- 13 Apr, 2018 40 commits
-
-
Nithin Sujir authored
[ Upstream commit 797a9364 ] In the loadbalance arp monitoring scheme, when a slave link change is detected, the slave->link is immediately updated and slave_state_changed is set. Later down the function, the rtnl_lock is acquired and the changes are committed, updating the bond link state. However, the acquisition of the rtnl_lock can fail. The next time the monitor runs, since slave->link is already updated, it determines that link is unchanged. This results in the bond link state permanently out of sync with the slave link. This patch modifies bond_loadbalance_arp_mon() to handle link changes identical to bond_ab_arp_{inspect/commit}(). The new link state is maintained in slave->new_link until we're ready to commit at which point it's copied into slave->link. NOTE: miimon_{inspect/commit}() has a more complex state machine requiring the use of the bond_{propose,commit}_link_state() functions which maintains the intermediate state in slave->link_new_state. The arp monitors don't require that. Testing: This bug is very easy to reproduce with the following steps. 1. In a loop, toggle a slave link of a bond slave interface. 2. In a separate loop, do ifconfig up/down of an unrelated interface to create contention for rtnl_lock. Within a few iterations, the bond link goes out of sync with the slave link. Signed-off-by:
Nithin Nayak Sujir <nsujir@tintri.com> Cc: Mahesh Bandewar <maheshb@google.com> Cc: Jay Vosburgh <jay.vosburgh@canonical.com> Acked-by:
Mahesh Bandewar <maheshb@google.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Sasha Levin <alexander.levin@microsoft.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
KT Liao authored
[ Upstream commit 4b3c7dbb ] Some old touchpad FWs need to have interrupt cleared before issuing reset command after updating firmware. We clear interrupt by attempting to read full report from the controller, and discarding any data read. Signed-off-by:
KT Liao <kt.liao@emc.com.tw> Signed-off-by:
Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by:
-
Roman Kapl authored
[ Upstream commit 7c3f1875 ] The default value for somaxconn is set in sysctl_core_net_init(), but this function is not called when kernel is configured without CONFIG_SYSCTL. This results in the kernel not being able to accept TCP connections, because the backlog has zero size. Usually, the user ends up with: "TCP: request_sock_TCP: Possible SYN flooding on port 7. Dropping request. Check SNMP counters." If SYN cookies are not enabled the connection is rejected. Before ef547f2a (tcp: remove max_qlen_log), the effects were less severe, because the backlog was always at least eight slots long. Signed-off-by:
Roman Kapl <roman.kapl@sysgo.com> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit d0e1a1b5 ] Paul Fiterau Brostean reported : <quote> Linux TCP stack we analyze exhibits behavior that seems odd to me. The scenario is as follows (all packets have empty payloads, no window scaling, rcv/snd window size should not be a factor): TEST HARNESS (CLIENT) LINUX SERVER 1. - LISTEN (server listen, then accepts) 2. - --> <SEQ=100><CTL=SYN> --> SYN-RECEIVED 3. - <-- <SEQ=300><ACK=101><CTL=SYN,ACK> <-- SYN-RECEIVED 4. - --> <SEQ=101><ACK=301><CTL=ACK> --> ESTABLISHED 5. - <-- <SEQ=301><ACK=101><CTL=FIN,ACK> <-- FIN WAIT-1 (server opts to close the data connection calling "close" on the connection socket) 6. - --> <SEQ=101><ACK=99999><CTL=FIN,ACK> --> CLOSING (client sends FIN,ACK with not yet sent acknowledgement number) 7. - <-- <SEQ=302><ACK=102><CTL=ACK> <-- CLOSING (ACK is 102 instead of 101, why?) ... (silence from CLIENT) 8. - <-- <SEQ=301><ACK=102><CTL=FIN,ACK> <-- CLOSING (retransmission, again ACK is 102) Now, note that packet 6 while having the expected sequence number, acknowledges something that wasn't sent by the server. So I would expect the packet to maybe prompt an ACK response from the server, and then be ignored. Yet it is not ignored and actually leads to an increase of the acknowledgement number in the server's retransmission of the FIN,ACK packet. The explanation I found is that the FIN in packet 6 was processed, despite the acknowledgement number being unacceptable. Further experiments indeed show that the server processes this FIN, transitioning to CLOSING, then on receiving an ACK for the FIN it had send in packet 5, the server (or better said connection) transitions from CLOSING to TIME_WAIT (as signaled by netstat). </quote> Indeed, tcp_rcv_state_process() calls tcp_ack() but does not exploit the @acceptable status but for TCP_SYN_RECV state. What we want here is to send a challenge ACK, if not in TCP_SYN_RECV state. TCP_FIN_WAIT1 state is not the only state we should fix. Add a FLAG_NO_CHALLENGE_ACK so that tcp_rcv_state_process() can choose to send a challenge ACK and discard the packet instead of wrongly change socket state. With help from Neal Cardwell. Signed-off-by:
Eric Dumazet <edumazet@google.com> Reported-by:
Paul Fiterau Brostean <p.fiterau-brostean@science.ru.nl> Cc: Neal Cardwell <ncardwell@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by:
-
Timmy Li authored
[ Upstream commit 717902cc ] Commit 093d24a2 ("arm64: PCI: Manage controller-specific data on per-controller basis") added code to allocate ACPI PCI root_ops dynamically on a per host bridge basis but failed to update the corresponding memory allocation failure path in pci_acpi_scan_root() leading to a potential memory leakage. Fix it by adding the required kfree call. Fixes: 093d24a2 ("arm64: PCI: Manage controller-specific data on per-controller basis") Reviewed-by:
Tomasz Nowicki <tn@semihalf.com> Signed-off-by:
Timmy Li <lixiaoping3@huawei.com> [lorenzo.pieralisi@arm.com: refactored code, rewrote commit log] Signed-off-by:
Lorenzo Pieralisi <lorenzo.pieralisi@arm.com> CC: Will Deacon <will.deacon@arm.com> CC: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by:
Catalin Marinas <catalin.marinas@arm.com> Signed-off-by:
-
Eryu Guan authored
[ Upstream commit 624327f8 ] ext4_find_unwritten_pgoff() is used to search for offset of hole or data in page range [index, end] (both inclusive), and the max number of pages to search should be at least one, if end == index. Otherwise the only page is missed and no hole or data is found, which is not correct. When block size is smaller than page size, this can be demonstrated by preallocating a file with size smaller than page size and writing data to the last block. E.g. run this xfs_io command on a 1k block size ext4 on x86_64 host. # xfs_io -fc "falloc 0 3k" -c "pwrite 2k 1k" \ -c "seek -d 0" /mnt/ext4/testfile wrote 1024/1024 bytes at offset 2048 1 KiB, 1 ops; 0.0000 sec (42.459 MiB/sec and 43478.2609 ops/sec) Whence Result DATA EOF Data at offset 2k was missed, and lseek(2) returned ENXIO. This is unconvered by generic/285 subtest 07 and 08 on ppc64 host, where pagesize is 64k. Because a recent change to generic/285 reduced the preallocated file size to smaller than 64k. Signed-off-by:
Eryu Guan <eguan@redhat.com> Signed-off-by:
Theodore Ts'o <tytso@mit.edu> Reviewed-by:
Jan Kara <jack@suse.cz> Signed-off-by:
-
Michael Schmitz authored
[ Upstream commit 9dfa7bba ] get_reg() can be reentered on architectures with prioritized interrupts (m68k in this case), causing f->reg_index to be incremented after the range check. Out of bounds memory access past the pt_regs struct results. This will go mostly undetected unless access is beyond end of memory. Prevent the race by disabling interrupts in get_reg(). Tested on m68k (Atari Falcon, and ARAnyM emulator). Kudos to Geert Uytterhoeven for helping to trace this race. Signed-off-by:
Michael Schmitz <schmitzmic@gmail.com> Signed-off-by:
-
Maurizio Lombardi authored
[ Upstream commit c2dd893a ] If multiple tasks attempt to read the stats, it may happen that the start_req_done completion is re-initialized while still being used by another task, causing a list corruption. This patch fixes the bug by adding a mutex to serialize the calls to bnx2fc_get_host_stats(). WARNING: at lib/list_debug.c:48 list_del+0x6e/0xa0() (Not tainted) Hardware name: PowerEdge R820 list_del corruption. prev->next should be ffff882035627d90, but was ffff884069541588 Pid: 40267, comm: perl Not tainted 2.6.32-642.3.1.el6.x86_64 #1 Call Trace: [<ffffffff8107c691>] ? warn_slowpath_common+0x91/0xe0 [<ffffffff8107c796>] ? warn_slowpath_fmt+0x46/0x60 [<ffffffff812ad16e>] ? list_del+0x6e/0xa0 [<ffffffff81547eed>] ? wait_for_common+0x14d/0x180 [<ffffffff8106c4a0>] ? default_wake_function+0x0/0x20 [<ffffffff81547fd3>] ? wait_for_completion_timeout+0x13/0x20 [<ffffffffa05410b1>] ? bnx2fc_get_host_stats+0xa1/0x280 [bnx2fc] [<ffffffffa04cf630>] ? fc_stat_show+0x90/0xc0 [scsi_transport_fc] [<ffffffffa04cf8b6>] ? show_fcstat_tx_frames+0x16/0x20 [scsi_transport_fc] [<ffffffff8137c647>] ? dev_attr_show+0x27/0x50 [<ffffffff8113b9be>] ? __get_free_pages+0xe/0x50 [<ffffffff812170e1>] ? sysfs_read_file+0x111/0x200 [<ffffffff8119a305>] ? vfs_read+0xb5/0x1a0 [<ffffffff8119b0b6>] ? fget_light_pos+0x16/0x50 [<ffffffff8119a651>] ? sys_read+0x51/0xb0 [<ffffffff810ee1fe>] ? __audit_syscall_exit+0x25e/0x290 [<ffffffff8100b0d2>] ? system_call_fastpath+0x16/0x1b Signed-off-by:
Maurizio Lombardi <mlombard@redhat.com> Acked-by:
Chad Dupuis <chad.dupuis@cavium.com> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by:
-
Kuninori Morimoto authored
[ Upstream commit 7819a942 ] commit 90431eb4 ("ASoC: rsnd: don't use PDTA bit for 24bit on SSI") fixups 24bit mode data alignment, but PIO was not cared. This patch fixes PIO mode 24bit data alignment Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by:
Mark Brown <broonie@kernel.org> Signed-off-by:
-
Dan Carpenter authored
[ Upstream commit 662f9a10 ] If xdr_inline_decode() fails then we end up returning ERR_PTR(0). The caller treats NULL returns as -ENOMEM so it doesn't really hurt runtime, but obviously we intended to set an error code here. Fixes: d67ae825 ("pnfs/flexfiles: Add the FlexFile Layout Driver") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by:
-
Liping Zhang authored
[ Upstream commit fefa9267 ] If nf_conntrack_htable_size was adjusted by the user during the ct dump operation, we may invoke nf_ct_put twice for the same ct, i.e. the "last" ct. This will cause the ct will be freed but still linked in hash buckets. It's very easy to reproduce the problem by the following commands: # while : ; do echo $RANDOM > /proc/sys/net/netfilter/nf_conntrack_buckets done # while : ; do conntrack -L done # iperf -s 127.0.0.1 & # iperf -c 127.0.0.1 -P 60 -t 36000 After a while, the system will hang like this: NMI watchdog: BUG: soft lockup - CPU#1 stuck for 22s! [bash:20184] NMI watchdog: BUG: soft lockup - CPU#0 stuck for 22s! [iperf:20382] ... So at last if we find cb->args[1] is equal to "last", this means hash resize happened, then we can set cb->args[1] to 0 to fix the above issue. Fixes: d205dc40 ("[NETFILTER]: ctnetlink: fix deadlock in table dumping") Signed-off-by:
Liping Zhang <zlpnobody@gmail.com> Signed-off-by:
Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by:
-
Milian Wolff authored
[ Upstream commit 1982ad48 ] As the documentation for dwfl_frame_pc says, frames that are no activation frames need to have their program counter decremented by one to properly find the function of the caller. This fixes many cases where perf report currently attributes the cost to the next line. I.e. I have code like this: ~~~~~~~~~~~~~~~ #include <thread> #include <chrono> using namespace std; int main() { this_thread::sleep_for(chrono::milliseconds(1000)); this_thread::sleep_for(chrono::milliseconds(100)); this_thread::sleep_for(chrono::milliseconds(10)); return 0; } ~~~~~~~~~~~~~~~ Now compile and record it: ~~~~~~~~~~~~~~~ g++ -std=c++11 -g -O2 test.cpp echo 1 | sudo tee /proc/sys/kernel/sched_schedstats perf record \ --event sched:sched_stat_sleep \ --event sched:sched_process_exit \ --event sched:sched_switch --call-graph=dwarf \ --output perf.data.raw \ ./a.out echo 0 | sudo tee /proc/sys/kernel/sched_schedstats perf inject --sched-stat --input perf.data.raw --output perf.data ~~~~~~~~~~~~~~~ Before this patch, the report clearly shows the off-by-one issue. Most notably, the last sleep invocation is incorrectly attributed to the "return 0;" line: ~~~~~~~~~~~~~~~ Overhead Source:Line ........ ........... 100.00% core.c:0 | ---__schedule core.c:0 schedule do_nanosleep hrtimer.c:0 hrtimer_nanosleep sys_nanosleep entry_SYSCALL_64_fastpath .tmp_entry_64.o:0 __nanosleep_nocancel .:0 std::this_thread::sleep_for<long, std::ratio<1l, 1000l> > thread:323 | |--90.08%--main test.cpp:9 | __libc_start_main | _start | |--9.01%--main test.cpp:10 | __libc_start_main | _start | --0.91%--main test.cpp:13 __libc_start_main _start ~~~~~~~~~~~~~~~ With this patch here applied, the issue is fixed. The report becomes much more usable: ~~~~~~~~~~~~~~~ Overhead Source:Line ........ ........... 100.00% core.c:0 | ---__schedule core.c:0 schedule do_nanosleep hrtimer.c:0 hrtimer_nanosleep sys_nanosleep entry_SYSCALL_64_fastpath .tmp_entry_64.o:0 __nanosleep_nocancel .:0 std::this_thread::sleep_for<long, std::ratio<1l, 1000l> > thread:323 | |--90.08%--main test.cpp:8 | __libc_start_main | _start | |--9.01%--main test.cpp:9 | __libc_start_main | _start | --0.91%--main test.cpp:10 __libc_start_main _start ~~~~~~~~~~~~~~~ Similarly it works for signal frames: ~~~~~~~~~~~~~~~ __noinline void bar(void) { volatile long cnt = 0; for (cnt = 0; cnt < 100000000; cnt++); } __noinline void foo(void) { bar(); } void sig_handler(int sig) { foo(); } int main(void) { signal(SIGUSR1, sig_handler); raise(SIGUSR1); foo(); return 0; } ~~~~~~~~~~~~~~~~ Before, the report wrongly points to `signal.c:29` after raise(): ~~~~~~~~~~~~~~~~ $ perf report --stdio --no-children -g srcline -s srcline ... 100.00% signal.c:11 | ---bar signal.c:11 | |--50.49%--main signal.c:29 | __libc_start_main | _start | --49.51%--0x33a8f raise .:0 main signal.c:29 __libc_start_main _start ~~~~~~~~~~~~~~~~ With this patch in, the issue is fixed and we instead get: ~~~~~~~~~~~~~~~~ 100.00% signal signal [.] bar | ---bar signal.c:11 | |--50.49%--main signal.c:29 | __libc_start_main | _start | --49.51%--0x33a8f raise .:0 main signal.c:27 __libc_start_main _start ~~~~~~~~~~~~~~~~ Note how this patch fixes this issue for both unwinding methods, i.e. both dwfl and libunwind. The former case is straight-forward thanks to dwfl_frame_pc(). For libunwind, we replace the functionality via unw_is_signal_frame() for any but the very first frame. Signed-off-by:
Milian Wolff <milian.wolff@kdab.com> Signed-off-by:
Namhyung Kim <namhyung@kernel.org> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Yao Jin <yao.jin@linux.intel.com> Cc: kernel-team@lge.com Link: http://lkml.kernel.org/r/20170524062129.32529-4-namhyung@kernel.orgSigned-off-by:
Ingo Molnar <mingo@kernel.org> Signed-off-by:
-
Dan Carpenter authored
[ Upstream commit 293dffaa ] If there is not enough space then ceph_decode_32_safe() does a goto bad. We need to return an error code in that situation. The current code returns ERR_PTR(0) which is NULL. The callers are not expecting that and it results in a NULL dereference. Fixes: f24e9980 ("ceph: OSD client") Signed-off-by:
Ilya Dryomov <idryomov@gmail.com> Signed-off-by:
-
Lin Zhang authored
[ Upstream commit a611c58b ] This patch fixes the kernel oops when release net_device reference in advance. In function raw_sendmsg(i think the dgram_sendmsg has the same problem), there is a race condition between dev_put and dev_queue_xmit when the device is gong that maybe lead to dev_queue_ximt to see an illegal net_device pointer. My test kernel is 3.13.0-32 and because i am not have a real 802154 device, so i change lowpan_newlink function to this: /* find and hold real wpan device */ real_dev = dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK])); if (!real_dev) return -ENODEV; // if (real_dev->type != ARPHRD_IEEE802154) { // dev_put(real_dev); // return -EINVAL; // } lowpan_dev_info(dev)->real_dev = real_dev; lowpan_dev_info(dev)->fragment_tag = 0; mutex_init(&lowpan_dev_info(dev)->dev_list_mtx); Also, in order to simulate preempt, i change the raw_sendmsg function to this: skb->dev = dev; skb->sk = sk; skb->protocol = htons(ETH_P_IEEE802154); dev_put(dev); //simulate preempt schedule_timeout_uninterruptible(30 * HZ); err = dev_queue_xmit(skb); if (err > 0) err = net_xmit_errno(err); and this is my userspace test code named test_send_data: int main(int argc, char **argv) { char buf[127]; int sockfd; sockfd = socket(AF_IEEE802154, SOCK_RAW, 0); if (sockfd < 0) { printf("create sockfd error: %s\n", strerror(errno)); return -1; } send(sockfd, buf, sizeof(buf), 0); return 0; } This is my test case: root@zhanglin-x-computer:~/develop/802154# uname -a Linux zhanglin-x-computer 3.13.0-32-generic #57-Ubuntu SMP Tue Jul 15 03:51:08 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux root@zhanglin-x-computer:~/develop/802154# ip link add link eth0 name lowpan0 type lowpan root@zhanglin-x-computer:~/develop/802154# //keep the lowpan0 device down root@zhanglin-x-computer:~/develop/802154# ./test_send_data & //wait a while root@zhanglin-x-computer:~/develop/802154# ip link del link dev lowpan0 //the device is gone //oops [381.303307] general protection fault: 0000 [#1]SMP [381.303407] Modules linked in: af_802154 6lowpan bnep rfcomm bluetooth nls_iso8859_1 snd_hda_codec_hdmi snd_hda_codec_realtek rts5139(C) snd_hda_intel snd_had_codec snd_hwdep snd_pcm snd_page_alloc snd_seq_midi snd_seq_midi_event snd_rawmidi snd_req intel_rapl snd_seq_device coretemp i915 kvm_intel kvm snd_timer snd crct10dif_pclmul crc32_pclmul ghash_clmulni_intel cypted drm_kms_helper drm i2c_algo_bit soundcore video mac_hid parport_pc ppdev ip parport hid_generic usbhid hid ahci r8169 mii libahdi [381.304286] CPU:1 PID: 2524 Commm: 1 Tainted: G C 0 3.13.0-32-generic [381.304409] Hardware name: Haier Haier DT Computer/Haier DT Codputer, BIOS FIBT19H02_X64 06/09/2014 [381.304546] tasks: ffff000096965fc0 ti: ffffB0013779c000 task.ti: ffffB8013779c000 [381.304659] RIP: 0010:[<ffffffff01621fe1>] [<ffffffff81621fe1>] __dev_queue_ximt+0x61/0x500 [381.304798] RSP: 0018:ffffB8013779dca0 EFLAGS: 00010202 [381.304880] RAX: 272b031d57565351 RBX: 0000000000000000 RCX: ffff8800968f1a00 [381.304987] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff8800968f1a00 [381.305095] RBP: ffff8e013773dce0 R08: 0000000000000266 R09: 0000000000000004 [381.305202] R10: 0000000000000004 R11: 0000000000000005 R12: ffff88013902e000 [381.305310] R13: 000000000000007f R14: 000000000000007f R15: ffff8800968f1a00 [381.305418] FS: 00007fc57f50f740(0000) GS: ffff88013fc80000(0000) knlGS: 0000000000000000 [381.305540] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [381.305627] CR2: 00007fad0841c000 CR3: 00000001368dd000 CR4: 00000000001007e0 [361.905734] Stack: [381.305768] 00000000002052d0 000000003facb30a ffff88013779dcc0 ffff880137764000 [381.305898] ffff88013779de70 000000000000007f 000000000000007f ffff88013902e000 [381.306026] ffff88013779dcf0 ffffffff81622490 ffff88013779dd39 ffffffffa03af9f1 [381.306155] Call Trace: [381.306202] [<ffffffff81622490>] dev_queue_xmit+0x10/0x20 [381.306294] [<ffffffffa03af9f1>] raw_sendmsg+0x1b1/0x270 [af_802154] [381.306396] [<ffffffffa03af054>] ieee802154_sock_sendmsg+0x14/0x20 [af_802154] [381.306512] [<ffffffff816079eb>] sock_sendmsg+0x8b/0xc0 [381.306600] [<ffffffff811d52a5>] ? __d_alloc+0x25/0x180 [381.306687] [<ffffffff811a1f56>] ? kmem_cache_alloc_trace+0x1c6/0x1f0 [381.306791] [<ffffffff81607b91>] SYSC_sendto+0x121/0x1c0 [381.306878] [<ffffffff8109ddf4>] ? vtime_account_user+x54/0x60 [381.306975] [<ffffffff81020d45>] ? syscall_trace_enter+0x145/0x250 [381.307073] [<ffffffff816086ae>] SyS_sendto+0xe/0x10 [381.307156] [<ffffffff8172c87f>] tracesys+0xe1/0xe6 [381.307233] Code: c6 a1 a4 ff 41 8b 57 78 49 8b 47 20 85 d2 48 8b 80 78 07 00 00 75 21 49 8b 57 18 48 85 d2 74 18 48 85 c0 74 13 8b 92 ac 01 00 00 <3b> 50 10 73 08 8b 44 90 14 41 89 47 78 41 f6 84 24 d5 00 00 00 [381.307801] RIP [<ffffffff81621fe1>] _dev_queue_xmit+0x61/0x500 [381.307901] RSP <ffff88013779dca0> [381.347512] Kernel panic - not syncing: Fatal exception in interrupt [381.347747] drm_kms_helper: panic occurred, switching back to text console In my opinion, there is always exist a chance that the device is gong before call dev_queue_xmit. I think the latest kernel is have the same problem and that dev_put should be behind of the dev_queue_xmit. Signed-off-by:
Lin Zhang <xiaolou4617@gmail.com> Acked-by:
Stefan Schmidt <stefan@osg.samsung.com> Signed-off-by:
Marcel Holtmann <marcel@holtmann.org> Signed-off-by:
-
Jesper Dangaard Brouer authored
[ Upstream commit 12e8b570 ] Masks for extracting part of the Completion Queue Entry (CQE) field rss_hash_type was swapped, namely CQE_RSS_HTYPE_IP and CQE_RSS_HTYPE_L4. The bug resulted in setting skb->l4_hash, even-though the rss_hash_type indicated that hash was NOT computed over the L4 (UDP or TCP) part of the packet. Added comments from the datasheet, to make it more clear what these masks are selecting. Signed-off-by:
Jesper Dangaard Brouer <brouer@redhat.com> Acked-by:
Saeed Mahameed <saeedm@mellanox.com> Signed-off-by:
-
Dan Carpenter authored
[ Upstream commit 7bd897cf ] We don't set an error code on this path. It means that we return NULL instead of an error pointer and the caller does a NULL dereference. Fixes: 6d1d8050 ("block, partition: add partition_meta_info to hd_struct") Signed-off-by:
Jens Axboe <axboe@fb.com> Signed-off-by:
-
Stephen Smalley authored
[ Upstream commit ccb54478 ] open permission is currently only defined for files in the kernel (COMMON_FILE_PERMS rather than COMMON_FILE_SOCK_PERMS). Construction of an artificial test case that tries to open a socket via /proc/pid/fd will generate a recvfrom avc denial because recvfrom and open happen to map to the same permission bit in socket vs file classes. open of a socket via /proc/pid/fd is not supported by the kernel regardless and will ultimately return ENXIO. But we hit the permission check first and can thus produce these odd/misleading denials. Omit the open check when operating on a socket. Signed-off-by:
Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by:
Paul Moore <paul@paul-moore.com> Signed-off-by:
-
Tariq Toukan authored
[ Upstream commit b665d98e ] Add tolerance to failures of irq_set_affinity_hint(). Its role is to give hints that optimizes performance, and should not block the driver load. In non-SMP systems, functionality is not available as there is a single core, and all these calls definitely fail. Hence, do not call the function and avoid the warning prints. Fixes: db058a18 ("net/mlx5_core: Set irq affinity hints") Signed-off-by:
Tariq Toukan <tariqt@mellanox.com> Cc: kernel-team@fb.com Signed-off-by:
-
Hans de Goede authored
[ Upstream commit 9a752b4c ] The Crystal Cove PMIC has 16 real GPIOs but the ACPI code for devices with this PMIC may address up to 95 GPIOs, these extra GPIOs are called virtual GPIOs and are used by the ACPI code as a method of accessing various non GPIO bits of PMIC. Commit dcdc3018 ("gpio: crystalcove: support virtual GPIO") added dummy support for these to avoid a bunch of ACPI errors, but instead of ignoring writes / reads to them by doing: if (gpio >= CRYSTALCOVE_GPIO_NUM) return 0; It accidentally introduced the following wrong check: if (gpio > CRYSTALCOVE_VGPIO_NUM) return 0; Which means that attempts by the ACPI code to access these gpios causes some arbitrary gpio to get touched through for example GPIO1P0CTLO + gpionr % 8. Since we do support input/output (but not interrupts) on the 0x5e virtual GPIO, this commit makes to_reg return -ENOTSUPP for unsupported virtual GPIOs so as to not have to check for (gpio >= CRYSTALCOVE_GPIO_NUM && gpio != 0x5e) everywhere and to make it easier to add support for more virtual GPIOs in the future. It then adds a check for to_reg returning an error to all callers where this may happen fixing the ACPI code accessing virtual GPIOs accidentally causing changes to real GPIOs. Fixes: dcdc3018 ("gpio: crystalcove: support virtual GPIO") Cc: Aaron Lu <aaron.lu@intel.com> Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Reviewed-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by:
Linus Walleij <linus.walleij@linaro.org> Signed-off-by:
-
Vlastimil Babka authored
[ Upstream commit 8655d549 ] A customer has reported a soft-lockup when running an intensive memory stress test, where the trace on multiple CPU's looks like this: RIP: 0010:[<ffffffff810c53fe>] [<ffffffff810c53fe>] native_queued_spin_lock_slowpath+0x10e/0x190 ... Call Trace: [<ffffffff81182d07>] queued_spin_lock_slowpath+0x7/0xa [<ffffffff811bc331>] change_protection_range+0x3b1/0x930 [<ffffffff811d4be8>] change_prot_numa+0x18/0x30 [<ffffffff810adefe>] task_numa_work+0x1fe/0x310 [<ffffffff81098322>] task_work_run+0x72/0x90 Further investigation showed that the lock contention here is pmd_lock(). The task_numa_work() function makes sure that only one thread is let to perform the work in a single scan period (via cmpxchg), but if there's a thread with mmap_sem locked for writing for several periods, multiple threads in task_numa_work() can build up a convoy waiting for mmap_sem for read and then all get unblocked at once. This patch changes the down_read() to the trylock version, which prevents the build up. For a workload experiencing mmap_sem contention, it's probably better to postpone the NUMA balancing work anyway. This seems to have fixed the soft lockups involving pmd_lock(), which is in line with the convoy theory. Signed-off-by:
Vlastimil Babka <vbabka@suse.cz> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by:
Rik van Riel <riel@redhat.com> Acked-by:
Mel Gorman <mgorman@techsingularity.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/20170515131316.21909-1-vbabka@suse.czSigned-off-by:
-
Dan Carpenter authored
[ Upstream commit 36cc2b92 ] We don't set an error code here which means that perf_event_alloc() returns ERR_PTR(0) (in other words NULL). The callers are not expecting that and would Oops. Signed-off-by:
-
Tin Huynh authored
[ Upstream commit aace34c0 ] The driver checks an incorrect flag of functionality of adapter. When a driver requires i2c_smbus_read_byte_data and i2c_smbus_write_byte_data, it should check I2C_FUNC_SMBUS_BYTE_DATA instead I2C_FUNC_I2C. This patch fixes the problem. Signed-off-by:
Tin Huynh <tnhuynh@apm.com> Signed-off-by:
Jacek Anaszewski <jacek.anaszewski@gmail.com> Signed-off-by:
-
Holger Brunck authored
[ Upstream commit 85deed56 ] sizeof(priv->ucc_pram) is 4 as it is the size of a pointer, but we want to reserve space for the struct ucc_hdlc_param. Signed-off-by:
Holger Brunck <holger.brunck@keymile.com> Cc: Zhao Qiang <qiang.zhao@nxp.com> Signed-off-by:
-
Kees Cook authored
[ Upstream commit e48d661e ] Using memcpy() from a buffer that is shorter than the length copied means the destination buffer is being filled with arbitrary data from the kernel rodata segment. In this case, the source was made longer, since it did not match the destination structure size. Additionally removes a needless cast. This was found with the future CONFIG_FORTIFY_SOURCE feature. Cc: Daniel Micay <danielmicay@gmail.com> Signed-off-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
Kalle Valo <kvalo@codeaurora.org> Signed-off-by:
-
Suman Anna authored
[ Upstream commit f97f0357 ] The DSP device on Davinci platforms does not have an MMU and requires specific DDR memory to boot. This memory is reserved using the rproc_mem kernel boot parameter and is assigned to the device on non-DT boots. The remoteproc core uses the DMA API and so will fall back to assigning random memory if this memory is not assigned to the device, but the DSP remote processor boot will not be successful in such cases. So, check that memory has been reserved and assigned to the device specifically before even creating the DSP device. Signed-off-by:
Suman Anna <s-anna@ti.com> Signed-off-by:
Sekhar Nori <nsekhar@ti.com> Signed-off-by:
-
Guoqing Jiang authored
[ Upstream commit 2dffdc07 ] The add_new_disk returns with communication locked if __sendmsg returns failure, fix it with call unlock_comm before return. Reported-by:
Guoqing Jiang <gqjiang@suse.com> Signed-off-by:
Shaohua Li <shli@fb.com> Signed-off-by:
-
Konstantin Khlebnikov authored
[ Upstream commit 9651e6b2 ] I've got another report about breaking ext4 by ENOMEM error returned from ext4_mb_load_buddy() caused by memory shortage in memory cgroup. This time inside ext4_discard_preallocations(). This patch replaces ext4_error() with ext4_warning() where errors returned from ext4_mb_load_buddy() are not fatal and handled by caller: * ext4_mb_discard_group_preallocations() - called before generating ENOSPC, we'll try to discard other group or return ENOSPC into user-space. * ext4_trim_all_free() - just stop trimming and return ENOMEM from ioctl. Some callers cannot handle errors, thus __GFP_NOFAIL is used for them: * ext4_discard_preallocations() * ext4_mb_discard_lg_preallocations() Fixes: adb7ef60 ("ext4: use __GFP_NOFAIL in ext4_free_blocks()") Signed-off-by:
Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Signed-off-by:
-
Mikko Koivunen authored
[ Upstream commit 12d74949 ] Set sensor measurement off after probe fail in pm_runtime_set_active() or iio_device_register(). Without this change sensor measurement stays on even though probe fails on these calls. This is maybe rare case, but causes constant power drain without any benefits when it happens. Power drain is 20-500uA, typically 180uA. Signed-off-by:
Mikko Koivunen <mikko.koivunen@fi.rohmeurope.com> Acked-by:
Daniel Baluta <daniel.baluta@nxp.com> Signed-off-by:
Jonathan Cameron <jic23@kernel.org> Signed-off-by:
-
Nikita Yushchenko authored
[ Upstream commit 61305664 ] Reset GPIO is active low. Currently driver uses gpiod_set_value(1) to clean reset, which depends on device tree to contain GPIO_ACTIVE_HIGH - that does not match reality. This fixes driver to use _raw version of gpiod_set_value() to enforce active-low semantics despite of what's written in device tree. Allowing device tree to override that only opens possibility for errors and does not add any value. Additionally, use _cansleep version to make things work with i2c-gpio and other sleeping gpio drivers. Signed-off-by:
Nikita Yushchenko <nikita.yoush@cogentembedded.com> Signed-off-by:
-
Nikita Yushchenko authored
[ Upstream commit ee19ac34 ] Currently, driver generates events for channels if new reading differs from previous one. This "previous value" is initialized to zero, which results into event if value is constant-one. Fix that by initializing "previous value" by reading at event enable time. This provides reliable sequence for userspace: - enable event, - AFTER THAT read current value, - AFTER THAT each event will correspond to change. Signed-off-by:
-
Stefan Agner authored
[ Upstream commit f746aa5e ] Initialize asoc_simple_card_init_mic with the correct struct asoc_simple_jack. Fixes: 9eac3618 ("ASoC: simple-card: add new asoc_simple_jack and use it") Signed-off-by:
Stefan Agner <stefan@agner.ch> Acked-by:
-
Antony Antony authored
[ Upstream commit a486cd23 ] During xfrm migration copy replay and preplay sequence numbers from the previous state. Here is a tcpdump output showing the problem. 10.0.10.46 is running vanilla kernel, is the IKE/IPsec responder. After the migration it sent wrong sequence number, reset to 1. The migration is from 10.0.0.52 to 10.0.0.53. IP 10.0.0.52.4500 > 10.0.10.46.4500: UDP-encap: ESP(spi=0x43ef462d,seq=0x7cf), length 136 IP 10.0.10.46.4500 > 10.0.0.52.4500: UDP-encap: ESP(spi=0xca1c282d,seq=0x7cf), length 136 IP 10.0.0.52.4500 > 10.0.10.46.4500: UDP-encap: ESP(spi=0x43ef462d,seq=0x7d0), length 136 IP 10.0.10.46.4500 > 10.0.0.52.4500: UDP-encap: ESP(spi=0xca1c282d,seq=0x7d0), length 136 IP 10.0.0.53.4500 > 10.0.10.46.4500: NONESP-encap: isakmp: child_sa inf2[I] IP 10.0.10.46.4500 > 10.0.0.53.4500: NONESP-encap: isakmp: child_sa inf2[R] IP 10.0.0.53.4500 > 10.0.10.46.4500: NONESP-encap: isakmp: child_sa inf2[I] IP 10.0.10.46.4500 > 10.0.0.53.4500: NONESP-encap: isakmp: child_sa inf2[R] IP 10.0.0.53.4500 > 10.0.10.46.4500: UDP-encap: ESP(spi=0x43ef462d,seq=0x7d1), length 136 NOTE: next sequence is wrong 0x1 IP 10.0.10.46.4500 > 10.0.0.53.4500: UDP-encap: ESP(spi=0xca1c282d,seq=0x1), length 136 IP 10.0.0.53.4500 > 10.0.10.46.4500: UDP-encap: ESP(spi=0x43ef462d,seq=0x7d2), length 136 IP 10.0.10.46.4500 > 10.0.0.53.4500: UDP-encap: ESP(spi=0xca1c282d,seq=0x2), length 136 Signed-off-by:
Antony Antony <antony@phenome.org> Reviewed-by:
Richard Guy Briggs <rgb@tricolour.ca> Signed-off-by:
Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by:
-
Michael Ellerman authored
[ Upstream commit fe06fe86 ] The tm-resched-dscr test has started failing sometimes, depending on what compiler it's built with, eg: test: tm_resched_dscr Check DSCR TM context switch: tm-resched-dscr: tm-resched-dscr.c:76: test_body: Assertion `rv' failed. !! child died by signal 6 When it fails we see that the compiler doesn't initialise rv to 1 before entering the inline asm block. Although that's counter intuitive, it is allowed because we tell the compiler that the inline asm will write to rv (using "=r"), meaning the original value is irrelevant. Marking it as a read/write parameter would presumably work, but it seems simpler to fix it by setting the initial value of rv in the inline asm. Fixes: 96d01610 ("powerpc: Correct DSCR during TM context switch") Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au> Acked-by:
Michael Neuling <mikey@neuling.org> Signed-off-by:
-
Colin Ian King authored
[ Upstream commit 8fed6823 ] The AR5K_EEPROM_READ macro returns with -EIO if a read error occurs causing a memory leak on the allocated buffer buf. Fix this by explicitly calling ath5k_hw_nvram_read and exiting on the via the freebuf label that performs the necessary free'ing of buf when a read error occurs. Detected by CoverityScan, CID#1248782 ("Resource Leak") Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by:
-
Michael Ellerman authored
[ Upstream commit e41e53cd ] virt_addr_valid() is supposed to tell you if it's OK to call virt_to_page() on an address. What this means in practice is that it should only return true for addresses in the linear mapping which are backed by a valid PFN. We are failing to properly check that the address is in the linear mapping, because virt_to_pfn() will return a valid looking PFN for more or less any address. That bug is actually caused by __pa(), used in virt_to_pfn(). eg: __pa(0xc000000000010000) = 0x10000 # Good __pa(0xd000000000010000) = 0x10000 # Bad! __pa(0x0000000000010000) = 0x10000 # Bad! This started happening after commit bdbc29c1 ("powerpc: Work around gcc miscompilation of __pa() on 64-bit") (Aug 2013), where we changed the definition of __pa() to work around a GCC bug. Prior to that we subtracted PAGE_OFFSET from the value passed to __pa(), meaning __pa() of a 0xd or 0x0 address would give you something bogus back. Until we can verify if that GCC bug is no longer an issue, or come up with another solution, this commit does the minimal fix to make virt_addr_valid() work, by explicitly checking that the address is in the linear mapping region. Fixes: bdbc29c1 ("powerpc: Work around gcc miscompilation of __pa() on 64-bit") Signed-off-by:
Paul Mackerras <paulus@ozlabs.org> Reviewed-by:
Balbir Singh <bsingharora@gmail.com> Tested-by:
Breno Leitao <breno.leitao@gmail.com> Signed-off-by:
-
Varun Prakash authored
[ Upstream commit a351e40b ] mbp pointer is passed to csio_hw_validate_caps() so call mempool_free() after calling csio_hw_validate_caps(). Signed-off-by:
Varun Prakash <varun@chelsio.com> Fixes: 541c571f ("csiostor:Use firmware version from cxgb4/t4fw_version.h") Reviewed-by:
Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by:
-
Ido Schimmel authored
[ Upstream commit c0e01eac ] In case we got an FDB notification for a port that doesn't exist we execute an FDB entry delete to prevent it from re-appearing the next time we poll for notifications. If the operation failed we would trigger a NULL pointer dereference as 'mlxsw_sp_port' is NULL. Fix it by reporting the error using the underlying bus device instead. Fixes: 12f1501e ("mlxsw: spectrum: remove FDB entry in case we get unknown object notification") Signed-off-by:
Ido Schimmel <idosch@mellanox.com> Signed-off-by:
Jiri Pirko <jiri@mellanox.com> Signed-off-by:
-
Geert Uytterhoeven authored
[ Upstream commit 5f5c5449 ] The MDIO initialization failure message is printed using the network device, before it has been registered, leading to: (null): failed to initialise MDIO Use the platform device instead to fix this: sh-eth ee700000.ethernet: failed to initialise MDIO Fixes: daacf03f ("sh_eth: Register MDIO bus before registering the network device") Signed-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Reviewed-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by:
-
Holger Brunck authored
[ Upstream commit c7f235a7 ] Add the bitmask for the two bit SYNL register according to the QUICK Engine Reference Manual. Signed-off-by:
-
Holger Brunck authored
[ Upstream commit 5b8aad93 ] We need space for the struct qe_bd and not for a pointer to this struct. Signed-off-by:
-
