user/sven/linux.git/include/trace, branch v4.19.59

rxrpc: Fix client call connect/disconnect race

2019-04-20T07:16:05Z

[ Upstream commit 930c9f9125c85b5134b3e711bc252ecc094708e3 ] rxrpc_disconnect_client_call() reads the call's connection ID protocol value (call->cid) as part of that function's variable declarations. This is bad because it's not inside the locked section and so may race with someone granting use of the channel to the call. This manifests as an assertion failure (see below) where the call in the presumed channel (0 because call->cid wasn't set when we read it) doesn't match the call attached to the channel we were actually granted (if 1, 2 or 3). Fix this by moving the read and dependent calculations inside of the channel_lock section. Also, only set the channel number and pointer variables if cid is not zero (ie. unset). This problem can be induced by injecting an occasional error in rxrpc_wait_for_channel() before the call to schedule(). Make two further changes also: (1) Add a trace for wait failure in rxrpc_connect_call(). (2) Drop channel_lock before BUG'ing in the case of the assertion failure. The failure causes a trace akin to the following: rxrpc: Assertion failed - 18446612685268945920(0xffff8880beab8c00) == 18446612685268621312(0xffff8880bea69800) is false ------------[ cut here ]------------ kernel BUG at net/rxrpc/conn_client.c:824! ... RIP: 0010:rxrpc_disconnect_client_call+0x2bf/0x99d ... Call Trace: rxrpc_connect_call+0x902/0x9b3 ? wake_up_q+0x54/0x54 rxrpc_new_client_call+0x3a0/0x751 ? rxrpc_kernel_begin_call+0x141/0x1bc ? afs_alloc_call+0x1b5/0x1b5 rxrpc_kernel_begin_call+0x141/0x1bc afs_make_call+0x20c/0x525 ? afs_alloc_call+0x1b5/0x1b5 ? __lock_is_held+0x40/0x71 ? lockdep_init_map+0xaf/0x193 ? lockdep_init_map+0xaf/0x193 ? __lock_is_held+0x40/0x71 ? yfs_fs_fetch_data+0x33b/0x34a yfs_fs_fetch_data+0x33b/0x34a afs_fetch_data+0xdc/0x3b7 afs_read_dir+0x52d/0x97f afs_dir_iterate+0xa0/0x661 ? iterate_dir+0x63/0x141 iterate_dir+0xa2/0x141 ksys_getdents64+0x9f/0x11b ? filldir+0x111/0x111 ? do_syscall_64+0x3e/0x1a0 __x64_sys_getdents64+0x16/0x19 do_syscall_64+0x7d/0x1a0 entry_SYSCALL_64_after_hwframe+0x49/0xbe Fixes: 45025bceef17 ("rxrpc: Improve management and caching of client connection objects") Signed-off-by: David Howells Reviewed-by: Marc Dionne Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

sunrpc: use-after-free in svc_process_common()

2019-01-16T21:04:37Z

commit d4b09acf924b84bae77cad090a9d108e70b43643 upstream. if node have NFSv41+ mounts inside several net namespaces it can lead to use-after-free in svc_process_common() svc_process_common() /* Setup reply header */ rqstp->rq_xprt->xpt_ops->xpo_prep_reply_hdr(rqstp); <<< HERE svc_process_common() can use incorrect rqstp->rq_xprt, its caller function bc_svc_process() takes it from serv->sv_bc_xprt. The problem is that serv is global structure but sv_bc_xprt is assigned per-netnamespace. According to Trond, the whole "let's set up rqstp->rq_xprt for the back channel" is nothing but a giant hack in order to work around the fact that svc_process_common() uses it to find the xpt_ops, and perform a couple of (meaningless for the back channel) tests of xpt_flags. All we really need in svc_process_common() is to be able to run rqstp->rq_xprt->xpt_ops->xpo_prep_reply_hdr() Bruce J Fields points that this xpo_prep_reply_hdr() call is an awfully roundabout way just to do "svc_putnl(resv, 0);" in the tcp case. This patch does not initialiuze rqstp->rq_xprt in bc_svc_process(), now it calls svc_process_common() with rqstp->rq_xprt = NULL. To adjust reply header svc_process_common() just check rqstp->rq_prot and calls svc_tcp_prep_reply_hdr() for tcp case. To handle rqstp->rq_xprt = NULL case in functions called from svc_process_common() patch intruduces net namespace pointer svc_rqst->rq_bc_net and adjust SVC_NET() definition. Some other function was also adopted to properly handle described case. Signed-off-by: Vasily Averin Cc: stable@vger.kernel.org Fixes: 23c20ecd4475 ("NFS: callback up - users counting cleanup") Signed-off-by: J. Bruce Fields v2: added lost extern svc_tcp_prep_reply_hdr() Signed-off-by: Vasily Averin Signed-off-by: Greg Kroah-Hartman

ext4: force inode writes when nfsd calls commit_metadata()

2019-01-09T16:38:43Z

commit fde872682e175743e0c3ef939c89e3c6008a1529 upstream. Some time back, nfsd switched from calling vfs_fsync() to using a new commit_metadata() hook in export_operations(). If the file system did not provide a commit_metadata() hook, it fell back to using sync_inode_metadata(). Unfortunately doesn't work on all file systems. In particular, it doesn't work on ext4 due to how the inode gets journalled --- the VFS writeback code will not always call ext4_write_inode(). So we need to provide our own ext4_nfs_commit_metdata() method which calls ext4_write_inode() directly. Google-Bug-Id: 121195940 Signed-off-by: Theodore Ts'o Cc: stable@kernel.org Signed-off-by: Greg Kroah-Hartman

sched, trace: Fix prev_state output in sched_switch tracepoint

2018-12-08T11:59:06Z

commit 3054426dc68e5d63aa6a6e9b91ac4ec78e3f3805 upstream. commit 3f5fe9fef5b2 ("sched/debug: Fix task state recording/printout") tried to fix the problem introduced by a previous commit efb40f588b43 ("sched/tracing: Fix trace_sched_switch task-state printing"). However the prev_state output in sched_switch is still broken. task_state_index() uses fls() which considers the LSB as 1. Left shifting 1 by this value gives an incorrect mapping to the task state. Fix this by decrementing the value returned by __get_task_state() before shifting. Link: http://lkml.kernel.org/r/1540882473-1103-1-git-send-email-pkondeti@codeaurora.org Cc: stable@vger.kernel.org Fixes: 3f5fe9fef5b2 ("sched/debug: Fix task state recording/printout") Signed-off-by: Pavankumar Kondeti Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Greg Kroah-Hartman

Merge tag 'rxrpc-fixes-20181008' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs

2018-10-11T05:27:38Z

David Howells says: ==================== rxrpc: Fix packet reception code Here are a set of patches that prepares for and fix problems in rxrpc's package reception code. There serious problems are: (A) There's a window between binding the socket and setting the data_ready hook in which packets can find their way into the UDP socket's receive queues. (B) The skb_recv_udp() will return an error (and clear the error state) if there was an error on the Tx side. rxrpc doesn't handle this. (C) The rxrpc data_ready handler doesn't fully drain the UDP receive queue. (D) The rxrpc data_ready handler assumes it is called in a non-reentrant state. The second patch fixes (A) - (C); the third patch renders (B) and (C) non-issues by using the recap_rcv hook instead of data_ready - and the final patch fixes (D). That last is the most complex. The preparatory patches are: (1) Fix some places that are doing things in the wrong net namespace. (2) Stop taking the rcu read lock as it's held by the IP input routine in the call chain. (3) Only end the Tx phase if *we* rotated the final packet out of the Tx buffer. (4) Don't assume that the call state won't change after dropping the call_state lock. (5) Only take receive window and MTU suze parameters from an ACK packet if it's the latest ACK packet. (6) Record connection-level abort information correctly. (7) Fix a trace line. And then there are three main patches - note that these are mixed in with the preparatory patches somewhat: (1) Fix the setup window (A), skb_recv_udp() error check (B) and packet drainage (C). (2) Switch to using the encap_rcv instead of data_ready to cut out the effects of the UDP read queues and get the packets delivered directly. (3) Add more locking into the various packet input paths to defend against re-entrance (D). ==================== Signed-off-by: David S. Miller

rxrpc: Fix the rxrpc_tx_packet trace line

2018-10-08T21:42:04Z

Fix the rxrpc_tx_packet trace line by storing the where parameter. Fixes: 4764c0da69dc ("rxrpc: Trace packet transmission") Signed-off-by: David Howells

Merge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2018-10-05T22:39:38Z

Ingo writes: "scheduler fixes: These fixes address a rather involved performance regression between v4.17->v4.19 in the sched/numa auto-balancing code. Since distros really need this fix we accelerated it to sched/urgent for a faster upstream merge. NUMA scheduling and balancing performance is now largely back to v4.17 levels, without reintroducing the NUMA placement bugs that v4.18 and v4.19 fixed. Many thanks to Srikar Dronamraju, Mel Gorman and Jirka Hladky, for reporting, testing, re-testing and solving this rather complex set of bugs." * 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: sched/numa: Migrate pages to local nodes quicker early in the lifetime of a task mm, sched/numa: Remove rate-limiting of automatic NUMA balancing migration sched/numa: Avoid task migration for small NUMA improvement mm/migrate: Use spin_trylock() while resetting rate limit sched/numa: Limit the conditions where scan period is reset sched/numa: Reset scan rate whenever task moves across nodes sched/numa: Pass destination CPU as a parameter to migrate_task_rq sched/numa: Stop multiple tasks from moving to the CPU at the same time

mm, sched/numa: Remove rate-limiting of automatic NUMA balancing migration

2018-10-02T09:31:14Z

Rate limiting of page migrations due to automatic NUMA balancing was introduced to mitigate the worst-case scenario of migrating at high frequency due to false sharing or slowly ping-ponging between nodes. Since then, a lot of effort was spent on correctly identifying these pages and avoiding unnecessary migrations and the safety net may no longer be required. Jirka Hladky reported a regression in 4.17 due to a scheduler patch that avoids spreading STREAM tasks wide prematurely. However, once the task was properly placed, it delayed migrating the memory due to rate limiting. Increasing the limit fixed the problem for him. Currently, the limit is hard-coded and does not account for the real capabilities of the hardware. Even if an estimate was attempted, it would not properly account for the number of memory controllers and it could not account for the amount of bandwidth used for normal accesses. Rather than fudging, this patch simply eliminates the rate limiting. However, Jirka reports that a STREAM configuration using multiple processes achieved similar performance to 4.16. In local tests, this patch improved performance of STREAM relative to the baseline but it is somewhat machine-dependent. Most workloads show little or not performance difference implying that there is not a heavily reliance on the throttling mechanism and it is safe to remove. STREAM on 2-socket machine 4.19.0-rc5 4.19.0-rc5 numab-v1r1 noratelimit-v1r1 MB/sec copy 43298.52 ( 0.00%) 44673.38 ( 3.18%) MB/sec scale 30115.06 ( 0.00%) 31293.06 ( 3.91%) MB/sec add 32825.12 ( 0.00%) 34883.62 ( 6.27%) MB/sec triad 32549.52 ( 0.00%) 34906.60 ( 7.24% Signed-off-by: Mel Gorman Reviewed-by: Rik van Riel Acked-by: Peter Zijlstra Cc: Jirka Hladky Cc: Linus Torvalds Cc: Linux-MM Cc: Srikar Dronamraju Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20181001100525.29789-2-mgorman@techsingularity.net Signed-off-by: Ingo Molnar

rxrpc: Fix error distribution

2018-09-28T09:33:17Z

Fix error distribution by immediately delivering the errors to all the affected calls rather than deferring them to a worker thread. The problem with the latter is that retries and things can happen in the meantime when we want to stop that sooner. To this end: (1) Stop the error distributor from removing calls from the error_targets list so that peer->lock isn't needed to synchronise against other adds and removals. (2) Require the peer's error_targets list to be accessed with RCU, thereby avoiding the need to take peer->lock over distribution. (3) Don't attempt to affect a call's state if it is already marked complete. Signed-off-by: David Howells

Merge branch 'for-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup

2018-08-24T20:19:27Z

Pull cgroup updates from Tejun Heo: "Just one commit from Steven to take out spin lock from trace event handlers" * 'for-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: cgroup/tracing: Move taking of spin lock out of trace event handlers