user/sven/linux.git/kernel, branch v4.19.112

signal: avoid double atomic counter increments for user accounting

2020-03-20T10:55:53Z

[ Upstream commit fda31c50292a5062332fa0343c084bd9f46604d9 ] When queueing a signal, we increment both the users count of pending signals (for RLIMIT_SIGPENDING tracking) and we increment the refcount of the user struct itself (because we keep a reference to the user in the signal structure in order to correctly account for it when freeing). That turns out to be fairly expensive, because both of them are atomic updates, and particularly under extreme signal handling pressure on big machines, you can get a lot of cache contention on the user struct. That can then cause horrid cacheline ping-pong when you do these multiple accesses. So change the reference counting to only pin the user for the _first_ pending signal, and to unpin it when the last pending signal is dequeued. That means that when a user sees a lot of concurrent signal queuing - which is the only situation when this matters - the only atomic access needed is generally the 'sigpending' count update. This was noticed because of a particularly odd timing artifact on a dual-socket 96C/192T Cascade Lake platform: when you get into bad contention, on that machine for some reason seems to be much worse when the contention happens in the upper 32-byte half of the cacheline. As a result, the kernel test robot will-it-scale 'signal1' benchmark had an odd performance regression simply due to random alignment of the 'struct user_struct' (and pointed to a completely unrelated and apparently nonsensical commit for the regression). Avoiding the double increments (and decrements on the dequeueing side, of course) makes for much less contention and hugely improved performance on that will-it-scale microbenchmark. Quoting Feng Tang: "It makes a big difference, that the performance score is tripled! bump from original 17000 to 54000. Also the gap between 5.0-rc6 and 5.0-rc6+Jiri's patch is reduced to around 2%" [ The "2% gap" is the odd cacheline placement difference on that platform: under the extreme contention case, the effect of which half of the cacheline was hot was 5%, so with the reduced contention the odd timing artifact is reduced too ] It does help in the non-contended case too, but is not nearly as noticeable. Reported-and-tested-by: Feng Tang Cc: Eric W. Biederman Cc: Huang, Ying Cc: Philip Li Cc: Andi Kleen Cc: Jiri Olsa Cc: Peter Zijlstra Signed-off-by: Linus Torvalds Signed-off-by: Sasha Levin

workqueue: don't use wq_select_unbound_cpu() for bound works

2020-03-18T06:14:20Z

commit aa202f1f56960c60e7befaa0f49c72b8fa11b0a8 upstream. wq_select_unbound_cpu() is designed for unbound workqueues only, but it's wrongly called when using a bound workqueue too. Fixing this ensures work queued to a bound workqueue with cpu=WORK_CPU_UNBOUND always runs on the local CPU. Before, that would happen only if wq_unbound_cpumask happened to include it (likely almost always the case), or was empty, or we got lucky with forced round-robin placement. So restricting /sys/devices/virtual/workqueue/cpumask to a small subset of a machine's CPUs would cause some bound work items to run unexpectedly there. Fixes: ef557180447f ("workqueue: schedule WORK_CPU_UNBOUND work on wq_unbound_cpumask CPUs") Cc: stable@vger.kernel.org # v4.5+ Signed-off-by: Hillf Danton [dj: massage changelog] Signed-off-by: Daniel Jordan Cc: Tejun Heo Cc: Lai Jiangshan Cc: linux-kernel@vger.kernel.org Signed-off-by: Tejun Heo Signed-off-by: Greg Kroah-Hartman

cgroup: Iterate tasks that did not finish do_exit()

2020-03-18T06:14:19Z

commit 9c974c77246460fa6a92c18554c3311c8c83c160 upstream. PF_EXITING is set earlier than actual removal from css_set when a task is exitting. This can confuse cgroup.procs readers who see no PF_EXITING tasks, however, rmdir is checking against css_set membership so it can transitionally fail with EBUSY. Fix this by listing tasks that weren't unlinked from css_set active lists. It may happen that other users of the task iterator (without CSS_TASK_ITER_PROCS) spot a PF_EXITING task before cgroup_exit(). This is equal to the state before commit c03cd7738a83 ("cgroup: Include dying leaders with live threads in PROCS iterations") but it may be reviewed later. Reported-by: Suren Baghdasaryan Fixes: c03cd7738a83 ("cgroup: Include dying leaders with live threads in PROCS iterations") Signed-off-by: Michal Koutný Signed-off-by: Tejun Heo Signed-off-by: Greg Kroah-Hartman

cgroup: cgroup_procs_next should increase position index

2020-03-18T06:14:19Z

commit 2d4ecb030dcc90fb725ecbfc82ce5d6c37906e0e upstream. If seq_file .next fuction does not change position index, read after some lseek can generate unexpected output: 1) dd bs=1 skip output of each 2nd elements $ dd if=/sys/fs/cgroup/cgroup.procs bs=8 count=1 2 3 4 5 1+0 records in 1+0 records out 8 bytes copied, 0,000267297 s, 29,9 kB/s [test@localhost ~]$ dd if=/sys/fs/cgroup/cgroup.procs bs=1 count=8 2 4 <<< NB! 3 was skipped 6 <<< ... and 5 too 8 <<< ... and 7 8+0 records in 8+0 records out 8 bytes copied, 5,2123e-05 s, 153 kB/s This happen because __cgroup_procs_start() makes an extra extra cgroup_procs_next() call 2) read after lseek beyond end of file generates whole last line. 3) read after lseek into middle of last line generates expected rest of last line and unexpected whole line once again. Additionally patch removes an extra position index changes in __cgroup_procs_start() Cc: stable@vger.kernel.org https://bugzilla.kernel.org/show_bug.cgi?id=206283 Signed-off-by: Vasily Averin Signed-off-by: Tejun Heo Signed-off-by: Greg Kroah-Hartman

cgroup: memcg: net: do not associate sock with unrelated cgroup

2020-03-18T06:14:14Z

[ Upstream commit e876ecc67db80dfdb8e237f71e5b43bb88ae549c ] We are testing network memory accounting in our setup and noticed inconsistent network memory usage and often unrelated cgroups network usage correlates with testing workload. On further inspection, it seems like mem_cgroup_sk_alloc() and cgroup_sk_alloc() are broken in irq context specially for cgroup v1. mem_cgroup_sk_alloc() and cgroup_sk_alloc() can be called in irq context and kind of assumes that this can only happen from sk_clone_lock() and the source sock object has already associated cgroup. However in cgroup v1, where network memory accounting is opt-in, the source sock can be unassociated with any cgroup and the new cloned sock can get associated with unrelated interrupted cgroup. Cgroup v2 can also suffer if the source sock object was created by process in the root cgroup or if sk_alloc() is called in irq context. The fix is to just do nothing in interrupt. WARNING: Please note that about half of the TCP sockets are allocated from the IRQ context, so, memory used by such sockets will not be accouted by the memcg. The stack trace of mem_cgroup_sk_alloc() from IRQ-context: CPU: 70 PID: 12720 Comm: ssh Tainted: 5.6.0-smp-DEV #1 Hardware name: ... Call Trace: dump_stack+0x57/0x75 mem_cgroup_sk_alloc+0xe9/0xf0 sk_clone_lock+0x2a7/0x420 inet_csk_clone_lock+0x1b/0x110 tcp_create_openreq_child+0x23/0x3b0 tcp_v6_syn_recv_sock+0x88/0x730 tcp_check_req+0x429/0x560 tcp_v6_rcv+0x72d/0xa40 ip6_protocol_deliver_rcu+0xc9/0x400 ip6_input+0x44/0xd0 ? ip6_protocol_deliver_rcu+0x400/0x400 ip6_rcv_finish+0x71/0x80 ipv6_rcv+0x5b/0xe0 ? ip6_sublist_rcv+0x2e0/0x2e0 process_backlog+0x108/0x1e0 net_rx_action+0x26b/0x460 __do_softirq+0x104/0x2a6 do_softirq_own_stack+0x2a/0x40 do_softirq.part.19+0x40/0x50 __local_bh_enable_ip+0x51/0x60 ip6_finish_output2+0x23d/0x520 ? ip6table_mangle_hook+0x55/0x160 __ip6_finish_output+0xa1/0x100 ip6_finish_output+0x30/0xd0 ip6_output+0x73/0x120 ? __ip6_finish_output+0x100/0x100 ip6_xmit+0x2e3/0x600 ? ipv6_anycast_cleanup+0x50/0x50 ? inet6_csk_route_socket+0x136/0x1e0 ? skb_free_head+0x1e/0x30 inet6_csk_xmit+0x95/0xf0 __tcp_transmit_skb+0x5b4/0xb20 __tcp_send_ack.part.60+0xa3/0x110 tcp_send_ack+0x1d/0x20 tcp_rcv_state_process+0xe64/0xe80 ? tcp_v6_connect+0x5d1/0x5f0 tcp_v6_do_rcv+0x1b1/0x3f0 ? tcp_v6_do_rcv+0x1b1/0x3f0 __release_sock+0x7f/0xd0 release_sock+0x30/0xa0 __inet_stream_connect+0x1c3/0x3b0 ? prepare_to_wait+0xb0/0xb0 inet_stream_connect+0x3b/0x60 __sys_connect+0x101/0x120 ? __sys_getsockopt+0x11b/0x140 __x64_sys_connect+0x1a/0x20 do_syscall_64+0x51/0x200 entry_SYSCALL_64_after_hwframe+0x44/0xa9 The stack trace of mem_cgroup_sk_alloc() from IRQ-context: Fixes: 2d7580738345 ("mm: memcontrol: consolidate cgroup socket tracking") Fixes: d979a39d7242 ("cgroup: duplicate cgroup reference when cloning sockets") Signed-off-by: Shakeel Butt Reviewed-by: Roman Gushchin Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

kprobes: Fix optimize_kprobe()/unoptimize_kprobe() cancellation logic

2020-03-11T13:14:47Z

[ Upstream commit e4add247789e4ba5e08ad8256183ce2e211877d4 ] optimize_kprobe() and unoptimize_kprobe() cancels if a given kprobe is on the optimizing_list or unoptimizing_list already. However, since the following commit: f66c0447cca1 ("kprobes: Set unoptimized flag after unoptimizing code") modified the update timing of the KPROBE_FLAG_OPTIMIZED, it doesn't work as expected anymore. The optimized_kprobe could be in the following states: - [optimizing]: Before inserting jump instruction op.kp->flags has KPROBE_FLAG_OPTIMIZED and op->list is not empty. - [optimized]: jump inserted op.kp->flags has KPROBE_FLAG_OPTIMIZED and op->list is empty. - [unoptimizing]: Before removing jump instruction (including unused optprobe) op.kp->flags has KPROBE_FLAG_OPTIMIZED and op->list is not empty. - [unoptimized]: jump removed op.kp->flags doesn't have KPROBE_FLAG_OPTIMIZED and op->list is empty. Current code mis-expects [unoptimizing] state doesn't have KPROBE_FLAG_OPTIMIZED, and that can cause incorrect results. To fix this, introduce optprobe_queued_unopt() to distinguish [optimizing] and [unoptimizing] states and fixes the logic in optimize_kprobe() and unoptimize_kprobe(). [ mingo: Cleaned up the changelog and the code a bit. ] Signed-off-by: Masami Hiramatsu Reviewed-by: Steven Rostedt (VMware) Cc: Alexei Starovoitov Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: bristot@redhat.com Fixes: f66c0447cca1 ("kprobes: Set unoptimized flag after unoptimizing code") Link: https://lkml.kernel.org/r/157840814418.7181.13478003006386303481.stgit@devnote2 Signed-off-by: Ingo Molnar Signed-off-by: Sasha Levin

audit: always check the netlink payload length in audit_receive_msg()

2020-03-05T15:42:23Z

[ Upstream commit 756125289285f6e55a03861bf4b6257aa3d19a93 ] This patch ensures that we always check the netlink payload length in audit_receive_msg() before we take any action on the payload itself. Cc: stable@vger.kernel.org Reported-by: syzbot+399c44bf1f43b8747403@syzkaller.appspotmail.com Reported-by: syzbot+e4b12d8d202701f08b6d@syzkaller.appspotmail.com Signed-off-by: Paul Moore Signed-off-by: Sasha Levin

kprobes: Set unoptimized flag after unoptimizing code

2020-03-05T15:42:22Z

commit f66c0447cca1281116224d474cdb37d6a18e4b5b upstream. Set the unoptimized flag after confirming the code is completely unoptimized. Without this fix, when a kprobe hits the intermediate modified instruction (the first byte is replaced by an INT3, but later bytes can still be a jump address operand) while unoptimizing, it can return to the middle byte of the modified code, which causes an invalid instruction exception in the kernel. Usually, this is a rare case, but if we put a probe on the function call while text patching, it always causes a kernel panic as below: # echo p text_poke+5 > kprobe_events # echo 1 > events/kprobes/enable # echo 0 > events/kprobes/enable invalid opcode: 0000 [#1] PREEMPT SMP PTI RIP: 0010:text_poke+0x9/0x50 Call Trace: arch_unoptimize_kprobe+0x22/0x28 arch_unoptimize_kprobes+0x39/0x87 kprobe_optimizer+0x6e/0x290 process_one_work+0x2a0/0x610 worker_thread+0x28/0x3d0 ? process_one_work+0x610/0x610 kthread+0x10d/0x130 ? kthread_park+0x80/0x80 ret_from_fork+0x3a/0x50 text_poke() is used for patching the code in optprobes. This can happen even if we blacklist text_poke() and other functions, because there is a small time window during which we show the intermediate code to other CPUs. [ mingo: Edited the changelog. ] Tested-by: Alexei Starovoitov Signed-off-by: Masami Hiramatsu Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Steven Rostedt Cc: Thomas Gleixner Cc: bristot@redhat.com Fixes: 6274de4984a6 ("kprobes: Support delayed unoptimizing") Link: https://lkml.kernel.org/r/157483422375.25881.13508326028469515760.stgit@devnote2 Signed-off-by: Ingo Molnar Signed-off-by: Greg Kroah-Hartman

sched/fair: Fix O(nr_cgroups) in the load balancing path

2020-03-05T15:42:21Z

commit 039ae8bcf7a5f4476f4487e6bf816885fb3fb617 upstream. This re-applies the commit reverted here: commit c40f7d74c741 ("sched/fair: Fix infinite loop in update_blocked_averages() by reverting a9e7f6544b9c") I.e. now that cfs_rq can be safely removed/added in the list, we can re-apply: commit a9e7f6544b9c ("sched/fair: Fix O(nr_cgroups) in load balance path") Signed-off-by: Vincent Guittot Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: sargun@sargun.me Cc: tj@kernel.org Cc: xiexiuqi@huawei.com Cc: xiezhipeng1@huawei.com Link: https://lkml.kernel.org/r/1549469662-13614-3-git-send-email-vincent.guittot@linaro.org Signed-off-by: Ingo Molnar Cc: Vishnu Rangayyan Signed-off-by: Greg Kroah-Hartman

sched/fair: Optimize update_blocked_averages()

2020-03-05T15:42:21Z

commit 31bc6aeaab1d1de8959b67edbed5c7a4b3cdbe7c upstream. Removing a cfs_rq from rq->leaf_cfs_rq_list can break the parent/child ordering of the list when it will be added back. In order to remove an empty and fully decayed cfs_rq, we must remove its children too, so they will be added back in the right order next time. With a normal decay of PELT, a parent will be empty and fully decayed if all children are empty and fully decayed too. In such a case, we just have to ensure that the whole branch will be added when a new task is enqueued. This is default behavior since : commit f6783319737f ("sched/fair: Fix insertion in rq->leaf_cfs_rq_list") In case of throttling, the PELT of throttled cfs_rq will not be updated whereas the parent will. This breaks the assumption made above unless we remove the children of a cfs_rq that is throttled. Then, they will be added back when unthrottled and a sched_entity will be enqueued. As throttled cfs_rq are now removed from the list, we can remove the associated test in update_blocked_averages(). Signed-off-by: Vincent Guittot Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: sargun@sargun.me Cc: tj@kernel.org Cc: xiexiuqi@huawei.com Cc: xiezhipeng1@huawei.com Link: https://lkml.kernel.org/r/1549469662-13614-2-git-send-email-vincent.guittot@linaro.org Signed-off-by: Ingo Molnar Cc: Vishnu Rangayyan Signed-off-by: Greg Kroah-Hartman