Ming Lei [Tue, 30 Jun 2020 10:25:01 +0000 (18:25 +0800)]
blk-mq: support batching dispatch in case of io
More and more drivers want to get batching requests queued from
block layer, such as mmc, and tcp based storage drivers. Also
current in-tree users have virtio-scsi, virtio-blk and nvme.
For none, we already support batching dispatch.
But for io scheduler, every time we just take one request from scheduler
and pass the single request to blk_mq_dispatch_rq_list(). This way makes
batching dispatch not possible when io scheduler is applied. One reason
is that we don't want to hurt sequential IO performance, becasue IO
merge chance is reduced if more requests are dequeued from scheduler
queue.
Try to support batching dispatch for io scheduler by starting with the
following simple approach:
1) still make sure we can get budget before dequeueing request
2) use hctx->dispatch_busy to evaluate if queue is busy, if it is busy
we fackback to non-batching dispatch, otherwise dequeue as many as
possible requests from scheduler, and pass them to blk_mq_dispatch_rq_list().
Wrt. 2), we use similar policy for none, and turns out that SCSI SSD
performance got improved much.
In future, maybe we can develop more intelligent algorithem for batching
dispatch.
Baolin has tested this patch and found that MMC performance is improved[3].
Ming Lei [Tue, 30 Jun 2020 10:24:59 +0000 (18:24 +0800)]
blk-mq: remove dead check from blk_mq_dispatch_rq_list
When BLK_STS_RESOURCE or BLK_STS_DEV_RESOURCE is returned from
.queue_rq, the 'list' variable always holds this rq which isn't
queued to LLD successfully.
So blk_mq_dispatch_rq_list() always returns false from the branch
of '!list_empty(list)'.
Ming Lei [Tue, 30 Jun 2020 10:24:58 +0000 (18:24 +0800)]
blk-mq: move getting driver tag and budget into one helper
Move code for getting driver tag and budget into one helper, so
blk_mq_dispatch_rq_list gets a bit simplified, and easier to read.
Meantime move updating of 'no_tag' and 'no_budget_available' into
the branch for handling partial dispatch because that is exactly
consumer of the two local variables.
Also rename the parameter of 'got_budget' as 'ask_budget'.
Ming Lei [Tue, 30 Jun 2020 10:24:57 +0000 (18:24 +0800)]
blk-mq: pass hctx to blk_mq_dispatch_rq_list
All requests in the 'list' of blk_mq_dispatch_rq_list belong to same
hctx, so it is better to pass hctx instead of request queue, because
blk-mq's dispatch target is hctx instead of request queue.
Ming Lei [Tue, 30 Jun 2020 10:24:56 +0000 (18:24 +0800)]
blk-mq: pass request queue into get/put budget callback
blk-mq budget is abstract from scsi's device queue depth, and it is
always per-request-queue instead of hctx.
It can be quite absurd to get a budget from one hctx, then dequeue a
request from scheduler queue, and this request may not belong to this
hctx, at least for bfq and deadline.
So fix the mess and always pass request queue to get/put budget
callback.
blkcg_bio_issue_check is a giant inline function that does three entirely
different things. Factor out the blk-cgroup related bio initalization
into a new helper, and the open code the sequence in the only caller,
relying on the fact that all the actual functionality is stubbed out for
non-cgroup builds.
block: bypass blkg_tryget_closest for the root_blkg
The root_blkg is only torn down at the very end of removing a queue.
So in the I/O submission path is always has a life reference and we
can just grab another one using blkg_get instead of doing a tryget
and parent walk that won't lead anywhere.
block: move bio_associate_blkg_from_page to mm/page_io.c
bio_associate_blkg_from_page is a special purpose helper for swap bios
that doesn't need access to bio internals. Move it to the swap code
instead of having it in bio.c.
block: really clone the block cgroup in bio_clone_blkg_association
bio_clone_blkg_association is supposed to clone the associatation, but
actually ends up doing a search with a tryget. As we know we have a
reference on the source cgroup just get an unconditional additional
reference to it and call it a day. That also removes the need for
a RCU critical section.
dm: use bio_uninit instead of bio_disassociate_blkg
bio_uninit is the proper API to clean up a BIO that has been allocated
on stack or inside a structure that doesn't come from the BIO allocator.
Switch dm to use that instead of bio_disassociate_blkg, which really is
an implementation detail. Note that the bio_uninit calls are also moved
to the two callers of __send_empty_flush, so that they better pair with
the bio_init calls used to initialize them.
Jan Kara [Wed, 17 Jun 2020 13:58:23 +0000 (15:58 +0200)]
blktrace: Provide event for request merging
Currently blk-mq does not report any event when two requests get merged
in the elevator. This then results in difficult to understand sequence
of events like:
and you can only guess (after quite some headscratching since the above
excerpt is intermixed with a lot of other IO) that request 215023544+56
got merged to request 215023504+40. Provide proper event for request
merging like we used to do in the legacy block layer.
block: reduce ifdef CONFIG_BLOCK madness in headers
Large part of bio.h, blkdev.h and genhd.h are under ifdef CONFIG_BLOCK
for no good reason. Only stub out function that are called from
code that is not dependent on CONFIG_BLOCK and leave the harmless
other declarations around.
Luis Chamberlain [Fri, 19 Jun 2020 20:47:30 +0000 (20:47 +0000)]
block: create the request_queue debugfs_dir on registration
We were only creating the request_queue debugfs_dir only
for make_request block drivers (multiqueue), but never for
request-based block drivers. We did this as we were only
creating non-blktrace additional debugfs files on that directory
for make_request drivers. However, since blktrace *always* creates
that directory anyway, we special-case the use of that directory
on blktrace. Other than this being an eye-sore, this exposes
request-based block drivers to the same debugfs fragile
race that used to exist with make_request block drivers
where if we start adding files onto that directory we can later
run a race with a double removal of dentries on the directory
if we don't deal with this carefully on blktrace.
Instead, just simplify things by always creating the request_queue
debugfs_dir on request_queue registration. Rename the mutex also to
reflect the fact that this is used outside of the blktrace context.
Luis Chamberlain [Fri, 19 Jun 2020 20:47:29 +0000 (20:47 +0000)]
blktrace: ensure our debugfs dir exists
We make an assumption that a debugfs directory exists, but since
this can fail ensure it exists before allowing blktrace setup to
complete. Otherwise we end up stuffing blktrace files on the debugfs
root directory. In the worst case scenario this *in theory* can create
an eventual panic *iff* in the future a similarly named file is created
prior on the debugfs root directory. This theoretical crash can happen
due to a recursive removal followed by a specific dentry removal.
This doesn't fix any known crash, however I have seen the files
go into the main debugfs root directory in cases where the debugfs
directory was not created due to other internal bugs with blktrace
now fixed.
blktrace is also completely useless without this directory, so
this ensures to userspace we only setup blktrace if the kernel
can stuff files where they are supposed to go into.
debugfs directory creations typically aren't checked for, and we have
maintainers doing sweep removals of these checks, but since we need this
check to ensure proper userspace blktrace functionality we make sure
to annotate the justification for the check.
Luis Chamberlain [Fri, 19 Jun 2020 20:47:28 +0000 (20:47 +0000)]
blktrace: fix debugfs use after free
On commit 6ac93117ab00 ("blktrace: use existing disk debugfs directory")
merged on v4.12 Omar fixed the original blktrace code for request-based
drivers (multiqueue). This however left in place a possible crash, if you
happen to abuse blktrace while racing to remove / add a device.
We used to use asynchronous removal of the request_queue, and with that
the issue was easier to reproduce. Now that we have reverted to
synchronous removal of the request_queue, the issue is still possible to
reproduce, its however just a bit more difficult.
We essentially run two instances of break-blktrace which add/remove
a loop device, and setup a blktrace and just never tear the blktrace
down. We do this twice in parallel. This is easily reproduced with the
script run_0004.sh from break-blktrace [0].
We can end up with two types of panics each reflecting where we
race, one a failed blktrace setup:
debugfs: Directory 'loop0' with parent 'block' already present
This crash happens because of how blktrace uses the debugfs directory
where it places its files. Upon init we always create the same directory
which would be needed by blktrace but we only do this for make_request
drivers (multiqueue) block drivers. When you race a removal of these
devices with a blktrace setup you end up in a situation where the
make_request recursive debugfs removal will sweep away the blktrace
files and then later blktrace will also try to remove individual
dentries which are already NULL. The inverse is also possible and hence
the two types of use after frees.
We don't create the block debugfs directory on init for these types of
block devices:
* request-based block driver block devices
* every possible partition
* scsi-generic
And so, this race should in theory only be possible with make_request
drivers.
We can fix the UAF by simply re-using the debugfs directory for
make_request drivers (multiqueue) and only creating the ephemeral
directory for the other type of block devices. The new clarifications
on relying on the q->blk_trace_mutex *and* also checking for q->blk_trace
*prior* to processing a blktrace ensures the debugfs directories are
only created if no possible directory name clashes are possible.
This goes tested with:
o nvme partitions
o ISCSI with tgt, and blktracing against scsi-generic with:
o block
o tape
o cdrom
o media changer
o blktests
This patch is part of the work which disputes the severity of
CVE-2019-19770 which shows this issue is not a core debugfs issue, but
a misuse of debugfs within blktace.
Luis Chamberlain [Fri, 19 Jun 2020 20:47:25 +0000 (20:47 +0000)]
block: revert back to synchronous request_queue removal
Commit dc9edc44de6c ("block: Fix a blk_exit_rl() regression") merged on
v4.12 moved the work behind blk_release_queue() into a workqueue after a
splat floated around which indicated some work on blk_release_queue()
could sleep in blk_exit_rl(). This splat would be possible when a driver
called blk_put_queue() or blk_cleanup_queue() (which calls blk_put_queue()
as its final call) from an atomic context.
blk_put_queue() decrements the refcount for the request_queue kobject, and
upon reaching 0 blk_release_queue() is called. Although blk_exit_rl() is
now removed through commit db6d99523560 ("block: remove request_list code")
on v5.0, we reserve the right to be able to sleep within
blk_release_queue() context.
The last reference for the request_queue must not be called from atomic
context. *When* the last reference to the request_queue reaches 0 varies,
and so let's take the opportunity to document when that is expected to
happen and also document the context of the related calls as best as
possible so we can avoid future issues, and with the hopes that the
synchronous request_queue removal sticks.
We revert back to synchronous request_queue removal because asynchronous
removal creates a regression with expected userspace interaction with
several drivers. An example is when removing the loopback driver, one
uses ioctls from userspace to do so, but upon return and if successful,
one expects the device to be removed. Likewise if one races to add another
device the new one may not be added as it is still being removed. This was
expected behavior before and it now fails as the device is still present
and busy still. Moving to asynchronous request_queue removal could have
broken many scripts which relied on the removal to have been completed if
there was no error. Document this expectation as well so that this
doesn't regress userspace again.
Using asynchronous request_queue removal however has helped us find
other bugs. In the future we can test what could break with this
arrangement by enabling CONFIG_DEBUG_KOBJECT_RELEASE.
While at it, update the docs with the context expectations for the
request_queue / gendisk refcount decrement, and make these
expectations explicit by using might_sleep().
Luis Chamberlain [Fri, 19 Jun 2020 20:47:24 +0000 (20:47 +0000)]
block: clarify context for refcount increment helpers
Let us clarify the context under which the helpers to increment the
refcount for the gendisk and request_queue can be called under. We
make this explicit on the places where we may sleep with might_sleep().
We don't address the decrement context yet, as that needs some extra
work and fixes, but will be addressed in the next patch.
blk-mq: add a new blk_mq_complete_request_remote API
This is a variant of blk_mq_complete_request_remote that only completes
the request if it needs to be bounced to another CPU or a softirq. If
the request can be completed locally the function returns false and lets
the driver complete it without requring and indirect function call.
blk-mq: remove the get_cpu/put_cpu pair in blk_mq_complete_request
We don't really care if we get migrated during the I/O completion.
In the worth case we either perform an IPI that wasn't required, or
complete the request on a CPU which we just migrated off.
blk-mq: move failure injection out of blk_mq_complete_request
Move the call to blk_should_fake_timeout out of blk_mq_complete_request
and into the drivers, skipping call sites that are obvious error
handlers, and remove the now superflous blk_mq_force_complete_rq helper.
This ensures we don't keep injecting errors into completions that just
terminate the Linux request after the hardware has been reset or the
command has been aborted.
blk-mq: merge the softirq vs non-softirq IPI logic
Both the softirq path for single queue devices and the multi-queue
completion handler share the same logic to figure out if we need an
IPI for the completion and eventually issue it. Merge the two
versions into a single unified code path.
Even for single queue devices there is no point in offloading a polled
completion to the softirq, given that blk_mq_force_complete_rq is called
from the polling thread in that case and thus there are no starvation
issues.
By open coding raise_blk_irq in the only caller, and replacing the
ifdef CONFIG_SMP with an IS_ENABLED check the flow in the caller
can be significantly simplified.
__blk_complete_request is only called from the blk-mq code, and
duplicates a lot of code from blk-mq.c. Move it there to prepare
for better code sharing and simplifications.
Linus Torvalds [Sun, 21 Jun 2020 22:41:24 +0000 (15:41 -0700)]
Merge tag 'selinux-pr-20200621' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux
Pull SELinux fixes from Paul Moore:
"Three small patches to fix problems in the SELinux code, all found via
clang.
Two patches fix potential double-free conditions and one fixes an
undefined return value"
* tag 'selinux-pr-20200621' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux:
selinux: fix undefined return of cond_evaluate_expr
selinux: fix a double free in cond_read_node()/cond_read_list()
selinux: fix double free
Linus Torvalds [Sun, 21 Jun 2020 20:04:57 +0000 (13:04 -0700)]
Merge tag 'pinctrl-v5.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl
Pull pin control fixes from Linus Walleij:
"Some early fixes collected during the first week after the merge
window, all pretty self-evident, with the details below. The revert is
the crucial thing.
- Fix a warning on the Qualcomm SPMI GPIO chip being instatiated
twice without a unique irqchip struct
- Use the noirq variants of the suspend and resume callbacks in the
Tegra driver
- Clean up the errorpath on the MCP23s08 driver
- Revert the use of devm_of_iomap() in the Freescale driver as it was
regressing the platform
- Add some missing pins in the Qualcomm IPQ6018 driver
- Fix a simple documentation bug in the pinctrl-single driver"
* tag 'pinctrl-v5.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
pinctrl: single: fix function name in documentation
pinctrl: qcom: ipq6018 Add missing pins in qpic pin group
Revert "pinctrl: freescale: imx: Use 'devm_of_iomap()' to avoid a resource leak in case of error in 'imx_pinctrl_probe()'"
pinctrl: mcp23s08: Split to three parts: fix ptr_ret.cocci warnings
pinctrl: tegra: Use noirq suspend/resume callbacks
pinctrl: qcom: spmi-gpio: fix warning about irq chip reusage
Linus Torvalds [Sun, 21 Jun 2020 19:44:52 +0000 (12:44 -0700)]
Merge tag 'kbuild-fixes-v5.8' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
Pull Kbuild fixes from Masahiro Yamada:
- fix -gz=zlib compiler option test for CONFIG_DEBUG_INFO_COMPRESSED
- improve cc-option in scripts/Kbuild.include to clean up temp files
- improve cc-option in scripts/Kconfig.include for more reliable
compile option test
- do not copy modules.builtin by 'make install' because it would break
existing systems
- use 'userprogs' syntax for watch_queue sample
* tag 'kbuild-fixes-v5.8' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
samples: watch_queue: build sample program for target architecture
Revert "Makefile: install modules.builtin even if CONFIG_MODULES=n"
scripts: Fix typo in headers_install.sh
kconfig: unify cc-option and as-option
kbuild: improve cc-option to clean up all temporary files
Makefile: Improve compressed debug info support detection
Linus Torvalds [Sun, 21 Jun 2020 17:02:53 +0000 (10:02 -0700)]
Merge tag 'powerpc-5.8-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
- One fix for the interrupt rework we did last release which broke
KVM-PR
- Three commits fixing some fallout from the READ_ONCE() changes
interacting badly with our 8xx 16K pages support, which uses a pte_t
that is a structure of 4 actual PTEs
- A cleanup of the 8xx pte_update() to use the newly added pmd_off()
- A fix for a crash when handling an oops if CONFIG_DEBUG_VIRTUAL is
enabled
- A minor fix for the SPU syscall generation
Thanks to Aneesh Kumar K.V, Christian Zigotzky, Christophe Leroy, Mike
Rapoport, Nicholas Piggin.
* tag 'powerpc-5.8-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/8xx: Provide ptep_get() with 16k pages
mm: Allow arches to provide ptep_get()
mm/gup: Use huge_ptep_get() in gup_hugepte()
powerpc/syscalls: Use the number when building SPU syscall table
powerpc/8xx: use pmd_off() to access a PMD entry in pte_update()
powerpc/64s: Fix KVM interrupt using wrong save area
powerpc: Fix kernel crash in show_instructions() w/DEBUG_VIRTUAL
Masahiro Yamada [Wed, 17 Jun 2020 02:08:38 +0000 (11:08 +0900)]
samples: watch_queue: build sample program for target architecture
This userspace program includes UAPI headers exported to usr/include/.
'make headers' always works for the target architecture (i.e. the same
architecture as the kernel), so the sample program should be built for
the target as well. Kbuild now supports 'userprogs' for that.
I also guarded the CONFIG option by 'depends on CC_CAN_LINK' because
$(CC) may not provide libc.
Masahiro Yamada [Fri, 19 Jun 2020 15:09:55 +0000 (00:09 +0900)]
Revert "Makefile: install modules.builtin even if CONFIG_MODULES=n"
This reverts commit e0b250b57dcf403529081e5898a9de717f96b76b,
which broke build systems that need to install files to a certain
path, but do not set INSTALL_MOD_PATH when invoking 'make install'.
While modules.builtin is useful also for CONFIG_MODULES=n, this change
in the behavior is quite unexpected. Maybe "make modules_install"
can install modules.builtin irrespective of CONFIG_MODULES as Jonas
originally suggested.
Linus Torvalds [Sun, 21 Jun 2020 02:23:13 +0000 (19:23 -0700)]
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"One minor fix and two patches reworking the ata dma drain for the
!CONFIG_LIBATA case. The latter is a 5.7 regression fix"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: Wire up ata_scsi_dma_need_drain for SAS HBA drivers
scsi: libata: Provide an ata_scsi_dma_need_drain stub for !CONFIG_ATA
scsi: ufs-bsg: Fix runtime PM imbalance on error
Linus Torvalds [Sun, 21 Jun 2020 02:18:27 +0000 (19:18 -0700)]
Merge branch 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fixes from Wolfram Sang:
- a small collection of remaining API conversion patches (all acked)
which allow to finally remove the deprecated API
- some documentation fixes and a MAINTAINERS addition
* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
MAINTAINERS: Add robert and myself as qcom i2c cci maintainers
i2c: smbus: Fix spelling mistake in the comments
Documentation/i2c: SMBus start signal is S not A
i2c: remove deprecated i2c_new_device API
Documentation: media: convert to use i2c_new_client_device()
video: backlight: tosa_lcd: convert to use i2c_new_client_device()
x86/platform/intel-mid: convert to use i2c_new_client_device()
drm: encoder_slave: use new I2C API
drm: encoder_slave: fix refcouting error for modules
Drew Fustini [Fri, 12 Jun 2020 11:27:58 +0000 (13:27 +0200)]
pinctrl: single: fix function name in documentation
Use the correct the function name in the documentation for
"pcs_parse_one_pinctrl_entry()".
"smux_parse_one_pinctrl_entry()" appears to be an artifact from the
development of a prior patch series ("simple pinmux driver") which
transformed into pinctrl-single.
Linus Torvalds [Sat, 20 Jun 2020 20:17:47 +0000 (13:17 -0700)]
Merge tag 'trace-v5.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing fixes from Steven Rostedt:
- Have recordmcount work with > 64K sections (to support LTO)
- kprobe RCU fixes
- Correct a kprobe critical section with missing mutex
- Remove redundant arch_disarm_kprobe() call
- Fix lockup when kretprobe triggers within kprobe_flush_task()
- Fix memory leak in fetch_op_data operations
- Fix sleep in atomic in ftrace trace array sample code
- Free up memory on failure in sample trace array code
- Fix incorrect reporting of function_graph fields in format file
- Fix quote within quote parsing in bootconfig
- Fix return value of bootconfig tool
- Add testcases for bootconfig tool
- Fix maybe uninitialized warning in ftrace pid file code
- Remove unused variable in tracing_iter_reset()
- Fix some typos
* tag 'trace-v5.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ftrace: Fix maybe-uninitialized compiler warning
tools/bootconfig: Add testcase for show-command and quotes test
tools/bootconfig: Fix to return 0 if succeeded to show the bootconfig
tools/bootconfig: Fix to use correct quotes for value
proc/bootconfig: Fix to use correct quotes for value
tracing: Remove unused event variable in tracing_iter_reset
tracing/probe: Fix memleak in fetch_op_data operations
trace: Fix typo in allocate_ftrace_ops()'s comment
tracing: Make ftrace packed events have align of 1
sample-trace-array: Remove trace_array 'sample-instance'
sample-trace-array: Fix sleeping function called from invalid context
kretprobe: Prevent triggering kretprobe from within kprobe_flush_task
kprobes: Remove redundant arch_disarm_kprobe() call
kprobes: Fix to protect kick_kprobe_optimizer() by kprobe_mutex
kprobes: Use non RCU traversal APIs on kprobe_tables if possible
kprobes: Suppress the suspicious RCU warning on kprobes
recordmcount: support >64k sections
Linus Torvalds [Sat, 20 Jun 2020 20:13:21 +0000 (13:13 -0700)]
Merge tag 'libnvdimm-for-5.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull libnvdimm updates from Dan Williams:
"A feature (papr_scm health retrieval) and a fix (sysfs attribute
visibility) for v5.8.
Vaibhav explains in the merge commit below why missing v5.8 would be
painful and I agreed to try a -rc2 pull because only cosmetics kept
this out of -rc1 and his initial versions were posted in more than
enough time for v5.8 consideration:
'These patches are tied to specific features that were committed to
customers in upcoming distros releases (RHEL and SLES) whose
time-lines are tied to 5.8 kernel release.
Being able to track the health of an nvdimm is critical for our
customers that are running workloads leveraging papr-scm nvdimms.
Missing the 5.8 kernel would mean missing the distro timelines and
shifting forward the availability of this feature in distro kernels
by at least 6 months'
Summary:
- Fix the visibility of the region 'align' attribute.
The new unit tests for region alignment handling caught a corner
case where the alignment cannot be specified if the region is
converted from static to dynamic provisioning at runtime.
- Add support for device health retrieval for the persistent memory
supported by the papr_scm driver.
This includes both the standard sysfs "health flags" that the nfit
persistent memory driver publishes and a mechanism for the ndctl
tool to retrieve a health-command payload"
* tag 'libnvdimm-for-5.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
nvdimm/region: always show the 'align' attribute
powerpc/papr_scm: Implement support for PAPR_PDSM_HEALTH
ndctl/papr_scm,uapi: Add support for PAPR nvdimm specific methods
powerpc/papr_scm: Improve error logging and handling papr_scm_ndctl()
powerpc/papr_scm: Fetch nvdimm health information from PHYP
seq_buf: Export seq_buf_printf
powerpc: Document details on H_SCM_HEALTH hcall
After commit 26d8cde5260b ("pinctrl: freescale: imx: add shared
input select reg support"). i.MX7D has two iomux controllers
iomuxc and iomuxc-lpsr which share select_input register for
daisy chain settings.
If use 'devm_of_iomap()', when probe the iomuxc-lpsr, will call
devm_request_mem_region() for the region <0x30330000-0x3033ffff>
for the first time. Then, next time when probe the iomuxc, API
devm_platform_ioremap_resource() will also use the API
devm_request_mem_region() for the share region <0x30330000-0x3033ffff>
again, then cause issue, log like below:
[ 0.179561] imx7d-pinctrl 302c0000.iomuxc-lpsr: initialized IMX pinctrl driver
[ 0.191742] imx7d-pinctrl 30330000.pinctrl: can't request region for resource [mem 0x30330000-0x3033ffff]
[ 0.191842] imx7d-pinctrl: probe of 30330000.pinctrl failed with error -16
Linus Torvalds [Sat, 20 Jun 2020 19:31:08 +0000 (12:31 -0700)]
Merge tag 's390-5.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 fixes from Vasily Gorbik:
- a few ptrace fixes mostly for strace and seccomp_bpf kernel tests
findings
- cleanup unused pm callbacks in virtio ccw
- replace kmalloc + memset with kzalloc in crypto
- use $(LD) for vDSO linkage to make clang happy
- fix vDSO clock_getres() to preserve the same behaviour as
posix_get_hrtimer_res()
- fix workqueue cpumask warning when NUMA=n and nr_node_ids=2
- reduce SLSB writes during input processing, improve warnings and
cleanup qdio_data usage in qdio
- a few fixes to use scnprintf() instead of snprintf()
* tag 's390-5.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
s390: fix syscall_get_error for compat processes
s390/qdio: warn about unexpected SLSB states
s390/qdio: clean up usage of qdio_data
s390/numa: let NODES_SHIFT depend on NEED_MULTIPLE_NODES
s390/vdso: fix vDSO clock_getres()
s390/vdso: Use $(LD) instead of $(CC) to link vDSO
s390/protvirt: use scnprintf() instead of snprintf()
s390: use scnprintf() in sys_##_prefix##_##_name##_show
s390/crypto: use scnprintf() instead of snprintf()
s390/zcrypt: use kzalloc
s390/virtio: remove unused pm callbacks
s390/qdio: reduce SLSB writes during Input Queue processing
selftests/seccomp: s390 shares the syscall and return value register
s390/ptrace: fix setting syscall number
s390/ptrace: pass invalid syscall numbers to tracing
s390/ptrace: return -ENOSYS when invalid syscall is supplied
s390/seccomp: pass syscall arguments via seccomp_data
s390/qdio: fine-tune SLSB update
Linus Torvalds [Sat, 20 Jun 2020 19:14:29 +0000 (12:14 -0700)]
Merge tag 'riscv-for-linus-5.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V fixes from Palmer Dabbelt:
- a workaround for a compiler surprise related to the "r" inline
assembly that allows LLVM to boot.
- a fix to avoid WX-only mappings, which the ISA does not allow. While
this probably manifests in many ways, the bug was found in stress-ng.
- a missing lock in set_direct_map_*(), which due to a recent lockdep
change started asserting.
* tag 'riscv-for-linus-5.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
RISC-V: Acquire mmap lock before invoking walk_page_range
RISC-V: Don't allow write+exec only page mapping request in mmap
riscv/atomic: Fix sign extension for RV64I
Linus Torvalds [Sat, 20 Jun 2020 19:10:09 +0000 (12:10 -0700)]
Merge tag 'linux-kselftest-5.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest
Pull kselftest cleanups from Shuah Khan:
- ftrace "requires:" list for simplifying and unifying requirement
checks for each test case, adding "requires:" line instead of
checking required ftrace interfaces in each test case.
- a minor spelling correction patch
* tag 'linux-kselftest-5.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
selftests/ftrace: Support ":README" suffix for requires
selftests/ftrace: Support ":tracer" suffix for requires
selftests/ftrace: Convert check_filter_file() with requires list
selftests/ftrace: Convert required interface checks into requires list
selftests/ftrace: Add "requires:" list support
selftests/ftrace: Return unsupported for the unconfigured features
selftests/ftrace: Allow ":" in description
tools: testing: ftrace: trigger: fix spelling mistake
David Howells [Fri, 19 Jun 2020 22:39:36 +0000 (23:39 +0100)]
afs: Fix hang on rmmod due to outstanding timer
The fileserver probe timer, net->fs_probe_timer, isn't cancelled when
the kafs module is being removed and so the count it holds on
net->servers_outstanding doesn't get dropped..
This causes rmmod to wait forever. The hung process shows a stack like:
(1) Attempting to cancel the probe timer and, if successful, drop the
count that the timer was holding.
(2) Make the timer function just drop the count and not schedule the
prober if the afs portion of net namespace is being destroyed.
Also, whilst we're at it, make the following changes:
(3) Initialise net->servers_outstanding to 1 and decrement it before
waiting on it so that it doesn't generate wake up events by being
decremented to 0 until we're cleaning up.
(4) Switch the atomic_dec() on ->servers_outstanding for ->fs_timer in
afs_purge_servers() to use the helper function for that.
Fixes: f6cbb368bcb0 ("afs: Actively poll fileservers to maintain NAT or firewall openings") Signed-off-by: David Howells <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
David Howells [Thu, 18 Jun 2020 23:01:28 +0000 (00:01 +0100)]
afs: Fix afs_do_lookup() to call correct fetch-status op variant
Fix afs_do_lookup()'s fallback case for when FS.InlineBulkStatus isn't
supported by the server.
In the fallback, it calls FS.FetchStatus for the specific vnode it's
meant to be looking up. Commit b6489a49f7b7 broke this by renaming one
of the two identically-named afs_fetch_status_operation descriptors to
something else so that one of them could be made non-static. The site
that used the renamed one, however, wasn't renamed and didn't produce
any warning because the other was declared in a header.
Fix this by making afs_do_lookup() use the renamed variant.
Note that there are two variants of the success method because one is
called from ->lookup() where we may or may not have an inode, but can't
call iget until after we've talked to the server - whereas the other is
called from within iget where we have an inode, but it may or may not be
initialised.
The latter variant expects there to be an inode, but because it's being
called from there former case, there might not be - resulting in an oops
like the following:
Christophe Leroy [Mon, 15 Jun 2020 12:57:59 +0000 (12:57 +0000)]
powerpc/8xx: Provide ptep_get() with 16k pages
READ_ONCE() now enforces atomic read, which leads to:
CC mm/gup.o
In file included from ./include/linux/kernel.h:11:0,
from mm/gup.c:2:
In function 'gup_hugepte.constprop',
inlined from 'gup_huge_pd.isra.79' at mm/gup.c:2465:8:
./include/linux/compiler.h:392:38: error: call to '__compiletime_assert_222' declared with attribute error: Unsupported access size for {READ,WRITE}_ONCE().
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:373:4: note: in definition of macro '__compiletime_assert'
prefix ## suffix(); \
^
./include/linux/compiler.h:392:2: note: in expansion of macro '_compiletime_assert'
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:405:2: note: in expansion of macro 'compiletime_assert'
compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
^
./include/linux/compiler.h:291:2: note: in expansion of macro 'compiletime_assert_rwonce_type'
compiletime_assert_rwonce_type(x); \
^
mm/gup.c:2428:8: note: in expansion of macro 'READ_ONCE'
pte = READ_ONCE(*ptep);
^
In function 'gup_get_pte',
inlined from 'gup_pte_range' at mm/gup.c:2228:9,
inlined from 'gup_pmd_range' at mm/gup.c:2613:15,
inlined from 'gup_pud_range' at mm/gup.c:2641:15,
inlined from 'gup_p4d_range' at mm/gup.c:2666:15,
inlined from 'gup_pgd_range' at mm/gup.c:2694:15,
inlined from 'internal_get_user_pages_fast' at mm/gup.c:2795:3:
./include/linux/compiler.h:392:38: error: call to '__compiletime_assert_219' declared with attribute error: Unsupported access size for {READ,WRITE}_ONCE().
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:373:4: note: in definition of macro '__compiletime_assert'
prefix ## suffix(); \
^
./include/linux/compiler.h:392:2: note: in expansion of macro '_compiletime_assert'
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:405:2: note: in expansion of macro 'compiletime_assert'
compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
^
./include/linux/compiler.h:291:2: note: in expansion of macro 'compiletime_assert_rwonce_type'
compiletime_assert_rwonce_type(x); \
^
mm/gup.c:2199:9: note: in expansion of macro 'READ_ONCE'
return READ_ONCE(*ptep);
^
make[2]: *** [mm/gup.o] Error 1
Christophe Leroy [Mon, 15 Jun 2020 12:57:58 +0000 (12:57 +0000)]
mm: Allow arches to provide ptep_get()
Since commit 9e343b467c70 ("READ_ONCE: Enforce atomicity for
{READ,WRITE}_ONCE() memory accesses") it is not possible anymore to
use READ_ONCE() to access complex page table entries like the one
defined for powerpc 8xx with 16k size pages.
Define a ptep_get() helper that architectures can override instead
of performing a READ_ONCE() on the page table entry pointer.
Dan Williams [Fri, 19 Jun 2020 21:18:51 +0000 (14:18 -0700)]
Merge branch 'for-5.8/papr_scm' into libnvdimm-for-next
Include the papr_scm health retrieval feature for v5.8-rc2. The
functionality was initially posted well in advance of the merge window,
but review comments and a late build-bot warning kept them out of the
v5.8-rc1 libnvdimm pull request.
Vaibhav notes:
These patches are tied to specific features that were committed to
customers in upcoming distros releases (RHEL and SLES) whose time-lines
are tied to 5.8 kernel release.
Being able to track the health of an nvdimm is critical for our
customers that are running workloads leveraging papr-scm nvdimms.
Missing the 5.8 kernel would mean missing the distro timelines and
shifting forward the availability of this feature in distro kernels by
at least 6 months.
Linus Torvalds [Fri, 19 Jun 2020 20:16:58 +0000 (13:16 -0700)]
Merge tag 'io_uring-5.8-2020-06-19' of git://git.kernel.dk/linux-block
Pull io_uring fixes from Jens Axboe:
- Catch a case where io_sq_thread() didn't do proper mm acquire
- Ensure poll completions are reaped on shutdown
- Async cancelation and run fixes (Pavel)
- io-poll race fixes (Xiaoguang)
- Request cleanup race fix (Xiaoguang)
* tag 'io_uring-5.8-2020-06-19' of git://git.kernel.dk/linux-block:
io_uring: fix possible race condition against REQ_F_NEED_CLEANUP
io_uring: reap poll completions while waiting for refs to drop on exit
io_uring: acquire 'mm' for task_work for SQPOLL
io_uring: add memory barrier to synchronize io_kiocb's result and iopoll_completed
io_uring: don't fail links for EAGAIN error in IOPOLL mode
io_uring: cancel by ->task not pid
io_uring: lazy get task
io_uring: batch cancel in io_uring_cancel_files()
io_uring: cancel all task's requests on exit
io-wq: add an option to cancel all matched reqs
io-wq: reorder cancellation pending -> running
io_uring: fix lazy work init
Linus Torvalds [Fri, 19 Jun 2020 20:11:26 +0000 (13:11 -0700)]
Merge tag 'block-5.8-2020-06-19' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
- Use import_uuid() where appropriate (Andy)
- bcache fixes (Coly, Mauricio, Zhiqiang)
- blktrace sparse warnings fix (Jan)
- blktrace concurrent setup fix (Luis)
- blkdev_get use-after-free fix (Jason)
- Ensure all blk-mq maps are updated (Weiping)
- Loop invalidate bdev fix (Zheng)
* tag 'block-5.8-2020-06-19' of git://git.kernel.dk/linux-block:
block: make function 'kill_bdev' static
loop: replace kill_bdev with invalidate_bdev
partitions/ldm: Replace uuid_copy() with import_uuid() where it makes sense
block: update hctx map when use multiple maps
blktrace: Avoid sparse warnings when assigning q->blk_trace
blktrace: break out of blktrace setup on concurrent calls
block: Fix use-after-free in blkdev_get()
trace/events/block.h: drop kernel-doc for dropped function parameter
blk-mq: Remove redundant 'return' statement
bcache: pr_info() format clean up in bcache_device_init()
bcache: use delayed kworker fo asynchronous devices registration
bcache: check and adjust logical block size for backing devices
bcache: fix potential deadlock problem in btree_gc_coalesce
Linus Torvalds [Fri, 19 Jun 2020 20:09:40 +0000 (13:09 -0700)]
Merge tag 'libata-5.8-2020-06-19' of git://git.kernel.dk/linux-block
Pull libata fixes from Jens Axboe:
"A few minor changes that should go into this release"
* tag 'libata-5.8-2020-06-19' of git://git.kernel.dk/linux-block:
libata: Use per port sync for detach
ata/libata: Fix usage of page address by page_address in ata_scsi_mode_select_xlat function
sata_rcar: handle pm_runtime_get_sync failure cases
Linus Torvalds [Fri, 19 Jun 2020 19:40:57 +0000 (12:40 -0700)]
Merge tag 'drm-fixes-2020-06-19' of git://anongit.freedesktop.org/drm/drm
Pull drm fixes from Dave Airlie:
"Just i915 and amd here.
i915 has some workaround movement so they get applied at the right
times, and a timeslicing fix, along with some display fixes.
AMD has a few display floating point fix and a devcgroup fix for
amdkfd.
i915:
- Fix for timeslicing and virtual engines/unpremptable requests (+ 1
dependency patch)
- Fixes into TypeC register programming and interrupt storm detecting
- Disable DIP on MST ports with the transcoder clock still on
- Avoid missing GT workarounds at reset for HSW and older gens
- Fix for unwinding multiple requests missing force restore
- Fix encoder type check for DDI vswing sequence
- Build warning fixes
amdgpu:
- Fix kvfree/kfree mixup
- Fix hawaii device id in powertune configuration
- Display FP fixes
- Documentation fixes
amdkfd:
- devcgroup check fix"
* tag 'drm-fixes-2020-06-19' of git://anongit.freedesktop.org/drm/drm: (23 commits)
drm/amdgpu: fix documentation around busy_percentage
drm/amdgpu/pm: update comment to clarify Overdrive interfaces
drm/amdkfd: Use correct major in devcgroup check
drm/i915/display: Fix the encoder type check
drm/i915/icl+: Fix hotplug interrupt disabling after storm detection
drm/i915/gt: Move gen4 GT workarounds from init_clock_gating to workarounds
drm/i915/gt: Move ilk GT workarounds from init_clock_gating to workarounds
drm/i915/gt: Move snb GT workarounds from init_clock_gating to workarounds
drm/i915/gt: Move vlv GT workarounds from init_clock_gating to workarounds
drm/i915/gt: Move ivb GT workarounds from init_clock_gating to workarounds
drm/i915/gt: Move hsw GT workarounds from init_clock_gating to workarounds
drm/i915/icl: Disable DIP on MST ports with the transcoder clock still on
drm/i915/gt: Incrementally check for rewinding
drm/i915/tc: fix the reset of ln0
drm/i915/gt: Prevent timeslicing into unpreemptable requests
drm/i915/selftests: Restore to default heartbeat
drm/i915: work around false-positive maybe-uninitialized warning
drm/i915/pmu: avoid an maybe-uninitialized warning
drm/i915/gt: Incorporate the virtual engine into timeslicing
drm/amd/display: Rework dsc to isolate FPU operations
...
Linus Torvalds [Fri, 19 Jun 2020 19:25:04 +0000 (12:25 -0700)]
Merge tag 'ceph-for-5.8-rc2' of git://github.com/ceph/ceph-client
Pull ceph fixes from Ilya Dryomov:
"An important follow-up for replica reads support that went into -rc1
and two target_copy() fixups"
* tag 'ceph-for-5.8-rc2' of git://github.com/ceph/ceph-client:
libceph: don't omit used_replica in target_copy()
libceph: don't omit recovery_deletes in target_copy()
libceph: move away from global osd_req_flags
Linus Torvalds [Fri, 19 Jun 2020 19:19:12 +0000 (12:19 -0700)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"Unfortunately, we still have a number of outstanding issues so there
will be more fixes to come, but this lot are a good start.
- Fix handling of watchpoints triggered by uaccess routines
- Fix initialisation of gigantic pages for CMA buffers
- Raise minimum clang version for BTI to avoid miscompilation
- Fix data race in SVE vector length configuration code
- Ensure address tags are ignored in kern_addr_valid()
- Dump register state on fatal BTI exception
- kexec_file() cleanup to use struct_size() macro"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: hw_breakpoint: Don't invoke overflow handler on uaccess watchpoints
arm64: kexec_file: Use struct_size() in kmalloc()
arm64: mm: reserve hugetlb CMA after numa_init
arm64: bti: Require clang >= 10.0.1 for in-kernel BTI support
arm64: sve: Fix build failure when ARM64_SVE=y and SYSCTL=n
arm64: pgtable: Clear the GP bit for non-executable kernel pages
arm64: mm: reset address tag set by kasan sw tagging
arm64: traps: Dump registers prior to panic() in bad_mode()
arm64/sve: Eliminate data races on sve_default_vl
docs/arm64: Fix typo'd #define in sve.rst
arm64: remove TEXT_OFFSET randomization
Linus Torvalds [Fri, 19 Jun 2020 18:45:03 +0000 (11:45 -0700)]
Merge tag 'overflow-v5.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux
Pull flex-array size helper from Kees Cook:
"During the treewide clean-ups of zero-length "flexible arrays", the
struct_size() helper was heavily used, but it was noticed that many
times it would have been nice to have an additional helper to get the
size of just the flexible array itself.
This need appears to be even more common when cleaning up the 1-byte
array "flexible arrays", so Gustavo implemented it.
I'd love to get this landed early so it can be used during the v5.9
dev cycle to ease the 1-byte array cleanups."
* tag 'overflow-v5.8-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
overflow.h: Add flex_array_size() helper
Linus Torvalds [Fri, 19 Jun 2020 18:39:57 +0000 (11:39 -0700)]
Merge tag 'perf-tools-fixes-2020-06-02' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux
Pull perf tooling fixes from Arnaldo Carvalho de Melo:
- Update various UAPI headers, some automatically adding support for a
new MSR and the faccess2 syscall.
- Fix corner case NULL deref in the histograms code.
- Fix corner case NULL deref in 'perf stat' aggregation code.
- Fix array pointer deref and old style declaration in the parsing of
events.
- Fix segfault when processing ZSTD compressed perf.data files in 'perf
script' due to lack of initialization of the ZSTD library.
- Handle __attribute__((user)) in libtraceevent fixing the parsing of
syscall tracepoints with user buffers.
- Make libtraevent aware of __builtin_expect() appearing in tracepoint
fields.
- Make the BPF prologue generation use bpf_probe_read_{user,kernel}().
- Fix the '@user' attribute parsing in kprobes variables in 'perf
probe'.
- Fix error message when asking for -fsanitize=address without required
libraries.
* tag 'perf-tools-fixes-2020-06-02' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (22 commits)
perf build: Fix error message when asking for -fsanitize=address without required libraries
tools lib traceevent: Add handler for __builtin_expect()
tools lib traceevent: Handle __attribute__((user)) in field names
tools lib traceevent: Add append() function helper for appending strings
tools headers UAPI: Sync linux/fs.h with the kernel sources
tools include UAPI: Sync linux/vhost.h with the kernel sources
tools arch x86: Sync the msr-index.h copy with the kernel sources
perf script: Initialize zstd_data
perf pmu: Remove unused declaration
perf parse-events: Fix an old style declaration
perf parse-events: Fix an incompatible pointer
perf bpf: Fix bpf prologue generation
perf probe: Fix user attribute access in kprobes
perf stat: Fix NULL pointer dereference
perf report: Fix NULL pointer dereference in hists__fprintf_nr_sample_events()
tools headers UAPI: Sync kvm.h headers with the kernel sources
tools headers UAPI: Sync drm/i915_drm.h with the kernel sources
tools headers UAPI: Sync linux/fscrypt.h with the kernel sources
perf beauty: Add support to STATX_MNT_ID in the 'statx' syscall 'mask' argument
tools headers uapi: Sync linux/stat.h with the kernel sources
...
Wolfram Sang [Mon, 15 Jun 2020 07:58:11 +0000 (09:58 +0200)]
drm: encoder_slave: use new I2C API
i2c_new_client() is deprecated, use the replacement
i2c_new_client_device(). Also, we have a helper to check if a driver is
bound. Use it to simplify the code. Note that this changes the errno for
a failed device creation from ENOMEM to ENODEV. No callers currently
interpret this errno, though, so we use this condensed error check.
Atish Patra [Wed, 17 Jun 2020 20:37:32 +0000 (13:37 -0700)]
RISC-V: Acquire mmap lock before invoking walk_page_range
As per walk_page_range documentation, mmap lock should be acquired by the
caller before invoking walk_page_range. mmap_assert_locked gets triggered
without that. The details can be found here.
Yash Shah [Tue, 16 Jun 2020 14:03:06 +0000 (19:33 +0530)]
RISC-V: Don't allow write+exec only page mapping request in mmap
As per the table 4.4 of version "20190608-Priv-MSU-Ratified" of the
RISC-V instruction set manual[0], the PTE permission bit combination of
"write+exec only" is reserved for future use. Hence, don't allow such
mapping request in mmap call.
An issue is been reported by David Abdurachmanov, that while running
stress-ng with "sysbadaddr" argument, RCU stalls are observed on RISC-V
specific kernel.
This issue arises when the stress-sysbadaddr request for pages with
"write+exec only" permission bits and then passes the address obtain
from this mmap call to various system call. For the riscv kernel, the
mmap call should fail for this particular combination of permission bits
since it's not valid.
Signed-off-by: Yash Shah <[email protected]> Reported-by: David Abdurachmanov <[email protected]>
[Palmer: Refer to the latest ISA specification at the only link I could
find, and update the terminology.] Signed-off-by: Palmer Dabbelt <[email protected]>
projects / linux.git / log