- 24 Sep, 2016 40 commits
-
-
Trond Myklebust authored
commit b88fa69e upstream. Ensure that the client conforms to the normative behaviour described in RFC5661 Section 12.7.2: "If a client believes its lease has expired, it MUST NOT send I/O to the storage device until it has validated its lease." So ensure that we wait for the lease to be validated before using the layout. Signed-off-by:
Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Trond Myklebust authored
commit 3dc14735 upstream. If the attempt to connect to a DS fails inside ff_layout_pg_init_read or ff_layout_pg_init_write, then we currently end up clearing the layout segment carried by the struct nfs_pageio_descriptor, causing an Oops when we later call into ff_layout_read_pagelist/ff_layout_write_pagelist. The fix is to ensure we return the layout and then retry. Fixes: 446ca219 ("pNFS/flexfiles: When initing reads or writes, we...") Signed-off-by:
-
Tejun Heo authored
commit df6a58c5 upstream. kernfs_notify_workfn() sends out file modified events for the scheduled kernfs_nodes. Because the modifications aren't from userland, it doesn't have the matching file struct at hand and can't use fsnotify_modify(). Instead, it looked up the inode and then used d_find_any_alias() to find the dentry and used fsnotify_parent() and fsnotify() directly to generate notifications. The assumption was that the relevant dentries would have been pinned if there are listeners, which isn't true as inotify doesn't pin dentries at all and watching the parent doesn't pin the child dentries even for dnotify. This led to, for example, inotify watchers not getting notifications if the system is under memory pressure and the matching dentries got reclaimed. It can also be triggered through /proc/sys/vm/drop_caches or a remount attempt which involves shrinking dcache. fsnotify_parent() only uses the dentry to access the parent inode, which kernfs can do easily. Update kernfs_notify_workfn() so that it uses fsnotify() directly for both the parent and target inodes without going through d_find_any_alias(). While at it, supply the target file name to fsnotify() from kernfs_node->name. Signed-off-by:
Tejun Heo <tj@kernel.org> Reported-by:
Evgeny Vereshchagin <evvers@ya.ru> Fixes: d911d987 ("kernfs: make kernfs_notify() trigger inotify events too") Cc: John McCutchan <john@johnmccutchan.com> Cc: Robert Love <rlove@rlove.org> Cc: Eric Paris <eparis@parisplace.org> Signed-off-by:
-
Gavin Shan authored
commit caa58f80 upstream. In pnv_ioda_free_pe(), the PE object (including the associated PE number) is cleared before resetting the corresponding bit in the PE allocation bitmap. It means PE#0 is always released to the bitmap wrongly. This fixes above issue by caching the PE number before the PE object is cleared. Fixes: 1e916772 ("powerpc/powernv: Use PE instead of number during setup and release" Signed-off-by:
Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au> Signed-off-by:
-
Paul Mackerras authored
commit f077aaf0 upstream. In commit c60ac569 ("powerpc: Update kernel VSID range", 2013-03-13) we lost a check on the region number (the top four bits of the effective address) for addresses below PAGE_OFFSET. That commit replaced a check that the top 18 bits were all zero with a check that bits 46 - 59 were zero (performed for all addresses, not just user addresses). This means that userspace can access an address like 0x1000_0xxx_xxxx_xxxx and we will insert a valid SLB entry for it. The VSID used will be the same as if the top 4 bits were 0, but the page size will be some random value obtained by indexing beyond the end of the mm_ctx_high_slices_psize array in the paca. If that page size is the same as would be used for region 0, then userspace just has an alias of the region 0 space. If the page size is different, then no HPTE will be found for the access, and the process will get a SIGSEGV (since hash_page_mm() will refuse to create a HPTE for the bogus address). The access beyond the end of the mm_ctx_high_slices_psize can be at most 5.5MB past the array, and so will be in RAM somewhere. Since the access is a load performed in real mode, it won't fault or crash the kernel. At most this bug could perhaps leak a little bit of information about blocks of 32 bytes of memory located at offsets of i * 512kB past the paca->mm_ctx_high_slices_psize array, for 1 <= i <= 11. Fixes: c60ac569 ("powerpc: Update kernel VSID range") Signed-off-by:
Paul Mackerras <paulus@ozlabs.org> Reviewed-by:
Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by:
-
Christophe Leroy authored
commit 41017a75 upstream. of_mm_gpiochip_add_data() calls mm_gc->save_regs() before setting the data. Therefore ->save_regs() cannot use gpiochip_get_data() [ 0.275940] Unable to handle kernel paging request for data at address 0x00000130 [ 0.283120] Faulting instruction address: 0xc01b44cc [ 0.288175] Oops: Kernel access of bad area, sig: 11 [#1] [ 0.293343] PREEMPT CMPC885 [ 0.296141] CPU: 0 PID: 1 Comm: swapper Not tainted 4.7.0-g65124df-dirty #68 [ 0.304131] task: c6074000 ti: c6080000 task.ti: c6080000 [ 0.309459] NIP: c01b44cc LR: c0011720 CTR: c0011708 [ 0.314372] REGS: c6081d90 TRAP: 0300 Not tainted (4.7.0-g65124df-dirty) [ 0.322267] MSR: 00009032 <EE,ME,IR,DR,RI> CR: 24000028 XER: 20000000 [ 0.328813] DAR: 00000130 DSISR: c0000000 GPR00: c01b6d0c c6081e40 c6074000 c6017000 c9028000 c601d028 c6081dd8 00000000 GPR08: c601d028 00000000 ffffffff 00000001 24000044 00000000 c0002790 00000000 GPR16: 00000000 00000000 00000000 00000000 00000000 00000000 c05643b0 00000083 GPR24: c04a1a6c c0560000 c04a8308 c04c6480 c0012498 c6017000 c7ffcc78 c6017000 [ 0.360806] NIP [c01b44cc] gpiochip_get_data+0x4/0xc [ 0.365684] LR [c0011720] cpm1_gpio16_save_regs+0x18/0x44 [ 0.370972] Call Trace: [ 0.373451] [c6081e50] [c01b6d0c] of_mm_gpiochip_add_data+0x70/0xdc [ 0.379624] [c6081e70] [c00124c0] cpm_init_par_io+0x28/0x118 [ 0.385238] [c6081e80] [c04a8ac0] do_one_initcall+0xb0/0x17c [ 0.390819] [c6081ef0] [c04a8cbc] kernel_init_freeable+0x130/0x1dc [ 0.396924] [c6081f30] [c00027a4] kernel_init+0x14/0x110 [ 0.402177] [c6081f40] [c000b424] ret_from_kernel_thread+0x5c/0x64 [ 0.408233] Instruction dump: [ 0.411168] 4182fafc 3f80c040 48234c6d 3bc0fff0 3b9c5ed0 4bfffaf4 81290020 712a0004 [ 0.418825] 4182fb34 48234c51 4bfffb2c 81230004 <80690130> 4e800020 7c0802a6 9421ffe0 [ 0.426763] ---[ end trace fe4113ee21d72ffa ]--- fixes: e65078f1 ("powerpc: sysdev: cpm1: use gpiochip data pointer") fixes: a14a2d48 ("powerpc: cpm_common: use gpiochip data pointer") Signed-off-by:
Christophe Leroy <christophe.leroy@c-s.fr> Reviewed-by:
Linus Walleij <linus.walleij@linaro.org> Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Mukesh Ojha authored
commit a9cbf0b2 upstream. In a situation, where Linux kernel gets notified about duplicate error log from OPAL, it is been observed that kernel fails to remove sysfs entries (/sys/firmware/opal/elog/0xXXXXXXXX) of such error logs. This is because, we currently search the error log/dump kobject in the kset list via 'kset_find_obj()' routine. Which eventually increment the reference count by one, once it founds the kobject. So, unless we decrement the reference count by one after it found the kobject, we would not be able to release the kobject properly later. This patch adds the 'kobject_put()' which was missing earlier. Signed-off-by:
Mukesh Ojha <mukesh02@linux.vnet.ibm.com> Reviewed-by:
Vasant Hegde <hegdevasant@linux.vnet.ibm.com> Signed-off-by:
-
Nicholas Piggin authored
commit cc7786d3 upstream. tabort_syscall runs with RI=1, so a nested recoverable machine check will load the paca into r13 and overwrite what we loaded it with, because exceptions returning to privileged mode do not restore r13. Fixes: b4b56f9e (powerpc/tm: Abort syscalls in active transactions) Signed-off-by:
Nick Piggin <npiggin@gmail.com> Signed-off-by:
-
Artem Germanov authored
[ Upstream commit db7196a0 ] Commit 76174004 (tcp: do not slow start when cwnd equals ssthresh ) introduced regression in TCP YeAH. Using 100ms delay 1% loss virtual ethernet link kernel 4.2 shows bandwidth ~500KB/s for single TCP connection and kernel 4.3 and above (including 4.8-rc4) shows bandwidth ~100KB/s. That is caused by stalled cwnd when cwnd equals ssthresh. This patch fixes it by proper increasing cwnd in this case. Signed-off-by:
Artem Germanov <agermanov@anchorfree.com> Acked-by:
Dmitry Adamushko <d.adamushko@anchorfree.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Gal Pressman authored
[ Upstream commit cd17d230 ] Currently vlan tagged packets were not parsed correctly and assumed to be regular IPv4/IPv6 packets. We should check for 802.1Q/802.1ad tags and update the lro header accordingly. This fixes the use case where LRO is on and rxvlan is off (vlan stripping is off). Fixes: e586b3b0 ('net/mlx5: Ethernet Datapath files') Signed-off-by:
Gal Pressman <galp@mellanox.com> Signed-off-by:
Saeed Mahameed <saeedm@mellanox.com> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit 76061f63 ] When DATA and/or FIN are carried in a SYN/ACK message or SYN message, we append an skb in socket receive queue, but we forget to call sk_forced_mem_schedule(). Effect is that the socket has a negative sk->sk_forward_alloc as long as the message is not read by the application. Josh Hunt fixed a similar issue in commit d22e1537 ("tcp: fix tcp fin memory accounting") Fixes: 168a8f58 ("tcp: TCP Fast Open Server - main code path") Signed-off-by:
Eric Dumazet <edumazet@google.com> Reviewed-by:
Josh Hunt <johunt@akamai.com> Signed-off-by:
-
Wei Yongjun authored
[ Upstream commit 751eb6b6 ] In general, when DAD detected IPv6 duplicate address, ifp->state will be set to INET6_IFADDR_STATE_ERRDAD and DAD is stopped by a delayed work, the call tree should be like this: ndisc_recv_ns -> addrconf_dad_failure <- missing ifp put -> addrconf_mod_dad_work -> schedule addrconf_dad_work() -> addrconf_dad_stop() <- missing ifp hold before call it addrconf_dad_failure() called with ifp refcont holding but not put. addrconf_dad_work() call addrconf_dad_stop() without extra holding refcount. This will not cause any issue normally. But the race between addrconf_dad_failure() and addrconf_dad_work() may cause ifp refcount leak and netdevice can not be unregister, dmesg show the following messages: IPv6: eth0: IPv6 duplicate address fe80::XX:XXXX:XXXX:XX detected! ... unregister_netdevice: waiting for eth0 to become free. Usage count = 1 Cc: stable@vger.kernel.org Fixes: c15b1cca ("ipv6: move DAD and addrconf_verify processing to workqueue") Signed-off-by:
Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by:
-
Michael Chan authored
[ Upstream commit 9d13744b ] There is a code path where we are calling __iowrite64_copy() on an address that is not 64-bit aligned. This causes an exception on some architectures such as arm64. Fix that code path by using __iowrite32_copy(). Reported-by:
JD Zheng <jiandong.zheng@broadcom.com> Signed-off-by:
Michael Chan <michael.chan@broadcom.com> Signed-off-by:
-
Dave Jones authored
[ Upstream commit 03c2778a ] Neither the failure or success paths of ping_v6_sendmsg release the dst it acquires. This leads to a flood of warnings from "net/core/dst.c:288 dst_release" on older kernels that don't have 8bf4ada2 backported. That patch optimistically hoped this had been fixed post 3.10, but it seems at least one case wasn't, where I've seen this triggered a lot from machines doing unprivileged icmp sockets. Cc: Martin Lau <kafai@fb.com> Signed-off-by:
Dave Jones <davej@codemonkey.org.uk> Acked-by:
Martin KaFai Lau <kafai@fb.com> Signed-off-by:
-
Linus Torvalds authored
[ Upstream commit 6e1ce3c3 ] Right now we use the 'readlock' both for protecting some of the af_unix IO path and for making the bind be single-threaded. The two are independent, but using the same lock makes for a nasty deadlock due to ordering with regards to filesystem locking. The bind locking would want to nest outside the VSF pathname locking, but the IO locking wants to nest inside some of those same locks. We tried to fix this earlier with commit c845acb3 ("af_unix: Fix splice-bind deadlock") which moved the readlock inside the vfs locks, but that caused problems with overlayfs that will then call back into filesystem routines that take the lock in the wrong order anyway. Splitting the locks means that we can go back to having the bind lock be the outermost lock, and we don't have any deadlocks with lock ordering. Acked-by:
Rainer Weikusat <rweikusat@cyberadapt.com> Acked-by:
Al Viro <viro@zeniv.linux.org.uk> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Acked-by:
Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by:
-
Linus Torvalds authored
[ Upstream commit 38f7bd94 ] This reverts commit c845acb3. It turns out that it just replaces one deadlock with another one: we can still get the wrong lock ordering with the readlock due to overlayfs calling back into the filesystem layer and still taking the vfs locks after the readlock. The proper solution ends up being to just split the readlock into two pieces: the bind lock (taken *outside* the vfs locks) and the IO lock (taken *inside* the filesystem locks). The two locks are independent anyway. Signed-off-by:
Shmulik Ladkani <shmulik.ladkani@gmail.com> Signed-off-by:
-
Mahesh Bandewar authored
[ Upstream commit 24b27fc4 ] Following few steps will crash kernel - (a) Create bonding master > modprobe bonding miimon=50 (b) Create macvlan bridge on eth2 > ip link add link eth2 dev mvl0 address aa:0:0:0:0:01 \ type macvlan (c) Now try adding eth2 into the bond > echo +eth2 > /sys/class/net/bond0/bonding/slaves <crash> Bonding does lots of things before checking if the device enslaved is busy or not. In this case when the notifier call-chain sends notifications, the bond_netdev_event() assumes that the rx_handler /rx_handler_data is registered while the bond_enslave() hasn't progressed far enough to register rx_handler for the new slave. This patch adds a rx_handler check that can be performed right at the beginning of the enslave code to avoid getting into this situation. Signed-off-by:
Mahesh Bandewar <maheshb@google.com> Acked-by:
-
WANG Cong authored
[ Upstream commit c0338aff ] Dmitry reported a double free on kcm socket, which could be easily reproduced by: #include <unistd.h> #include <sys/syscall.h> int main() { int fd = syscall(SYS_socket, 0x29ul, 0x5ul, 0x0ul, 0, 0, 0); syscall(SYS_ioctl, fd, 0x89e2ul, 0x20a98000ul, 0, 0, 0); return 0; } This is because on the error path, after we install the new socket file, we call sock_release() to clean up the socket, which leaves the fd pointing to a freed socket. Fix this by calling sys_close() on that fd directly. Fixes: ab7ac4eb ("kcm: Kernel Connection Multiplexor module") Reported-by:
Dmitry Vyukov <dvyukov@google.com> Cc: Tom Herbert <tom@herbertland.com> Signed-off-by:
Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by:
-
Davide Caratti authored
[ Upstream commit 9264251e ] commit bc8c20ac ("bridge: multicast: treat igmpv3 report with INCLUDE and no sources as a leave") seems to have accidentally reverted commit 47cc84ce ("bridge: fix parsing of MLDv2 reports"). This commit brings back a change to br_ip6_multicast_mld2_report() where parsing of MLDv2 reports stops when the first group is successfully added to the MDB cache. Fixes: bc8c20ac ("bridge: multicast: treat igmpv3 report with INCLUDE and no sources as a leave") Signed-off-by:
Davide Caratti <dcaratti@redhat.com> Acked-by:
Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by:
Thadeu Lima de Souza Cascardo <cascardo@redhat.com> Signed-off-by:
-
Russell King authored
[ Upstream commit 2fb04fdf ] Commit b70661c7 ("net: smc91x: use run-time configuration on all ARM machines") broke some ARM platforms through several mistakes. Firstly, the access size must correspond to the following rule: (a) at least one of 16-bit or 8-bit access size must be supported (b) 32-bit accesses are optional, and may be enabled in addition to the above. Secondly, it provides no emulation of 16-bit accesses, instead blindly making 16-bit accesses even when the platform specifies that only 8-bit is supported. Reorganise smc91x.h so we can make use of the existing 16-bit access emulation already provided - if 16-bit accesses are supported, use 16-bit accesses directly, otherwise if 8-bit accesses are supported, use the provided 16-bit access emulation. If neither, BUG(). This exactly reflects the driver behaviour prior to the commit being fixed. Since the conversion incorrectly cut down the available access sizes on several platforms, we also need to go through every platform and fix up the overly-restrictive access size: Arnd assumed that if a platform can perform 32-bit, 16-bit and 8-bit accesses, then only a 32-bit access size needed to be specified - not so, all available access sizes must be specified. This likely fixes some performance regressions in doing this: if a platform does not support 8-bit accesses, 8-bit accesses have been emulated by performing a 16-bit read-modify-write access. Tested on the Intel Assabet/Neponset platform, which supports only 8-bit accesses, which was broken by the original commit. Fixes: b70661c7 ("net: smc91x: use run-time configuration on all ARM machines") Signed-off-by:
Russell King <rmk+kernel@armlinux.org.uk> Tested-by:
Robert Jarzmik <robert.jarzmik@free.fr> Signed-off-by:
-
Xander Huff authored
[ Upstream commit c3e70edd ] This reverts: commit 33c133cc ("phy: IRQ cannot be shared") On hardware with multiple PHY devices hooked up to the same IRQ line, allow them to share it. Sergei Shtylyov says: "I'm not sure now what was the reason I concluded that the IRQ sharing was impossible... most probably I thought that the kernel IRQ handling code exited the loop over the IRQ actions once IRQ_HANDLED was returned -- which is obviously not so in reality..." Signed-off-by:
Xander Huff <xander.huff@ni.com> Signed-off-by:
Nathan Sullivan <nathan.sullivan@ni.com> Signed-off-by:
-
Florian Fainelli authored
[ Upstream commit 4f101c47 ] We kept shadow copies of which interrupt sources we have enabled and disabled, but due to an order bug in how intrl2_mask_clear was defined, we could run into the following scenario: CPU0 CPU1 intrl2_1_mask_clear(..) sets INTRL2_CPU_MASK_CLEAR bcm_sf2_switch_1_isr read INTRL2_CPU_STATUS and masks with stale irq1_mask value updates irq1_mask value Which would make us loop again and again trying to process and interrupt we are not clearing since our copy of whether it was enabled before still indicates it was not. Fix this by updating the shadow copy first, and then unasking at the HW level. Fixes: 246d7f77 ("net: dsa: add Broadcom SF2 switch driver") Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Soheil Hassas Yeganeh authored
[ Upstream commit 7b996243 ] Instead of using sock_tx_timestamp, use skb_tx_timestamp to record software transmit timestamp of a packet. sock_tx_timestamp resets and overrides the tx_flags of the skb. The function is intended to be called from within the protocol layer when creating the skb, not from a device driver. This is inconsistent with other drivers and will cause issues for TCP. In TCP, we intend to sample the timestamps for the last byte for each sendmsg/sendpage. For that reason, tcp_sendmsg calls tcp_tx_timestamp only with the last skb that it generates. For example, if a 128KB message is split into two 64KB packets we want to sample the SND timestamp of the last packet. The current code in the tun driver, however, will result in sampling the SND timestamp for both packets. Also, when the last packet is split into smaller packets for retranmission (see tcp_fragment), the tun driver will record timestamps for all of the retransmitted packets and not only the last packet. Fixes: eda29772 (tun: Support software transmit time stamping.) Signed-off-by:
Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by:
Francis Yan <francisyyan@google.com> Acked-by:
-
Lance Richardson authored
[ Upstream commit 232cb53a ] The function sctp_diag_dump_one() currently performs a memcpy() of 64 bytes from a 16 byte field into another 16 byte field. Fix by using correct size, use sizeof to obtain correct size instead of using a hard-coded constant. Fixes: 8f840e47 ("sctp: add the sctp_diag.c file") Signed-off-by:
Lance Richardson <lrichard@redhat.com> Reviewed-by:
Xin Long <lucien.xin@gmail.com> Acked-by:
Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit 20a2b49f ] When sending an ack in SYN_RECV state, we must scale the offered window if wscale option was negotiated and accepted. Tested: Following packetdrill test demonstrates the issue : 0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0 +0 bind(3, ..., ...) = 0 +0 listen(3, 1) = 0 // Establish a connection. +0 < S 0:0(0) win 20000 <mss 1000,sackOK,wscale 7, nop, TS val 100 ecr 0> +0 > S. 0:0(0) ack 1 win 28960 <mss 1460,sackOK, TS val 100 ecr 100, nop, wscale 7> +0 < . 1:11(10) ack 1 win 156 <nop,nop,TS val 99 ecr 100> // check that window is properly scaled ! +0 > . 1:1(0) ack 1 win 226 <nop,nop,TS val 200 ecr 100> Signed-off-by:
Yuchung Cheng <ycheng@google.com> Acked-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit e83c6744 ] Laura tracked poll() [and friends] regression caused by commit e6afc8ac ("udp: remove headers from UDP packets before queueing") udp_poll() needs to know if there is a valid packet in receive queue, even if its payload length is 0. Change first_packet_length() to return an signed int, and use -1 as the indication of an empty queue. Fixes: e6afc8ac ("udp: remove headers from UDP packets before queueing") Reported-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
-
Jamal Hadi Salim authored
[ Upstream commit 28a10c42 ] Encoding of the metadata was using the padded length as opposed to the real length of the data which is a bug per specification. This has not been an issue todate because all metadatum specified so far has been 32 bit where aligned and data length are the same width. This also includes a bug fix for validating the length of a u16 field. But since there is no metadata of size u16 yes we are fine to include it here. While at it get rid of magic numbers. Fixes: ef6980b6 ("net sched: introduce IFE action") Signed-off-by:
Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
-
Hadar Hen Zion authored
[ Upstream commit 1dbd0d37 ] The wrong key is used when extracting the address type field set by the flower offload code. We have to use the control key and not the basic key, fix that. Fixes: e3a2b7ed ('net/mlx5e: Support offload cls_flower with drop action') Signed-off-by:
Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by:
-
Paul Blakey authored
[ Upstream commit 2c0f8ce1 ] Set and verify signature calculates the signature for each of the mailbox nodes, even for those that are unused (from cache). Added a missing length check to set and verify only those which are used. While here, also moved the setting of msg's nodes token to where we already go over them. This saves a pass because checksum is disabled, and the only useful thing remaining that set signature does is setting the token. Fixes: e126ba97 ('mlx5: Add driver for Mellanox Connect-IB adapters') Signed-off-by:
Paul Blakey <paulb@mellanox.com> Signed-off-by:
-
Mohamad Haj Yahia authored
[ Upstream commit 1061c90f ] When PCI error is detected we should save the state of the pci prior to disabling it. Also when receiving pci slot reset call we need to verify that the device is responsive. Fixes: 89d44f0a ('net/mlx5_core: Add pci error handlers to mlx5_core driver') Signed-off-by:
Mohamad Haj Yahia <mohamad@mellanox.com> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit bb1fceca ] When tcp_sendmsg() allocates a fresh and empty skb, it puts it at the tail of the write queue using tcp_add_write_queue_tail() Then it attempts to copy user data into this fresh skb. If the copy fails, we undo the work and remove the fresh skb. Unfortunately, this undo lacks the change done to tp->highest_sack and we can leave a dangling pointer (to a freed skb) Later, tcp_xmit_retransmit_queue() can dereference this pointer and access freed memory. For regular kernels where memory is not unmapped, this might cause SACK bugs because tcp_highest_sack_seq() is buggy, returning garbage instead of tp->snd_nxt, but with various debug features like CONFIG_DEBUG_PAGEALLOC, this can crash the kernel. This bug was found by Marco Grassi thanks to syzkaller. Fixes: 6859d494 ("[TCP]: Abstract tp->highest_sack accessing & point to next skb") Reported-by:
Marco Grassi <marco.gra@gmail.com> Signed-off-by:
-
Vegard Nossum authored
[ Upstream commit d2fbdf76 ] tipc_msg_create() can return a NULL skb and if so, we shouldn't try to call tipc_node_xmit_skb() on it. general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 3 PID: 30298 Comm: trinity-c0 Not tainted 4.7.0-rc7+ #19 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014 task: ffff8800baf09980 ti: ffff8800595b8000 task.ti: ffff8800595b8000 RIP: 0010:[<ffffffff830bb46b>] [<ffffffff830bb46b>] tipc_node_xmit_skb+0x6b/0x140 RSP: 0018:ffff8800595bfce8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 000000003023b0e0 RDX: 0000000000000000 RSI: dffffc0000000000 RDI: ffffffff83d12580 RBP: ffff8800595bfd78 R08: ffffed000b2b7f32 R09: 0000000000000000 R10: fffffbfff0759725 R11: 0000000000000000 R12: 1ffff1000b2b7f9f R13: ffff8800595bfd58 R14: ffffffff83d12580 R15: dffffc0000000000 FS: 00007fcdde242700(0000) GS:ffff88011af80000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fcddde1db10 CR3: 000000006874b000 CR4: 00000000000006e0 DR0: 00007fcdde248000 DR1: 00007fcddd73d000 DR2: 00007fcdde248000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000090602 Stack: 0000000000000018 0000000000000018 0000000041b58ab3 ffffffff83954208 ffffffff830bb400 ffff8800595bfd30 ffffffff8309d767 0000000000000018 0000000000000018 ffff8800595bfd78 ffffffff8309da1a 00000000810ee611 Call Trace: [<ffffffff830c84a3>] tipc_shutdown+0x553/0x880 [<ffffffff825b4a3b>] SyS_shutdown+0x14b/0x170 [<ffffffff8100334c>] do_syscall_64+0x19c/0x410 [<ffffffff83295ca5>] entry_SYSCALL64_slow_path+0x25/0x25 Code: 90 00 b4 0b 83 c7 00 f1 f1 f1 f1 4c 8d 6d e0 c7 40 04 00 00 00 f4 c7 40 08 f3 f3 f3 f3 48 89 d8 48 c1 e8 03 c7 45 b4 00 00 00 00 <80> 3c 30 00 75 78 48 8d 7b 08 49 8d 75 c0 48 b8 00 00 00 00 00 RIP [<ffffffff830bb46b>] tipc_node_xmit_skb+0x6b/0x140 RSP <ffff8800595bfce8> ---[ end trace 57b0484e351e71f1 ]--- I feel like we should maybe return -ENOMEM or -ENOBUFS, but I'm not sure userspace is equipped to handle that. Anyway, this is better than a GPF and looks somewhat consistent with other tipc_msg_create() callers. Signed-off-by:
Vegard Nossum <vegard.nossum@oracle.com> Acked-by:
Ying Xue <ying.xue@windriver.com> Acked-by:
Jon Maloy <jon.maloy@ericsson.com> Signed-off-by:
-
Mike Manning authored
[ Upstream commit bc561632 ] If IPv6 is disabled when the option is set to keep IPv6 addresses on link down, userspace is unaware of this as there is no such indication via netlink. The solution is to remove the IPv6 addresses in this case, which results in netlink messages indicating removal of addresses in the usual manner. This fix also makes the behavior consistent with the case of having IPv6 disabled first, which stops IPv6 addresses from being added. Fixes: f1705ec1 ("net: ipv6: Make address flushing on ifdown optional") Signed-off-by:
Mike Manning <mmanning@brocade.com> Acked-by:
David Ahern <dsa@cumulusnetworks.com> Signed-off-by:
-
Vegard Nossum authored
[ Upstream commit 54236ab0 ] sctp_transport_seq_start() does not currently clear iter->start_fail on success, but relies on it being zero when it is allocated (by seq_open_net()). This can be a problem in the following sequence: open() // allocates iter (and implicitly sets iter->start_fail = 0) read() - iter->start() // fails and sets iter->start_fail = 1 - iter->stop() // doesn't call sctp_transport_walk_stop() (correct) read() again - iter->start() // succeeds, but doesn't change iter->start_fail - iter->stop() // doesn't call sctp_transport_walk_stop() (wrong) We should initialize sctp_ht_iter::start_fail to zero if ->start() succeeds, otherwise it's possible that we leave an old value of 1 there, which will cause ->stop() to not call sctp_transport_walk_stop(), which causes all sorts of problems like not calling rcu_read_unlock() (and preempt_enable()), eventually leading to more warnings like this: BUG: sleeping function called from invalid context at mm/slab.h:388 in_atomic(): 0, irqs_disabled(): 0, pid: 16551, name: trinity-c2 Preemption disabled at:[<ffffffff819bceb6>] rhashtable_walk_start+0x46/0x150 [<ffffffff81149abb>] preempt_count_add+0x1fb/0x280 [<ffffffff83295892>] _raw_spin_lock+0x12/0x40 [<ffffffff819bceb6>] rhashtable_walk_start+0x46/0x150 [<ffffffff82ec665f>] sctp_transport_walk_start+0x2f/0x60 [<ffffffff82edda1d>] sctp_transport_seq_start+0x4d/0x150 [<ffffffff81439e50>] traverse+0x170/0x850 [<ffffffff8143aeec>] seq_read+0x7cc/0x1180 [<ffffffff814f996c>] proc_reg_read+0xbc/0x180 [<ffffffff813d0384>] do_loop_readv_writev+0x134/0x210 [<ffffffff813d2a95>] do_readv_writev+0x565/0x660 [<ffffffff813d6857>] vfs_readv+0x67/0xa0 [<ffffffff813d6c16>] do_preadv+0x126/0x170 [<ffffffff813d710c>] SyS_preadv+0xc/0x10 [<ffffffff8100334c>] do_syscall_64+0x19c/0x410 [<ffffffff83296225>] return_from_SYSCALL_64+0x0/0x6a [<ffffffffffffffff>] 0xffffffffffffffff Notice that this is a subtly different stacktrace from the one in commit 5fc382d8 ("net/sctp: terminate rhashtable walk correctly"). Cc: Xin Long <lucien.xin@gmail.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by:
Neil Horman <nhorman@tuxdriver.com> Acked-by:
-
Vegard Nossum authored
[ Upstream commit 5ba092ef ] If iriap_register_lsap() fails to allocate memory, self->lsap is set to NULL. However, none of the callers handle the failure and irlmp_connect_request() will happily dereference it: iriap_register_lsap: Unable to allocated LSAP! ================================================================================ UBSAN: Undefined behaviour in net/irda/irlmp.c:378:2 member access within null pointer of type 'struct lsap_cb' CPU: 1 PID: 15403 Comm: trinity-c0 Not tainted 4.8.0-rc1+ #81 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.9.3-0-ge2fc41e-prebuilt.qemu-project.org 04/01/2014 0000000000000000 ffff88010c7e78a8 ffffffff82344f40 0000000041b58ab3 ffffffff84f98000 ffffffff82344e94 ffff88010c7e78d0 ffff88010c7e7880 ffff88010630ad00 ffffffff84a5fae0 ffffffff84d3f5c0 000000000000017a Call Trace: [<ffffffff82344f40>] dump_stack+0xac/0xfc [<ffffffff8242f5a8>] ubsan_epilogue+0xd/0x8a [<ffffffff824302bf>] __ubsan_handle_type_mismatch+0x157/0x411 [<ffffffff83b7bdbc>] irlmp_connect_request+0x7ac/0x970 [<ffffffff83b77cc0>] iriap_connect_request+0xa0/0x160 [<ffffffff83b77f48>] state_s_disconnect+0x88/0xd0 [<ffffffff83b78904>] iriap_do_client_event+0x94/0x120 [<ffffffff83b77710>] iriap_getvaluebyclass_request+0x3e0/0x6d0 [<ffffffff83ba6ebb>] irda_find_lsap_sel+0x1eb/0x630 [<ffffffff83ba90c8>] irda_connect+0x828/0x12d0 [<ffffffff833c0dfb>] SYSC_connect+0x22b/0x340 [<ffffffff833c7e09>] SyS_connect+0x9/0x10 [<ffffffff81007bd3>] do_syscall_64+0x1b3/0x4b0 [<ffffffff845f946a>] entry_SYSCALL64_slow_path+0x25/0x25 ================================================================================ The bug seems to have been around since forever. There's more problems with missing error checks in iriap_init() (and indeed all of irda_init()), but that's a bigger problem that needs very careful review and testing. This patch will fix the most serious bug (as it's easily reached from unprivileged userspace). I have tested my patch with a reproducer. Signed-off-by:
-
Daniel Borkmann authored
[ Upstream commit 0ed661d5 ] Fix the bpf_try_make_writable() helper and all call sites we have in BPF, it's currently defect with regards to skbs when the write_len spans into non-linear parts, no matter if cloned or not. There are multiple issues at once. First, using skb_store_bits() is not correct since even if we have a cloned skb, page frags can still be shared. To really make them private, we need to pull them in via __pskb_pull_tail() first, which also gets us a private head via pskb_expand_head() implicitly. This is for helpers like bpf_skb_store_bytes(), bpf_l3_csum_replace(), bpf_l4_csum_replace(). Really, the only thing reasonable and working here is to call skb_ensure_writable() before any write operation. Meaning, via pskb_may_pull() it makes sure that parts we want to access are pulled in and if not does so plus unclones the skb implicitly. If our write_len still fits the headlen and we're cloned and our header of the clone is not writable, then we need to make a private copy via pskb_expand_head(). skb_store_bits() is a bit misleading and only safe to store into non-linear data in different contexts such as 357b40a1 ("[IPV6]: IPV6_CHECKSUM socket option can corrupt kernel memory"). For above BPF helper functions, it means after fixed bpf_try_make_writable(), we've pulled in enough, so that we operate always based on skb->data. Thus, the call to skb_header_pointer() and skb_store_bits() becomes superfluous. In bpf_skb_store_bytes(), the len check is unnecessary too since it can only pass in maximum of BPF stack size, so adding offset is guaranteed to never overflow. Also bpf_l3/4_csum_replace() helpers must test for proper offset alignment since they use __sum16 pointer for writing resulting csum. The remaining helpers that change skb data not discussed here yet are bpf_skb_vlan_push(), bpf_skb_vlan_pop() and bpf_skb_change_proto(). The vlan helpers internally call either skb_ensure_writable() (pop case) and skb_cow_head() (push case, for head expansion), respectively. Similarly, bpf_skb_proto_xlat() takes care to not mangle page frags. Fixes: 608cd71a ("tc: bpf: generalize pedit action") Fixes: 91bc4822 ("tc: bpf: add checksum helpers") Fixes: 3697649f ("bpf: try harder on clones when writing into skb") Signed-off-by:
Daniel Borkmann <daniel@iogearbox.net> Acked-by:
Alexei Starovoitov <ast@kernel.org> Signed-off-by:
-
Lance Richardson authored
[ Upstream commit a5d0dc81 ] When executing the script included below, the netns delete operation hangs with the following message (repeated at 10 second intervals): kernel:unregister_netdevice: waiting for lo to become free. Usage count = 1 This occurs because a reference to the lo interface in the "secure" netns is still held by a dst entry in the xfrm bundle cache in the init netns. Address this problem by garbage collecting the tunnel netns flow cache when a cross-namespace vti interface receives a NETDEV_DOWN notification. A more detailed description of the problem scenario (referencing commands in the script below): (1) ip link add vti_test type vti local 1.1.1.1 remote 1.1.1.2 key 1 The vti_test interface is created in the init namespace. vti_tunnel_init() attaches a struct ip_tunnel to the vti interface's netdev_priv(dev), setting the tunnel net to &init_net. (2) ip link set vti_test netns secure The vti_test interface is moved to the "secure" netns. Note that the associated struct ip_tunnel still has tunnel->net set to &init_net. (3) ip netns exec secure ping -c 4 -i 0.02 -I 192.168.100.1 192.168.200.1 The first packet sent using the vti device causes xfrm_lookup() to be called as follows: dst = xfrm_lookup(tunnel->net, skb_dst(skb), fl, NULL, 0); Note that tunnel->net is the init namespace, while skb_dst(skb) references the vti_test interface in the "secure" namespace. The returned dst references an interface in the init namespace. Also note that the first parameter to xfrm_lookup() determines which flow cache is used to store the computed xfrm bundle, so after xfrm_lookup() returns there will be a cached bundle in the init namespace flow cache with a dst referencing a device in the "secure" namespace. (4) ip netns del secure Kernel begins to delete the "secure" namespace. At some point the vti_test interface is deleted, at which point dst_ifdown() changes the dst->dev in the cached xfrm bundle flow from vti_test to lo (still in the "secure" namespace however). Since nothing has happened to cause the init namespace's flow cache to be garbage collected, this dst remains attached to the flow cache, so the kernel loops waiting for the last reference to lo to go away. <Begin script> ip link add br1 type bridge ip link set dev br1 up ip addr add dev br1 1.1.1.1/8 ip netns add secure ip link add vti_test type vti local 1.1.1.1 remote 1.1.1.2 key 1 ip link set vti_test netns secure ip netns exec secure ip link set vti_test up ip netns exec secure ip link s lo up ip netns exec secure ip addr add dev lo 192.168.100.1/24 ip netns exec secure ip route add 192.168.200.0/24 dev vti_test ip xfrm policy flush ip xfrm state flush ip xfrm policy add dir out tmpl src 1.1.1.1 dst 1.1.1.2 \ proto esp mode tunnel mark 1 ip xfrm policy add dir in tmpl src 1.1.1.2 dst 1.1.1.1 \ proto esp mode tunnel mark 1 ip xfrm state add src 1.1.1.1 dst 1.1.1.2 proto esp spi 1 \ mode tunnel enc des3_ede 0x112233445566778811223344556677881122334455667788 ip xfrm state add src 1.1.1.2 dst 1.1.1.1 proto esp spi 1 \ mode tunnel enc des3_ede 0x112233445566778811223344556677881122334455667788 ip netns exec secure ping -c 4 -i 0.02 -I 192.168.100.1 192.168.200.1 ip netns del secure <End script> Reported-by:
Hangbin Liu <haliu@redhat.com> Reported-by:
Jan Tluka <jtluka@redhat.com> Signed-off-by:
-
David Forster authored
[ Upstream commit 94d9f1c5 ] Panic occurs when issuing "cat /proc/net/route" whilst populating FIB with > 1M routes. Use of cached node pointer in fib_route_get_idx is unsafe. BUG: unable to handle kernel paging request at ffffc90001630024 IP: [<ffffffff814cf6a0>] leaf_walk_rcu+0x10/0xe0 PGD 11b08d067 PUD 11b08e067 PMD dac4b067 PTE 0 Oops: 0000 [#1] SMP Modules linked in: nfsd auth_rpcgss oid_registry nfs_acl nfs lockd grace fscac snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep virti acpi_cpufreq button parport_pc ppdev lp parport autofs4 ext4 crc16 mbcache jbd tio_ring virtio floppy uhci_hcd ehci_hcd usbcore usb_common libata scsi_mod CPU: 1 PID: 785 Comm: cat Not tainted 4.2.0-rc8+ #4 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2007 task: ffff8800da1c0bc0 ti: ffff88011a05c000 task.ti: ffff88011a05c000 RIP: 0010:[<ffffffff814cf6a0>] [<ffffffff814cf6a0>] leaf_walk_rcu+0x10/0xe0 RSP: 0018:ffff88011a05fda0 EFLAGS: 00010202 RAX: ffff8800d8a40c00 RBX: ffff8800da4af940 RCX: ffff88011a05ff20 RDX: ffffc90001630020 RSI: 0000000001013531 RDI: ffff8800da4af950 RBP: 0000000000000000 R08: ffff8800da1f9a00 R09: 0000000000000000 R10: ffff8800db45b7e4 R11: 0000000000000246 R12: ffff8800da4af950 R13: ffff8800d97a74c0 R14: 0000000000000000 R15: ffff8800d97a7480 FS: 00007fd3970e0700(0000) GS:ffff88011fd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: ffffc90001630024 CR3: 000000011a7e4000 CR4: 00000000000006e0 Stack: ffffffff814d00d3 0000000000000000 ffff88011a05ff20 ffff8800da1f9a00 ffffffff811dd8b9 0000000000000800 0000000000020000 00007fd396f35000 ffffffff811f8714 0000000000003431 ffffffff8138dce0 0000000000000f80 Call Trace: [<ffffffff814d00d3>] ? fib_route_seq_start+0x93/0xc0 [<ffffffff811dd8b9>] ? seq_read+0x149/0x380 [<ffffffff811f8714>] ? fsnotify+0x3b4/0x500 [<ffffffff8138dce0>] ? process_echoes+0x70/0x70 [<ffffffff8121cfa7>] ? proc_reg_read+0x47/0x70 [<ffffffff811bb823>] ? __vfs_read+0x23/0xd0 [<ffffffff811bbd42>] ? rw_verify_area+0x52/0xf0 [<ffffffff811bbe61>] ? vfs_read+0x81/0x120 [<ffffffff811bcbc2>] ? SyS_read+0x42/0xa0 [<ffffffff81549ab2>] ? entry_SYSCALL_64_fastpath+0x16/0x75 Code: 48 85 c0 75 d8 f3 c3 31 c0 c3 f3 c3 66 66 66 66 66 66 2e 0f 1f 84 00 00 a 04 89 f0 33 02 44 89 c9 48 d3 e8 0f b6 4a 05 49 89 RIP [<ffffffff814cf6a0>] leaf_walk_rcu+0x10/0xe0 RSP <ffff88011a05fda0> CR2: ffffc90001630024 Signed-off-by:
Dave Forster <dforster@brocade.com> Acked-by:
Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by:
-
Jakub Kicinski authored
[ Upstream commit 1f415a74 ] Using per-register incrementing ID can lead to find_good_pkt_pointers() confusing registers which have completely different values. Consider example: 0: (bf) r6 = r1 1: (61) r8 = *(u32 *)(r6 +76) 2: (61) r0 = *(u32 *)(r6 +80) 3: (bf) r7 = r8 4: (07) r8 += 32 5: (2d) if r8 > r0 goto pc+9 R0=pkt_end R1=ctx R6=ctx R7=pkt(id=0,off=0,r=32) R8=pkt(id=0,off=32,r=32) R10=fp 6: (bf) r8 = r7 7: (bf) r9 = r7 8: (71) r1 = *(u8 *)(r7 +0) 9: (0f) r8 += r1 10: (71) r1 = *(u8 *)(r7 +1) 11: (0f) r9 += r1 12: (07) r8 += 32 13: (2d) if r8 > r0 goto pc+1 R0=pkt_end R1=inv56 R6=ctx R7=pkt(id=0,off=0,r=32) R8=pkt(id=1,off=32,r=32) R9=pkt(id=1,off=0,r=32) R10=fp 14: (71) r1 = *(u8 *)(r9 +16) 15: (b7) r7 = 0 16: (bf) r0 = r7 17: (95) exit We need to get a UNKNOWN_VALUE with imm to force id generation so lines 0-5 make r7 a valid packet pointer. We then read two different bytes from the packet and add them to copies of the constructed packet pointer. r8 (line 9) and r9 (line 11) will get the same id of 1, independently. When either of them is validated (line 13) - find_good_pkt_pointers() will also mark the other as safe. This leads to access on line 14 being mistakenly considered safe. Fixes: 969bf05e ("bpf: direct packet access") Signed-off-by:
Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by:
-
Rob Clark authored
commit d78d383a upstream. An evil userspace could try to cause deadlock by passing an unfaulted-in GEM bo as submit->bos (or submit->cmds) table. Which will trigger msm_gem_fault() while we already hold struct_mutex. See: https://github.com/freedreno/msmtest/blob/master/evilsubmittest.cSigned-off-by:
Rob Clark <robdclark@gmail.com> Signed-off-by:
-
