU: 1 PID: 1880 Comm: python Not tainted 4.14.0-edison+ #118 [ 100.868085] Hardware name: Intel Corporation Merrifield/BODEGA BAY, BIOS 542 2015.01.21:18.19.48 [ 100.868093] Call Trace: [ 100.868122] dump_stack+0x47/0x62 [ 100.868156] ___might_sleep+0xfd/0x110 [ 100.868182] __might_sleep+0x68/0x70 [ 100.868217] kmem_cache_alloc_trace+0x4b/0x200 [ 100.868248] ? dwc3_gadget_ep_alloc_request+0x24/0xe0 [dwc3] [ 100.868302] dwc3_gadget_ep_alloc_request+0x24/0xe0 [dwc3] [ 100.868343] usb_ep_alloc_request+0x16/0xc0 [udc_core] [ 100.868386] ffs_epfile_io.isra.17+0x444/0x590 [usb_f_fs] [ 100.868424] ? _raw_spin_unlock_irqrestore+0x27/0x40 [ 100.868457] ? kiocb_set_cancel_fn+0x57/0x60 [ 100.868477] ? ffs_ep0_poll+0xc0/0xc0 [usb_f_fs] [ 100.868512] ffs_epfile_read_iter+0xfe/0x157 [usb_f_fs] [ 100.868551] ? security_file_permission+0x9c/0xd0 [ 100.868587] ? rw_verify_area+0xac/0x120 [ 100.868633] aio_read+0x9d/0x100 [ 100.868692] ? __fget+0xa2/0xd0 [ 100.868727] ? __might_sleep+0x68/0x70 [ 100.868763] SyS_io_submit+0x471/0x680 [ 100.868878] do_int80_syscall_32+0x4e/0xd0 [ 100.868921] entry_INT80_32+0x2a/0x2a [ 100.868932] EIP: 0xb7fbb676 [ 100.868941] EFLAGS: 00000292 CPU: 1 [ 100.868951] EAX: ffffffda EBX: b7aa2000 ECX: 00000002 EDX: b7af8368 [ 100.868961] ESI: b7fbb660 EDI: b7aab000 EBP: bfb6c658 ESP: bfb6c638 [ 100.868973] DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 007b Signed-off-by:
Vincent Pelletier <plr.vincent@gmail.com> Signed-off-by:
Felipe Balbi <felipe.balbi@linux.intel.com> Signed-off-by:
-
Masamitsu Yamazaki authored
commit 4f7f5551 upstream. System may crash after unloading ipmi_si.ko module because a timer may remain and fire after the module cleaned up resources. cleanup_one_si() contains the following processing. /* * Make sure that interrupts, the timer and the thread are * stopped and will not run again. */ if (to_clean->irq_cleanup) to_clean->irq_cleanup(to_clean); wait_for_timer_and_thread(to_clean); /* * Timeouts are stopped, now make sure the interrupts are off * in the BMC. Note that timers and CPU interrupts are off, * so no need for locks. */ while (to_clean->curr_msg || (to_clean->si_state != SI_NORMAL)) { poll(to_clean); schedule_timeout_uninterruptible(1); } si_state changes as following in the while loop calling poll(to_clean). SI_GETTING_MESSAGES => SI_CHECKING_ENABLES => SI_SETTING_ENABLES => SI_GETTING_EVENTS => SI_NORMAL As written in the code comments above, timers are expected to stop before the polling loop and not to run again. But the timer is set again in the following process when si_state becomes SI_SETTING_ENABLES. => poll => smi_event_handler => handle_transaction_done // smi_info->si_state == SI_SETTING_ENABLES => start_getting_events => start_new_msg => smi_mod_timer => mod_timer As a result, before the timer set in start_new_msg() expires, the polling loop may see si_state becoming SI_NORMAL and the module clean-up finishes. For example, hard LOCKUP and panic occurred as following. smi_timeout was called after smi_event_handler, kcs_event and hangs at port_inb() trying to access I/O port after release. [exception RIP: port_inb+19] RIP: ffffffffc0473053 RSP: ffff88069fdc3d80 RFLAGS: 00000006 RAX: ffff8806800f8e00 RBX: ffff880682bd9400 RCX: 0000000000000000 RDX: 0000000000000ca3 RSI: 0000000000000ca3 RDI: ffff8806800f8e40 RBP: ffff88069fdc3d80 R8: ffffffff81d86dfc R9: ffffffff81e36426 R10: 00000000000509f0 R11: 0000000000100000 R12: 0000000000]:000000 R13: 0000000000000000 R14: 0000000000000246 R15: ffff8806800f8e00 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0000 --- <NMI exception stack> --- To fix the problem I defined a flag, timer_can_start, as member of struct smi_info. The flag is enabled immediately after initializing the timer and disabled immediately before waiting for timer deletion. Fixes: 0cfec916 ("ipmi: Start the timer and thread on internal msgs") Signed-off-by:
Yamazaki Masamitsu <m-yamazaki@ah.jp.nec.com> [Some fairly major changes went into the IPMI driver in 4.15, so this required a backport as the code had changed and moved to a different file.] Signed-off-by:
Corey Minyard <cminyard@mvista.com> Signed-off-by:
-
Xin Long authored
[ Upstream commit a8dd3979 ] Commit d04adf1b ("sctp: reset owner sk for data chunks on out queues when migrating a sock") made a mistake that using 'list' as the param of list_for_each_entry to traverse the retransmit, sacked and abandoned queues, while chunks are using 'transmitted_list' to link into these queues. It could cause NULL dereference panic if there are chunks in any of these queues when peeling off one asoc. So use the chunk member 'transmitted_list' instead in this patch. Fixes: d04adf1b ("sctp: reset owner sk for data chunks on out queues when migrating a sock") Signed-off-by:
Xin Long <lucien.xin@gmail.com> Acked-by:
Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Acked-by:
Neil Horman <nhorman@tuxdriver.com> Signed-off-by:
-
Jakub Kicinski authored
[ Upstream commit 25415cec ] When cls_bpf offload was added it seemed like a good idea to call cls_bpf_delete_prog() instead of extending the error handling path, since the software state is fully initialized at that point. This handling of errors without jumping to the end of the function is error prone, as proven by later commit missing that extra call to __cls_bpf_delete_prog(). __cls_bpf_delete_prog() is now expected to be invoked with a reference on exts->net or the field zeroed out. The call on the offload's error patch does not fullfil this requirement, leading to each error stealing a reference on net namespace. Create a function undoing what cls_bpf_set_parms() did and use it from __cls_bpf_delete_prog() and the error path. Fixes: aae2c35e ("cls_bpf: use tcf_exts_get_net() before call_rcu()") Signed-off-by:
Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by:
Simon Horman <simon.horman@netronome.com> Acked-by:
Daniel Borkmann <daniel@iogearbox.net> Acked-by:
Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by:
-
Gustavo A. R. Silva authored
[ Upstream commit 2734166e ] gso_type is being used in binary AND operations together with SKB_GSO_UDP. The issue is that variable gso_type is of type unsigned short and SKB_GSO_UDP expands to more than 16 bits: SKB_GSO_UDP = 1 << 16 this makes any binary AND operation between gso_type and SKB_GSO_UDP to be always zero, hence making some code unreachable and likely causing undesired behavior. Fix this by changing the data type of variable gso_type to unsigned int. Addresses-Coverity-ID: 1462223 Fixes: 0c19f846 ("net: accept UFO datagrams from tuntap and packet") Signed-off-by:
Gustavo A. R. Silva <garsilva@embeddedor.com> Acked-by:
Willem de Bruijn <willemb@google.com> Signed-off-by:
-
Willem de Bruijn authored
[ Upstream commit 0c19f846 ] Tuntap and similar devices can inject GSO packets. Accept type VIRTIO_NET_HDR_GSO_UDP, even though not generating UFO natively. Processes are expected to use feature negotiation such as TUNSETOFFLOAD to detect supported offload types and refrain from injecting other packets. This process breaks down with live migration: guest kernels do not renegotiate flags, so destination hosts need to expose all features that the source host does. Partially revert the UFO removal from 182e0b6b~1..d9d30adf. This patch introduces nearly(*) no new code to simplify verification. It brings back verbatim tuntap UFO negotiation, VIRTIO_NET_HDR_GSO_UDP insertion and software UFO segmentation. It does not reinstate protocol stack support, hardware offload (NETIF_F_UFO), SKB_GSO_UDP tunneling in SKB_GSO_SOFTWARE or reception of VIRTIO_NET_HDR_GSO_UDP packets in tuntap. To support SKB_GSO_UDP reappearing in the stack, also reinstate logic in act_csum and openvswitch. Achieve equivalence with v4.13 HEAD by squashing in commit 93991221 ("net: skb_needs_check() removes CHECKSUM_UNNECESSARY check for tx.") and reverting commit 8d63bee6 ("net: avoid skb_warn_bad_offload false positives on UFO"). (*) To avoid having to bring back skb_shinfo(skb)->ip6_frag_id, ipv6_proxy_select_ident is changed to return a __be32 and this is assigned directly to the frag_hdr. Also, SKB_GSO_UDP is inserted at the end of the enum to minimize code churn. Tested Booted a v4.13 guest kernel with QEMU. On a host kernel before this patch `ethtool -k eth0` shows UFO disabled. After the patch, it is enabled, same as on a v4.13 host kernel. A UFO packet sent from the guest appears on the tap device: host: nc -l -p -u 8000 & tcpdump -n -i tap0 guest: dd if=/dev/zero of=payload.txt bs=1 count=2000 nc -u 192.16.1.1 8000 < payload.txt Direct tap to tap transmission of VIRTIO_NET_HDR_GSO_UDP succeeds, packets arriving fragmented: ./with_tap_pair.sh ./tap_send_ufo tap0 tap1 (from https://github.com/wdebruij/kerneltools/tree/master/tests) Changes v1 -> v2 - simplified set_offload change (review comment) - documented test procedure Link: http://lkml.kernel.org/r/<CAF=yD-LuUeDuL9YWPJD9ykOZ0QCjNeznPDr6whqZ9NGMNF12Mw@mail.gmail.com> Fixes: fb652fdf ("macvlan/macvtap: Remove NETIF_F_UFO advertisement.") Reported-by:
Michal Kubecek <mkubecek@suse.cz> Signed-off-by:
Jason Wang <jasowang@redhat.com> Signed-off-by:
-
Xin Long authored
[ Upstream commit 654d5738 ] rcu_read_lock in tun_build_skb is used to rcu_dereference tun->xdp_prog safely, rcu_read_unlock should be done in every return path. Now I could see one place missing it, where it returns NULL in switch-case XDP_REDIRECT, another palce using rcu_read_lock wrongly, where it returns NULL in if (xdp_xmit) chunk. So fix both in this patch. Fixes: 761876c8 ("tap: XDP support") Signed-off-by:
-
David Ahern authored
[ Upstream commit 98d11291 ] Florian reported a breakage with anycast routes due to commit 4832c30d ("net: ipv6: put host and anycast routes on device with address"). Prior to this commit anycast routes were added against the loopback device causing repetitive route entries with no insight into why they existed. e.g.: $ ip -6 ro ls table local type anycast anycast 2001:db8:1:: dev lo proto kernel metric 0 pref medium anycast 2001:db8:2:: dev lo proto kernel metric 0 pref medium anycast fe80:: dev lo proto kernel metric 0 pref medium anycast fe80:: dev lo proto kernel metric 0 pref medium The point of commit 4832c30d is to add the routes using the device with the address which is causing the route to be added. e.g.,: $ ip -6 ro ls table local type anycast anycast 2001:db8:1:: dev eth1 proto kernel metric 0 pref medium anycast 2001:db8:2:: dev eth2 proto kernel metric 0 pref medium anycast fe80:: dev eth2 proto kernel metric 0 pref medium anycast fe80:: dev eth1 proto kernel metric 0 pref medium For traffic to work as it did before, the dst device needs to be switched to the loopback when the copy is created similar to local routes. Fixes: 4832c30d ("net: ipv6: put host and anycast routes on device with address") Signed-off-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
-
Wei Xu authored
[ Upstream commit c33ee15b ] tun_recvmsg() supports accepting skb by msg_control after commit ac77cfd4 ("tun: support receiving skb through msg_control"), the skb if presented should be freed no matter how far it can go along, otherwise it would be leaked. This patch fixes several missed cases. Signed-off-by:
Wei Xu <wexu@redhat.com> Reported-by:
Matthew Rosato <mjrosato@linux.vnet.ibm.com> Acked-by:
Michael S. Tsirkin <mst@redhat.com> Signed-off-by:
-
Neal Cardwell authored
[ Upstream commit ed66dfaf ] Fix the TLP scheduling logic so that when scheduling a TLP probe, we ensure that the estimated time at which an RTO would fire accounts for the fact that ACKs indicating forward progress should push back RTO times. After the following fix: df92c839 ("tcp: fix xmit timer to only be reset if data ACKed/SACKed") we had an unintentional behavior change in the following kind of scenario: suppose the RTT variance has been very low recently. Then suppose we send out a flight of N packets and our RTT is 100ms: t=0: send a flight of N packets t=100ms: receive an ACK for N-1 packets The response before df92c839 that was: -> schedule a TLP for now + RTO_interval The response after df92c839 is: -> schedule a TLP for t=0 + RTO_interval Since RTO_interval = srtt + RTT_variance, this means that we have scheduled a TLP timer at a point in the future that only accounts for RTT_variance. If the RTT_variance term is small, this means that the timer fires soon. Before df92c839 this would not happen, because in that code, when we receive an ACK for a prefix of flight, we did: 1) Near the top of tcp_ack(), switch from TLP timer to RTO at write_queue_head->paket_tx_time + RTO_interval: if (icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) tcp_rearm_rto(sk); 2) In tcp_clean_rtx_queue(), update the RTO to now + RTO_interval: if (flag & FLAG_ACKED) { tcp_rearm_rto(sk); 3) In tcp_ack() after tcp_fastretrans_alert() switch from RTO to TLP at now + RTO_interval: if (icsk->icsk_pending == ICSK_TIME_RETRANS) tcp_schedule_loss_probe(sk); In df92c839 we removed that 3-phase dance, and instead directly set the TLP timer once: we set the TLP timer in cases like this to write_queue_head->packet_tx_time + RTO_interval. So if the RTT variance is small, then this means that this is setting the TLP timer to fire quite soon. This means if the ACK for the tail of the flight takes longer than an RTT to arrive (often due to delayed ACKs), then the TLP timer fires too quickly. Fixes: df92c839 ("tcp: fix xmit timer to only be reset if data ACKed/SACKed") Signed-off-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
Yuchung Cheng <ycheng@google.com> Signed-off-by:
Eric Dumazet <edumazet@google.com> Acked-by:
Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by:
-
Wei Xu authored
[ Upstream commit 61d78537 ] tap_recvmsg() supports accepting skb by msg_control after commit 3b4ba04a ("tap: support receiving skb from msg_control"), the skb if presented should be freed within the function, otherwise it would be leaked. Signed-off-by:
-
Jiri Pirko authored
[ Upstream commit d51aae68 ] q->link.block is not initialized, that leads to EINVAL when one tries to add filter there. So initialize it properly. This can be reproduced by: $ tc qdisc add dev eth0 root handle 1: cbq avpkt 1000 rate 1000Mbit bandwidth 1000Mbit $ tc filter add dev eth0 parent 1: protocol ip prio 100 u32 match ip protocol 0 0x00 flowid 1:1 Reported-by:
Jaroslav Aster <jaster@redhat.com> Reported-by:
Ivan Vecera <ivecera@redhat.com> Fixes: 6529eaba ("net: sched: introduce tcf block infractructure") Signed-off-by:
Jiri Pirko <jiri@mellanox.com> Acked-by:
Eelco Chaudron <echaudro@redhat.com> Reviewed-by:
-
Eric Dumazet authored
[ Upstream commit 86323850 ] When I switched rcv_rtt_est to high resolution timestamps, I forgot that tp->tcp_mstamp needed to be refreshed in tcp_rcv_space_adjust() Using an old timestamp leads to autotuning lags. Fixes: 645f4c6f ("tcp: switch rcv_rtt_est and rcvq_space to high resolution timestamps") Signed-off-by:
-
Tommi Rantala authored
[ Upstream commit c7799c06 ] Remove the second tipc_rcv() call in tipc_udp_recv(). We have just checked that the bearer is not up, and calling tipc_rcv() with a bearer that is not up leads to a TIPC div-by-zero crash in tipc_node_calculate_timer(). The crash is rare in practice, but can happen like this: We're enabling a bearer, but it's not yet up and fully initialized. At the same time we receive a discovery packet, and in tipc_udp_recv() we end up calling tipc_rcv() with the not-yet-initialized bearer, causing later the div-by-zero crash in tipc_node_calculate_timer(). Jon Maloy explains the impact of removing the second tipc_rcv() call: "link setup in the worst case will be delayed until the next arriving discovery messages, 1 sec later, and this is an acceptable delay." As the tipc_rcv() call is removed, just leave the function via the rcu_out label, so that we will kfree_skb(). [ 12.590450] Own node address <1.1.1>, network identity 1 [ 12.668088] divide error: 0000 [#1] SMP [ 12.676952] CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.14.2-dirty #1 [ 12.679225] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-2.fc27 04/01/2014 [ 12.682095] task: ffff8c2a761edb80 task.stack: ffffa41cc0cac000 [ 12.684087] RIP: 0010:tipc_node_calculate_timer.isra.12+0x45/0x60 [tipc] [ 12.686486] RSP: 0018:ffff8c2a7fc838a0 EFLAGS: 00010246 [ 12.688451] RAX: 0000000000000000 RBX: ffff8c2a5b382600 RCX: 0000000000000000 [ 12.691197] RDX: 0000000000000000 RSI: ffff8c2a5b382600 RDI: ffff8c2a5b382600 [ 12.693945] RBP: ffff8c2a7fc838b0 R08: 0000000000000001 R09: 0000000000000001 [ 12.696632] R10: 0000000000000000 R11: 0000000000000000 R12: ffff8c2a5d8949d8 [ 12.699491] R13: ffffffff95ede400 R14: 0000000000000000 R15: ffff8c2a5d894800 [ 12.702338] FS: 0000000000000000(0000) GS:ffff8c2a7fc80000(0000) knlGS:0000000000000000 [ 12.705099] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 12.706776] CR2: 0000000001bb9440 CR3: 00000000bd009001 CR4: 00000000003606e0 [ 12.708847] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 12.711016] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 12.712627] Call Trace: [ 12.713390] <IRQ> [ 12.714011] tipc_node_check_dest+0x2e8/0x350 [tipc] [ 12.715286] tipc_disc_rcv+0x14d/0x1d0 [tipc] [ 12.716370] tipc_rcv+0x8b0/0xd40 [tipc] [ 12.717396] ? minmax_running_min+0x2f/0x60 [ 12.718248] ? dst_alloc+0x4c/0xa0 [ 12.718964] ? tcp_ack+0xaf1/0x10b0 [ 12.719658] ? tipc_udp_is_known_peer+0xa0/0xa0 [tipc] [ 12.720634] tipc_udp_recv+0x71/0x1d0 [tipc] [ 12.721459] ? dst_alloc+0x4c/0xa0 [ 12.722130] udp_queue_rcv_skb+0x264/0x490 [ 12.722924] __udp4_lib_rcv+0x21e/0x990 [ 12.723670] ? ip_route_input_rcu+0x2dd/0xbf0 [ 12.724442] ? tcp_v4_rcv+0x958/0xa40 [ 12.725039] udp_rcv+0x1a/0x20 [ 12.725587] ip_local_deliver_finish+0x97/0x1d0 [ 12.726323] ip_local_deliver+0xaf/0xc0 [ 12.726959] ? ip_route_input_noref+0x19/0x20 [ 12.727689] ip_rcv_finish+0xdd/0x3b0 [ 12.728307] ip_rcv+0x2ac/0x360 [ 12.728839] __netif_receive_skb_core+0x6fb/0xa90 [ 12.729580] ? udp4_gro_receive+0x1a7/0x2c0 [ 12.730274] __netif_receive_skb+0x1d/0x60 [ 12.730953] ? __netif_receive_skb+0x1d/0x60 [ 12.731637] netif_receive_skb_internal+0x37/0xd0 [ 12.732371] napi_gro_receive+0xc7/0xf0 [ 12.732920] receive_buf+0x3c3/0xd40 [ 12.733441] virtnet_poll+0xb1/0x250 [ 12.733944] net_rx_action+0x23e/0x370 [ 12.734476] __do_softirq+0xc5/0x2f8 [ 12.734922] irq_exit+0xfa/0x100 [ 12.735315] do_IRQ+0x4f/0xd0 [ 12.735680] common_interrupt+0xa2/0xa2 [ 12.736126] </IRQ> [ 12.736416] RIP: 0010:native_safe_halt+0x6/0x10 [ 12.736925] RSP: 0018:ffffa41cc0cafe90 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff4d [ 12.737756] RAX: 0000000000000000 RBX: ffff8c2a761edb80 RCX: 0000000000000000 [ 12.738504] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 [ 12.739258] RBP: ffffa41cc0cafe90 R08: 0000014b5b9795e5 R09: ffffa41cc12c7e88 [ 12.740118] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000002 [ 12.740964] R13: ffff8c2a761edb80 R14: 0000000000000000 R15: 0000000000000000 [ 12.741831] default_idle+0x2a/0x100 [ 12.742323] arch_cpu_idle+0xf/0x20 [ 12.742796] default_idle_call+0x28/0x40 [ 12.743312] do_idle+0x179/0x1f0 [ 12.743761] cpu_startup_entry+0x1d/0x20 [ 12.744291] start_secondary+0x112/0x120 [ 12.744816] secondary_startup_64+0xa5/0xa5 [ 12.745367] Code: b9 f4 01 00 00 48 89 c2 48 c1 ea 02 48 3d d3 07 00 00 48 0f 47 d1 49 8b 0c 24 48 39 d1 76 07 49 89 14 24 48 89 d1 31 d2 48 89 df <48> f7 f1 89 c6 e8 81 6e ff ff 5b 41 5c 5d c3 66 90 66 2e 0f 1f [ 12.747527] RIP: tipc_node_calculate_timer.isra.12+0x45/0x60 [tipc] RSP: ffff8c2a7fc838a0 [ 12.748555] ---[ end trace 1399ab83390650fd ]--- [ 12.749296] Kernel panic - not syncing: Fatal exception in interrupt [ 12.750123] Kernel Offset: 0x13200000 from 0xffffffff82000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff) [ 12.751215] Rebooting in 60 seconds.. Fixes: c9b64d49 ("tipc: add replicast peer discovery") Signed-off-by:
Tommi Rantala <tommi.t.rantala@nokia.com> Cc: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by:
-
David Ahern authored
[ Usptream commit b4d1605a ] After this fix : ("tcp: add tcp_v4_fill_cb()/tcp_v4_restore_cb()"), socket lookups happen while skb->cb[] has not been mangled yet by TCP. Fixes: a04a480d ("net: Require exact match for TCP socket lookups if dif is l3mdev") Signed-off-by:
-
Julian Wiedmann authored
[ Upstream commit 6d69b1f1 ] Using GSO with small MTUs currently results in a substantial throughput regression - which is caused by how qeth needs to map non-linear skbs into its IO buffer elements: compared to a linear skb, each GSO-segmented skb effectively consumes twice as many buffer elements (ie two instead of one) due to the additional header-only part. This causes the Output Queue to be congested with low-utilized IO buffers. Fix this as follows: If the MSS is low enough so that a non-SG GSO segmentation produces order-0 skbs (currently ~3500 byte), opt out from NETIF_F_SG. This is where we anticipate the biggest savings, since an SG-enabled GSO segmentation produces skbs that always consume at least two buffer elements. Larger MSS values continue to get a SG-enabled GSO segmentation, since 1) the relative overhead of the additional header-only buffer element becomes less noticeable, and 2) the linearization overhead increases. With the throughput regression fixed, re-enable NETIF_F_SG by default to reap the significant CPU savings of GSO. Fixes: 5722963a ("qeth: do not turn on SG per default") Reported-by:
Nils Hoppmann <niho@de.ibm.com> Signed-off-by:
Julian Wiedmann <jwi@linux.vnet.ibm.com> Signed-off-by:
-
Julian Wiedmann authored
[ Upsteam commit bc3ab705 ] Commit 5f78e29c ("qeth: optimize IP handling in rx_mode callback") reworked how secondary addresses are managed for qeth devices. Instead of dropping & subsequently re-adding all addresses on every ndo_set_rx_mode() call, qeth now keeps track of the addresses that are currently registered with the HW. On a ndo_set_rx_mode(), we thus only need to do (de-)registration requests for the addresses that have actually changed. On L3 devices, the lookup for IPv4 Multicast addresses checks the wrong hashtable - and thus never finds a match. As a result, we first delete *all* such addresses, and then re-add them again. So each set_rx_mode() causes a short period where the IPv4 Multicast addresses are not registered, and the card stops forwarding inbound traffic for them. Fix this by setting the ->is_multicast flag on the lookup object, thus enabling qeth_l3_ip_from_hash() to search the correct hashtable and find a match there. Fixes: 5f78e29c ("qeth: optimize IP handling in rx_mode callback") Signed-off-by:
-
Julian Wiedmann authored
[ Upstream commit 0cbff6d4 ] The current GSO skb size limit was copy&pasted over from the L3 path, where it is needed due to a TSO limitation. As L2 devices don't offer TSO support (and thus all GSO skbs are segmented before they reach the driver), there's no reason to restrict the stack in how large it may build the GSO skbs. Fixes: d52aec97 ("qeth: enable scatter/gather in layer 2 mode") Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit cfac7f83 ] Maciej Żenczykowski reported some panics in tcp_twsk_destructor() that might be caused by the following bug. timewait timer is pinned to the cpu, because we want to transition timwewait refcount from 0 to 4 in one go, once everything has been initialized. At the time commit ed2e9239 ("tcp/dccp: fix timewait races in timer handling") was merged, TCP was always running from BH habdler. After commit 5413d1ba ("net: do not block BH while processing socket backlog") we definitely can run tcp_time_wait() from process context. We need to block BH in the critical section so that the pinned timer has still its purpose. This bug is more likely to happen under stress and when very small RTO are used in datacenter flows. Fixes: 5413d1ba ("net: do not block BH while processing socket backlog") Signed-off-by:
Maciej Żenczykowski <maze@google.com> Acked-by:
-
Lars Persson authored
[ Upstream commit 45ab4b13 ] The mss variable tracks the last max segment size sent to the TSO engine. We do not update the hardware as long as we receive skb:s with the same value in gso_size. During a network device down/up cycle (mapped to stmmac_release() and stmmac_open() callbacks) we issue a reset to the hardware and it forgets the setting for mss. However we did not zero out our mss variable so the next transmission of a gso packet happens with an undefined hardware setting. This triggers a hang in the TSO engine and eventuelly the netdev watchdog will bark. Fixes: f748be53 ("stmmac: support new GMAC4") Signed-off-by:
Lars Persson <larper@axis.com> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit d7efc6c1 ] Alexander Potapenko reported use of uninitialized memory [1] This happens when inserting a request socket into TCP ehash, in __sk_nulls_add_node_rcu(), since sk_reuseport is not initialized. Bug was added by commit d894ba18 ("soreuseport: fix ordering for mixed v4/v6 sockets") Note that d296ba60 ("soreuseport: Resolve merge conflict for v4/v6 ordering fix") missed the opportunity to get rid of hlist_nulls_add_tail_rcu() : Both UDP sockets and TCP/DCCP listeners no longer use __sk_nulls_add_node_rcu() for their hash insertion. Since all other sockets have unique 4-tuple, the reuseport status has no special meaning, so we can always use hlist_nulls_add_head_rcu() for them and save few cycles/instructions. [1] ================================================================== BUG: KMSAN: use of uninitialized memory in inet_ehash_insert+0xd40/0x1050 CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.13.0+ #3288 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011 Call Trace: <IRQ> __dump_stack lib/dump_stack.c:16 dump_stack+0x185/0x1d0 lib/dump_stack.c:52 kmsan_report+0x13f/0x1c0 mm/kmsan/kmsan.c:1016 __msan_warning_32+0x69/0xb0 mm/kmsan/kmsan_instr.c:766 __sk_nulls_add_node_rcu ./include/net/sock.h:684 inet_ehash_insert+0xd40/0x1050 net/ipv4/inet_hashtables.c:413 reqsk_queue_hash_req net/ipv4/inet_connection_sock.c:754 inet_csk_reqsk_queue_hash_add+0x1cc/0x300 net/ipv4/inet_connection_sock.c:765 tcp_conn_request+0x31e7/0x36f0 net/ipv4/tcp_input.c:6414 tcp_v4_conn_request+0x16d/0x220 net/ipv4/tcp_ipv4.c:1314 tcp_rcv_state_process+0x42a/0x7210 net/ipv4/tcp_input.c:5917 tcp_v4_do_rcv+0xa6a/0xcd0 net/ipv4/tcp_ipv4.c:1483 tcp_v4_rcv+0x3de0/0x4ab0 net/ipv4/tcp_ipv4.c:1763 ip_local_deliver_finish+0x6bb/0xcb0 net/ipv4/ip_input.c:216 NF_HOOK ./include/linux/netfilter.h:248 ip_local_deliver+0x3fa/0x480 net/ipv4/ip_input.c:257 dst_input ./include/net/dst.h:477 ip_rcv_finish+0x6fb/0x1540 net/ipv4/ip_input.c:397 NF_HOOK ./include/linux/netfilter.h:248 ip_rcv+0x10f6/0x15c0 net/ipv4/ip_input.c:488 __netif_receive_skb_core+0x36f6/0x3f60 net/core/dev.c:4298 __netif_receive_skb net/core/dev.c:4336 netif_receive_skb_internal+0x63c/0x19c0 net/core/dev.c:4497 napi_skb_finish net/core/dev.c:4858 napi_gro_receive+0x629/0xa50 net/core/dev.c:4889 e1000_receive_skb drivers/net/ethernet/intel/e1000/e1000_main.c:4018 e1000_clean_rx_irq+0x1492/0x1d30 drivers/net/ethernet/intel/e1000/e1000_main.c:4474 e1000_clean+0x43aa/0x5970 drivers/net/ethernet/intel/e1000/e1000_main.c:3819 napi_poll net/core/dev.c:5500 net_rx_action+0x73c/0x1820 net/core/dev.c:5566 __do_softirq+0x4b4/0x8dd kernel/softirq.c:284 invoke_softirq kernel/softirq.c:364 irq_exit+0x203/0x240 kernel/softirq.c:405 exiting_irq+0xe/0x10 ./arch/x86/include/asm/apic.h:638 do_IRQ+0x15e/0x1a0 arch/x86/kernel/irq.c:263 common_interrupt+0x86/0x86 Fixes: d894ba18 ("soreuseport: fix ordering for mixed v4/v6 sockets") Fixes: d296ba60 ("soreuseport: Resolve merge conflict for v4/v6 ordering fix") Signed-off-by:
Alexander Potapenko <glider@google.com> Acked-by:
Craig Gallek <kraig@google.com> Signed-off-by:
-
Bjørn Mork authored
[ Upstream commit a4abd7a8 ] The qmi_wwan minidriver support a 'raw-ip' mode where frames are received without any ethernet header. This causes alignment issues because the skbs allocated by usbnet are "IP aligned". Fix by allowing minidrivers to disable the additional alignment offset. This is implemented using a per-device flag, since the same minidriver also supports 'ethernet' mode. Fixes: 32f7adf6 ("net: qmi_wwan: support "raw IP" mode") Reported-and-tested-by:
Jay Foster <jay@systech.com> Signed-off-by:
Bjørn Mork <bjorn@mork.no> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit 3016dad7 ] tcp_v6_send_reset() expects to receive an skb with skb->cb[] layout as used in TCP stack. MD5 lookup uses tcp_v6_iif() and tcp_v6_sdif() and thus TCP_SKB_CB(skb)->header.h6 This patch probably fixes RST packets sent on behalf of a timewait md5 ipv6 socket. Before Florian patch, tcp_v6_restore_cb() was needed before jumping to no_tcp_socket label. Fixes: 271c3b9b ("tcp: honour SO_BINDTODEVICE for TW_RST case too") Signed-off-by:
Florian Westphal <fw@strlen.de> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit 15fe076e ] syzbot reported crashes [1] and provided a C repro easing bug hunting. When/if packet_do_bind() calls __unregister_prot_hook() and releases po->bind_lock, another thread can run packet_notifier() and process an NETDEV_UP event. This calls register_prot_hook() and hooks again the socket right before first thread is able to grab again po->bind_lock. Fixes this issue by temporarily setting po->num to 0, as suggested by David Miller. [1] dev_remove_pack: ffff8801bf16fa80 not found ------------[ cut here ]------------ kernel BUG at net/core/dev.c:7945! ( BUG_ON(!list_empty(&dev->ptype_all)); ) invalid opcode: 0000 [#1] SMP KASAN Dumping ftrace buffer: (ftrace buffer empty) Modules linked in: device syz0 entered promiscuous mode CPU: 0 PID: 3161 Comm: syzkaller404108 Not tainted 4.14.0+ #190 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 task: ffff8801cc57a500 task.stack: ffff8801cc588000 RIP: 0010:netdev_run_todo+0x772/0xae0 net/core/dev.c:7945 RSP: 0018:ffff8801cc58f598 EFLAGS: 00010293 RAX: ffff8801cc57a500 RBX: dffffc0000000000 RCX: ffffffff841f75b2 RDX: 0000000000000000 RSI: 1ffff100398b1ede RDI: ffff8801bf1f8810 device syz0 entered promiscuous mode RBP: ffff8801cc58f898 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: ffff8801bf1f8cd8 R13: ffff8801cc58f870 R14: ffff8801bf1f8780 R15: ffff8801cc58f7f0 FS: 0000000001716880(0000) GS:ffff8801db400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020b13000 CR3: 0000000005e25000 CR4: 00000000001406f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: rtnl_unlock+0xe/0x10 net/core/rtnetlink.c:106 tun_detach drivers/net/tun.c:670 [inline] tun_chr_close+0x49/0x60 drivers/net/tun.c:2845 __fput+0x333/0x7f0 fs/file_table.c:210 ____fput+0x15/0x20 fs/file_table.c:244 task_work_run+0x199/0x270 kernel/task_work.c:113 exit_task_work include/linux/task_work.h:22 [inline] do_exit+0x9bb/0x1ae0 kernel/exit.c:865 do_group_exit+0x149/0x400 kernel/exit.c:968 SYSC_exit_group kernel/exit.c:979 [inline] SyS_exit_group+0x1d/0x20 kernel/exit.c:977 entry_SYSCALL_64_fastpath+0x1f/0x96 RIP: 0033:0x44ad19 Fixes: 30f7ea1c ("packet: race condition in packet_bind") Signed-off-by:
syzbot <syzkaller@googlegroups.com> Cc: Francesco Ruggeri <fruggeri@aristanetworks.com> Signed-off-by:
-
Mike Maloney authored
syzkaller found a race condition fanout_demux_rollover() while removing a packet socket from a fanout group. po->rollover is read and operated on during packet_rcv_fanout(), via fanout_demux_rollover(), but the pointer is currently cleared before the synchronization in packet_release(). It is safer to delay the cleanup until after synchronize_net() has been called, ensuring all calls to packet_rcv_fanout() for this socket have finished. To further simplify synchronization around the rollover structure, set po->rollover in fanout_add() only if there are no errors. This removes the need for rcu in the struct and in the call to packet_getsockopt(..., PACKET_ROLLOVER_STATS, ...). Crashing stack trace: fanout_demux_rollover+0xb6/0x4d0 net/packet/af_packet.c:1392 packet_rcv_fanout+0x649/0x7c8 net/packet/af_packet.c:1487 dev_queue_xmit_nit+0x835/0xc10 net/core/dev.c:1953 xmit_one net/core/dev.c:2975 [inline] dev_hard_start_xmit+0x16b/0xac0 net/core/dev.c:2995 __dev_queue_xmit+0x17a4/0x2050 net/core/dev.c:3476 dev_queue_xmit+0x17/0x20 net/core/dev.c:3509 neigh_connected_output+0x489/0x720 net/core/neighbour.c:1379 neigh_output include/net/neighbour.h:482 [inline] ip6_finish_output2+0xad1/0x22a0 net/ipv6/ip6_output.c:120 ip6_finish_output+0x2f9/0x920 net/ipv6/ip6_output.c:146 NF_HOOK_COND include/linux/netfilter.h:239 [inline] ip6_output+0x1f4/0x850 net/ipv6/ip6_output.c:163 dst_output include/net/dst.h:459 [inline] NF_HOOK.constprop.35+0xff/0x630 include/linux/netfilter.h:250 mld_sendpack+0x6a8/0xcc0 net/ipv6/mcast.c:1660 mld_send_initial_cr.part.24+0x103/0x150 net/ipv6/mcast.c:2072 mld_send_initial_cr net/ipv6/mcast.c:2056 [inline] ipv6_mc_dad_complete+0x99/0x130 net/ipv6/mcast.c:2079 addrconf_dad_completed+0x595/0x970 net/ipv6/addrconf.c:4039 addrconf_dad_work+0xac9/0x1160 net/ipv6/addrconf.c:3971 process_one_work+0xbf0/0x1bc0 kernel/workqueue.c:2113 worker_thread+0x223/0x1990 kernel/workqueue.c:2247 kthread+0x35e/0x430 kernel/kthread.c:231 ret_from_fork+0x2a/0x40 arch/x86/entry/entry_64.S:432 Fixes: 0648ab70 ("packet: rollover prepare: per-socket state") Fixes: 509c7a1e ("packet: avoid panic in packet_getsockopt()") Reported-by:
Mike Maloney <maloney@google.com> Reviewed-by:
-
Eric Dumazet authored
[ Upstream commit eeea10b8 ] James Morris reported kernel stack corruption bug [1] while running the SELinux testsuite, and bisected to a recent commit bffa72cf ("net: sk_buff rbnode reorg") We believe this commit is fine, but exposes an older bug. SELinux code runs from tcp_filter() and might send an ICMP, expecting IP options to be found in skb->cb[] using regular IPCB placement. We need to defer TCP mangling of skb->cb[] after tcp_filter() calls. This patch adds tcp_v4_fill_cb()/tcp_v4_restore_cb() in a very similar way we added them for IPv6. [1] [ 339.806024] SELinux: failure in selinux_parse_skb(), unable to parse packet [ 339.822505] Kernel panic - not syncing: stack-protector: Kernel stack is corrupted in: ffffffff81745af5 [ 339.822505] [ 339.852250] CPU: 4 PID: 3642 Comm: client Not tainted 4.15.0-rc1-test #15 [ 339.868498] Hardware name: LENOVO 10FGS0VA1L/30BC, BIOS FWKT68A 01/19/2017 [ 339.885060] Call Trace: [ 339.896875] <IRQ> [ 339.908103] dump_stack+0x63/0x87 [ 339.920645] panic+0xe8/0x248 [ 339.932668] ? ip_push_pending_frames+0x33/0x40 [ 339.946328] ? icmp_send+0x525/0x530 [ 339.958861] ? kfree_skbmem+0x60/0x70 [ 339.971431] __stack_chk_fail+0x1b/0x20 [ 339.984049] icmp_send+0x525/0x530 [ 339.996205] ? netlbl_skbuff_err+0x36/0x40 [ 340.008997] ? selinux_netlbl_err+0x11/0x20 [ 340.021816] ? selinux_socket_sock_rcv_skb+0x211/0x230 [ 340.035529] ? security_sock_rcv_skb+0x3b/0x50 [ 340.048471] ? sk_filter_trim_cap+0x44/0x1c0 [ 340.061246] ? tcp_v4_inbound_md5_hash+0x69/0x1b0 [ 340.074562] ? tcp_filter+0x2c/0x40 [ 340.086400] ? tcp_v4_rcv+0x820/0xa20 [ 340.098329] ? ip_local_deliver_finish+0x71/0x1a0 [ 340.111279] ? ip_local_deliver+0x6f/0xe0 [ 340.123535] ? ip_rcv_finish+0x3a0/0x3a0 [ 340.135523] ? ip_rcv_finish+0xdb/0x3a0 [ 340.147442] ? ip_rcv+0x27c/0x3c0 [ 340.158668] ? inet_del_offload+0x40/0x40 [ 340.170580] ? __netif_receive_skb_core+0x4ac/0x900 [ 340.183285] ? rcu_accelerate_cbs+0x5b/0x80 [ 340.195282] ? __netif_receive_skb+0x18/0x60 [ 340.207288] ? process_backlog+0x95/0x140 [ 340.218948] ? net_rx_action+0x26c/0x3b0 [ 340.230416] ? __do_softirq+0xc9/0x26a [ 340.241625] ? do_softirq_own_stack+0x2a/0x40 [ 340.253368] </IRQ> [ 340.262673] ? do_softirq+0x50/0x60 [ 340.273450] ? __local_bh_enable_ip+0x57/0x60 [ 340.285045] ? ip_finish_output2+0x175/0x350 [ 340.296403] ? ip_finish_output+0x127/0x1d0 [ 340.307665] ? nf_hook_slow+0x3c/0xb0 [ 340.318230] ? ip_output+0x72/0xe0 [ 340.328524] ? ip_fragment.constprop.54+0x80/0x80 [ 340.340070] ? ip_local_out+0x35/0x40 [ 340.350497] ? ip_queue_xmit+0x15c/0x3f0 [ 340.361060] ? __kmalloc_reserve.isra.40+0x31/0x90 [ 340.372484] ? __skb_clone+0x2e/0x130 [ 340.382633] ? tcp_transmit_skb+0x558/0xa10 [ 340.393262] ? tcp_connect+0x938/0xad0 [ 340.403370] ? ktime_get_with_offset+0x4c/0xb0 [ 340.414206] ? tcp_v4_connect+0x457/0x4e0 [ 340.424471] ? __inet_stream_connect+0xb3/0x300 [ 340.435195] ? inet_stream_connect+0x3b/0x60 [ 340.445607] ? SYSC_connect+0xd9/0x110 [ 340.455455] ? __audit_syscall_entry+0xaf/0x100 [ 340.466112] ? syscall_trace_enter+0x1d0/0x2b0 [ 340.476636] ? __audit_syscall_exit+0x209/0x290 [ 340.487151] ? SyS_connect+0xe/0x10 [ 340.496453] ? do_syscall_64+0x67/0x1b0 [ 340.506078] ? entry_SYSCALL64_slow_path+0x25/0x25 Fixes: 971f10ec ("tcp: better TCP_SKB_CB layout to reduce cache line misses") Signed-off-by:
James Morris <james.l.morris@oracle.com> Tested-by:
Casey Schaufler <casey@schaufler-ca.com> Signed-off-by:
-
