Tomi Valkeinen [Tue, 24 Sep 2019 13:17:02 +0000 (16:17 +0300)]
drm/bridge: tc358767: fix max_tu_symbol value
max_tu_symbol was programmed to TU_SIZE_RECOMMENDED - 1, which is not
what the spec says. The spec says:
roundup ((input active video bandwidth in bytes/output active video
bandwidth in bytes) * tu_size)
It is not quite clear what the above means, but calculating
max_tu_symbol = (input Bps / output Bps) * tu_size seems to work and
fixes the issues seen.
This fixes artifacts in some videomodes (e.g. 1024x768@60 on 2-lanes &
1.62Gbps was pretty bad for me).
Halil Pasic [Mon, 30 Sep 2019 15:38:02 +0000 (17:38 +0200)]
s390/cio: fix virtio-ccw DMA without PV
Commit 37db8985b211 ("s390/cio: add basic protected virtualization
support") breaks virtio-ccw devices with VIRTIO_F_IOMMU_PLATFORM for non
Protected Virtualization (PV) guests. The problem is that the dma_mask
of the ccw device, which is used by virtio core, gets changed from 64 to
31 bit, because some of the DMA allocations do require 31 bit
addressable memory. For PV the only drawback is that some of the virtio
structures must end up in ZONE_DMA because we have the bounce the
buffers mapped via DMA API anyway.
But for non PV guests we have a problem: because of the 31 bit mask
guests bigger than 2G are likely to try bouncing buffers. The swiotlb
however is only initialized for PV guests, because we don't want to
bounce anything for non PV guests. The first such map kills the guest.
Since the DMA API won't allow us to specify for each allocation whether
we need memory from ZONE_DMA (31 bit addressable) or any DMA capable
memory will do, let us use coherent_dma_mask (which is used for
allocations) to force allocating form ZONE_DMA while changing dma_mask
to DMA_BIT_MASK(64) so that at least the streaming API will regard
the whole memory DMA capable.
Keith Busch [Wed, 9 Oct 2019 15:38:13 +0000 (00:38 +0900)]
null_blk: Fix zoned command return code
The return code from null_handle_zoned() sets the cmd->error value.
Returning OK status when an error occured overwrites the intended
cmd->error. Return the appropriate error code instead of setting the
error in the cmd.
Fixes: fceb5d1b19cbe626 ("null_blk: create a helper for zoned devices") Cc: Chaitanya Kulkarni <[email protected]> Signed-off-by: Keith Busch <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
Chris Wilson [Wed, 9 Oct 2019 16:09:06 +0000 (17:09 +0100)]
drm/i915/gt: execlists->active is serialised by the tasklet
The active/pending execlists is no longer protected by the
engine->active.lock, but is serialised by the tasklet instead. Update
the locking around the debug and stats to follow suit.
v2: local_bh_disable() to prevent recursing into the tasklet in case we
trigger a softirq (Tvrtko)
Chris Wilson [Wed, 9 Oct 2019 10:09:54 +0000 (11:09 +0100)]
drm/i915/execlists: Protect peeking at execlists->active
Now that we dropped the engine->active.lock serialisation from around
process_csb(), direct submission can run concurrently to the interrupt
handler. As such execlists->active may be advanced as we dequeue,
dropping the reference to the request. We need to employ our RCU request
protection to ensure that the request is not freed too early.
Chris Wilson [Mon, 23 Sep 2019 15:28:43 +0000 (16:28 +0100)]
drm/i915: Fixup preempt-to-busy vs reset of a virtual request
Due to the nature of preempt-to-busy the execlists active tracking and
the schedule queue may become temporarily desync'ed (between resubmission
to HW and its ack from HW). This means that we may have unwound a
request and passed it back to the virtual engine, but it is still
inflight on the HW and may even result in a GPU hang. If we detect that
GPU hang and try to reset, the hanging request->engine will no longer
match the current engine, which means that the request is not on the
execlists active list and we should not try to find an older incomplete
request. Given that we have deduced this must be a request on a virtual
engine, it is the single active request in the context and so must be
guilty (as the context is still inflight, it is prevented from being
executed on another engine as we process the reset).
Chris Wilson [Mon, 23 Sep 2019 11:00:55 +0000 (12:00 +0100)]
drm/i915: Only enqueue already completed requests
If we are asked to submit a completed request, just move it onto the
active-list without modifying it's payload. If we try to emit the
modified payload of a completed request, we risk racing with the
ring->head update during retirement which may advance the head past our
breadcrumb and so we generate a warning for the emission being behind
the RING_HEAD.
v2: Commentary for the sneaky, shared responsibility between functions.
v3: Spelling mistakes and bonus assertion
Chris Wilson [Mon, 23 Sep 2019 11:00:54 +0000 (12:00 +0100)]
drm/i915/execlists: Drop redundant list_del_init(&rq->sched.link)
Since amalgamating the queued and active lists in commit 422d7df4f090
("drm/i915: Replace engine->timeline with a plain list"), performing a
i915_request_submit() will remove the request from the execlists
priority queue.
Matt Roper [Mon, 16 Sep 2019 23:32:51 +0000 (16:32 -0700)]
drm/i915/cml: Add second PCH ID for CMP
The CMP PCH ID we have in the driver is correct for the CML-U machines we have
in our CI system, but the CML-S and CML-H CI machines appear to use a
different PCH ID, leading our driver to detect no PCH for them.
Commit ac7c3e4ff401 ("compiler: enable CONFIG_OPTIMIZE_INLINING
forcibly") allows compiler to uninline functions marked as 'inline'.
In cace of __xchg this would cause to reference function
__xchg_called_with_bad_pointer, which is an error case
for catching bugs and will not happen for correct code, if
__xchg is inlined.
iio: Fix an undefied reference error in noa1305_probe
I hit the following error when compile the kernel.
drivers/iio/light/noa1305.o: In function `noa1305_probe':
noa1305.c:(.text+0x65): undefined reference to `__devm_regmap_init_i2c'
make: *** [vmlinux] Error 1
David Frey [Thu, 19 Sep 2019 22:54:18 +0000 (15:54 -0700)]
iio: light: opt3001: fix mutex unlock race
When an end-of-conversion interrupt is received after performing a
single-shot reading of the light sensor, the driver was waking up the
result ready queue before checking opt->ok_to_ignore_lock to determine
if it should unlock the mutex. The problem occurred in the case where
the other thread woke up and changed the value of opt->ok_to_ignore_lock
to false prior to the interrupt thread performing its read of the
variable. In this case, the mutex would be unlocked twice.
Marco Felsch [Tue, 17 Sep 2019 16:09:23 +0000 (18:09 +0200)]
iio: adc: ad799x: fix probe error handling
Since commit 0f7ddcc1bff1 ("iio:adc:ad799x: Write default config on probe
and reset alert status on probe") the error path is wrong since it
leaves the vref regulator on. Fix this by disabling both regulators.
Fixes: 0f7ddcc1bff1 ("iio:adc:ad799x: Write default config on probe and reset alert status on probe") Signed-off-by: Marco Felsch <[email protected]> Reviewed-by: Alexandru Ardelean <[email protected]> Cc: <[email protected]> Signed-off-by: Jonathan Cameron <[email protected]>
Marco Felsch [Tue, 17 Sep 2019 14:56:37 +0000 (16:56 +0200)]
iio: light: add missing vcnl4040 of_compatible
Commit 5a441aade5b3 ("iio: light: vcnl4000 add support for the VCNL4040
proximity and light sensor") added the support for the vcnl4040 but
forgot to add the of_compatible. Fix this by adding it now.
Signed-off-by: Marco Felsch <[email protected]> Fixes: 5a441aade5b3 ("iio: light: vcnl4000 add support for the VCNL4040 proximity and light sensor") Reviewed-by: Angus Ainslie (Purism) [email protected] Cc: <[email protected]> Signed-off-by: Jonathan Cameron <[email protected]>
Marco Felsch [Tue, 17 Sep 2019 14:56:36 +0000 (16:56 +0200)]
iio: light: fix vcnl4000 devicetree hooks
Since commit ebd457d55911 ("iio: light: vcnl4000 add devicetree hooks")
the of_match_table is supported but the data shouldn't be a string.
Instead it shall be one of 'enum vcnl4000_device_ids'. Also the matching
logic for the vcnl4020 was wrong. Since the data retrieve mechanism is
still based on the i2c_device_id no failures did appeared till now.
Lorenzo Bianconi [Mon, 16 Sep 2019 22:01:29 +0000 (00:01 +0200)]
iio: imu: st_lsm6dsx: fix waitime for st_lsm6dsx i2c controller
i2c controller available in st_lsm6dsx series performs i2c slave
configuration using accel clock as trigger.
st_lsm6dsx_shub_wait_complete routine is used to wait the controller has
carried out the requested configuration. However if the accel sensor is not
enabled we should not use its configured odr to estimate a proper timeout
Fixes: c91c1c844ebd ("iio: imu: st_lsm6dsx: add i2c embedded controller support") Signed-off-by: Lorenzo Bianconi <[email protected]> Signed-off-by: Jonathan Cameron <[email protected]>
Hans de Goede [Sun, 15 Sep 2019 18:53:42 +0000 (20:53 +0200)]
iio: adc: axp288: Override TS pin bias current for some models
Since commit 9bcf15f75cac ("iio: adc: axp288: Fix TS-pin handling") we
preserve the bias current set by the firmware at boot. This fixes issues
we were seeing on various models, but it seems our old hardcoded 80ųA bias
current was working around a firmware bug on at least one model laptop.
In order to both have our cake and eat it, this commit adds a dmi based
list of models where we need to override the firmware set bias current and
adds the one model we now know needs this to it: The Lenovo Ideapad 100S
(11 inch version).
iio: adc: stm32-adc: fix a race when using several adcs with dma and irq
End of conversion may be handled by using IRQ or DMA. There may be a
race when two conversions complete at the same time on several ADCs.
EOC can be read as 'set' for several ADCs, with:
- an ADC configured to use IRQs. EOCIE bit is set. The handler is normally
called in this case.
- an ADC configured to use DMA. EOCIE bit isn't set. EOC triggers the DMA
request instead. It's then automatically cleared by DMA read. But the
handler gets called due to status bit is temporarily set (IRQ triggered
by the other ADC).
So both EOC status bit in CSR and EOCIE control bit must be checked
before invoking the interrupt handler (e.g. call ISR only for
IRQ-enabled ADCs).
Move STM32 ADC registers definitions to common header.
This is precursor patch to:
- iio: adc: stm32-adc: fix a race when using several adcs with dma and irq
It keeps registers definitions as a whole block, to ease readability and
allow simple access path to EOC bits (readl) in stm32-adc-core driver.
Stefan Popa [Tue, 10 Sep 2019 14:44:46 +0000 (17:44 +0300)]
iio: accel: adxl372: Perform a reset at start up
We need to perform a reset a start up to make sure that the chip is in a
consistent state. This reset also disables all the interrupts which
should only be enabled together with the iio buffer. Not doing this, was
sometimes causing unwanted interrupts to trigger.
Stefan Popa [Tue, 10 Sep 2019 14:44:21 +0000 (17:44 +0300)]
iio: accel: adxl372: Fix push to buffers lost samples
One in two sample sets was lost by multiplying fifo_set_size with
sizeof(u16). Also, the double number of available samples were pushed to
the iio buffers.
Stefan Popa [Tue, 10 Sep 2019 14:43:32 +0000 (17:43 +0300)]
iio: accel: adxl372: Fix/remove limitation for FIFO samples
Currently, the driver sets the FIFO_SAMPLES register with the number of
sample sets (maximum of 170 for 3 axis data, 256 for 2-axis and 512 for
single axis). However, the FIFO_SAMPLES register should store the number
of samples, regardless of how the FIFO format is configured.
Andreas Klinger [Mon, 9 Sep 2019 12:37:21 +0000 (14:37 +0200)]
iio: adc: hx711: fix bug in sampling of data
Fix bug in sampling function hx711_cycle() when interrupt occures while
PD_SCK is high. If PD_SCK is high for at least 60 us power down mode of
the sensor is entered which in turn leads to a wrong measurement.
Switch off interrupts during a PD_SCK high period and move query of DOUT
to the latest point of time which is at the end of PD_SCK low period.
This bug exists in the driver since it's initial addition. The more
interrupts on the system the higher is the probability that it happens.
Linus Torvalds [Wed, 9 Oct 2019 16:46:46 +0000 (09:46 -0700)]
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
Pull rdma fixes from Jason Gunthorpe:
"The usual collection of driver bug fixes, and a few regressions from
the merge window. Nothing particularly worrisome.
- Various missed memory frees and error unwind bugs
- Fix regressions in a few iwarp drivers from 5.4 patches
- A few regressions added in past kernels
- Squash a number of races in mlx5 ODP code"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
RDMA/mlx5: Add missing synchronize_srcu() for MW cases
RDMA/mlx5: Put live in the correct place for ODP MRs
RDMA/mlx5: Order num_pending_prefetch properly with synchronize_srcu
RDMA/odp: Lift umem_mutex out of ib_umem_odp_unmap_dma_pages()
RDMA/mlx5: Fix a race with mlx5_ib_update_xlt on an implicit MR
RDMA/mlx5: Do not allow rereg of a ODP MR
IB/core: Fix wrong iterating on ports
RDMA/nldev: Reshuffle the code to avoid need to rebind QP in error path
RDMA/cxgb4: Do not dma memory off of the stack
RDMA/cm: Fix memory leak in cm_add/remove_one
RDMA/core: Fix an error handling path in 'res_get_common_doit()'
RDMA/i40iw: Associate ibdev to netdev before IB device registration
RDMA/iwcm: Fix a lock inversion issue
RDMA/iw_cxgb4: fix SRQ access from dump_qp()
RDMA/hfi1: Prevent memory leak in sdma_init
RDMA/core: Fix use after free and refcnt leak on ndev in_device in iwarp_query_port
RDMA/siw: Fix serialization issue in write_space()
RDMA/vmw_pvrdma: Free SRQ only once
Linus Torvalds [Wed, 9 Oct 2019 16:27:22 +0000 (09:27 -0700)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"A larger-than-usual batch of arm64 fixes for -rc3.
The bulk of the fixes are dealing with a bunch of issues with the
build system from the compat vDSO, which unfortunately led to some
significant Makefile rework to manage the horrible combinations of
toolchains that we can end up needing to drive simultaneously.
We came close to disabling the thing entirely, but Vincenzo was quick
to spin up some patches and I ended up picking up most of the bits
that were left [*]. Future work will look at disentangling the header
files properly.
Other than that, we have some important fixes all over, including one
papering over the miscompilation fallout from forcing
CONFIG_OPTIMIZE_INLINING=y, which I'm still unhappy about. Harumph.
We've still got a couple of open issues, so I'm expecting to have some
more fixes later this cycle.
Summary:
- Numerous fixes to the compat vDSO build system, especially when
combining gcc and clang
- Fix parsing of PAR_EL1 in spurious kernel fault detection
- Partial workaround for Neoverse-N1 erratum #1542419
- Fix IRQ priority masking on entry from compat syscalls
- Fix advertisment of FRINT HWCAP to userspace
- Attempt to workaround inlining breakage with '__always_inline'
- Fix accidental freeing of parent SVE state on fork() error path
- Add some missing NULL pointer checks in instruction emulation init
- Some formatting and comment fixes"
[*] Will's final fixes were
Reviewed-by: Vincenzo Frascino <[email protected]> Tested-by: Vincenzo Frascino <[email protected]>
but they were already in linux-next by then and he didn't rebase
just to add those.
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (21 commits)
arm64: armv8_deprecated: Checking return value for memory allocation
arm64: Kconfig: Make CONFIG_COMPAT_VDSO a proper Kconfig option
arm64: vdso32: Rename COMPATCC to CC_COMPAT
arm64: vdso32: Pass '--target' option to clang via VDSO_CAFLAGS
arm64: vdso32: Don't use KBUILD_CPPFLAGS unconditionally
arm64: vdso32: Move definition of COMPATCC into vdso32/Makefile
arm64: Default to building compat vDSO with clang when CONFIG_CC_IS_CLANG
lib: vdso: Remove CROSS_COMPILE_COMPAT_VDSO
arm64: vdso32: Remove jump label config option in Makefile
arm64: vdso32: Detect binutils support for dmb ishld
arm64: vdso: Remove stale files from old assembly implementation
arm64: vdso32: Fix broken compat vDSO build warnings
arm64: mm: fix spurious fault detection
arm64: ftrace: Ensure synchronisation in PLT setup for Neoverse-N1 #1542419
arm64: Fix incorrect irqflag restore for priority masking for compat
arm64: mm: avoid virt_to_phys(init_mm.pgd)
arm64: cpufeature: Effectively expose FRINT capability to userspace
arm64: Mark functions using explicit register variables as '__always_inline'
docs: arm64: Fix indentation and doc formatting
arm64/sve: Fix wrong free for task->thread.sve_state
...
Brian Foster [Mon, 7 Oct 2019 19:54:16 +0000 (12:54 -0700)]
xfs: move local to extent inode logging into bmap helper
The callers of xfs_bmap_local_to_extents_empty() log the inode
external to the function, yet this function is where the on-disk
format value is updated. Push the inode logging down into the
function itself to help prevent future mistakes.
Note that internal bmap callers track the inode logging flags
independently and thus may log the inode core twice due to this
change. This is harmless, so leave this code around for consistency
with the other attr fork conversion functions.
Brian Foster [Mon, 7 Oct 2019 19:54:15 +0000 (12:54 -0700)]
xfs: remove broken error handling on failed attr sf to leaf change
xfs_attr_shortform_to_leaf() attempts to put the shortform fork back
together after a failed attempt to convert from shortform to leaf
format. While this code reallocates and copies back the shortform
attr fork data, it never resets the inode format field back to local
format. Further, now that the inode is properly logged after the
initial switch from local format, any error that triggers the
recovery code will eventually abort the transaction and shutdown the
fs. Therefore, remove the broken and unnecessary error handling
code.
Brian Foster [Mon, 7 Oct 2019 19:54:15 +0000 (12:54 -0700)]
xfs: log the inode on directory sf to block format change
When a directory changes from shortform (sf) to block format, the sf
format is copied to a temporary buffer, the inode format is modified
and the updated format filled with the dentries from the temporary
buffer. If the inode format is modified and attempt to grow the
inode fails (due to I/O error, for example), it is possible to
return an error while leaving the directory in an inconsistent state
and with an otherwise clean transaction. This results in corruption
of the associated directory and leads to xfs_dabuf_map() errors as
subsequent lookups cannot accurately determine the format of the
directory. This problem is reproduced occasionally by generic/475.
The fundamental problem is that xfs_dir2_sf_to_block() changes the
on-disk inode format without logging the inode. The inode is
eventually logged by the bmapi layer in the common case, but error
checking introduces the possibility of failing the high level
request before this happens.
Update both of the dir2 and attr callers of
xfs_bmap_local_to_extents_empty() to log the inode core as
consistent with the bmap local to extent format change codepath.
This ensures that any subsequent errors after the format has changed
cause the transaction to abort.
NFS: Fix O_DIRECT accounting of number of bytes read/written
When a series of O_DIRECT reads or writes are truncated, either due to
eof or due to an error, then we should return the number of contiguous
bytes that were received/sent starting at the offset specified by the
application.
Currently, we are failing to correctly check contiguity, and so we're
failing the generic/465 in xfstests when the race between the read
and write RPCs causes the file to get extended while the 2 reads are
outstanding. If the first read RPC call wins the race and returns with
eof set, we should treat the second read RPC as being truncated.
Tom Lendacky [Thu, 1 Aug 2019 18:57:41 +0000 (18:57 +0000)]
perf/x86/amd: Change/fix NMI latency mitigation to use a timestamp
It turns out that the NMI latency workaround from commit:
6d3edaae16c6 ("x86/perf/amd: Resolve NMI latency issues for active PMCs")
ends up being too conservative and results in the perf NMI handler claiming
NMIs too easily on AMD hardware when the NMI watchdog is active.
This has an impact, for example, on the hpwdt (HPE watchdog timer) module.
This module can produce an NMI that is used to reset the system. It
registers an NMI handler for the NMI_UNKNOWN type and relies on the fact
that nothing has claimed an NMI so that its handler will be invoked when
the watchdog device produces an NMI. After the referenced commit, the
hpwdt module is unable to process its generated NMI if the NMI watchdog is
active, because the current NMI latency mitigation results in the NMI
being claimed by the perf NMI handler.
Update the AMD perf NMI latency mitigation workaround to, instead, use a
window of time. Whenever a PMC is handled in the perf NMI handler, set a
timestamp which will act as a perf NMI window. Any NMIs arriving within
that window will be claimed by perf. Anything outside that window will
not be claimed by perf. The value for the NMI window is set to 100 msecs.
This is a conservative value that easily covers any NMI latency in the
hardware. While this still results in a window in which the hpwdt module
will not receive its NMI, the window is now much, much smaller.
Song Liu [Tue, 8 Oct 2019 16:59:49 +0000 (09:59 -0700)]
perf/core: Fix corner case in perf_rotate_context()
In perf_rotate_context(), when the first cpu flexible event fail to
schedule, cpu_rotate is 1, while cpu_event is NULL. Since cpu_event is
NULL, perf_rotate_context will _NOT_ call cpu_ctx_sched_out(), thus
cpuctx->ctx.is_active will have EVENT_FLEXIBLE set. Then, the next
perf_event_sched_in() will skip all cpu flexible events because of the
EVENT_FLEXIBLE bit.
In the next call of perf_rotate_context(), cpu_rotate stays 1, and
cpu_event stays NULL, so this process repeats. The end result is, flexible
events on this cpu will not be scheduled (until another event being added
to the cpuctx).
Here is an easy repro of this issue. On Intel CPUs, where ref-cycles
could only use one counter, run one pinned event for ref-cycles, one
flexible event for ref-cycles, and one flexible event for cycles. The
flexible ref-cycles is never scheduled, which is expected. However,
because of this issue, the cycles event is never scheduled either.
$ perf stat -e ref-cycles:D,ref-cycles,cycles -C 5 -I 1000
To fix this, when the flexible_active list is empty, try rotate the
first event in the flexible_groups. Also, rename ctx_first_active() to
ctx_event_to_rotate(), which is more accurate.
Song Liu [Wed, 4 Sep 2019 21:46:18 +0000 (14:46 -0700)]
perf/core: Rework memory accounting in perf_mmap()
perf_mmap() always increases user->locked_vm. As a result, "extra" could
grow bigger than "user_extra", which doesn't make sense. Here is an
example case:
sched/vtime: Fix guest/system mis-accounting on task switch
vtime_account_system() assumes that the target task to account cputime
to is always the current task. This is most often true indeed except on
task switch where we call:
Here prev is the scheduling-out task where we account the cputime to. It
doesn't match current that is already the scheduling-in task at this
stage of the context switch.
So we end up checking the wrong task flags to determine if we are
accounting guest or system time to the previous task.
As a result the wrong task is used to check if the target is running in
guest mode. We may then spuriously account or leak either system or
guest time on task switch.
Fix this assumption and also turn vtime_guest_enter/exit() to use the
task passed in parameter as well to avoid future similar issues.
If the quota/period ratio was changed during this scaling due to
precision loss, it will cause inconsistency between parent and child
task groups.
See below example:
A userspace container manager (kubelet) does three operations:
1) Create a parent cgroup, set quota to 1,000us and period to 10,000us.
2) Create a few children cgroups.
3) Set quota to 1,000us and period to 10,000us on a child cgroup.
These operations are expected to succeed. However, if the scaling of
147/128 happens before step 3, quota and period of the parent cgroup
will be changed:
Jordan Niethe [Fri, 4 Oct 2019 02:53:17 +0000 (12:53 +1000)]
powerpc/kvm: Fix kvmppc_vcore->in_guest value in kvmhv_switch_to_host
kvmhv_switch_to_host() in arch/powerpc/kvm/book3s_hv_rmhandlers.S
needs to set kvmppc_vcore->in_guest to 0 to signal secondary CPUs to
continue. This happens after resetting the PCR. Before commit 13c7bb3c57dc ("powerpc/64s: Set reserved PCR bits"), r0 would always
be 0 before it was stored to kvmppc_vcore->in_guest. However because
of this change in the commit:
We are no longer comparing r0 against 0 and loading it with 0 if it
contains something else. Hence when we store r0 to
kvmppc_vcore->in_guest, it might not be 0. This means that secondary
CPUs will not be signalled to continue. Those CPUs get stuck and
errors like the following are logged:
KVM: CPU 1 seems to be stuck
KVM: CPU 2 seems to be stuck
KVM: CPU 3 seems to be stuck
KVM: CPU 4 seems to be stuck
KVM: CPU 5 seems to be stuck
KVM: CPU 6 seems to be stuck
KVM: CPU 7 seems to be stuck
This can be reproduced with:
$ for i in `seq 1 7` ; do chcpu -d $i ; done ;
$ taskset -c 0 qemu-system-ppc64 -smp 8,threads=8 \
-M pseries,accel=kvm,kvm-type=HV -m 1G -nographic -vga none \
-kernel vmlinux -initrd initrd.cpio.xz
Fix by making sure r0 is 0 before storing it to
kvmppc_vcore->in_guest.
selftests/powerpc: Fix compile error on tlbie_test due to newer gcc
Newer versions of GCC (>= 9) demand that the size of the string to be
copied must be explicitly smaller than the size of the destination.
Thus, the NULL char has to be taken into account on strncpy.
This will avoid the following compiling error:
tlbie_test.c: In function 'main':
tlbie_test.c:639:4: error: 'strncpy' specified bound 100 equals destination size
strncpy(logdir, optarg, LOGDIR_NAME_SIZE);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
Since commit 1211ee61b4a8 ("powerpc/pseries: Read TLB Block Invalidate
Characteristics"), a warning message is displayed when booting a guest
on top of KVM:
Pavel Shilovsky [Mon, 30 Sep 2019 17:06:20 +0000 (10:06 -0700)]
CIFS: Force reval dentry if LOOKUP_REVAL flag is set
Mark inode for force revalidation if LOOKUP_REVAL flag is set.
This tells the client to actually send a QueryInfo request to
the server to obtain the latest metadata in case a directory
or a file were changed remotely. Only do that if the client
doesn't have a lease for the file to avoid unneeded round
trips to the server.
Pavel Shilovsky [Mon, 30 Sep 2019 17:06:19 +0000 (10:06 -0700)]
CIFS: Force revalidate inode when dentry is stale
Currently the client indicates that a dentry is stale when inode
numbers or type types between a local inode and a remote file
don't match. If this is the case attributes is not being copied
from remote to local, so, it is already known that the local copy
has stale metadata. That's why the inode needs to be marked for
revalidation in order to tell the VFS to lookup the dentry again
before openning a file. This prevents unexpected stale errors
to be returned to the user space when openning a file.
Steve French [Tue, 8 Oct 2019 05:27:14 +0000 (00:27 -0500)]
smb3: Fix regression in time handling
Fixes: cb7a69e60590 ("cifs: Initialize filesystem timestamp ranges")
Only very old servers (e.g. OS/2 and DOS) did not support
DCE TIME (100 nanosecond granularity). Fix the checks used
to set minimum and maximum times.
Steve French [Sat, 5 Oct 2019 15:53:58 +0000 (10:53 -0500)]
smb3: remove noisy debug message and minor cleanup
Message was intended only for developer temporary build
In addition cleanup two minor warnings noticed by Coverity
and a trivial change to workaround a sparse warning
Linus Torvalds [Tue, 8 Oct 2019 22:36:04 +0000 (15:36 -0700)]
Merge tag 'led-fixes-for-5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds
Pull LED fixes from Jacek Anaszewski:
- fix a leftover from earlier stage of development in the documentation
of recently added led_compose_name() and fix old mistake in the
documentation of led_set_brightness_sync() parameter name.
- MAINTAINERS: add pointer to Pavel Machek's linux-leds.git tree.
Pavel is going to take over LED tree maintainership from myself.
* tag 'led-fixes-for-5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds:
Add my linux-leds branch to MAINTAINERS
leds: core: Fix leds.h structure documentation
Linus Torvalds [Tue, 8 Oct 2019 17:55:22 +0000 (10:55 -0700)]
Merge tag 'gpio-v5.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio
Pull GPIO fixes from Linus Walleij:
- don't clear FLAG_IS_OUT when emulating open drain/source in gpiolib
- fix up the usage of nonexclusive GPIO descriptors from device trees
- fix the incorrect IEC offset when toggling trigger edge in the
Spreadtrum driver
- use the correct unit for debounce settings in the MAX77620 driver
* tag 'gpio-v5.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
gpio: max77620: Use correct unit for debounce times
gpio: eic: sprd: Fix the incorrect EIC offset when toggling
gpio: fix getting nonexclusive gpiods from DT
gpiolib: don't clear FLAG_IS_OUT when emulating open-drain/open-source
Linus Torvalds [Tue, 8 Oct 2019 17:51:37 +0000 (10:51 -0700)]
Merge tag 'selinux-pr-20191007' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux
Pull selinuxfix from Paul Moore:
"One patch to ensure we don't copy bad memory up into userspace"
* tag 'selinux-pr-20191007' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux:
selinux: fix context string corruption in convert_context()
Linus Torvalds [Tue, 8 Oct 2019 17:49:05 +0000 (10:49 -0700)]
Merge tag 'linux-kselftest-5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest
Pull Kselftest fixes from Shuah Khan:
"Fixes for existing tests and the framework.
Cristian Marussi's patches add the ability to skip targets (tests) and
exclude tests that didn't build from run-list. These patches improve
the Kselftest results. Ability to skip targets helps avoid running
tests that aren't supported in certain environments. As an example,
bpf tests from mainline aren't supported on stable kernels and have
dependency on bleeding edge llvm. Being able to skip bpf on systems
that can't meet this llvm dependency will be helpful.
Kselftest can be built and installed from the main Makefile. This
change help simplify Kselftest use-cases which addresses request from
users.
Kees Cook added per test timeout support to limit individual test
run-time"
* tag 'linux-kselftest-5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
selftests: watchdog: Add command line option to show watchdog_info
selftests: watchdog: Validate optional file argument
selftests/kselftest/runner.sh: Add 45 second timeout per test
kselftest: exclude failed TARGETS from runlist
kselftest: add capability to skip chosen TARGETS
selftests: Add kselftest-all and kselftest-install targets
Masahiro Yamada [Tue, 8 Oct 2019 03:10:09 +0000 (12:10 +0900)]
doc: move namespaces.rst from kbuild/ to core-api/
We discussed a better location for this file, and agreed that
core-api/ is a good fit. Rename it to symbol-namespaces.rst
for disambiguation, and also add it to index.rst and MAINTAINERS.
Arvind Sankar [Mon, 7 Oct 2019 22:00:02 +0000 (18:00 -0400)]
lib/string: Make memzero_explicit() inline instead of external
With the use of the barrier implied by barrier_data(), there is no need
for memzero_explicit() to be extern. Making it inline saves the overhead
of a function call, and allows the code to be reused in arch/*/purgatory
without having to duplicate the implementation.
As per "AMD64 Architecture Programmer's Manual Volume 3: General-Purpose
and System Instructions", MWAITX EAX[7:4]+1 specifies the optional hint
of the optimized C-state. For C0 state, EAX[7:4] should be set to 0xf.
Currently, a value of 0xf is set for EAX[3:0] instead of EAX[7:4]. Fix
this by changing MWAITX_DISABLE_CSTATES from 0xf to 0xf0.
This hasn't had any implications so far because setting reserved bits in
EAX is simply ignored by the CPU.
[ bp: Fixup comment in delay_mwaitx() and massage. ]
Austin Kim [Tue, 3 Sep 2019 03:30:19 +0000 (12:30 +0900)]
btrfs: silence maybe-uninitialized warning in clone_range
GCC throws warning message as below:
‘clone_src_i_size’ may be used uninitialized in this function
[-Wmaybe-uninitialized]
#define IS_ALIGNED(x, a) (((x) & ((typeof(x))(a) - 1)) == 0)
^
fs/btrfs/send.c:5088:6: note: ‘clone_src_i_size’ was declared here
u64 clone_src_i_size;
^
The clone_src_i_size is only used as call-by-reference
in a call to get_inode_info().
Silence the warning by initializing clone_src_i_size to 0.
Note that the warning is a false positive and reported by older versions
of GCC (eg. 7.x) but not eg 9.x. As there have been numerous people, the
patch is applied. Setting clone_src_i_size to 0 does not otherwise make
sense and would not do any action in case the code changes in the future.
The panel-tpo-td043mtea1 driver incorrectly includes the OF vendor
prefix in its SPI alias. Fix it, and move the manual alias to an SPI
module device table.
The panel-sony-acx565akm driver incorrectly includes the OF vendor
prefix in its SPI alias. Fix it, and move the manual alias to an SPI
module device table.
The panel-nec-nl8048hl11 driver incorrectly includes the OF vendor
prefix in its SPI alias. Fix it, and move the manual alias to an SPI
module device table.
The panel-lg-lb035q02 driver incorrectly includes the OF vendor prefix
in its SPI alias. Fix it, and move the manual alias to an SPI module
device table.
Pavel Begunkov [Mon, 7 Oct 2019 23:18:42 +0000 (02:18 +0300)]
io_uring: remove wait loop spurious wakeups
Any changes interesting to tasks waiting in io_cqring_wait() are
commited with io_cqring_ev_posted(). However, io_ring_drop_ctx_refs()
also tries to do that but with no reason, that means spurious wakeups
every io_free_req() and io_uring_enter().
Linus Torvalds [Mon, 7 Oct 2019 23:04:19 +0000 (16:04 -0700)]
Merge branch 'akpm' (patches from Andrew)
Merge misc fixes from Andrew Morton:
"The usual shower of hotfixes.
Chris's memcg patches aren't actually fixes - they're mature but a few
niggling review issues were late to arrive.
The ocfs2 fixes are quite old - those took some time to get reviewer
attention.
Subsystems affected by this patch series: ocfs2, hotfixes, mm/memcg,
mm/slab-generic"
* emailed patches from Andrew Morton <[email protected]>:
mm, sl[aou]b: guarantee natural alignment for kmalloc(power-of-two)
mm, sl[ou]b: improve memory accounting
mm, memcg: make scan aggression always exclude protection
mm, memcg: make memory.emin the baseline for utilisation determination
mm, memcg: proportional memory.{low,min} reclaim
mm/vmpressure.c: fix a signedness bug in vmpressure_register_event()
mm/page_alloc.c: fix a crash in free_pages_prepare()
mm/z3fold.c: claim page in the beginning of free
kernel/sysctl.c: do not override max_threads provided by userspace
memcg: only record foreign writebacks with dirty pages when memcg is not disabled
mm: fix -Wmissing-prototypes warnings
writeback: fix use-after-free in finish_writeback_work()
mm/memremap: drop unused SECTION_SIZE and SECTION_MASK
panic: ensure preemption is disabled during panic()
fs: ocfs2: fix a possible null-pointer dereference in ocfs2_info_scan_inode_alloc()
fs: ocfs2: fix a possible null-pointer dereference in ocfs2_write_end_nolock()
fs: ocfs2: fix possible null-pointer dereferences in ocfs2_xa_prepare_entry()
ocfs2: clear zero in unaligned direct IO
Vlastimil Babka [Mon, 7 Oct 2019 00:58:45 +0000 (17:58 -0700)]
mm, sl[aou]b: guarantee natural alignment for kmalloc(power-of-two)
In most configurations, kmalloc() happens to return naturally aligned
(i.e. aligned to the block size itself) blocks for power of two sizes.
That means some kmalloc() users might unknowingly rely on that
alignment, until stuff breaks when the kernel is built with e.g.
CONFIG_SLUB_DEBUG or CONFIG_SLOB, and blocks stop being aligned. Then
developers have to devise workaround such as own kmem caches with
specified alignment [1], which is not always practical, as recently
evidenced in [2].
The topic has been discussed at LSF/MM 2019 [3]. Adding a
'kmalloc_aligned()' variant would not help with code unknowingly relying
on the implicit alignment. For slab implementations it would either
require creating more kmalloc caches, or allocate a larger size and only
give back part of it. That would be wasteful, especially with a generic
alignment parameter (in contrast with a fixed alignment to size).
Ideally we should provide to mm users what they need without difficult
workarounds or own reimplementations, so let's make the kmalloc()
alignment to size explicitly guaranteed for power-of-two sizes under all
configurations. What this means for the three available allocators?
* SLAB object layout happens to be mostly unchanged by the patch. The
implicitly provided alignment could be compromised with
CONFIG_DEBUG_SLAB due to redzoning, however SLAB disables redzoning for
caches with alignment larger than unsigned long long. Practically on at
least x86 this includes kmalloc caches as they use cache line alignment,
which is larger than that. Still, this patch ensures alignment on all
arches and cache sizes.
* SLUB layout is also unchanged unless redzoning is enabled through
CONFIG_SLUB_DEBUG and boot parameter for the particular kmalloc cache.
With this patch, explicit alignment is guaranteed with redzoning as
well. This will result in more memory being wasted, but that should be
acceptable in a debugging scenario.
* SLOB has no implicit alignment so this patch adds it explicitly for
kmalloc(). The potential downside is increased fragmentation. While
pathological allocation scenarios are certainly possible, in my testing,
after booting a x86_64 kernel+userspace with virtme, around 16MB memory
was consumed by slab pages both before and after the patch, with
difference in the noise.
Vlastimil Babka [Mon, 7 Oct 2019 00:58:42 +0000 (17:58 -0700)]
mm, sl[ou]b: improve memory accounting
Patch series "guarantee natural alignment for kmalloc()", v2.
This patch (of 2):
SLOB currently doesn't account its pages at all, so in /proc/meminfo the
Slab field shows zero. Modifying a counter on page allocation and
freeing should be acceptable even for the small system scenarios SLOB is
intended for. Since reclaimable caches are not separated in SLOB,
account everything as unreclaimable.
SLUB currently doesn't account kmalloc() and kmalloc_node() allocations
larger than order-1 page, that are passed directly to the page
allocator. As they also don't appear in /proc/slabinfo, it might look
like a memory leak. For consistency, account them as well. (SLAB
doesn't actually use page allocator directly, so no change there).
Ideally SLOB and SLUB would be handled in separate patches, but due to
the shared kmalloc_order() function and different kfree()
implementations, it's easier to patch both at once to prevent
inconsistencies.
Chris Down [Mon, 7 Oct 2019 00:58:38 +0000 (17:58 -0700)]
mm, memcg: make scan aggression always exclude protection
This patch is an incremental improvement on the existing
memory.{low,min} relative reclaim work to base its scan pressure
calculations on how much protection is available compared to the current
usage, rather than how much the current usage is over some protection
threshold.
This change doesn't change the experience for the user in the normal
case too much. One benefit is that it replaces the (somewhat arbitrary)
100% cutoff with an indefinite slope, which makes it easier to ballpark
a memory.low value.
As well as this, the old methodology doesn't quite apply generically to
machines with varying amounts of physical memory. Let's say we have a
top level cgroup, workload.slice, and another top level cgroup,
system-management.slice. We want to roughly give 12G to
system-management.slice, so on a 32GB machine we set memory.low to 20GB
in workload.slice, and on a 64GB machine we set memory.low to 52GB.
However, because these are relative amounts to the total machine size,
while the amount of memory we want to generally be willing to yield to
system.slice is absolute (12G), we end up putting more pressure on
system.slice just because we have a larger machine and a larger workload
to fill it, which seems fairly unintuitive. With this new behaviour, we
don't end up with this unintended side effect.
Previously the way that memory.low protection works is that if you are
50% over a certain baseline, you get 50% of your normal scan pressure.
This is certainly better than the previous cliff-edge behaviour, but it
can be improved even further by always considering memory under the
currently enforced protection threshold to be out of bounds. This means
that we can set relatively low memory.low thresholds for variable or
bursty workloads while still getting a reasonable level of protection,
whereas with the previous version we may still trivially hit the 100%
clamp. The previous 100% clamp is also somewhat arbitrary, whereas this
one is more concretely based on the currently enforced protection
threshold, which is likely easier to reason about.
There is also a subtle issue with the way that proportional reclaim
worked previously -- it promotes having no memory.low, since it makes
pressure higher during low reclaim. This happens because we base our
scan pressure modulation on how far memory.current is between memory.min
and memory.low, but if memory.low is unset, we only use the overage
method. In most cromulent configurations, this then means that we end
up with *more* pressure than with no memory.low at all when we're in low
reclaim, which is not really very usable or expected.
With this patch, memory.low and memory.min affect reclaim pressure in a
more understandable and composable way. For example, from a user
standpoint, "protected" memory now remains untouchable from a reclaim
aggression standpoint, and users can also have more confidence that
bursty workloads will still receive some amount of guaranteed
protection.
Chris Down [Mon, 7 Oct 2019 00:58:35 +0000 (17:58 -0700)]
mm, memcg: make memory.emin the baseline for utilisation determination
Roman points out that when when we do the low reclaim pass, we scale the
reclaim pressure relative to position between 0 and the maximum
protection threshold.
However, if the maximum protection is based on memory.elow, and
memory.emin is above zero, this means we still may get binary behaviour
on second-pass low reclaim. This is because we scale starting at 0, not
starting at memory.emin, and since we don't scan at all below emin, we
end up with cliff behaviour.
This should be a fairly uncommon case since usually we don't go into the
second pass, but it makes sense to scale our low reclaim pressure
starting at emin.
You can test this by catting two large sparse files, one in a cgroup
with emin set to some moderate size compared to physical RAM, and
another cgroup without any emin. In both cgroups, set an elow larger
than 50% of physical RAM. The one with emin will have less page
scanning, as reclaim pressure is lower.
Rebase on top of and apply the same idea as what was applied to handle
cgroup_memory=disable properly for the original proportional patch
http://lkml.kernel.org/r/20190201045711[email protected] ("mm,
memcg: Handle cgroup_disable=memory when getting memcg protection").
Chris Down [Mon, 7 Oct 2019 00:58:32 +0000 (17:58 -0700)]
mm, memcg: proportional memory.{low,min} reclaim
cgroup v2 introduces two memory protection thresholds: memory.low
(best-effort) and memory.min (hard protection). While they generally do
what they say on the tin, there is a limitation in their implementation
that makes them difficult to use effectively: that cliff behaviour often
manifests when they become eligible for reclaim. This patch implements
more intuitive and usable behaviour, where we gradually mount more
reclaim pressure as cgroups further and further exceed their protection
thresholds.
This cliff edge behaviour happens because we only choose whether or not
to reclaim based on whether the memcg is within its protection limits
(see the use of mem_cgroup_protected in shrink_node), but we don't vary
our reclaim behaviour based on this information. Imagine the following
timeline, with the numbers the lruvec size in this zone:
1. memory.low=1000000, memory.current=999999. 0 pages may be scanned.
2. memory.low=1000000, memory.current=1000000. 0 pages may be scanned.
3. memory.low=1000000, memory.current=1000001. 1000001* pages may be
scanned. (?!)
* Of course, we won't usually scan all available pages in the zone even
without this patch because of scan control priority, over-reclaim
protection, etc. However, as shown by the tests at the end, these
techniques don't sufficiently throttle such an extreme change in input,
so cliff-like behaviour isn't really averted by their existence alone.
Here's an example of how this plays out in practice. At Facebook, we are
trying to protect various workloads from "system" software, like
configuration management tools, metric collectors, etc (see this[0] case
study). In order to find a suitable memory.low value, we start by
determining the expected memory range within which the workload will be
comfortable operating. This isn't an exact science -- memory usage deemed
"comfortable" will vary over time due to user behaviour, differences in
composition of work, etc, etc. As such we need to ballpark memory.low,
but doing this is currently problematic:
1. If we end up setting it too low for the workload, it won't have
*any* effect (see discussion above). The group will receive the full
weight of reclaim and won't have any priority while competing with the
less important system software, as if we had no memory.low configured
at all.
2. Because of this behaviour, we end up erring on the side of setting
it too high, such that the comfort range is reliably covered. However,
protected memory is completely unavailable to the rest of the system,
so we might cause undue memory and IO pressure there when we *know* we
have some elasticity in the workload.
3. Even if we get the value totally right, smack in the middle of the
comfort zone, we get extreme jumps between no pressure and full
pressure that cause unpredictable pressure spikes in the workload due
to the current binary reclaim behaviour.
With this patch, we can set it to our ballpark estimation without too much
worry. Any undesirable behaviour, such as too much or too little reclaim
pressure on the workload or system will be proportional to how far our
estimation is off. This means we can set memory.low much more
conservatively and thus waste less resources *without* the risk of the
workload falling off a cliff if we overshoot.
As a more abstract technical description, this unintuitive behaviour
results in having to give high-priority workloads a large protection
buffer on top of their expected usage to function reliably, as otherwise
we have abrupt periods of dramatically increased memory pressure which
hamper performance. Having to set these thresholds so high wastes
resources and generally works against the principle of work conservation.
In addition, having proportional memory reclaim behaviour has other
benefits. Most notably, before this patch it's basically mandatory to set
memory.low to a higher than desirable value because otherwise as soon as
you exceed memory.low, all protection is lost, and all pages are eligible
to scan again. By contrast, having a gradual ramp in reclaim pressure
means that you now still get some protection when thresholds are exceeded,
which means that one can now be more comfortable setting memory.low to
lower values without worrying that all protection will be lost. This is
important because workingset size is really hard to know exactly,
especially with variable workloads, so at least getting *some* protection
if your workingset size grows larger than you expect increases user
confidence in setting memory.low without a huge buffer on top being
needed.
Thanks a lot to Johannes Weiner and Tejun Heo for their advice and
assistance in thinking about how to make this work better.
In testing these changes, I intended to verify that:
1. Changes in page scanning become gradual and proportional instead of
binary.
To test this, I experimented stepping further and further down
memory.low protection on a workload that floats around 19G workingset
when under memory.low protection, watching page scan rates for the
workload cgroup:
As you can see, the test kernel (with a kernel containing this
patch) ramps up page scanning significantly more gradually than the
control kernel (without this patch).
2. More gradual ramp up in reclaim aggression doesn't result in
premature OOMs.
To test this, I wrote a script that slowly increments the number of
pages held by stress(1)'s --vm-keep mode until a production system
entered severe overall memory contention. This script runs in a highly
protected slice taking up the majority of available system memory.
Watching vmstat revealed that page scanning continued essentially
nominally between test and control, without causing forward reclaim
progress to become arrested.
Qian Cai [Mon, 7 Oct 2019 00:58:25 +0000 (17:58 -0700)]
mm/page_alloc.c: fix a crash in free_pages_prepare()
On architectures like s390, arch_free_page() could mark the page unused
(set_page_unused()) and any access later would trigger a kernel panic.
Fix it by moving arch_free_page() after all possible accessing calls.
In the past, only kernel_poison_pages() would trigger this but it needs
"page_poison=on" kernel cmdline, and I suspect nobody tested that on
s390. Recently, kernel_init_free_pages() (commit 6471384af2a6 ("mm:
security: introduce init_on_alloc=1 and init_on_free=1 boot options"))
was added and could trigger this as well.
Vitaly Wool [Mon, 7 Oct 2019 00:58:22 +0000 (17:58 -0700)]
mm/z3fold.c: claim page in the beginning of free
There's a really hard to reproduce race in z3fold between z3fold_free()
and z3fold_reclaim_page(). z3fold_reclaim_page() can claim the page
after z3fold_free() has checked if the page was claimed and
z3fold_free() will then schedule this page for compaction which may in
turn lead to random page faults (since that page would have been
reclaimed by then).
Fix that by claiming page in the beginning of z3fold_free() and not
forgetting to clear the claim in the end.
Michal Hocko [Mon, 7 Oct 2019 00:58:19 +0000 (17:58 -0700)]
kernel/sysctl.c: do not override max_threads provided by userspace
Partially revert 16db3d3f1170 ("kernel/sysctl.c: threads-max observe
limits") because the patch is causing a regression to any workload which
needs to override the auto-tuning of the limit provided by kernel.
set_max_threads is implementing a boot time guesstimate to provide a
sensible limit of the concurrently running threads so that runaways will
not deplete all the memory. This is a good thing in general but there
are workloads which might need to increase this limit for an application
to run (reportedly WebSpher MQ is affected) and that is simply not
possible after the mentioned change. It is also very dubious to
override an admin decision by an estimation that doesn't have any direct
relation to correctness of the kernel operation.
Fix this by dropping set_max_threads from sysctl_max_threads so any
value is accepted as long as it fits into MAX_THREADS which is important
to check because allowing more threads could break internal robust futex
restriction. While at it, do not use MIN_THREADS as the lower boundary
because it is also only a heuristic for automatic estimation and admin
might have a good reason to stop new threads to be created even when
below this limit.
This became more severe when we switched x86 from 4k to 8k kernel
stacks. Starting since 6538b8ea886e ("x86_64: expand kernel stack to
16K") (3.16) we use THREAD_SIZE_ORDER = 2 and that halved the auto-tuned
value.
In the particular case
3.12
kernel.threads-max = 515561
4.4
kernel.threads-max = 200000
Neither of the two values is really insane on 32GB machine.
I am not sure we want/need to tune the max_thread value further. If
anything the tuning should be removed altogether if proven not useful in
general. But we definitely need a way to override this auto-tuning.
Baoquan He [Mon, 7 Oct 2019 00:58:15 +0000 (17:58 -0700)]
memcg: only record foreign writebacks with dirty pages when memcg is not disabled
In kdump kernel, memcg usually is disabled with 'cgroup_disable=memory'
for saving memory. Now kdump kernel will always panic when dump vmcore
to local disk:
And this will corrupt the 1st kernel too with 'cgroup_disable=memory'.
Via the trace and with debugging, it is pointing to commit 97b27821b485
("writeback, memcg: Implement foreign dirty flushing") which introduced
this regression. Disabling memcg causes the null pointer dereference at
uninitialized data in function mem_cgroup_track_foreign_dirty_slowpath().
Fix it by returning directly if memcg is disabled, but not trying to
record the foreign writebacks with dirty pages.
Yi Wang [Mon, 7 Oct 2019 00:58:12 +0000 (17:58 -0700)]
mm: fix -Wmissing-prototypes warnings
We get two warnings when build kernel W=1:
mm/shuffle.c:36:12: warning: no previous prototype for `shuffle_show' [-Wmissing-prototypes]
mm/sparse.c:220:6: warning: no previous prototype for `subsection_mask_set' [-Wmissing-prototypes]
Tejun Heo [Mon, 7 Oct 2019 00:58:09 +0000 (17:58 -0700)]
writeback: fix use-after-free in finish_writeback_work()
finish_writeback_work() reads @done->waitq after decrementing
@done->cnt. However, once @done->cnt reaches zero, @done may be freed
(from stack) at any moment and @done->waitq can contain something
unrelated by the time finish_writeback_work() tries to read it. This
led to the following crash.
mm/memremap: drop unused SECTION_SIZE and SECTION_MASK
SECTION_SIZE and SECTION_MASK macros are not getting used anymore. But
they do conflict with existing definitions on arm64 platform causing
following warning during build. Lets drop these unused macros.
mm/memremap.c:16: warning: "SECTION_MASK" redefined
#define SECTION_MASK ~((1UL << PA_SECTION_SHIFT) - 1)
arch/arm64/include/asm/pgtable-hwdef.h:79: note: this is the location of the previous definition
#define SECTION_MASK (~(SECTION_SIZE-1))
mm/memremap.c:17: warning: "SECTION_SIZE" redefined
#define SECTION_SIZE (1UL << PA_SECTION_SHIFT)
arch/arm64/include/asm/pgtable-hwdef.h:78: note: this is the location of the previous definition
#define SECTION_SIZE (_AC(1, UL) << SECTION_SHIFT)
Will Deacon [Mon, 7 Oct 2019 00:58:00 +0000 (17:58 -0700)]
panic: ensure preemption is disabled during panic()
Calling 'panic()' on a kernel with CONFIG_PREEMPT=y can leave the
calling CPU in an infinite loop, but with interrupts and preemption
enabled. From this state, userspace can continue to be scheduled,
despite the system being "dead" as far as the kernel is concerned.
This is easily reproducible on arm64 when booting with "nosmp" on the
command line; a couple of shell scripts print out a periodic "Ping"
message whilst another triggers a crash by writing to
/proc/sysrq-trigger:
Jia Guo [Mon, 7 Oct 2019 00:57:47 +0000 (17:57 -0700)]
ocfs2: clear zero in unaligned direct IO
Unused portion of a part-written fs-block-sized block is not set to zero
in unaligned append direct write.This can lead to serious data
inconsistencies.
Ocfs2 manage disk with cluster size(for example, 1M), part-written in
one cluster will change the cluster state from UN-WRITTEN to WRITTEN,
VFS(function dio_zero_block) doesn't do the cleaning because bh's state
is not set to NEW in function ocfs2_dio_wr_get_block when we write a
WRITTEN cluster. For example, the cluster size is 1M, file size is 8k
and we direct write from 14k to 15k, then 12k~14k and 15k~16k will
contain dirty data.
We have to deal with two cases:
1.The starting position of direct write is outside the file.
2.The starting position of direct write is located in the file.
We need set bh's state to NEW in the first case. In the second case, we
need mapped twice because bh's state of area out file should be set to
NEW while area in file not.
Boris Ostrovsky [Mon, 30 Sep 2019 20:44:41 +0000 (16:44 -0400)]
x86/xen: Return from panic notifier
Currently execution of panic() continues until Xen's panic notifier
(xen_panic_event()) is called at which point we make a hypercall that
never returns.
This means that any notifier that is supposed to be called later as
well as significant part of panic() code (such as pstore writes from
kmsg_dump()) is never executed.
There is no reason for xen_panic_event() to be this last point in
execution since panic()'s emergency_restart() will call into
xen_emergency_restart() from where we can perform our hypercall.
Nevertheless, we will provide xen_legacy_crash boot option that will
preserve original behavior during crash. This option could be used,
for example, if running kernel dumper (which happens after panic
notifiers) is undesirable.
Vincent Chen [Mon, 23 Sep 2019 00:45:16 +0000 (08:45 +0800)]
riscv: Correct the handling of unexpected ebreak in do_trap_break()
For the kernel space, all ebreak instructions are determined at compile
time because the kernel space debugging module is currently unsupported.
Hence, it should be treated as a bug if an ebreak instruction which does
not belong to BUG_TRAP_TYPE_WARN or BUG_TRAP_TYPE_BUG is executed in
kernel space. For the userspace, debugging module or user problem may
intentionally insert an ebreak instruction to trigger a SIGTRAP signal.
To approach the above two situations, the do_trap_break() will direct
the BUG_TRAP_TYPE_NONE ebreak exception issued in kernel space to die()
and will send a SIGTRAP to the trapped process only when the ebreak is
in userspace.
Vincent Chen [Mon, 23 Sep 2019 00:45:15 +0000 (08:45 +0800)]
riscv: avoid sending a SIGTRAP to a user thread trapped in WARN()
On RISC-V, when the kernel runs code on behalf of a user thread, and the
kernel executes a WARN() or WARN_ON(), the user thread will be sent
a bogus SIGTRAP. Fix the RISC-V kernel code to not send a SIGTRAP when
a WARN()/WARN_ON() is executed.
Vincent Chen [Mon, 23 Sep 2019 00:45:14 +0000 (08:45 +0800)]
riscv: avoid kernel hangs when trapped in BUG()
When the CONFIG_GENERIC_BUG is disabled by disabling CONFIG_BUG, if a
kernel thread is trapped by BUG(), the whole system will be in the
loop that infinitely handles the ebreak exception instead of entering the
die function. To fix this problem, the do_trap_break() will always call
the die() to deal with the break exception as the type of break is
BUG_TRAP_TYPE_BUG.
Linus Torvalds [Mon, 7 Oct 2019 19:56:48 +0000 (12:56 -0700)]
uaccess: implement a proper unsafe_copy_to_user() and switch filldir over to it
In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to
unsafe_put_user()") I made filldir() use unsafe_put_user(), which
improves code generation on x86 enormously.
But because we didn't have a "unsafe_copy_to_user()", the dirent name
copy was also done by hand with unsafe_put_user() in a loop, and it
turns out that a lot of other architectures didn't like that, because
unlike x86, they have various alignment issues.
Most non-x86 architectures trap and fix it up, and some (like xtensa)
will just fail unaligned put_user() accesses unconditionally. Which
makes that "copy using put_user() in a loop" not work for them at all.
I could make that code do explicit alignment etc, but the architectures
that don't like unaligned accesses also don't really use the fancy
"user_access_begin/end()" model, so they might just use the regular old
__copy_to_user() interface.
So this commit takes that looping implementation, turns it into the x86
version of "unsafe_copy_to_user()", and makes other architectures
implement the unsafe copy version as __copy_to_user() (the same way they
do for the other unsafe_xyz() accessor functions).
Note that it only does this for the copying _to_ user space, and we
still don't have a unsafe version of copy_from_user().
That's partly because we have no current users of it, but also partly
because the copy_from_user() case is slightly different and cannot
efficiently be implemented in terms of a unsafe_get_user() loop (because
gcc can't do asm goto with outputs).
It would be trivial to do this using "rep movsb", which would work
really nicely on newer x86 cores, but really badly on some older ones.
Al Viro is looking at cleaning up all our user copy routines to make
this all a non-issue, but for now we have this simple-but-stupid version
for x86 that works fine for the dirent name copy case because those
names are short strings and we simply don't need anything fancier.
Chris Wilson [Fri, 20 Sep 2019 12:18:21 +0000 (13:18 +0100)]
drm/i915: Mark contents as dirty on a write fault
Since dropping the set-to-gtt-domain in commit a679f58d0510 ("drm/i915:
Flush pages on acquisition"), we no longer mark the contents as dirty on
a write fault. This has the issue of us then not marking the pages as
dirty on releasing the buffer, which means the contents are not written
out to the swap device (should we ever pick that buffer as a victim).
Notably, this is visible in the dumb buffer interface used for cursors.
Having updated the cursor contents via mmap, and swapped away, if the
shrinker should evict the old cursor, upon next reuse, the cursor would
be invisible.
E.g. echo 80 > /proc/sys/kernel/sysrq ; echo f > /proc/sysrq-trigger
Chris Wilson [Mon, 23 Sep 2019 15:28:44 +0000 (16:28 +0100)]
drm/i915: Prevent bonded requests from overtaking each other on preemption
Force bonded requests to run on distinct engines so that they cannot be
shuffled onto the same engine where timeslicing will reverse the order.
A bonded request will often wait on a semaphore signaled by its master,
creating an implicit dependency -- if we ignore that implicit dependency
and allow the bonded request to run on the same engine and before its
master, we will cause a GPU hang. [Whether it will hang the GPU is
debatable, we should keep on timeslicing and each timeslice should be
"accidentally" counted as forward progress, in which case it should run
but at one-half to one-third speed.]
We can prevent this inversion by restricting which engines we allow
ourselves to jump to upon preemption, i.e. baking in the arrangement
established at first execution. (We should also consider capturing the
implicit dependency using i915_sched_add_dependency(), but first we need
to think about the constraints that requires on the execution/retirement
ordering.)
Ville Syrjälä [Thu, 5 Sep 2019 13:50:43 +0000 (16:50 +0300)]
drm/i915: Bump skl+ max plane width to 5k for linear/x-tiled
The officially validated plane width limit is 4k on skl+, however
we already had people using 5k displays before we started to enforce
the limit. Also it seems Windows allows 5k resolutions as well
(though not sure if they do it with one plane or two).
According to hw folks 5k should work with the possible
exception of the following features:
- Ytile (already limited to 4k)
- FP16 (already limited to 4k)
- render compression (already limited to 4k)
- KVMR sprite and cursor (don't care)
- horizontal panning (need to verify this)
- pipe and plane scaling (need to verify this)
So apart from last two items on that list we are already
fine. We should really verify what happens with those last
two items but I don't have a 5k display on hand atm so it'll
have to wait.
In the meantime let's just bump the limit back up to 5k since
several users have already been using it without apparent issues.
At least we'll be no worse off than we were prior to lowering
the limits.
Chris Wilson [Wed, 18 Sep 2019 14:54:50 +0000 (15:54 +0100)]
drm/i915: Verify the engine after acquiring the active.lock
When using virtual engines, the rq->engine is not stable until we hold
the engine->active.lock (as the virtual engine may be exchanged with the
sibling). Since commit 22b7a426bbe1 ("drm/i915/execlists: Preempt-to-busy")
we may retire a request concurrently with resubmitting it to HW, we need
to be extra careful to verify we are holding the correct lock for the
request's active list. This is similar to the issue we saw with
rescheduling the virtual requests, see sched_lock_engine().
projects / linux.git / log