aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2018-05-31btrfs: Add unprivileged ioctl which returns subvolume informationTomohiro Misono2-0/+183
Add new unprivileged ioctl BTRFS_IOC_GET_SUBVOL_INFO which returns the information of subvolume containing this inode. (i.e. returns the information in ROOT_ITEM and ROOT_BACKREF.) Reviewed-by: Gu Jinxiang <[email protected]> Tested-by: Gu Jinxiang <[email protected]> Signed-off-by: Tomohiro Misono <[email protected]> [ minor style fixes, update struct comments ] Signed-off-by: David Sterba <[email protected]>
2018-05-31xfrm Fix potential error pointer dereference in xfrm_bundle_create.Steffen Klassert1-3/+2
We may derference an invalid pointer in the error path of xfrm_bundle_create(). Fix this by returning this error pointer directly instead of assigning it to xdst0. Fixes: 45b018beddb6 ("ipsec: Create and use new helpers for dst child access.") Signed-off-by: Steffen Klassert <[email protected]>
2018-05-31m68k: Set default dma mask for platform devicesFinn Thain1-0/+10
This avoids a WARNING splat when loading the macsonic or macmace driver. Please see commit 205e1b7f51e4 ("dma-mapping: warn when there is no coherent_dma_mask"). This implementation of arch_setup_pdev_archdata() differs from the powerpc one, in that this one avoids clobbering a device dma mask which has already been initialized. Cc: Christoph Hellwig <[email protected]> Cc: Greg Ungerer <[email protected]> Signed-off-by: Finn Thain <[email protected]> Acked-by: Greg Ungerer <[email protected]> Signed-off-by: Geert Uytterhoeven <[email protected]>
2018-05-30fs: clear writeback errors in inode_init_alwaysDarrick J. Wong1-0/+1
In inode_init_always(), we clear the inode mapping flags, which clears any retained error (AS_EIO, AS_ENOSPC) bits. Unfortunately, we do not also clear wb_err, which means that old mapping errors can leak through to new inodes. This is crucial for the XFS inode allocation path because we recycle old in-core inodes and we do not want error state from an old file to leak into the new file. This bug was discovered by running generic/036 and generic/047 in a loop and noticing that the EIOs generated by the collision of direct and buffered writes in generic/036 would survive the remount between 036 and 047, and get reported to the fsyncs (on different files!) in generic/047. Signed-off-by: Darrick J. Wong <[email protected]> Reviewed-by: Jeff Layton <[email protected]> Reviewed-by: Brian Foster <[email protected]>
2018-05-30smb3: add tracepoints for smb2/smb3 openSteve French2-1/+92
add two tracepoints for open completion. One for error one for completion (open_done). Sample output below TASK-PID CPU# |||| TIMESTAMP FUNCTION | | | |||| | | bash-15348 [007] .... 42441.027492: smb3_enter: cifs_lookup: xid=45 bash-15348 [007] .... 42441.028214: smb3_cmd_err: sid=0x6173e4ce tid=0xa05150e6 cmd=5 mid=105 status=0xc0000034 rc=-2 bash-15348 [007] .... 42441.028219: smb3_open_err: xid=45 sid=0x6173e4ce tid=0xa05150e6 cr_opts=0x0 des_access=0x80 rc=-2 bash-15348 [007] .... 42441.028225: smb3_exit_done: cifs_lookup: xid=45 fop777-24560 [002] .... 42442.627617: smb3_enter: cifs_revalidate_dentry_attr: xid=46 fop777-24560 [003] .... 42442.628301: smb3_cmd_err: sid=0x6173e4ce tid=0xa05150e6 cmd=5 mid=106 status=0xc0000034 rc=-2 fop777-24560 [003] .... 42442.628319: smb3_open_err: xid=46 sid=0x6173e4ce tid=0xa05150e6 cr_opts=0x0 des_access=0x80 rc=-2 fop777-24560 [003] .... 42442.628335: smb3_enter: cifs_atomic_open: xid=47 fop777-24560 [003] .... 42442.629587: smb3_cmd_done: sid=0x6173e4ce tid=0xa05150e6 cmd=5 mid=107 fop777-24560 [003] .... 42442.629592: smb3_open_done: xid=47 sid=0x6173e4ce tid=0xa05150e6 fid=0xb8a0984d cr_opts=0x40 des_access=0x40000080 Signed-off-by: Steve French <[email protected]>
2018-05-30block: fixup bioset_integrity_create() callJens Axboe1-1/+1
Missed converting the bioset_integrity_create() bounce bio set call. Fixes: 338aa96d5661 ("block: convert bounce, q->bio_split to bioset_init()/mempool_init()") Signed-off-by: Jens Axboe <[email protected]>
2018-05-30cifs: add debug output to show nocase mount optionSteve French1-0/+2
For smb1 nocase can be specified on mount. Allow displaying it in debug data. Signed-off-by: Steve French <[email protected]>
2018-05-30smb3: add define for id for posix create context and corresponding structSteve French1-1/+10
Signed-off-by: Steve French <[email protected]>
2018-05-31Merge tag 'drm-misc-fixes-2018-05-30' of ↵Dave Airlie3-13/+6
git://anongit.freedesktop.org/drm/drm-misc into drm-fixes dw-hdmi: Fix Oops regression from rc1 (Neil) Cc: Neil Armstrong <[email protected]> * tag 'drm-misc-fixes-2018-05-30' of git://anongit.freedesktop.org/drm/drm-misc: drm/bridge/synopsys: dw-hdmi: fix dw_hdmi_setup_rx_sense
2018-05-30cifs: update smb2_check_message to handle PDUs without a 4 byte length headerRonnie Sahlberg1-30/+20
Signed-off-by: Ronnie Sahlberg <[email protected]> Signed-off-by: Steve French <[email protected]>
2018-05-30Merge tag 'for-linus-20180530' of git://git.kernel.dk/linux-blockLinus Torvalds1-1/+1
Pull block fix from Jens Axboe: "Just a single fix that should make it into this release, fixing a regression with T10-DIF on NVMe" * tag 'for-linus-20180530' of git://git.kernel.dk/linux-block: nvme: fix extended data LBA supported setting
2018-05-30Merge tag 'selinux-pr-20180530' of ↵Linus Torvalds1-1/+1
git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux Pull SELinux fix from Paul Moore: "One more small fix for SELinux: a small string length fix found by KASAN. I dislike sending patches this late in the release cycle, but this patch fixes a legitimate problem, is very small, limited in scope, and well understood. There are two threads with more information on the problem, the latest is linked below: https://marc.info/?t=152723737400001&r=1&w=2 Stephen points out in the thread linked above: 'Such a setxattr() call can only be performed by a process with CAP_MAC_ADMIN that is also allowed mac_admin permission in SELinux policy. Consequently, this is never possible on Android (no process is allowed mac_admin permission, always enforcing) and is only possible in Fedora/RHEL for a few domains (if enforcing)'" * tag 'selinux-pr-20180530' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux: selinux: KASAN: slab-out-of-bounds in xattr_getsecurity
2018-05-30block: Drop bioset_create()Kent Overstreet2-52/+15
All users have been converted to bioset_init(), kill off the old API. Reviewed-by: Christoph Hellwig <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30xfs: convert to bioset_init()/mempool_init()Kent Overstreet3-8/+7
Convert XFS to embedded bio sets. Acked-by: Darrick J. Wong <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30btrfs: convert to bioset_init()/mempool_init()Kent Overstreet1-14/+11
Convert btrfs to embedded bio sets. Acked-by: Chris Mason <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30fs: convert block_dev.c to bioset_init()Kent Overstreet1-6/+3
Convert block DIO code to embedded bio sets. Reviewed-by: Christoph Hellwig <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30target: convert to bioset_init()/mempool_init()Kent Overstreet2-9/+7
Convert the target code to embedded bio sets. Reviewed-by: Christoph Hellwig <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30dm: convert to bioset_init()/mempool_init()Kent Overstreet17-206/+197
Convert dm to embedded bio sets. Acked-by: Mike Snitzer <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30md: convert to bioset_init()/mempool_init()Kent Overstreet15-181/+159
Convert md to embedded bio sets. Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30bcache: convert to bioset_init()/mempool_init()Kent Overstreet7-52/+37
Convert bcache to embedded bio sets. Reviewed-by: Coly Li <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30lightnvm: convert to bioset_init()/mempool_init()Kent Overstreet6-65/+65
Convert lightnvm to embedded bio sets. Reviewed-by: Javier González <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30pktcdvd: convert to bioset_init()/mempool_init()Kent Overstreet2-26/+26
Convert pktcdvd to embedded bio sets. Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30drbd: convert to bioset_init()/mempool_init()Kent Overstreet6-59/+38
Convert drbd to embedded bio sets and mempools. Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30block: convert bounce, q->bio_split to bioset_init()/mempool_init()Kent Overstreet7-33/+41
Convert the core block functionality to embedded bio sets. Reviewed-by: Christoph Hellwig <[email protected]> Signed-off-by: Kent Overstreet <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30Merge branch 'linus' of ↵Linus Torvalds1-2/+2
git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6 Pull crypto fix from Herbert Xu: "This fixes a potential kernel panic in the inside-secure driver" * 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: crypto: inside-secure - do not use memset on MMIO
2018-05-30smb3: allow "posix" mount option to enable new SMB311 protocol extensionsSteve French3-1/+36
If "posix" (or synonym "unix" for backward compatibility) specified on mount, and server advertises support for SMB3.11 POSIX negotiate context, then enable the new posix extensions on the tcon. This can be viewed by looking for "posix" in the mount options displayed by /proc/mounts for that mount (ie if posix extensions allowed by server and the experimental POSIX extensions also requested on the mount by specifying "posix" at mount time). Also add check to warn user if conflicting unix/nounix or posix/noposix specified on mount. Signed-off-by: Steve French <[email protected]>
2018-05-30smb3: add support for posix negotiate contextSteve French4-6/+36
Unlike CIFS where UNIX/POSIX extensions had been negotiatable, SMB3 did not have POSIX extensions yet. Add the new SMB3.11 POSIX negotiate context to ask the server whether it can support POSIX (and thus whether we can send the new POSIX open context). Signed-off-by: Steve French <[email protected]>
2018-05-30cifs: allow disabling less secure legacy dialectsSteve French3-9/+29
To improve security it may be helpful to have additional ways to restrict the ability to override the default dialects (SMB2.1, SMB3 and SMB3.02) on mount with old dialects (CIFS/SMB1 and SMB2) since vers=1.0 (CIFS/SMB1) and vers=2.0 are weaker and less secure. Add a module parameter "disable_legacy_dialects" (/sys/module/cifs/parameters/disable_legacy_dialects) which can be set to 1 (or equivalently Y) to forbid use of vers=1.0 or vers=2.0 on mount. Also cleans up a few build warnings about globals for various module parms. Signed-off-by: Steve French <[email protected]>
2018-05-30cifs: make minor clarifications to module params for cifs.koSteve French2-3/+5
Note which ones of the module params are cifs dialect only (N/A for default dialect now that has moved to SMB2.1 or later) Signed-off-by: Steve French <[email protected]> Reviewed-by: Ronnie Sahlberg <[email protected]>
2018-05-30cifs: show the "w" bit for writeable /proc/fs/cifs/* filesRonnie Sahlberg1-14/+14
RHBZ: 1539612 Lets show the "w" bit for those files have a .write interface to set/enable/... the feature. Reported-by: Xiaoli Feng <[email protected]> Signed-off-by: Ronnie Sahlberg <[email protected]> Signed-off-by: Steve French <[email protected]>
2018-05-30smb3: add module alias for smb3 to cifs.koSteve French1-1/+19
We really don't want to be encouraging people to use the old (less secure) cifs dialect (SMB1) and it can be confusing for them with SMB3 (or later) being recommended but the module name is cifs. Add a module alias for "smb3" to cifs.ko to make this less confusing. Signed-off-by: Steve French <[email protected]>
2018-05-30cifs: return error on invalid value written to cifsFYIRonnie Sahlberg1-0/+2
RHBZ: 1539617 Check that, if it is not a boolean, the value the user tries to write to /proc/fs/cifs/cifsFYI is valid and return an error if not. Signed-off-by: Ronnie Sahlberg <[email protected]> Signed-off-by: Steve French <[email protected]> Reported-by: Xiaoli Feng <[email protected]>
2018-05-30cifs: invalidate cache when we truncate a fileRonnie Sahlberg1-4/+9
RHBZ: 1566345 When truncating a file we always do this synchronously to the server. Thus we need to make sure that the cached inode metadata is marked as stale so that on next getattr we will refresh the metadata. In this particular bug we want to ensure that both ctime and mtime are updated and become visible to the application after a truncate. Signed-off-by: Ronnie Sahlberg <[email protected]> Signed-off-by: Steve French <[email protected]> Reported-by: Xiaoli Feng <[email protected]>
2018-05-30smb3: print tree id in debugdata in proc to be able to help loggingSteve French1-0/+2
When loooking at the logs for the new trace-cmd tracepoints for cifs, it would help to know which tid is for which share (UNC name) so update /proc/fs/cifs/DebugData to display the tid. Also display Maximal Access which was missing as well. Now the entry for typical entry for a tcon (in proc/fs/cifs/) looks like: 1) \\localhost\test Mounts: 1 DevInfo: 0x20 Attributes: 0x1006f PathComponentMax: 255 Status: 1 type: DISK Share Capabilities: None Aligned, Partition Aligned, Share Flags: 0x0 tid: 0xe0632a55 Optimal sector size: 0x200 Maximal Access: 0x1f01ff Signed-off-by: Steve French <[email protected]>
2018-05-30smb3: add additional ftrace entry points for entry/exit to cifs.koSteve French3-35/+85
Signed-off-by: Steve French <[email protected]>
2018-05-30smb3: fix various xid leaksSteve French1-19/+44
Fix a few cases where we were not freeing the xid which led to active requests being non-zero at unmount time. Signed-off-by: Steve French <[email protected]> CC: Stable <[email protected]> Reviewed-by: Ronnie Sahlberg <[email protected]>
2018-05-30CIFS: Introduce offset for the 1st page in data transfer structuresLong Li2-0/+4
When direct I/O is used, the data buffer may not always align to page boundaries. Introduce a page offset in transport data structures to describe the location of the buffer within the page. Also change the function to pass the page offset when sending data to transport. Signed-off-by: Long Li <[email protected]> Signed-off-by: Steve French <[email protected]>
2018-05-30Btrfs: clean up error handling in btrfs_truncate()Omar Sandoval1-19/+14
btrfs_truncate() uses two variables for error handling, ret and err (if this sounds familiar, it's because btrfs_truncate_inode_items() did something similar). This is error prone, as was made evident by "Btrfs: fix error handling in btrfs_truncate()". We only have err because we don't want to mask an error if we call btrfs_update_inode() and btrfs_end_transaction(), so let's make that its own scoped return variable and use ret everywhere else. Reviewed-by: Nikolay Borisov <[email protected]> Signed-off-by: Omar Sandoval <[email protected]> Signed-off-by: David Sterba <[email protected]>
2018-05-30platform/chrome: Use to_cros_ec_dev more broadlyGwendal Grignou4-22/+12
Move to_cros_ec_dev macro to cros_ec.h and use it when the private ec object is needed from device object. Signed-off-by: Gwendal Grignou <[email protected]> Reviewed-by: Enric Balletbo i Serra <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-05-30blk-throttle: return proper bool type to caller instead of 0/1Chengguang Xu1-5/+5
Change to return true/false only for bool type return code. Signed-off-by: Chengguang Xu <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30perf tools: Fix perf.data format description of NRCPUS headerArnaldo Carvalho de Melo1-1/+1
In the perf.data HEADER_CPUDESC feadure header we store first the number of available CPUs in the system, then the number of CPUs at the time of writing the header, not the other way around. Reported-by: Thomas-Mich Richter <[email protected]> Acked-by: Andi Kleen <[email protected]> Cc: Adrian Hunter <[email protected]> Cc: David Ahern <[email protected]> Cc: He Kuang <[email protected]> Cc: Hendrik Brueckner <[email protected]> Cc: Jin Yao <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Kim Phillips <[email protected]> Cc: Lakshman Annadorai <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Simon Que <[email protected]> Cc: Stephane Eranian <[email protected]> Cc: Wang Nan <[email protected]> Link: https://lkml.kernel.org/n/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2018-05-30perf script python: Add addr into perf sample dictLeo Yan1-0/+2
ARM CoreSight auxtrace uses 'sample->addr' to record the target address for branch instructions, so the data of 'sample->addr' is required for tracing data analysis. This commit collects data of 'sample->addr' into perf sample dict, finally can be used for python script for parsing event. Signed-off-by: Leo Yan <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Jonathan Corbet <[email protected]> Cc: Mathieu Poirier <[email protected]> Cc: Mike Leach <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Robert Walker <[email protected]> Cc: Tor Jeremiassen <[email protected]> Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2018-05-30perf data: Update documentation section on cpu topologyThomas Richter1-0/+8
Add an explanation of each cpu's core and socket identifier to the perf.data file format documentation. Signed-off-by: Thomas Richter <[email protected]> Cc: Heiko Carstens <[email protected]> Cc: Hendrik Brueckner <[email protected]> Cc: Martin Schwidefsky <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2018-05-30perf cs-etm: Fix indexing for decoder packet queueMathieu Poirier1-2/+10
The tail of a queue is supposed to be pointing to the next available slot in a queue. In this implementation the tail is incremented before it is used and as such points to the last used element, something that has the immense advantage of centralizing tail management at a single location and eliminating a lot of redundant code. But this needs to be taken into consideration on the dequeueing side where the head also needs to be incremented before it is used, or the first available element of the queue will be skipped. Signed-off-by: Mathieu Poirier <[email protected]> Tested-by: Leo Yan <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Robert Walker <[email protected]> Cc: [email protected] Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2018-05-30perf bpf: Fix NULL return handling in bpf__prepare_load()YueHaibing1-3/+3
bpf_object__open()/bpf_object__open_buffer can return error pointer or NULL, check the return values with IS_ERR_OR_NULL() in bpf__prepare_load and bpf__prepare_load_buffer Signed-off-by: YueHaibing <[email protected]> Acked-by: Daniel Borkmann <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: [email protected] Link: https://lkml.kernel.org/n/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2018-05-30drm/bridge/synopsys: dw-hdmi: fix dw_hdmi_setup_rx_senseNeil Armstrong3-13/+6
The dw_hdmi_setup_rx_sense exported function should not use struct device to recover the dw-hdmi context using drvdata, but take struct dw_hdmi directly like other exported functions. This caused a regression using Meson DRM on S905X since v4.17-rc1 : Internal error: Oops: 96000007 [#1] PREEMPT SMP [...] CPU: 0 PID: 124 Comm: irq/32-dw_hdmi_ Not tainted 4.17.0-rc7 #2 Hardware name: Libre Technology CC (DT) [...] pc : osq_lock+0x54/0x188 lr : __mutex_lock.isra.0+0x74/0x530 [...] Process irq/32-dw_hdmi_ (pid: 124, stack limit = 0x00000000adf418cb) Call trace: osq_lock+0x54/0x188 __mutex_lock_slowpath+0x10/0x18 mutex_lock+0x30/0x38 __dw_hdmi_setup_rx_sense+0x28/0x98 dw_hdmi_setup_rx_sense+0x10/0x18 dw_hdmi_top_thread_irq+0x2c/0x50 irq_thread_fn+0x28/0x68 irq_thread+0x10c/0x1a0 kthread+0x128/0x130 ret_from_fork+0x10/0x18 Code: 34000964 d00050a2 51000484 9135c042 (f864d844) ---[ end trace 945641e1fbbc07da ]--- note: irq/32-dw_hdmi_[124] exited with preempt_count 1 genirq: exiting task "irq/32-dw_hdmi_" (124) is an active IRQ thread (irq 32) Fixes: eea034af90c6 ("drm/bridge/synopsys: dw-hdmi: don't clobber drvdata") Signed-off-by: Neil Armstrong <[email protected]> Tested-by: Koen Kooi <[email protected]> Signed-off-by: Sean Paul <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2018-05-30blk-mq: only iterate over inflight requests in blk_mq_tagset_busy_iterChristoph Hellwig5-21/+4
We already check for started commands in all callbacks, but we should also protect against already completed commands. Do this by taking the checks to common code. Acked-by: Josef Bacik <[email protected]> Signed-off-by: Christoph Hellwig <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30nbd: clear DISCONNECT_REQUESTED flag once disconnection occurs.Kevin Vigor1-6/+15
When a userspace client requests a NBD device be disconnected, the DISCONNECT_REQUESTED flag is set. While this flag is set, the driver will not inform userspace when a connection is closed. Unfortunately the flag was never cleared, so once a disconnect was requested the driver would thereafter never tell userspace about a closed connection. Thus when connections failed due to timeout, no attempt to reconnect was made and eventually the device would fail. Fix by clearing the DISCONNECT_REQUESTED flag (and setting the DISCONNECTED flag) once all connections are closed. Reviewed-by: Josef Bacik <[email protected]> Signed-off-by: Kevin Vigor <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-05-30vhost_net: flush batched heads before trying to busy pollingJason Wang1-13/+24
After commit e2b3b35eb989 ("vhost_net: batch used ring update in rx"), we tend to batch updating used heads. But it doesn't flush batched heads before trying to do busy polling, this will cause vhost to wait for guest TX which waits for the used RX. Fixing by flush batched heads before busy loop. 1 byte TCP_RR performance recovers from 13107.83 to 50402.65. Fixes: e2b3b35eb989 ("vhost_net: batch used ring update in rx") Signed-off-by: Jason Wang <[email protected]> Acked-by: Michael S. Tsirkin <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2018-05-30btrfs: Factor out write portion of btrfs_get_blocks_directNikolay Borisov1-99/+108
Now that the read side is extracted into its own function, do the same to the write side. This leaves btrfs_get_blocks_direct_write with the sole purpose of handling common locking required. Also flip the condition in btrfs_get_blocks_direct_write so that the write case comes first and we check for if (Create) rather than if (!create). This is purely subjective but I believe makes reading a bit more "linear". No functional changes. Signed-off-by: Nikolay Borisov <[email protected]> Signed-off-by: David Sterba <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.