- 24 Jun, 2016 15 commits
-
-
Yuchung Cheng authored
[ Upstream commit ce3cf4ec ] The v6 tcp stats scan do not provide TLP and ER timer information correctly like the v4 version . This patch fixes that. Fixes: 6ba8a3b1 ("tcp: Tail loss probe (TLP)") Fixes: eed530b6 ("tcp: early retransmit") Signed-off-by:
Yuchung Cheng <ycheng@google.com> Signed-off-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Chen Haiquan authored
[ Upstream commit ce577668 ] When create a new vxlan link, example: ip link add vtap mtu 1440 type vxlan vni 1 dev eth0 The argument "mtu" has no effect, because it is not set to conf->mtu. The default value is used in vxlan_dev_configure function. This problem was introduced by commit 0dfbdf41 (vxlan: Factor out device configuration). Fixes: 0dfbdf41 (vxlan: Factor out device configuration) Signed-off-by:
Chen Haiquan <oc@yunify.com> Acked-by:
Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by:
-
Ivan Vecera authored
[ Upstream commit f6988cb6 ] The team_device_event() notifier calls team_compute_features() to fix vlan_features under team->lock to protect team->port_list. The problem is that subsequent __team_compute_features() calls netdev_change_features() to propagate vlan_features to upper vlan devices while team->lock is still taken. This can lead to deadlock when NETIF_F_LRO is modified on lower devices or team device itself. Example: The team0 as active backup with eth0 and eth1 NICs. Both eth0 & eth1 are LRO capable and LRO is enabled. Thus LRO is also enabled on team0. The command 'ethtool -K team0 lro off' now hangs due to this deadlock: dev_ethtool() -> ethtool_set_features() -> __netdev_update_features(team) -> netdev_sync_lower_features() -> netdev_update_features(lower_1) -> __netdev_update_features(lower_1) -> netdev_features_change(lower_1) -> call_netdevice_notifiers(...) -> team_device_event(lower_1) -> team_compute_features(team) [TAKES team->lock] -> netdev_change_features(team) -> __netdev_update_features(team) -> netdev_sync_lower_features() -> netdev_update_features(lower_2) -> __netdev_update_features(lower_2) -> netdev_features_change(lower_2) -> call_netdevice_notifiers(...) -> team_device_event(lower_2) -> team_compute_features(team) [DEADLOCK] The bug is present in team from the beginning but it appeared after the commit fd867d51 (net/core: generic support for disabling netdev features down stack) that adds synchronization of features with lower devices. Fixes: fd867d51 (net/core: generic support for disabling netdev features down stack) Cc: Jiri Pirko <jiri@resnulli.us> Signed-off-by:
Ivan Vecera <ivecera@redhat.com> Signed-off-by:
Jiri Pirko <jiri@mellanox.com> Signed-off-by:
-
Edward Cree authored
[ Upstream commit c0795bf6 ] Otherwise, if we fail to allocate new PIO buffers, our TXQs will try to use the old ones, which aren't there any more. Fixes: 183233be "sfc: Allocate and link PIO buffers; map them with write-combining" Signed-off-by:
Edward Cree <ecree@solarflare.com> Signed-off-by:
-
Daniel Borkmann authored
[ Upstream commit 612bacad ] Follow-up to commit e27f4a94 ("bpf: Use mount_nodev not mount_ns to mount the bpf filesystem"), which removes the FS_USERNS_MOUNT flag. The original idea was to have a per mountns instance instead of a single global fs instance, but that didn't work out and we had to switch to mount_nodev() model. The intent of that middle ground was that we avoid users who don't play nice to create endless instances of bpf fs which are difficult to control and discover from an admin point of view, but at the same time it would have allowed us to be more flexible with regard to namespaces. Therefore, since we now did the switch to mount_nodev() as a fix where individual instances are created, we also need to remove userns mount flag along with it to avoid running into mentioned situation. I don't expect any breakage at this early point in time with removing the flag and we can revisit this later should the requirement for this come up with future users. This and commit e27f4a94 have been split to facilitate tracking should any of them run into the unlikely case of causing a regression. Fixes: b2197755 ("bpf: add support for persistent maps/progs") Signed-off-by:
Daniel Borkmann <daniel@iogearbox.net> Acked-by:
Hannes Frederic Sowa <hannes@stressinduktion.org> Acked-by:
Alexei Starovoitov <ast@kernel.org> Signed-off-by:
-
Nicolas Dichtel authored
[ Upstream commit f0a3fdca ] These structures are defined only if __USE_MISC is set in glibc net/if.h headers, ie when _BSD_SOURCE or _SVID_SOURCE are defined. CC: Jan Engelhardt <jengelh@inai.de> CC: Josh Boyer <jwboyer@fedoraproject.org> CC: Stephen Hemminger <shemming@brocade.com> CC: Waldemar Brodkorb <mail@waldemar-brodkorb.de> CC: Gabriel Laskar <gabriel@lse.epita.fr> CC: Mikko Rapeli <mikko.rapeli@iki.fi> Fixes: 4a91cb61 ("uapi glibc compat: fix compile errors when glibc net/if.h included before linux/if.h") Signed-off-by:
Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by:
-
Hannes Frederic Sowa authored
[ Upstream commit e5aed006 ] In case we find a socket with encapsulation enabled we should call the encap_recv function even if just a udp header without payload is available. The callbacks are responsible for correctly verifying and dropping the packets. Also, in case the header validation fails for geneve and vxlan we shouldn't put the skb back into the socket queue, no one will pick them up there. Instead we can simply discard them in the respective encap_recv functions. Signed-off-by:
-
Eric W. Biederman authored
[ Upstream commit e27f4a94 ] While reviewing the filesystems that set FS_USERNS_MOUNT I spotted the bpf filesystem. Looking at the code I saw a broken usage of mount_ns with current->nsproxy->mnt_ns. As the code does not acquire a reference to the mount namespace it can not possibly be correct to store the mount namespace on the superblock as it does. Replace mount_ns with mount_nodev so that each mount of the bpf filesystem returns a distinct instance, and the code is not buggy. In discussion with Hannes Frederic Sowa it was reported that the use of mount_ns was an attempt to have one bpf instance per mount namespace, in an attempt to keep resources that pin resources from hiding. That intent simply does not work, the vfs is not built to allow that kind of behavior. Which means that the bpf filesystem really is buggy both semantically and in it's implemenation as it does not nor can it implement the original intent. This change is userspace visible, but my experience with similar filesystems leads me to believe nothing will break with a model of each mount of the bpf filesystem is distinct from all others. Fixes: b2197755 ("bpf: add support for persistent maps/progs") Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Acked-by:
"Eric W. Biederman" <ebiederm@xmission.com> Acked-by:
-
Jason Wang authored
[ Upstream commit addf8fc4 ] We used to check dev->reg_state against NETREG_REGISTERED after each time we are woke up. But after commit 9e641bdc ("net-tun: restructure tun_do_read for better sleep/wakeup efficiency"), it uses skb_recv_datagram() which does not check dev->reg_state. This will result if we delete a tun/tap device after a process is blocked in the reading. The device will wait for the reference count which was held by that process for ever. Fixes this by using RCV_SHUTDOWN which will be checked during sk_recv_datagram() before trying to wake up the process during uninit. Fixes: 9e641bdc ("net-tun: restructure tun_do_read for better sleep/wakeup efficiency") Cc: Eric Dumazet <edumazet@google.com> Cc: Xi Wang <xii@google.com> Cc: Michael S. Tsirkin <mst@redhat.com> Signed-off-by:
Jason Wang <jasowang@redhat.com> Acked-by:
Eric Dumazet <edumazet@google.com> Acked-by:
Michael S. Tsirkin <mst@redhat.com> Signed-off-by:
-
Jiri Pirko authored
[ Upstream commit da4ed551 ] The problem is that fib_info->nh is [0] so the struct fib_info allocation size depends on number of nexthops. If we just copy fib_info, we do not copy the nexthops info and driver accesses memory which is not ours. Given the fact that fib4 does not defer operations and therefore it does not need copy, just pass the pointer down to drivers as it was done before. Fixes: 850d0cbc ("switchdev: remove pointers from switchdev objects") Signed-off-by:
-
Richard Alpe authored
[ Upstream commit 03aaaa9b ] The publication field of the old netlink API should contain the publication key and not the publication reference. Fixes: 44a8ae94 (tipc: convert legacy nl name table dump to nl compat) Signed-off-by:
Richard Alpe <richard.alpe@ericsson.com> Acked-by:
Jon Maloy <jon.maloy@ericsson.com> Signed-off-by:
-
Herbert Xu authored
[ Upstream commit 92964c79 ] When we free cb->skb after a dump, we do it after releasing the lock. This means that a new dump could have started in the time being and we'll end up freeing their skb instead of ours. This patch saves the skb and module before we unlock so we free the right memory. Fixes: 16b304f3 ("netlink: Eliminate kmalloc in netlink dump operation.") Reported-by:
Baozeng Ding <sploving1@gmail.com> Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Acked-by:
-
Richard Alpe authored
[ Upstream commit 45e093ae ] Make sure the socket for which the user is listing publication exists before parsing the socket netlink attributes. Prior to this patch a call without any socket caused a NULL pointer dereference in tipc_nl_publ_dump(). Tested-and-reported-by:
Jon Maloy <jon.maloy@ericsson.cm> Signed-off-by:
-
Ewan D. Milne authored
commit fbd83006 upstream. Linux fails to boot as a guest with a QEMU CD-ROM: [ 4.439488] ata2.00: ATAPI: QEMU CD-ROM, 0.8.2, max UDMA/100 [ 4.443649] ata2.00: configured for MWDMA2 [ 4.450267] scsi 1:0:0:0: CD-ROM QEMU QEMU CD-ROM 0.8. PQ: 0 ANSI: 5 [ 4.464317] ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen [ 4.464319] ata2.00: BMDMA stat 0x5 [ 4.464339] ata2.00: cmd a0/01:00:00:00:01/00:00:00:00:00/a0 tag 0 dma 16640 in [ 4.464339] Inquiry 12 01 00 00 ff 00res 48/20:02:00:24:00/00:00:00:00:00/a0 Emask 0x2 (HSM violation) [ 4.464341] ata2.00: status: { DRDY DRQ } [ 4.465864] ata2: soft resetting link [ 4.625971] ata2.00: configured for MWDMA2 [ 4.628290] ata2: EH complete [ 4.646670] ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen [ 4.646671] ata2.00: BMDMA stat 0x5 [ 4.646683] ata2.00: cmd a0/01:00:00:00:01/00:00:00:00:00/a0 tag 0 dma 16640 in [ 4.646683] Inquiry 12 01 00 00 ff 00res 48/20:02:00:24:00/00:00:00:00:00/a0 Emask 0x2 (HSM violation) [ 4.646685] ata2.00: status: { DRDY DRQ } [ 4.648193] ata2: soft resetting link ... Fix this by suppressing VPD inquiry for this device. Signed-off-by:
Ewan D. Milne <emilne@redhat.com> Reported-by:
Jan Stancek <jstancek@redhat.com> Tested-by:
Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by:
-
James Bottomley authored
commit a621bac3 upstream. When SCSI was written, all commands coming from the filesystem (REQ_TYPE_FS commands) had data. This meant that our signal for needing to complete the command was the number of bytes completed being equal to the number of bytes in the request. Unfortunately, with the advent of flush barriers, we can now get zero length REQ_TYPE_FS commands, which confuse this logic because they satisfy the condition every time. This means they never get retried even for retryable conditions, like UNIT ATTENTION because we complete them early assuming they're done. Fix this by special casing the early completion condition to recognise zero length commands with errors and let them drop through to the retry code. Reported-by:
Sebastian Parschauer <s.parschauer@gmx.de> Signed-off-by:
James E.J. Bottomley <jejb@linux.vnet.ibm.com> Tested-by:
Jack Wang <jinpu.wang@profitbricks.com> Signed-off-by:
-
- 08 Jun, 2016 25 commits
-
-
Greg Kroah-Hartman authored
-
Dave Chinner authored
commit 7d6a13f0 upstream. When we do dquot readahead in log recovery, we do not use a verifier as the underlying buffer may not have dquots in it. e.g. the allocation operation hasn't yet been replayed. Hence we do not want to fail recovery because we detect an operation to be replayed has not been run yet. This problem was addressed for inodes in commit d8914002 ("xfs: inode buffers may not be valid during recovery readahead") but the problem was not recognised to exist for dquots and their buffers as the dquot readahead did not have a verifier. The result of not using a verifier is that when the buffer is then next read to replay a dquot modification, the dquot buffer verifier will only be attached to the buffer if *readahead is not complete*. Hence we can read the buffer, replay the dquot changes and then add it to the delwri submission list without it having a verifier attached to it. This then generates warnings in xfs_buf_ioapply(), which catches and warns about this case. Fix this and make it handle the same readahead verifier error cases as for inode buffers by adding a new readahead verifier that has a write operation as well as a read operation that marks the buffer as not done if any corruption is detected. Also make sure we don't run readahead if the dquot buffer has been marked as cancelled by recovery. This will result in readahead either succeeding and the buffer having a valid write verifier, or readahead failing and the buffer state requiring the subsequent read to resubmit the IO with the new verifier. In either case, this will result in the buffer always ending up with a valid write verifier on it. Note: we also need to fix the inode buffer readahead error handling to mark the buffer with EIO. Brian noticed the code I copied from there wrong during review, so fix it at the same time. Add comments linking the two functions that handle readahead verifier errors together so we don't forget this behavioural link in future. Signed-off-by:
Dave Chinner <dchinner@redhat.com> Reviewed-by:
Brian Foster <bfoster@redhat.com> Signed-off-by:
Dave Chinner <david@fromorbit.com> Signed-off-by:
-
Eric Sandeen authored
commit 233135b7 upstream. This adds a name to each buf_ops structure, so that if a verifier fails we can print the type of verifier that failed it. Should be a slight debugging aid, I hope. Signed-off-by:
Eric Sandeen <sandeen@redhat.com> Reviewed-by:
-
Dave Chinner authored
commit 7d3aa7fe upstream. We don't write back stale inodes so we should skip them in xfs_iflush_cluster, too. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Dave Chinner authored
commit 51b07f30 upstream. Some careless idiot(*) wrote crap code in commit 1a3e8f3d ("xfs: convert inode cache lookups to use RCU locking") back in late 2010, and so xfs_iflush_cluster checks the wrong inode for whether it is still valid under RCU protection. Fix it to lock and check the correct inode. (*) Careless-idiot: Dave Chinner <dchinner@redhat.com> Discovered-by:
-
Dave Chinner authored
commit b1438f47 upstream. When a failure due to an inode buffer occurs, the error handling fails to abort the inode writeback correctly. This can result in the inode being reclaimed whilst still in the AIL, leading to use-after-free situations as well as filesystems that cannot be unmounted as the inode log items left in the AIL never get removed. Fix this by ensuring fatal errors from xfs_imap_to_bp() result in the inode flush being aborted correctly. Reported-by:
Shyam Kaushik <shyam@zadarastorage.com> Diagnosed-by:
-
Dave Chinner authored
commit ad747e3b upstream. Commit 96f859d5 ("libxfs: pack the agfl header structure so XFS_AGFL_SIZE is correct") allowed the freelist to use the empty slot at the end of the freelist on 64 bit systems that was not being used due to sizeof() rounding up the structure size. This has caused versions of xfs_repair prior to 4.5.0 (which also has the fix) to report this as a corruption once the filesystem has been grown. Older kernels can also have problems (seen from a whacky container/vm management environment) mounting filesystems grown on a system with a newer kernel than the vm/container it is deployed on. To avoid this problem, change the initial free list indexes not to wrap across the end of the AGFL, hence avoiding the initialisation of agf_fllast to the last index in the AGFL. Signed-off-by:
Carlos Maiolino <cmaiolino@redhat.com> Signed-off-by:
-
Eric Sandeen authored
commit d0a58e83 upstream. Today, a kernel which refuses to mount a filesystem read-write due to unknown ro-compat features can still transition to read-write via the remount path. The old kernel is most likely none the wiser, because it's unaware of the new feature, and isn't using it. However, writing to the filesystem may well corrupt metadata related to that new feature, and moving to a newer kernel which understand the feature will have problems. Right now the only ro-compat feature we have is the free inode btree, which showed up in v3.16. It would be good to push this back to all the active stable kernels, I think, so that if anyone is using newer mkfs (which enables the finobt feature) with older kernel releases, they'll be protected. Signed-off-by:
Bill O'Donnell <billodo@redhat.com> Reviewed-by:
-
Arnd Bergmann authored
commit c87bf431 upstream. Enabling CONFIG_GCOV_PROFILE_ALL produces us a lot of warnings like lib/lz4/lz4hc_compress.c: In function 'lz4_compresshcctx': lib/lz4/lz4hc_compress.c:514:1: warning: the frame size of 1504 bytes is larger than 1024 bytes [-Wframe-larger-than=] After some investigation, I found that this behavior started with gcc-4.9, and opened https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69702. A suggested workaround for it is to use the -fno-tree-loop-im flag that turns off one of the optimization stages in gcc, so the code runs a little slower but does not use excessive amounts of stack. We could make this conditional on the gcc version, but I could not find an easy way to do this in Kbuild and the benefit would be fairly small, given that most of the gcc version in production are affected now. I'm marking this for 'stable' backports because it addresses a bug with code generation in gcc that exists in all kernel versions with the affected gcc releases. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Acked-by:
Peter Oberparleiter <oberpar@linux.vnet.ibm.com> Signed-off-by:
Michal Marek <mmarek@suse.com> Signed-off-by:
-
Srinivas Pandruvada authored
commit 65a9f31c upstream. After commit 21a59991 ("scripts/package/Makefile: rpmbuild is needed for rpm targets"), it is no longer possible to specify RPMOPTS. For example, we can no longer able to control _topdir using the following make command. make RPMOPTS="--define '_topdir /home/xyz/workspace/'" binrpm-pkg Fixes: 21a59991 ("scripts/package/Makefile: rpmbuild is needed for rpm targets") Signed-off-by:
Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> Signed-off-by:
-
Ville Syrjälä authored
commit 3017cd63 upstream. With netconsole (at least) the pr_err("... disablingn") call can recurse back into the dma-debug code, where it'll try to grab free_entries_lock again. Avoid the problem by doing the printk after dropping the lock. Link: http://lkml.kernel.org/r/1463678421-18683-1-git-send-email-ville.syrjala@linux.intel.comSigned-off-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Rafael J. Wysocki authored
commit 3a17fb32 upstream. Grygorii Strashko reports: The PM runtime will be left disabled for the device if its .suspend_late() callback fails and async suspend is not allowed for this device. In this case device will not be added in dpm_late_early_list and dpm_resume_early() will ignore this device, as result PM runtime will be disabled for it forever (side effect: after 8 subsequent failures for the same device the PM runtime will be reenabled due to disable_depth overflow). To fix this problem, add devices to dpm_late_early_list regardless of whether or not device_suspend_late() returns errors for them. That will ensure failures in there to be handled consistently for all devices regardless of their async suspend/resume status. Reported-by:
Grygorii Strashko <grygorii.strashko@ti.com> Tested-by:
Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by:
-
Nicolai Stange authored
commit 935244cd upstream. Currently, in ext4_mb_init(), there's a loop like the following: do { ... offset += 1 << (sb->s_blocksize_bits - i); i++; } while (i <= sb->s_blocksize_bits + 1); Note that the updated offset is used in the loop's next iteration only. However, at the last iteration, that is at i == sb->s_blocksize_bits + 1, the shift count becomes equal to (unsigned)-1 > 31 (c.f. C99 6.5.7(3)) and UBSAN reports UBSAN: Undefined behaviour in fs/ext4/mballoc.c:2621:15 shift exponent 4294967295 is too large for 32-bit type 'int' [...] Call Trace: [<ffffffff818c4d25>] dump_stack+0xbc/0x117 [<ffffffff818c4c69>] ? _atomic_dec_and_lock+0x169/0x169 [<ffffffff819411ab>] ubsan_epilogue+0xd/0x4e [<ffffffff81941cac>] __ubsan_handle_shift_out_of_bounds+0x1fb/0x254 [<ffffffff81941ab1>] ? __ubsan_handle_load_invalid_value+0x158/0x158 [<ffffffff814b6dc1>] ? kmem_cache_alloc+0x101/0x390 [<ffffffff816fc13b>] ? ext4_mb_init+0x13b/0xfd0 [<ffffffff814293c7>] ? create_cache+0x57/0x1f0 [<ffffffff8142948a>] ? create_cache+0x11a/0x1f0 [<ffffffff821c2168>] ? mutex_lock+0x38/0x60 [<ffffffff821c23ab>] ? mutex_unlock+0x1b/0x50 [<ffffffff814c26ab>] ? put_online_mems+0x5b/0xc0 [<ffffffff81429677>] ? kmem_cache_create+0x117/0x2c0 [<ffffffff816fcc49>] ext4_mb_init+0xc49/0xfd0 [...] Observe that the mentioned shift exponent, 4294967295, equals (unsigned)-1. Unless compilers start to do some fancy transformations (which at least GCC 6.0.0 doesn't currently do), the issue is of cosmetic nature only: the such calculated value of offset is never used again. Silence UBSAN by introducing another variable, offset_incr, holding the next increment to apply to offset and adjust that one by right shifting it by one position per loop iteration. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=114701 Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=112161Signed-off-by:
Nicolai Stange <nicstange@gmail.com> Signed-off-by:
Theodore Ts'o <tytso@mit.edu> Signed-off-by:
-
Nicolai Stange authored
commit b5cb316c upstream. Currently, in mb_find_order_for_block(), there's a loop like the following: while (order <= e4b->bd_blkbits + 1) { ... bb += 1 << (e4b->bd_blkbits - order); } Note that the updated bb is used in the loop's next iteration only. However, at the last iteration, that is at order == e4b->bd_blkbits + 1, the shift count becomes negative (c.f. C99 6.5.7(3)) and UBSAN reports UBSAN: Undefined behaviour in fs/ext4/mballoc.c:1281:11 shift exponent -1 is negative [...] Call Trace: [<ffffffff818c4d35>] dump_stack+0xbc/0x117 [<ffffffff818c4c79>] ? _atomic_dec_and_lock+0x169/0x169 [<ffffffff819411bb>] ubsan_epilogue+0xd/0x4e [<ffffffff81941cbc>] __ubsan_handle_shift_out_of_bounds+0x1fb/0x254 [<ffffffff81941ac1>] ? __ubsan_handle_load_invalid_value+0x158/0x158 [<ffffffff816e93a0>] ? ext4_mb_generate_from_pa+0x590/0x590 [<ffffffff816502c8>] ? ext4_read_block_bitmap_nowait+0x598/0xe80 [<ffffffff816e7b7e>] mb_find_order_for_block+0x1ce/0x240 [...] Unless compilers start to do some fancy transformations (which at least GCC 6.0.0 doesn't currently do), the issue is of cosmetic nature only: the such calculated value of bb is never used again. Silence UBSAN by introducing another variable, bb_incr, holding the next increment to apply to bb and adjust that one by right shifting it by one position per loop iteration. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=114701 Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=112161Signed-off-by:
-
Jan Kara authored
commit 74177f55 upstream. When filesystem is corrupted in the right way, it can happen ext4_mark_iloc_dirty() in ext4_orphan_add() returns error and we subsequently remove inode from the in-memory orphan list. However this deletion is done with list_del(&EXT4_I(inode)->i_orphan) and thus we leave i_orphan list_head with a stale content. Later we can look at this content causing list corruption, oops, or other issues. The reported trace looked like: WARNING: CPU: 0 PID: 46 at lib/list_debug.c:53 __list_del_entry+0x6b/0x100() list_del corruption, 0000000061c1d6e0->next is LIST_POISON1 0000000000100100) CPU: 0 PID: 46 Comm: ext4.exe Not tainted 4.1.0-rc4+ #250 Stack: 60462947 62219960 602ede24 62219960 602ede24 603ca293 622198f0 602f02eb 62219950 6002c12c 62219900 601b4d6b Call Trace: [<6005769c>] ? vprintk_emit+0x2dc/0x5c0 [<602ede24>] ? printk+0x0/0x94 [<600190bc>] show_stack+0xdc/0x1a0 [<602ede24>] ? printk+0x0/0x94 [<602ede24>] ? printk+0x0/0x94 [<602f02eb>] dump_stack+0x2a/0x2c [<6002c12c>] warn_slowpath_common+0x9c/0xf0 [<601b4d6b>] ? __list_del_entry+0x6b/0x100 [<6002c254>] warn_slowpath_fmt+0x94/0xa0 [<602f4d09>] ? __mutex_lock_slowpath+0x239/0x3a0 [<6002c1c0>] ? warn_slowpath_fmt+0x0/0xa0 [<60023ebf>] ? set_signals+0x3f/0x50 [<600a205a>] ? kmem_cache_free+0x10a/0x180 [<602f4e88>] ? mutex_lock+0x18/0x30 [<601b4d6b>] __list_del_entry+0x6b/0x100 [<601177ec>] ext4_orphan_del+0x22c/0x2f0 [<6012f27c>] ? __ext4_journal_start_sb+0x2c/0xa0 [<6010b973>] ? ext4_truncate+0x383/0x390 [<6010bc8b>] ext4_write_begin+0x30b/0x4b0 [<6001bb50>] ? copy_from_user+0x0/0xb0 [<601aa840>] ? iov_iter_fault_in_readable+0xa0/0xc0 [<60072c4f>] generic_perform_write+0xaf/0x1e0 [<600c4166>] ? file_update_time+0x46/0x110 [<60072f0f>] __generic_file_write_iter+0x18f/0x1b0 [<6010030f>] ext4_file_write_iter+0x15f/0x470 [<60094e10>] ? unlink_file_vma+0x0/0x70 [<6009b180>] ? unlink_anon_vmas+0x0/0x260 [<6008f169>] ? free_pgtables+0xb9/0x100 [<600a6030>] __vfs_write+0xb0/0x130 [<600a61d5>] vfs_write+0xa5/0x170 [<600a63d6>] SyS_write+0x56/0xe0 [<6029fcb0>] ? __libc_waitpid+0x0/0xa0 [<6001b698>] handle_syscall+0x68/0x90 [<6002633d>] userspace+0x4fd/0x600 [<6002274f>] ? save_registers+0x1f/0x40 [<60028bd7>] ? arch_prctl+0x177/0x1b0 [<60017bd5>] fork_handler+0x85/0x90 Fix the problem by using list_del_init() as we always should with i_orphan list. Reported-by:
Vegard Nossum <vegard.nossum@oracle.com> Signed-off-by:
Jan Kara <jack@suse.cz> Signed-off-by:
-
Theodore Ts'o authored
commit 7827a7f6 upstream. Instead of just printing warning messages, if the orphan list is corrupted, declare the file system is corrupted. If there are any reserved inodes in the orphaned inode list, declare the file system corrupted and stop right away to avoid doing more potential damage to the file system. Signed-off-by:
-
Theodore Ts'o authored
commit c9eb13a9 upstream. If the orphaned inode list contains inode #5, ext4_iget() returns a bad inode (since the bootloader inode should never be referenced directly). Because of the bad inode, we end up processing the inode repeatedly and this hangs the machine. This can be reproduced via: mke2fs -t ext4 /tmp/foo.img 100 debugfs -w -R "ssv last_orphan 5" /tmp/foo.img mount -o loop /tmp/foo.img /mnt (But don't do this if you are using an unpatched kernel if you care about the system staying functional. :-) This bug was found by the port of American Fuzzy Lop into the kernel to find file system problems[1]. (Since it *only* happens if inode #5 shows up on the orphan list --- 3, 7, 8, etc. won't do it, it's not surprising that AFL needed two hours before it found it.) [1] http://events.linuxfoundation.org/sites/events/files/slides/AFL%20filesystem%20fuzzing%2C%20Vault%202016_0.pdf Reported by: Vegard Nossum <vegard.nossum@oracle.com> Signed-off-by:
-
Philipp Zabel authored
commit 310944d1 upstream. The component master driver imx-drm-core matches component devices using their of_node. Since commit 950b410dd1ab ("gpu: ipu-v3: Fix imx-ipuv3-crtc module autoloading"), the imx-ipuv3-crtc dev->of_node is not set during probing. Before that, of_node was set and caused an of: modalias to be used instead of the platform: modalias, which broke module autoloading. On the other hand, if dev->of_node is not set yet when the imx-ipuv3-crtc probe function calls component_add, component matching in imx-drm-core fails. While dev->of_node will be set once the next component tries to bring up the component master, imx-drm-core component binding will never succeed if one of the crtc devices is probed last. Add of_node to the component platform data and match against the pdata->of_node instead of dev->of_node in imx-drm-core to work around this problem. Fixes: 950b410dd1ab ("gpu: ipu-v3: Fix imx-ipuv3-crtc module autoloading") Signed-off-by:
Philipp Zabel <p.zabel@pengutronix.de> Tested-by:
Fabio Estevam <fabio.estevam@nxp.com> Tested-by:
Lothar Waßmann <LW@KARO-electronics.de> Tested-by:
Heiko Schocher <hs@denx.de> Tested-by:
Chris Ruehl <chris.ruehl@gtsys.com.hk> Signed-off-by:
-
Ville Syrjälä authored
commit 7045c368 upstream. When we read out the watermark state from the hardware we're supposed to transfer that into the active watermarks, but currently we fail to any part of the active watermarks that isn't explicitly written. Let's clear it all upfront. Looks like this has been like this since the beginning, when I added the readout. No idea why I didn't clear it up. Cc: Matt Roper <matthew.d.roper@intel.com> Fixes: 243e6a44 ("drm/i915: Init HSW watermark tracking in intel_modeset_setup_hw_state()") Signed-off-by:
Matt Roper <matthew.d.roper@intel.com> Signed-off-by:
Jani Nikula <jani.nikula@intel.com> Signed-off-by:
-
Lyude authored
Unfortunately since we don't have Dave's connector refcounting patch here yet, it's very possible that drm_atomic_state_default_clear() could get called by intel_display_resume() when intel_dp_mst_destroy_connector() isn't completely finished destroying an mst connector, but has already finished setting connector->funcs to NULL. As such, we need to treat the connector like it's already been destroyed and just skip it, otherwise we'll end up dereferencing a NULL pointer. This fix is only required for 4.6 and below. David Airlie's patchseries for 4.7 to add connector reference counting provides a more proper fix for this. Changes since v1: - Fix leftover whitespace Upstream fix: 0552f765 ("drm/i915/mst: use reference counted connectors. (v3)") Reviewed-by:
Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by:
Lyude <cpaul@redhat.com>
-
Lyude authored
commit 255f0e7c upstream. During boot, MST hotplugs are generally expected (even if no physical hotplugging occurs) and result in DRM's connector topology changing. This means that using num_connector from the current mode configuration can lead to the number of connectors changing under us. This can lead to some nasty scenarios in fbcon: - We allocate an array to the size of dev->mode_config.num_connectors. - MST hotplug occurs, dev->mode_config.num_connectors gets incremented. - We try to loop through each element in the array using the new value of dev->mode_config.num_connectors, and end up going out of bounds since dev->mode_config.num_connectors is now larger then the array we allocated. fb_helper->connector_count however, will always remain consistent while we do a modeset in fb_helper. Note: This is just polish for 4.7, Dave Airlie's drm_connector refcounting fixed these bugs for real. But it's good enough duct-tape for stable kernel backporting, since backporting the refcounting changes is way too invasive. Signed-off-by:
-
Lyude authored
commit 14a3842a upstream. During boot time, MST devices usually send a ton of hotplug events irregardless of whether or not any physical hotplugs actually occurred. Hotplugs mean connectors being created/destroyed, and the number of DRM connectors changing under us. This isn't a problem if we use fb_helper->connector_count since we only set it once in the code, however if we use num_connector from struct drm_mode_config we risk it's value changing under us. On top of that, there's even a chance that dev->mode_config.num_connector != fb_helper->connector_count. If the number of connectors happens to increase under us, we'll end up using the wrong array size for memcpy and start writing beyond the actual length of the array, occasionally resulting in kernel panics. Note: This is just polish for 4.7, Dave Airlie's drm_connector refcounting fixed these bugs for real. But it's good enough duct-tape for stable kernel backporting, since backporting the refcounting changes is way too invasive. Signed-off-by:
-
Mario Kleiner authored
commit 9d746ab6 upstream. When porting the hdmi deep color detection code from radeon-kms to amdgpu-kms apparently some kind of copy and paste error happened, attaching an else branch to the wrong if statement. The result is that hdmi deep color mode is always disabled, regardless of gpu and display capabilities and user wishes, as the code mistakenly thinks that the display doesn't provide the required max_tmds_clock limit and falls back to 8 bpc. This patch fixes deep color support, as tested on a R9 380 Tonga Pro + suitable display, and should be backported to all kernels with amdgpu-kms support. Signed-off-by:
Mario Kleiner <mario.kleiner.de@gmail.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
-
Alex Deucher authored
commit 6b8812eb upstream. This is a port of radeon commit: 3d2d98ee drm/radeon: use drm_mode_vrefresh() rather than mode->vrefresh to amdgpu. Signed-off-by:
-
Sinclair Yeh authored
commit 7851496a upstream. mode->hdisplay * (var->bits_per_pixel + 7) gets evaluated before the division, potentially making the pitch larger than it should be. Since the original intention is to do a div-round-up, just use the macro instead. Signed-off-by:
Sinclair Yeh <syeh@vmware.com> Reviewed-by:
Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by:
-
