Linus Torvalds [Sun, 9 Dec 2018 18:15:13 +0000 (10:15 -0800)]
Merge tag '4.20-rc5-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6
Pull cifs fixes from Steve French:
"Three small fixes: a fix for smb3 direct i/o, a fix for CIFS DFS for
stable and a minor cifs Kconfig fix"
* tag '4.20-rc5-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6:
CIFS: Avoid returning EBUSY to upper layer VFS
cifs: Fix separator when building path from dentry
cifs: In Kconfig CONFIG_CIFS_POSIX needs depends on legacy (insecure cifs)
Linus Torvalds [Sun, 9 Dec 2018 17:54:04 +0000 (09:54 -0800)]
Merge tag 'dax-fixes-4.20-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull dax fixes from Dan Williams:
"The last of the known regression fixes and fallout from the Xarray
conversion of the filesystem-dax implementation.
On the path to debugging why the dax memory-failure injection test
started failing after the Xarray conversion a couple more fixes for
the dax_lock_mapping_entry(), now called dax_lock_page(), surfaced.
Those plus the bug that started the hunt are now addressed. These
patches have appeared in a -next release with no issues reported.
Note the touches to mm/memory-failure.c are just the conversion to the
new function signature for dax_lock_page().
Summary:
- Fix the Xarray conversion of fsdax to properly handle
dax_lock_mapping_entry() in the presense of pmd entries
- Fix inode destruction racing a new lock request"
* tag 'dax-fixes-4.20-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
dax: Fix unlock mismatch with updated API
dax: Don't access a freed inode
dax: Check page->mapping isn't NULL
Linus Torvalds [Sun, 9 Dec 2018 17:46:54 +0000 (09:46 -0800)]
Merge tag 'libnvdimm-fixes-4.20-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull libnvdimm fixes from Dan Williams:
"A regression fix for the Address Range Scrub implementation, yes
another one, and support for platforms that misalign persistent memory
relative to the Linux memory hotplug section constraint. Longer term,
support for sub-section memory hotplug would alleviate alignment
waste, but until then this hack allows a 'struct page' memmap to be
established for these misaligned memory regions.
These have all appeared in a -next release, and thanks to Patrick for
reporting and testing the alignment padding fix.
Summary:
- Unless and until the core mm handles memory hotplug units smaller
than a section (128M), persistent memory namespaces must be padded
to section alignment.
The libnvdimm core already handled section collision with "System
RAM", but some configurations overlap independent "Persistent
Memory" ranges within a section, so additional padding injection is
added for that case.
- The recent reworks of the ARS (address range scrub) state machine
to reduce the number of state flags inadvertantly missed a
conversion of acpi_nfit_ars_rescan() call sites. Fix the regression
whereby user-requested ARS results in a "short" scrub rather than a
"long" scrub.
- Fixup the unit tests to handle / test the 128M section alignment of
mocked test resources.
* tag 'libnvdimm-fixes-4.20-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
acpi/nfit: Fix user-initiated ARS to be "ARS-long" rather than "ARS-short"
libnvdimm, pfn: Pad pfn namespaces relative to other regions
tools/testing/nvdimm: Align test resources to 128M
Tarick Bedeir [Fri, 7 Dec 2018 08:30:26 +0000 (00:30 -0800)]
net/mlx4_core: Correctly set PFC param if global pause is turned off.
rx_ppp and tx_ppp can be set between 0 and 255, so don't clamp to 1.
Fixes: 6e8814ceb7e8 ("net/mlx4_en: Fix mixed PFC and Global pause user control requests") Signed-off-by: Tarick Bedeir <tarick@google.com> Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Sat, 8 Dec 2018 19:44:04 +0000 (11:44 -0800)]
Merge tag 'asm-generic-4.20' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic
Pull asm-generic fix from Arnd Bergmann:
"Multiple people reported a bug I introduced in asm-generic/unistd.h in
4.20, this is the obvious bugfix to get glibc and others to correctly
build again on new architectures that no longer provide the old
fstatat64() family of system calls"
* tag 'asm-generic-4.20' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic:
asm-generic: unistd.h: fixup broken macro include.
Linus Torvalds [Sat, 8 Dec 2018 19:33:26 +0000 (11:33 -0800)]
Merge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk fixes from Stephen Boyd:
"A few clk driver fixes this time:
- Introduce protected-clock DT binding to fix breakage on qcom
sdm845-mtp boards where the qspi clks introduced this merge window
cause the firmware on those boards to take down the system if we
try to read the clk registers
- Fix a couple off-by-one errors found by Dan Carpenter
- Handle failure in zynq fixed factor clk driver to avoid using
uninitialized data"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: zynqmp: Off by one in zynqmp_is_valid_clock()
clk: mmp: Off by one in mmp_clk_add()
clk: mvebu: Off by one bugs in cp110_of_clk_get()
arm64: dts: qcom: sdm845-mtp: Mark protected gcc clocks
clk: qcom: Support 'protected-clocks' property
dt-bindings: clk: Introduce 'protected-clocks' property
clk: zynqmp: handle fixed factor param query error
Linus Torvalds [Sat, 8 Dec 2018 19:25:02 +0000 (11:25 -0800)]
Merge tag 'xfs-4.20-fixes-3' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Pull xfs fixes from Darrick Wong:
"Here are hopefully the last set of fixes for 4.20.
There's a fix for a longstanding statfs reporting problem with project
quotas, a correction for page cache invalidation behaviors when
fallocating near EOF, and a fix for a broken metadata verifier return
code.
Finally, the most important fix is to the pipe splicing code (aka the
generic copy_file_range fallback) to avoid pointless short directio
reads by only asking the filesystem for as much data as there are
available pages in the pipe buffer. Our previous fix (simulated short
directio reads because the number of pages didn't match the length of
the read requested) caused subtle problems on overlayfs, so that part
is reverted.
Anyhow, this series passes fstests -g all on xfs and overlay+xfs, and
has passed 17 billion fsx operations problem-free since I started
testing
Summary:
- Fix broken project quota inode counts
- Fix incorrect PAGE_MASK/PAGE_SIZE usage
- Fix incorrect return value in btree verifier
- Fix WARN_ON remap flags false positive
- Fix splice read overflows"
* tag 'xfs-4.20-fixes-3' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
iomap: partially revert 4721a601099 (simulated directio short read on EFAULT)
splice: don't read more than available pipe space
vfs: allow some remap flags to be passed to vfs_clone_file_range
xfs: fix inverted return from xfs_btree_sblock_verify_crc
xfs: fix PAGE_MASK usage in xfs_free_file_space
fs/xfs: fix f_ffree value for statfs when project quota is set
This should have been done as part of 2f0799a0ffc0 ("mm, thp: restore
node-local hugepage allocations"). The movement of the thp allocation
policy from alloc_pages_vma() to alloc_hugepage_direct_gfpmask() was
intended to only set __GFP_THISNODE for mempolicies that are not
MPOL_BIND whereas the revert could set this regardless of mempolicy.
While the check for MPOL_BIND between alloc_hugepage_direct_gfpmask()
and alloc_pages_vma() was racy, that has since been removed since the
revert. What is left is the possibility to use __GFP_THISNODE in
policy_node() when it is unexpected because the special handling for
hugepages in alloc_pages_vma() was removed as part of the consolidation.
Secondly, prior to 89c83fb539f9, alloc_pages_vma() implemented a somewhat
different policy for hugepage allocations, which were allocated through
alloc_hugepage_vma(). For hugepage allocations, if the allocating
process's node is in the set of allowed nodes, allocate with
__GFP_THISNODE for that node (for MPOL_PREFERRED, use that node with
__GFP_THISNODE instead). This was changed for shmem_alloc_hugepage() to
allow fallback to other nodes in 89c83fb539f9 as it did for new_page() in
mm/mempolicy.c which is functionally different behavior and removes the
requirement to only allocate hugepages locally.
So this commit does a full revert of 89c83fb539f9 instead of the partial
revert that was done in 2f0799a0ffc0. The result is the same thp
allocation policy for 4.20 that was in 4.19.
Fixes: 89c83fb539f9 ("mm, thp: consolidate THP gfp handling into alloc_hugepage_direct_gfpmask") Fixes: 2f0799a0ffc0 ("mm, thp: restore node-local hugepage allocations") Signed-off-by: David Rientjes <rientjes@google.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Michal Hocko <mhocko@kernel.org> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is completely bogus. The STACR register has two formats, old
and new, depending on the version of the IP block used. There's a pair of
device-tree properties that can be used to specify the format used:
has-inverted-stacr-oc
has-new-stacr-staopc
What this commit did was to change the bit definition used with the old
parts to match the new parts. This of course breaks the driver on all
the old ones.
Instead, the author should have set the appropriate properties in the
device-tree for the variant used on his board.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Dan Carpenter [Sat, 8 Dec 2018 05:06:12 +0000 (22:06 -0700)]
scsi: Fix a harmless double shift bug
Smatch generates a warning:
drivers/scsi/scsi_lib.c:1656 scsi_mq_done() warn: test_bit() takes a bit number
The problem is that SCMD_STATE_COMPLETE is supposed to be bit number 0
and not a mask like "(1 << 0)". It is used like this:
if (test_and_set_bit(SCMD_STATE_COMPLETE, &scmd->state))
The test_and_set_bit() has a shift built in so it's a double left shift
and uses bit number 1 instead of number 0. This bug is harmless because
it's done consistently and it doesn't clash with any other flags.
Fixes: f1342709d18a ("scsi: Do not rely on blk-mq for double completions") Reviewed-by: Keith Busch <keith.busch@intel.com> Acked-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Keith Busch [Tue, 27 Nov 2018 16:40:57 +0000 (09:40 -0700)]
nvme: implement Enhanced Command Retry
A controller may have an internal state that is not able to successfully
process commands for a short duration. In such states, an immediate
command requeue is expected to fail. The driver may exceed its max
retry count, which permanently ends the command in failure when the same
command would succeed after waiting for the controller to be ready.
NVMe ratified TP 4033 provides a delay hint in the completion status
code for failed commands. Implement the retry delay based on the command
completion status and the controller's requested delay.
Note that requeued commands are handled per request_queue, not per
individual request. If multiple commands fail, the controller should
consistently report the desired delay time for retryable commands in
all CQEs, otherwise the requeue list may be kicked too soon.
Signed-off-by: Keith Busch <keith.busch@intel.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
This patch adds unlikely in the nvmet request completion path for the
status check in the low level function __nvmet_req_complete.
This is helpful in the scenario where host and target connection is
working smoothly.
Sagi Grimberg [Mon, 19 Nov 2018 22:11:12 +0000 (14:11 -0800)]
nvmet: support fabrics sq flow control
Technical proposal 8005 "fabrics SQ flow control" introduces a mode
where a host and controller agree to omit sq_head pointer updates
when sending nvme completions.
In case the host indicated desire to operate in this mode (connect attribute)
the controller will return back a connect completion with sq_head value
of 0xffff as indication that it will omit sq_head pointer updates.
This mode saves us an atomic update in the I/O path.
Reviewed-by: Hannes Reinecke <hare@suse.com>
[hch: suggested better implementation] Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
James Smart [Wed, 14 Nov 2018 23:57:46 +0000 (15:57 -0800)]
nvmet-fc: remove the IN_ISR deferred scheduling options
All target lldd's call the cmd receive and op completions in non-isr
thread contexts. As such the IN_ISR options are not necessary.
Remove the functionality and flags, which also removes cpu assignments
to queues.
Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jay Sternberg [Mon, 12 Nov 2018 21:56:40 +0000 (13:56 -0800)]
nvmet: enable Discovery Controller AENs
Add functions to find connections requesting Discovery Change events
and send a notification to hosts that maintain an explicit persistent
connection and have and active Asynchronous Event Request pending.
Only Hosts that have access to the Subsystem effected by the change
will receive notifications of Discovery Change event.
Call these functions each time there is a configfs change that effects
the Discover Log Pages.
Set the OAES field in the Identify Controller response to advertise the
support for Asynchronous Event Notifications.
Signed-off-by: Jay Sternberg <jay.e.sternberg@intel.com> Reviewed-by: Phil Cayton <phil.cayton@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Sagi Grimberg [Mon, 12 Nov 2018 21:56:39 +0000 (13:56 -0800)]
nvmet: allow host connect even if no allowed subsystems are exported
It is perfectly valid that a host connects to a discovery subsystem
and gets an empty discovery log page since no subsystems are
provisioned to it. No reason to disallow connecting to the discovery
subsystem all together.
Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Jay Sternberg <jay.e.sternberg@intel.com> Reviewed-by: Phil Cayton <phil.cayton@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jay Sternberg [Mon, 12 Nov 2018 21:56:36 +0000 (13:56 -0800)]
nvmet: make kato and AEN processing for use by other controllers
Make common process of get/set features available to other controllers by
making simple functions static inline and others not static and prototypes
in nvmet.h file
Also remove static from nvmet_execute_async_event and add prototype to
nvmet.h to allow used by other controllers
Signed-off-by: Jay Sternberg <jay.e.sternberg@intel.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jay Sternberg [Mon, 12 Nov 2018 21:56:35 +0000 (13:56 -0800)]
nvmet: allow Keep Alive for Discovery controller
Per change to specification allowing Discovery controllers to have
explicit persistent connections, remove restriction on Discovery
controllers allowing kato on connect.
Signed-off-by: Jay Sternberg <jay.e.sternberg@intel.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jay Sternberg [Mon, 12 Nov 2018 21:56:34 +0000 (13:56 -0800)]
nvmet: change aen mask functions to use bit numbers
Functions nvmet_aen_disabled and nvmet_clear_aen were using
values not bit numbers ie 1 << 9 not 9 for bit function clear_bit
and test_and_set_bit.
Signed-off-by: Jay Sternberg <jay.e.sternberg@intel.com> Reviewed-by: Phil Cayton <phil.cayton@intel.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
This patch optimizes read command behavior when file-ns configured
with buffered I/O. Instead of offloading the buffered I/O read operations
to the worker threads, we first issue the read operation with IOCB_NOWAIT
and try and access the data from the cache. Here we only offload the
request to the worker thread and complete the request in the worker
thread context when IOCB_NOWAIT request fails.
Sagi Grimberg [Fri, 2 Nov 2018 17:28:13 +0000 (10:28 -0700)]
nvmet: support for traffic based keep-alive
A controller that supports traffic based keep-alive can restart the keep
alive timer even when no keep-alive was not received in the kato period
as long as other admin or I/O commands were received. For each command
set ctrl->cmd_seen to true, and when keep-alive timer expires, if any
commands were seen, resched ka_work instead of escalating to a fatal
error.
Sagi Grimberg [Fri, 2 Nov 2018 17:28:15 +0000 (10:28 -0700)]
nvme: support traffic based keep-alive
If the controller supports traffic based keep alive, we restart the keep
alive timer if any admin or io commands was completed during the kato
period. This prevents a possible starvation of keep alive commands in
the presence of heavy traffic as in such case, we already have a health
indication from the host perspective.
Only set a comp_seen indicator in case the controller supports keep
alive to minimize the overhead for pci controllers.
Hannes Reinecke [Fri, 16 Nov 2018 08:22:29 +0000 (09:22 +0100)]
nvme: add a numa_node field to struct nvme_ctrl
Instead of directly poking into the struct device add a new numa_node
field to struct nvme_ctrl. This allows fabrics drivers where ctrl->dev
is a virtual device to support NUMA affinity as well.
Also expose the field as a sysfs attribute, and populate it for the
RDMA and FC transports.
Signed-off-by: Hannes Reinecke <hare@suse.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
nvme: consolidate memset calls in the nvme_setup_cmd path
In function nvme_setup_cmd() we call command specific setup function
for flush, rw, and discard. Instead of calling memset in each function
lets call it once in the parent function.
This is purely code cleanup patch and it does not change any existing
functionality.
Signed-off-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 7 Dec 2018 03:03:53 +0000 (11:03 +0800)]
blk-mq: re-build queue map in case of kdump kernel
Now almost all .map_queues() implementation based on managed irq
affinity doesn't update queue mapping and it just retrieves the
old built mapping, so if nr_hw_queues is changed, the mapping talbe
includes stale mapping. And only blk_mq_map_queues() may rebuild
the mapping talbe.
One case is that we limit .nr_hw_queues as 1 in case of kdump kernel.
However, drivers often builds queue mapping before allocating tagset
via pci_alloc_irq_vectors_affinity(), but set->nr_hw_queues can be set
as 1 in case of kdump kernel, so wrong queue mapping is used, and
kernel panic[1] is observed during booting.
This patch fixes the kernel panic triggerd on nvme by rebulding the
mapping table via blk_mq_map_queues().
Cc: Christoph Hellwig <hch@lst.de> Cc: linux-nvme@lists.infradead.org Cc: David Milburn <dmilburn@redhat.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Josef Bacik [Tue, 4 Dec 2018 17:59:02 +0000 (12:59 -0500)]
block: add rq_qos_wait to rq_qos
Originally when I split out the common code from blk-wbt into rq_qos I
left the wbt_wait() where it was and simply copied and modified it
slightly to work for io-latency. However they are both basically the
same thing, and as time has gone on wbt_wait() has ended up much smarter
and kinder than it was when I copied it into io-latency, which means
io-latency has lost out on these improvements.
Since they are the same thing essentially except for a few minor things,
create rq_qos_wait() that replicates what wbt_wait() currently does with
callbacks that can be passed in for the snowflakes to do their own thing
as appropriate.
Signed-off-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:39 +0000 (12:10 -0500)]
blkcg: rename blkg_try_get() to blkg_tryget()
blkg reference counting now uses percpu_ref rather than atomic_t. Let's
make this consistent with css_tryget. This renames blkg_try_get to
blkg_tryget and now returns a bool rather than the blkg or %NULL.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:38 +0000 (12:10 -0500)]
blkcg: change blkg reference counting to use percpu_ref
Every bio is now associated with a blkg putting blkg_get, blkg_try_get,
and blkg_put on the hot path. Switch over the refcnt in blkg to use
percpu_ref.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Acked-by: Tejun Heo <tj@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:37 +0000 (12:10 -0500)]
blkcg: remove bio_disassociate_task()
Now that a bio only holds a blkg reference, so clean up is simply
putting back that reference. Remove bio_disassociate_task() as it just
calls bio_disassociate_blkg() and call the latter directly.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Acked-by: Tejun Heo <tj@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:36 +0000 (12:10 -0500)]
blkcg: remove additional reference to the css
The previous patch in this series removed carrying around a pointer to
the css in blkg. However, the blkg association logic still relied on
taking a reference on the css to ensure we wouldn't fail in getting a
reference for the blkg.
Here the implicit dependency on the css is removed. The association
continues to rely on the tryget logic walking up the blkg tree. This
streamlines the three ways that association can happen: normal, swap,
and writeback.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Acked-by: Tejun Heo <tj@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:35 +0000 (12:10 -0500)]
blkcg: remove bio->bi_css and instead use bio->bi_blkg
Prior patches ensured that any bio that interacts with a request_queue
is properly associated with a blkg. This makes bio->bi_css unnecessary
as blkg maintains a reference to blkcg already.
This removes the bio field bi_css and transfers corresponding uses to
access via bi_blkg.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:34 +0000 (12:10 -0500)]
blkcg: associate writeback bios with a blkg
One of the goals of this series is to remove a separate reference to
the css of the bio. This can and should be accessed via bio_blkcg(). In
this patch, wbc_init_bio() now requires a bio to have a device
associated with it.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:33 +0000 (12:10 -0500)]
blkcg: associate a blkg for pages being evicted by swap
A prior patch in this series added blkg association to bios issued by
cgroups. There are two other paths that we want to attribute work back
to the appropriate cgroup: swap and writeback. Here we modify the way
swap tags bios to include the blkg. Writeback will be tackle in the next
patch.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:32 +0000 (12:10 -0500)]
blkcg: consolidate bio_issue_init() to be a part of core
bio_issue_init among other things initializes the timestamp for an IO.
Rather than have this logic handled by policies, this consolidates it to
be on the init paths (normal, clone, bounce clone).
Signed-off-by: Dennis Zhou <dennis@kernel.org> Acked-by: Tejun Heo <tj@kernel.org> Reviewed-by: Liu Bo <bo.liu@linux.alibaba.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:31 +0000 (12:10 -0500)]
blkcg: associate blkg when associating a device
Previously, blkg association was handled by controller specific code in
blk-throttle and blk-iolatency. However, because a blkg represents a
relationship between a blkcg and a request_queue, it makes sense to keep
the blkg->q and bio->bi_disk->queue consistent.
This patch moves association into the bio_set_dev macro(). This should
cover the majority of cases where the device is set/changed keeping the
two pointers consistent. Fallback code is added to
blkcg_bio_issue_check() to catch any missing paths.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:30 +0000 (12:10 -0500)]
dm: set the static flush bio device on demand
The next patch changes the macro bio_set_dev() to associate a bio with a
blkg based on the device set. However, dm creates a static bio to be
used as the basis for cloning empty flush bios on creation. The
bio_set_dev() call in alloc_dev() will cause problems with the next
patch adding association to bio_set_dev() because the call is before the
bdev is associated with a gendisk (bd_disk is %NULL). To get around
this, set the device on the static bio every time and use that to clone
to the other bios.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Acked-by: Mike Snitzer <snitzer@redhat.com> Cc: Alasdair Kergon <agk@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:29 +0000 (12:10 -0500)]
blkcg: introduce common blkg association logic
There are 3 ways blkg association can happen: association with the
current css, with the page css (swap), or from the wbc css (writeback).
This patch handles how association is done for the first case where we
are associating bsaed on the current css. If there is already a blkg
associated, the css will be reused and association will be redone as the
request_queue may have changed.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:28 +0000 (12:10 -0500)]
blkcg: convert blkg_lookup_create() to find closest blkg
There are several scenarios where blkg_lookup_create() can fail such as
the blkcg dying, request_queue is dying, or simply being OOM. Most
handle this by simply falling back to the q->root_blkg and calling it a
day.
This patch implements the notion of closest blkg. During
blkg_lookup_create(), if it fails to create, return the closest blkg
found or the q->root_blkg. blkg_try_get_closest() is introduced and used
during association so a bio is always attached to a blkg.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Acked-by: Tejun Heo <tj@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:27 +0000 (12:10 -0500)]
blkcg: update blkg_lookup_create() to do locking
To know when to create a blkg, the general pattern is to do a
blkg_lookup() and if that fails, lock and do the lookup again, and if
that fails finally create. It doesn't make much sense for everyone who
wants to do creation to write this themselves.
This changes blkg_lookup_create() to do locking and implement this
pattern. The old blkg_lookup_create() is renamed to
__blkg_lookup_create(). If a call site wants to do its own error
handling or already owns the queue lock, they can use
__blkg_lookup_create(). This will be used in upcoming patches.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Acked-by: Tejun Heo <tj@kernel.org> Reviewed-by: Liu Bo <bo.liu@linux.alibaba.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dennis Zhou [Wed, 5 Dec 2018 17:10:26 +0000 (12:10 -0500)]
blkcg: fix ref count issue with bio_blkcg() using task_css
The bio_blkcg() function turns out to be inconsistent and consequently
dangerous to use. The first part returns a blkcg where a reference is
owned by the bio meaning it does not need to be rcu protected. However,
the third case, the last line, is problematic:
This can race against task migration and the cgroup dying. It is also
semantically different as it must be called rcu protected and is
susceptible to failure when trying to get a reference to it.
This patch adds association ahead of calling bio_blkcg() rather than
after. This makes association a required and explicit step along the
code paths for calling bio_blkcg(). In blk-iolatency, association is
moved above the bio_blkcg() call to ensure it will not return %NULL.
BFQ uses the old bio_blkcg() function, but I do not want to address it
in this series due to the complexity. I have created a private version
documenting the inconsistency and noting not to use it.
Signed-off-by: Dennis Zhou <dennis@kernel.org> Acked-by: Tejun Heo <tj@kernel.org> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 5 Dec 2018 13:50:40 +0000 (06:50 -0700)]
blk-mq: remove QUEUE_FLAG_POLL from default MQ flags
We only support polling if we have poll queues now, but the flag is
being set by default. Remove the default QUEUE_FLAG_POLL setting, we'll
set it in blk_mq_init_allocated_queue() if we have poll queues available
for this device.
Fixes: 6544d229bf43 ("block: enable polling by default if a poll map is initalized") Reported-by: Kirill Tkhai <ktkhai@virtuozzo.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
David S. Miller [Sat, 8 Dec 2018 00:24:40 +0000 (16:24 -0800)]
Merge branch 'skb-headroom-slab-out-of-bounds'
Stefano Brivio says:
====================
Fix slab out-of-bounds on insufficient headroom for IPv6 packets
Patch 1/2 fixes a slab out-of-bounds occurring with short SCTP packets over
IPv4 over L2TP over IPv6 on a configuration with relatively low HEADER_MAX.
Patch 2/2 makes sure we avoid writing before the allocated buffer in
neigh_hh_output() in case the headroom is enough for the unaligned hardware
header size, but not enough for the aligned one, and that we warn if we hit
this condition.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 6 Dec 2018 18:30:37 +0000 (19:30 +0100)]
neighbour: Avoid writing before skb->head in neigh_hh_output()
While skb_push() makes the kernel panic if the skb headroom is less than
the unaligned hardware header size, it will proceed normally in case we
copy more than that because of alignment, and we'll silently corrupt
adjacent slabs.
In the case fixed by the previous patch,
"ipv6: Check available headroom in ip6_xmit() even without options", we
end up in neigh_hh_output() with 14 bytes headroom, 14 bytes hardware
header and write 16 bytes, starting 2 bytes before the allocated buffer.
Always check we're not writing before skb->head and, if the headroom is
not enough, warn and drop the packet.
v2:
- instead of panicking with BUG_ON(), WARN_ON_ONCE() and drop the packet
(Eric Dumazet)
- if we avoid the panic, though, we need to explicitly check the headroom
before the memcpy(), otherwise we'll have corrupted slabs on a running
kernel, after we warn
- use __skb_push() instead of skb_push(), as the headroom check is
already implemented here explicitly (Eric Dumazet)
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 6 Dec 2018 18:30:36 +0000 (19:30 +0100)]
ipv6: Check available headroom in ip6_xmit() even without options
Even if we send an IPv6 packet without options, MAX_HEADER might not be
enough to account for the additional headroom required by alignment of
hardware headers.
On a configuration without HYPERV_NET, WLAN, AX25, and with IPV6_TUNNEL,
sending short SCTP packets over IPv4 over L2TP over IPv6, we start with
100 bytes of allocated headroom in sctp_packet_transmit(), end up with 54
bytes after l2tp_xmit_skb(), and 14 bytes in ip6_finish_output2().
Those would be enough to append our 14 bytes header, but we're going to
align that to 16 bytes, and write 2 bytes out of the allocated slab in
neigh_hh_output().
Eric Dumazet [Thu, 6 Dec 2018 17:58:24 +0000 (09:58 -0800)]
tcp: lack of available data can also cause TSO defer
tcp_tso_should_defer() can return true in three different cases :
1) We are cwnd-limited
2) We are rwnd-limited
3) We are application limited.
Neal pointed out that my recent fix went too far, since
it assumed that if we were not in 1) case, we must be rwnd-limited
Fix this by properly populating the is_cwnd_limited and
is_rwnd_limited booleans.
After this change, we can finally move the silly check for FIN
flag only for the application-limited case.
The same move for EOR bit will be handled in net-next,
since commit 1c09f7d073b1 ("tcp: do not try to defer skbs
with eor mark (MSG_EOR)") is scheduled for linux-4.21
Tested by running 200 concurrent netperf -t TCP_RR -- -r 60000,100
and checking none of them was rwnd_limited in the chrono_stat
output from "ss -ti" command.
Fixes: 41727549de3e ("tcp: Do not underestimate rwnd_limited") Signed-off-by: Eric Dumazet <edumazet@google.com> Suggested-by: Neal Cardwell <ncardwell@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Reviewed-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Fri, 7 Dec 2018 22:34:10 +0000 (14:34 -0800)]
Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost
Pull vhost/virtio fixes from Michael Tsirkin:
"A couple of last-minute fixes"
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
vhost/vsock: fix use-after-free in network stack callers
virtio/s390: fix race in ccw_io_helper()
virtio/s390: avoid race on vcdev->config
vhost/vsock: fix reset orphans race with close timeout
Linus Torvalds [Fri, 7 Dec 2018 22:18:49 +0000 (14:18 -0800)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fix from Catalin Marinas:
"Avoid sending IPIs with interrupts disabled"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: hibernate: Avoid sending cross-calling with interrupts disabled
* tag 'gcc-plugins-v4.20-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
stackleak: Register the 'stackleak_cleanup' pass before the '*free_cfg' pass
stackleak: Mark stackleak_track_stack() as notrace
Linus Torvalds [Fri, 7 Dec 2018 21:07:10 +0000 (13:07 -0800)]
Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Pull crypto fixes from Herbert Xu:
- Disable the new crypto stats interface as it's still being changed
- Fix potential uses-after-free in cbc/cfb/pcbc.
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: user - Disable statistics interface
crypto: do not free algorithm before using
Linus Torvalds [Fri, 7 Dec 2018 20:58:34 +0000 (12:58 -0800)]
Merge tag 'pci-v4.20-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fixes from Bjorn Helgaas:
"Revert ASPM change that caused a regression"
* tag 'pci-v4.20-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
Revert "PCI/ASPM: Do not initialize link state when aspm_disabled is set"
Shmulik Ladkani [Fri, 7 Dec 2018 07:50:17 +0000 (09:50 +0200)]
ipv6: sr: properly initialize flowi6 prior passing to ip6_route_output
In 'seg6_output', stack variable 'struct flowi6 fl6' was missing
initialization.
Fixes: 6c8702c60b88 ("ipv6: sr: add support for SRH encapsulation and injection with lwtunnels") Signed-off-by: Shmulik Ladkani <shmulik.ladkani@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Fri, 7 Dec 2018 18:40:37 +0000 (10:40 -0800)]
Merge tag 'for-linus-20181207' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
"Let's try this again...
We're finally happy with the DM livelock issue, and it's also passed
overnight testing and the corruption regression test. The end result
is much nicer now too, which is great.
Outside of that fix, there's a pull request for NVMe with two small
fixes, and a regression fix for BFQ from this merge window. The BFQ
fix looks bigger than it is, it's 90% comment updates"
* tag 'for-linus-20181207' of git://git.kernel.dk/linux-block:
blk-mq: punt failed direct issue to dispatch list
nvmet-rdma: fix response use after free
nvme: validate controller state before rescheduling keep alive
block, bfq: fix decrement of num_active_groups
Linus Torvalds [Fri, 7 Dec 2018 18:31:31 +0000 (10:31 -0800)]
Merge branch 'i2c/for-current-fixed' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fixes from Wolfram Sang:
"A set of driver bugfixes for the I2C subsystem"
* 'i2c/for-current-fixed' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: uniphier-f: fix violation of tLOW requirement for Fast-mode
i2c: uniphier: fix violation of tLOW requirement for Fast-mode
i2c: uniphier-f: fill TX-FIFO only in IRQ handler for repeated START
i2c: uniphier-f: fix timeout error after reading 8 bytes
i2c: scmi: Fix probe error on devices with an empty SMB0001 ACPI device node
i2c: axxia: properly handle master timeout
i2c: rcar: check bus state before reinitializing
i2c: nvidia-gpu: limit reads also for combined messages
i2c: nvidia-gpu: adhere to I2C fault codes
Linus Torvalds [Fri, 7 Dec 2018 17:58:34 +0000 (09:58 -0800)]
Merge tag 'dmaengine-fix-4.20-rc6' of git://git.infradead.org/users/vkoul/slave-dma
Pull dmaengine fixes from Vinod Koul:
"Another pull request for dmaengine. We got bunch of fixes early this
week and all are tagged to stable. Hope this is last fix for this
cycle:
- Fix imx-sdma handling of channel terminations, this involves
reverting two commits and implement async termination
- Fix cppi dma channel deletion from pending list on stop
- Fix FIFO size for dw controller in Intel Merrifield"
* tag 'dmaengine-fix-4.20-rc6' of git://git.infradead.org/users/vkoul/slave-dma:
dmaengine: dw: Fix FIFO size for Intel Merrifield
dmaengine: cppi41: delete channel from pending list when stop channel
dmaengine: imx-sdma: use GFP_NOWAIT for dma descriptor allocations
dmaengine: imx-sdma: implement channel termination via worker
Revert "dmaengine: imx-sdma: alloclate bd memory from dma pool"
Revert "dmaengine: imx-sdma: Use GFP_NOWAIT for dma allocations"
x86/vdso: Drop implicit common-page-size linker flag
GNU linker's -z common-page-size's default value is based on the target
architecture. arch/x86/entry/vdso/Makefile sets it to the architecture
default, which is implicit and redundant. Drop it.
Fixes: 2aae950b21e4 ("x86_64: Add vDSO for x86-64 with gettimeofday/clock_gettime/getcpu") Reported-by: Dmitry Golovin <dima@golovin.in> Reported-by: Bill Wendling <morbo@google.com> Suggested-by: Dmitry Golovin <dima@golovin.in> Suggested-by: Rui Ueyama <ruiu@google.com> Signed-off-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Borislav Petkov <bp@suse.de> Acked-by: Andy Lutomirski <luto@kernel.org> Cc: Andi Kleen <andi@firstfloor.org> Cc: Fangrui Song <maskray@google.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: x86-ml <x86@kernel.org> Link: https://lkml.kernel.org/r/20181206191231.192355-1-ndesaulniers@google.com Link: https://bugs.llvm.org/show_bug.cgi?id=38774 Link: https://github.com/ClangBuiltLinux/linux/issues/31
Will Deacon [Fri, 7 Dec 2018 12:47:10 +0000 (12:47 +0000)]
arm64: hibernate: Avoid sending cross-calling with interrupts disabled
Since commit 3b8c9f1cdfc50 ("arm64: IPI each CPU after invalidating the
I-cache for kernel mappings"), a call to flush_icache_range() will use
an IPI to cross-call other online CPUs so that any stale instructions
are flushed from their pipelines. This triggers a WARN during the
hibernation resume path, where flush_icache_range() is called with
interrupts disabled and is therefore prone to deadlock:
Since all secondary CPUs have been taken offline prior to invalidating
the I-cache, there's actually no need for an IPI and we can simply call
__flush_icache_range() instead.
Cc: <stable@vger.kernel.org> Fixes: 3b8c9f1cdfc50 ("arm64: IPI each CPU after invalidating the I-cache for kernel mappings") Reported-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com> Tested-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com> Tested-by: James Morse <james.morse@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Jens Axboe [Fri, 7 Dec 2018 05:17:44 +0000 (22:17 -0700)]
blk-mq: punt failed direct issue to dispatch list
After the direct dispatch corruption fix, we permanently disallow direct
dispatch of non read/write requests. This works fine off the normal IO
path, as they will be retried like any other failed direct dispatch
request. But for the blk_insert_cloned_request() that only DM uses to
bypass the bottom level scheduler, we always first attempt direct
dispatch. For some types of requests, that's now a permanent failure,
and no amount of retrying will make that succeed. This results in a
livelock.
Instead of making special cases for what we can direct issue, and now
having to deal with DM solving the livelock while still retaining a BUSY
condition feedback loop, always just add a request that has been through
->queue_rq() to the hardware queue dispatch list. These are safe to use
as no merging can take place there. Additionally, if requests do have
prepped data from drivers, we aren't dependent on them not sharing space
in the request structure to safely add them to the IO scheduler lists.
This basically reverts ffe81d45322c and is based on a patch from Ming,
but with the list insert case covered as well.
Fixes: ffe81d45322c ("blk-mq: fix corruption with direct issue") Cc: stable@vger.kernel.org Suggested-by: Ming Lei <ming.lei@redhat.com> Reported-by: Bart Van Assche <bvanassche@acm.org> Tested-by: Ming Lei <ming.lei@redhat.com> Acked-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Israel Rukshin [Wed, 5 Dec 2018 16:54:57 +0000 (16:54 +0000)]
nvmet-rdma: fix response use after free
nvmet_rdma_release_rsp() may free the response before using it at error
flow.
Fixes: 8407879 ("nvmet-rdma: fix possible bogus dereference under heavy load") Signed-off-by: Israel Rukshin <israelr@mellanox.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Max Gurtovoy <maxg@mellanox.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
James Smart [Wed, 28 Nov 2018 01:04:44 +0000 (17:04 -0800)]
nvme: validate controller state before rescheduling keep alive
Delete operations are seeing NULL pointer references in call_timer_fn.
Tracking these back, the timer appears to be the keep alive timer.
nvme_keep_alive_work() which is tied to the timer that is cancelled
by nvme_stop_keep_alive(), simply starts the keep alive io but doesn't
wait for it's completion. So nvme_stop_keep_alive() only stops a timer
when it's pending. When a keep alive is in flight, there is no timer
running and the nvme_stop_keep_alive() will have no affect on the keep
alive io. Thus, if the io completes successfully, the keep alive timer
will be rescheduled. In the failure case, delete is called, the
controller state is changed, the nvme_stop_keep_alive() is called while
the io is outstanding, and the delete path continues on. The keep
alive happens to successfully complete before the delete paths mark it
as aborted as part of the queue termination, so the timer is restarted.
The delete paths then tear down the controller, and later on the timer
code fires and the timer entry is now corrupt.
Fix by validating the controller state before rescheduling the keep
alive. Testing with the fix has confirmed the condition above was hit.
Signed-off-by: James Smart <jsmart2021@gmail.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de>
Paolo Valente [Thu, 6 Dec 2018 18:18:18 +0000 (19:18 +0100)]
block, bfq: fix decrement of num_active_groups
Since commit '2d29c9f89fcd ("block, bfq: improve asymmetric scenarios
detection")', if there are process groups with I/O requests waiting for
completion, then BFQ tags the scenario as 'asymmetric'. This detection
is needed for preserving service guarantees (for details, see comments
on the computation * of the variable asymmetric_scenario in the
function bfq_better_to_idle).
Unfortunately, commit '2d29c9f89fcd ("block, bfq: improve asymmetric
scenarios detection")' contains an error exactly in the updating of
the number of groups with I/O requests waiting for completion: if a
group has more than one descendant process, then the above number of
groups, which is renamed from num_active_groups to a more appropriate
num_groups_with_pending_reqs by this commit, may happen to be wrongly
decremented multiple times, namely every time one of the descendant
processes gets all its pending I/O requests completed.
A correct, complete solution should work as follows. Consider a group
that is inactive, i.e., that has no descendant process with pending
I/O inside BFQ queues. Then suppose that num_groups_with_pending_reqs
is still accounting for this group, because the group still has some
descendant process with some I/O request still in
flight. num_groups_with_pending_reqs should be decremented when the
in-flight request of the last descendant process is finally completed
(assuming that nothing else has changed for the group in the meantime,
in terms of composition of the group and active/inactive state of
child groups and processes). To accomplish this, an additional
pending-request counter must be added to entities, and must be
updated correctly.
To avoid this additional field and operations, this commit resorts to
the following tradeoff between simplicity and accuracy: for an
inactive group that is still counted in num_groups_with_pending_reqs,
this commit decrements num_groups_with_pending_reqs when the first
descendant process of the group remains with no request waiting for
completion.
This simplified scheme provides a fix to the unbalanced decrements
introduced by 2d29c9f89fcd. Since this error was also caused by lack
of comments on this non-trivial issue, this commit also adds related
comments.
Long Li [Thu, 6 Dec 2018 04:51:06 +0000 (04:51 +0000)]
CIFS: Avoid returning EBUSY to upper layer VFS
EBUSY is not handled by VFS, and will be passed to user-mode. This is not
correct as we need to wait for more credits.
This patch also fixes a bug where rsize or wsize is used uninitialized when
the call to server->ops->wait_mtu_credits() fails.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com>
Linus Torvalds [Fri, 7 Dec 2018 03:35:50 +0000 (19:35 -0800)]
Merge tag 'drm-fixes-2018-12-07' of git://anongit.freedesktop.org/drm/drm
Pull drm fixes from Dave Airlie:
"There's a bit more in here than I'd like, and I'm hoping things calm
down when I'm out.
msm:
- a bunch of display fixes for the new DPU
- a couple of command submission fixes
omap:
- some DSI fixes
ast:
- driver unload crash fix
core:
- fix the lease uevent so userspace can distinguish it
amd:
- fix a bpc regression
- fix lru handling regression
- fixed firmware support for new GPUs
- power management fixes for vega20"
* tag 'drm-fixes-2018-12-07' of git://anongit.freedesktop.org/drm/drm: (37 commits)
drm/ast: Fix connector leak during driver unload
drm/amdgpu/vcn: Update vcn.cur_state during suspend
drm/amd/display: Fix overflow/truncation from strncpy.
drm/amd/powerplay: improve OD code robustness
drm/amdgpu: enlarge maximum waiting time of KIQ
drm/fb-helper: Fix typo in parameter description
drm/amd/powerplay: support SoftMin/Max setting for some specific DPM
drm/amd/powerplay: issue pre-display settings for display change event
drm/amd/powerplay: support new pptable upload on Vega20
drm/amdgpu/gmc8: always load MC firmware in the driver
drm/amdgpu/gmc8: update MC firmware for polaris
drm/amdgpu: update mc firmware image for polaris12 variants
drm/msm: Fix error return checking
drm/msm/dpu: Ignore alpha for XBGR8888 format
drm/msm: dpu: Fix "WARNING: invalid free of devm_ allocated data"
drm/msm/hdmi: Drop pointless static qualifier in msm_hdmi_bind()
drm/msm: Move fence put to where failure occurs
drm/msm: dpu: Don't set legacy plane->crtc pointer
drm/msm/gpu: Don't map command buffers with nr_relocs equal to 0
drm/msm/hdmi: Enable HPD after HDMI IRQ is set up
...
Linus Torvalds [Fri, 7 Dec 2018 02:57:04 +0000 (18:57 -0800)]
Merge tag 'nfs-for-4.20-5' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client bugfixes from Trond Myklebust:
"This is mainly fallout from the updates to the SUNRPC code that is
being triggered from less common combinations of NFS mount options.
Highlights include:
Stable fixes:
- Fix a page leak when using RPCSEC_GSS/krb5p to encrypt data.
Bugfixes:
- Fix a regression that causes the RPC receive code to hang
- Fix call_connect_status() so that it handles tasks that got
transmitted while queued waiting for the socket lock.
- Fix a memory leak in call_encode()
- Fix several other connect races.
- Fix receive code error handling.
- Use the discard iterator rather than MSG_TRUNC for compatibility
with AF_UNIX/AF_LOCAL sockets.
- nfs: don't dirty kernel pages read by direct-io
- pnfs/Flexfiles fix to enforce per-mirror stateid only for NFSv4
data servers"
* tag 'nfs-for-4.20-5' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
SUNRPC: Don't force a redundant disconnection in xs_read_stream()
SUNRPC: Fix up socket polling
SUNRPC: Use the discard iterator rather than MSG_TRUNC
SUNRPC: Treat EFAULT as a truncated message in xs_read_stream_request()
SUNRPC: Fix up handling of the XDRBUF_SPARSE_PAGES flag
SUNRPC: Fix RPC receive hangs
SUNRPC: Fix a potential race in xprt_connect()
SUNRPC: Fix a memory leak in call_encode()
SUNRPC: Fix leak of krb5p encode pages
SUNRPC: call_connect_status() must handle tasks that got transmitted
nfs: don't dirty kernel pages read by direct-io
flexfiles: enforce per-mirror stateid only for v4 DSes
Linus Torvalds [Fri, 7 Dec 2018 00:45:36 +0000 (16:45 -0800)]
Merge branch 'spectre' of git://git.armlinux.org.uk/~rmk/linux-arm
Pull ARM spectre fix from Russell King:
"Exynos folk noticed that CPU hotplug wasn't working with their kernel
configuration, and have tested this as fixing the problem"
* 'spectre' of git://git.armlinux.org.uk/~rmk/linux-arm:
ARM: ensure that processor vtables is not lost after boot
Masahiro Yamada [Thu, 6 Dec 2018 03:55:26 +0000 (12:55 +0900)]
i2c: uniphier-f: fill TX-FIFO only in IRQ handler for repeated START
- For a repeated START condition, this controller starts data transfer
immediately after the slave address is written to the TX-FIFO.
- Once the TX-FIFO empty interrupt is asserted, the controller makes
a pause even if additional data are written to the TX-FIFO.
Given those circumstances, the data after a repeated START may not be
transferred if the interrupt is asserted while the TX-FIFO is being
filled up. A more reliable way is to append TX data only in the
interrupt handler.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Masahiro Yamada [Thu, 6 Dec 2018 03:55:25 +0000 (12:55 +0900)]
i2c: uniphier-f: fix timeout error after reading 8 bytes
I was totally screwed up in commit eaba68785c2d ("i2c: uniphier-f:
fix race condition when IRQ is cleared"). Since that commit, if the
number of read bytes is multiple of the FIFO size (8, 16, 24... bytes),
the STOP condition could be issued twice, depending on the timing.
If this happens, the controller will go wrong, resulting in the timeout
error.
It was more than 3 years ago when I wrote this driver, so my memory
about this hardware was vague. Please let me correct the description
in the commit log of eaba68785c2d.
Clearing the IRQ status on exiting the IRQ handler is absolutely
fine. This controller makes a pause while any IRQ status is asserted.
If the IRQ status is cleared first, the hardware may start the next
transaction before the IRQ handler finishes what it supposed to do.
This partially reverts the bad commit with clear comments so that I
will never repeat this mistake.
I also investigated what is happening at the last moment of the read
mode. The UNIPHIER_FI2C_INT_RF interrupt is asserted a bit earlier
(by half a period of the clock cycle) than UNIPHIER_FI2C_INT_RB.
I consulted a hardware engineer, and I got the following information:
UNIPHIER_FI2C_INT_RF
asserted at the falling edge of SCL at the 8th bit.
UNIPHIER_FI2C_INT_RB
asserted at the rising edge of SCL at the 9th (ACK) bit.
In order to avoid calling uniphier_fi2c_stop() twice, check the latter
interrupt. I also commented this because it is obscure hardware internal.
Fixes: eaba68785c2d ("i2c: uniphier-f: fix race condition when IRQ is cleared") Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Hans de Goede [Wed, 21 Nov 2018 09:19:55 +0000 (10:19 +0100)]
i2c: scmi: Fix probe error on devices with an empty SMB0001 ACPI device node
Some AMD based HP laptops have a SMB0001 ACPI device node which does not
define any methods.
This leads to the following error in dmesg:
[ 5.222731] cmi: probe of SMB0001:00 failed with error -5
This commit makes acpi_smbus_cmi_add() return -ENODEV instead in this case
silencing the error. In case of a failure of the i2c_add_adapter() call
this commit now propagates the error from that call instead of -EIO.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
According to Intel (R) Axxia TM Lionfish Communication Processor
Peripheral Subsystem Hardware Reference Manual, the AXXIA I2C module
have a programmable Master Wait Timer, which among others, checks the
time between commands send in manual mode. When a timeout (25ms) passes,
TSS bit is set in Master Interrupt Status register and a Stop command is
issued by the hardware.
The axxia_i2c_xfer(), does not properly handle this situation, however.
For each message a separate axxia_i2c_xfer_msg() is called and this
function incorrectly assumes that any interrupt might happen only when
waiting for completion. This is mostly correct but there is one
exception - a master timeout can trigger if enough time has passed
between individual transfers. It will, by definition, happen between
transfers when the interrupts are disabled by the code. If that happens,
the hardware issues Stop command.
The interrupt indicating timeout will not be triggered as soon as we
enable them since the Master Interrupt Status is cleared when master
mode is entered again (which happens before enabling irqs) meaning this
error is lost and the transfer is continued even though the Stop was
issued on the bus. The subsequent operations completes without error but
a bogus value (0xFF in case of read) is read as the client device is
confused because aborted transfer. No error is returned from
master_xfer() making caller believe that a valid value was read.
To fix the problem, the TSS bit (indicating timeout) in Master Interrupt
Status register is checked before each transfer. If it is set, there was
a timeout before this transfer and (as described above) the hardware
already issued Stop command so the transaction should be aborted thus
-ETIMEOUT is returned from the master_xfer() callback. In order to be
sure no timeout was issued we can't just read the status just before
starting new transaction as there will always be a small window of time
(few CPU cycles at best) where this might still happen. For this reason
we have to temporally disable the timer before checking for TSS bit.
Disabling it will, however, clear the TSS bit so in order to preserve
that information, we have to read it in ISR so we have to ensure that
the TSS interrupt is not masked between transfers of one transaction.
There is no need to call bus recovery or controller reinitialization if
that happens so it's skipped.
Signed-off-by: Krzysztof Adamski <krzysztof.adamski@nokia.com> Reviewed-by: Alexander Sverdlin <alexander.sverdlin@nokia.com> Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
David S. Miller [Thu, 6 Dec 2018 21:31:09 +0000 (13:31 -0800)]
Merge branch 'mlxsw-Various-fixes'
Ido Schimmel says:
====================
mlxsw: Various fixes
Patches #1 and #2 fix two VxLAN related issues. The first patch removes
warnings that can currently be triggered from user space. Second patch
avoids leaking a FID in an error path.
Patch #3 fixes a too strict check that causes certain host routes not to
be promoted to perform GRE decapsulation in hardware.
Last patch avoids a use-after-free when deleting a VLAN device via an
ioctl when it is enslaved to a bridge. I have a patchset for net-next
that reworks this code and makes the driver more robust.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Ido Schimmel [Thu, 6 Dec 2018 17:44:53 +0000 (17:44 +0000)]
mlxsw: spectrum_switchdev: Fix VLAN device deletion via ioctl
When deleting a VLAN device using an ioctl the netdev is unregistered
before the VLAN filter is updated via ndo_vlan_rx_kill_vid(). It can
lead to a use-after-free in mlxsw in case the VLAN device is deleted
while being enslaved to a bridge.
The reason for the above is that when mlxsw receives the CHANGEUPPER
event, it wrongly assumes that the VLAN device is no longer its upper
and thus destroys the internal representation of the bridge port despite
the reference count being non-zero.
Fix this by checking if the VLAN device is our upper using its real
device. In net-next I'm going to remove this trick and instead make
mlxsw completely agnostic to the order of the events.
Fixes: c57529e1d5d8 ("mlxsw: spectrum: Replace vPorts with Port-VLAN") Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Petr Machata <petrm@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Nir Dotan [Thu, 6 Dec 2018 17:44:52 +0000 (17:44 +0000)]
mlxsw: spectrum_router: Relax GRE decap matching check
GRE decap offload is configured when local routes prefix correspond to the
local address of one of the offloaded GRE tunnels. The matching check was
found to be too strict, such that for a flat GRE configuration, in which
the overlay and underlay traffic share the same non-default VRF, decap flow
was not offloaded.
Relax the check for decap flow offloading. A match occurs if the local
address of the tunnel matches the local route address while both share the
same VRF table.
Fixes: 4607f6d26950 ("mlxsw: spectrum_router: Support IPv4 underlay decap") Signed-off-by: Nir Dotan <nird@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
It should never be possible for a user to set a VNI on a FID in case one
is already set. The driver therefore returns an error, but fails to drop
the reference count taken earlier when calling
mlxsw_sp_fid_8021d_lookup().
Drop the reference when this unlikely error is hit.
Fixes: 1c30d1836aeb ("mlxsw: spectrum: Enable VxLAN enslavement to bridges") Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
It is possible to trigger a warning in mlxsw in case a flood entry which
mlxsw is not aware of is deleted from the VxLAN device. This is because
mlxsw expects to find a singly linked list where the flood entry is
present in.
Fix by removing these warnings for now.
Will re-add them in the next release after we teach mlxsw to ask for a
dump of FDB entries from the VxLAN device, once it is enslaved to a
bridge mlxsw cares about.
Fixes: 6e6030bd5412 ("mlxsw: spectrum_nve: Implement common NVE core") Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Petr Machata <petrm@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefan Hajnoczi [Mon, 5 Nov 2018 10:35:47 +0000 (10:35 +0000)]
vhost/vsock: fix use-after-free in network stack callers
If the network stack calls .send_pkt()/.cancel_pkt() during .release(),
a struct vhost_vsock use-after-free is possible. This occurs because
.release() does not wait for other CPUs to stop using struct
vhost_vsock.
Switch to an RCU-enabled hashtable (indexed by guest CID) so that
.release() can wait for other CPUs by calling synchronize_rcu(). This
also eliminates vhost_vsock_lock acquisition in the data path so it
could have a positive effect on performance.
This is CVE-2018-14625 "kernel: use-after-free Read in vhost_transport_send_pkt".
Cc: stable@vger.kernel.org Reported-and-tested-by: syzbot+bd391451452fb0b93039@syzkaller.appspotmail.com Reported-by: syzbot+e3e074963495f92a89ed@syzkaller.appspotmail.com Reported-by: syzbot+d5a0a170c5069658b141@syzkaller.appspotmail.com Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com>
Halil Pasic [Wed, 26 Sep 2018 16:48:30 +0000 (18:48 +0200)]
virtio/s390: fix race in ccw_io_helper()
While ccw_io_helper() seems like intended to be exclusive in a sense that
it is supposed to facilitate I/O for at most one thread at any given
time, there is actually nothing ensuring that threads won't pile up at
vcdev->wait_q. If they do, all threads get woken up and see the status
that belongs to some other request than their own. This can lead to bugs.
For an example see:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1788432
This race normally does not cause any problems. The operations provided
by struct virtio_config_ops are usually invoked in a well defined
sequence, normally don't fail, and are normally used quite infrequent
too.
Yet, if some of the these operations are directly triggered via sysfs
attributes, like in the case described by the referenced bug, userspace
is given an opportunity to force races by increasing the frequency of the
given operations.
Let us fix the problem by ensuring, that for each device, we finish
processing the previous request before starting with a new one.
Signed-off-by: Halil Pasic <pasic@linux.ibm.com> Reported-by: Colin Ian King <colin.king@canonical.com> Cc: stable@vger.kernel.org
Message-Id: <20180925121309.58524-3-pasic@linux.ibm.com> Signed-off-by: Cornelia Huck <cohuck@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Halil Pasic [Wed, 26 Sep 2018 16:48:29 +0000 (18:48 +0200)]
virtio/s390: avoid race on vcdev->config
Currently we have a race on vcdev->config in virtio_ccw_get_config() and
in virtio_ccw_set_config().
This normally does not cause problems, as these are usually infrequent
operations. However, for some devices writing to/reading from the config
space can be triggered through sysfs attributes. For these, userspace can
force the race by increasing the frequency.
Signed-off-by: Halil Pasic <pasic@linux.ibm.com> Cc: stable@vger.kernel.org
Message-Id: <20180925121309.58524-2-pasic@linux.ibm.com> Signed-off-by: Cornelia Huck <cohuck@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Stefan Hajnoczi [Thu, 6 Dec 2018 19:14:34 +0000 (19:14 +0000)]
vhost/vsock: fix reset orphans race with close timeout
If a local process has closed a connected socket and hasn't received a
RST packet yet, then the socket remains in the table until a timeout
expires.
When a vhost_vsock instance is released with the timeout still pending,
the socket is never freed because vhost_vsock has already set the
SOCK_DONE flag.
Check if the close timer is pending and let it close the socket. This
prevents the race which can leak sockets.
Reported-by: Maximilian Riemensberger <riemensberger@cadami.net> Cc: Graham Whaley <graham.whaley@gmail.com> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>