Add the SoC specific compatible for SM8450 implementing
arm,mmu-500.
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20211201073943.3969549-2-vkoul@kernel.org
Signed-off-by: Will Deacon <will@kernel.org>
It is a 64b register, lets not lose the upper bits.
Fixes: ab5df7b953 ("iommu/arm-smmu-qcom: Add an adreno-smmu-priv callback to get pagefault info")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20211108171724.470973-1-robdclark@gmail.com
Signed-off-by: Will Deacon <will@kernel.org>
ttm->num_pages is uint32_t which was causing very large buffers to
only populate a truncated size.
This fixes gem_create@create-clear igt test on large memory systems.
Fixes: 7ae034590c ("drm/i915/ttm: add tt shmem backend")
Signed-off-by: Robert Beckett <bob.beckett@collabora.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211210195005.2582884-1-bob.beckett@collabora.com
In case of BPF_CORE_TYPE_ID_LOCAL we fill out target result explicitly.
But targ_res itself isn't initialized in such a case, and subsequent
call to bpf_core_patch_insn() might read uninitialized field (like
fail_memsz_adjust in this case). So ensure that targ_res is
zero-initialized for BPF_CORE_TYPE_ID_LOCAL case.
This was reported by Coverity static analyzer.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20211214010032.3843804-1-andrii@kernel.org
Add binding for reference clock PAD modes of the i.MX8 PCIe PHY.
Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Reviewed-by: Tim Harvey <tharvey@gateworks.com>
Tested-by: Tim Harvey <tharvey@gateworks.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/1638432158-4119-2-git-send-email-hongxing.zhu@nxp.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The commit referenced below added fixup_map_timer support (to create a
BPF map containing timers), but failed to increase the size of the
map_fds array, leading to out of bounds write. Fix this by changing
MAX_NR_MAPS to 22.
Fixes: e60e6962c5 ("selftests/bpf: Add tests for restricted helpers")
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20211214014800.78762-1-memxor@gmail.com
Do not sleep in poll() when the need_wakeup flag is set. When this
flag is set, the application needs to explicitly wake up the driver
with a syscall (poll, recvmsg, sendmsg, etc.) to guarantee that Rx
and/or Tx processing will be processed promptly. But the current code
in poll(), sleeps first then wakes up the driver. This means that no
driver processing will occur (baring any interrupts) until the timeout
has expired.
Fix this by checking the need_wakeup flag first and if set, wake the
driver and return to the application. Only if need_wakeup is not set
should the process sleep if there is a timeout set in the poll() call.
Fixes: 77cd0d7b3f ("xsk: add support for need_wakeup flag in AF_XDP rings")
Reported-by: Keith Wiles <keith.wiles@intel.com>
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/bpf/20211214102607.7677-1-magnus.karlsson@gmail.com
As the first thing this function does is to call
v4l2_device_register(), it should call v4l2_device_unregister()
if an error occurs, the same way as done at saa7146_vv_release().
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
In hexium_attach(dev, info), saa7146_vv_init() is called to allocate
a new memory for dev->vv_data. saa7146_vv_release() will be called on
failure of saa7146_register_device(). There is a dereference of
dev->vv_data in saa7146_vv_release(), which could lead to a NULL
pointer dereference on failure of saa7146_vv_init().
Fix this bug by adding a check of saa7146_vv_init().
This bug was found by a static analyzer. The analysis employs
differential checking to identify inconsistent security operations
(e.g., checks or kfrees) between two code paths and confirms that the
inconsistent operations are not recovered in the current function or
the callers, so they constitute bugs.
Note that, as a bug found by static analysis, it can be a false
positive or hard to trigger. Multiple researchers have cross-reviewed
the bug.
Builds with CONFIG_VIDEO_HEXIUM_GEMINI=m show no new warnings,
and our static analyzer no longer warns about this code.
Link: https://lore.kernel.org/linux-media/20211203154030.111210-1-zhou1615@umn.edu
Signed-off-by: Zhou Qingyang <zhou1615@umn.edu>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
'ent_enum->bmap' is a bitmap. So use 'bitmap_zalloc()' to simplify
code, improve the semantic and avoid some open-coded arithmetic in
allocator arguments.
Also change the corresponding 'kfree()' into 'bitmap_free()' to keep
consistency.
While at it, remove a useless 'bitmap_zero()'.
Link: https://lore.kernel.org/linux-media/b11f646dda189f490c06bf671f64a2cc0af4d45c.1638397089.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The only usage of max96712_subdev_ops is to pass its address to
v4l2_i2c_subdev_init() which takes a pointer to const struct
v4l2_subdev_ops as argument. Make it const to allow the compiler to put
it in read-only memory.
Link: https://lore.kernel.org/linux-media/20211127094945.27985-1-rikard.falkeborn@gmail.com
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Reviewed-by: Niklas S\xF6derlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
I got a null-ptr-deref report:
BUG: kernel NULL pointer dereference, address: 0000000000000060
...
RIP: 0010:v4l2_ctrl_auto_cluster+0x57/0x270
...
Call Trace:
msi001_probe+0x13b/0x24b [msi001]
spi_probe+0xeb/0x130
...
do_syscall_64+0x35/0xb0
In msi001_probe(), if the creation of control for bandwidth_auto
fails, there will be a null-ptr-deref issue when it is used in
v4l2_ctrl_auto_cluster().
Check dev->hdl.error before v4l2_ctrl_auto_cluster() to fix this bug.
Link: https://lore.kernel.org/linux-media/20211026112348.2878040-1-wanghai38@huawei.com
Fixes: 93203dd6c7 ("[media] msi001: Mirics MSi001 silicon tuner driver")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wang Hai <wanghai38@huawei.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
While testing Rockchip RK3399 with both ISPs enabled, a dmesg error
was observed:
```
[ 15.559141] debugfs: Directory 'rkisp1' with parent '/' already present!
```
Fix it by using the device name for the debugfs subdirectory name
instead of the driver name, thus preventing name collision.
Link: https://lore.kernel.org/linux-media/20211010175457.438627-1-mike.rudenko@gmail.com
Signed-off-by: Mikhail Rudenko <mike.rudenko@gmail.com>
Reviewed-by: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
pt3_probe() did not free one of IO mappings in case when one of them was
successful while another one failed. The patch fixed that by using
functions pcim_*. Also, it simplifies error handling through switching
to devm_* functions.
Found by Linux Driver Verification project (linuxtesting.org).
Link: https://lore.kernel.org/linux-media/20210814155742.11392-1-novikov@ispras.ru
Signed-off-by: Evgeny Novikov <novikov@ispras.ru>
Co-developed-by: Kirill Shilimanov <kirill.shilimanov@huawei.com>
Signed-off-by: Kirill Shilimanov <kirill.shilimanov@huawei.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Tested-by: Akihiro Tsukada <tskd08@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
dvb_usb_device_init stores parts of properties at d->props
and d->desc and uses it on dvb_usb_device_exit.
Free of properties on module probe leads to use after free.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=204597
The patch makes properties static instead of allocated on heap to prevent
memleak and use after free.
Also fixes s421_properties.devices initialization to have 2 element
instead of 6 copied from p7500_properties.
[mchehab: fix function call alignments]
Link: https://lore.kernel.org/linux-media/20190822104147.4420-1-vasilyev@ispras.ru
Signed-off-by: Anton Vasilyev <vasilyev@ispras.ru>
Fixes: 299c7007e9 ("media: dw2102: Fix memleak on sequence of probes")
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Step 5 expects that the sensor is in LP11 mode. Use the new
pre_streamon callback to signal the sensor that it should switch into
LP11.
Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
GPIO library does copy the of_node from the parent device of
the GPIO chip, there is no need to repeat this in the individual
drivers. Remove assignment here.
For the details one may look into the of_gpio_dev_init() implementation.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The lirc core already introduces gaps, so there is no need for this in
the driver.
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
These struct members do not serve any purpose.
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Remove superfluous messages which add no information.
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
The usb pid/vid can be found elsewhere, the idVendor/idProduct usb sysfs
files for example.
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
There is no reason to have a reset after an IR timeout.
Calling ir_raw_event_handle() twice for the same interrupt has no
affect.
Fixes: 56b0ec30c4 ("[media] rc/streamzap: fix reporting response times")
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Without this, some IR will be missing mid-stream and we might decode
something which never really occurred.
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
An IR reset is only used when the IR hardware reports an error.
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
An IR reset is only used when the IR hardware reports an error.
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Without timeout reports, it is impossible to decode many protocols since
it is not known when the transmission ends. timeout reports are sent by
default, but can be turned off. There is no reason to turn them off, and
I cannot find any software which does this, so we can safely remove it.
This makes the ioctl LIRC_SET_REC_TIMEOUT_REPORTS a no-op.
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Previously we passed a struct pci_dev * to mtip_check_surprise_removal(),
which immediately looked up the driver_data. But all callers already have
the driver_data pointer, so just pass it directly and skip the extra
lookup. No functional change intended.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20211208192449.146076-3-helgaas@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The .suspend() and .resume() methods are only called after the .probe()
method (mtip_pci_probe()) has set the drvdata and returned success.
Therefore, if we get to mtip_pci_suspend() or mtip_pci_resume(), the
drvdata must be valid. Drop the unnecessary checking.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20211208192449.146076-2-helgaas@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The donation calculation logic assumes that the donor has non-zero
after-donation hweight, so the lowest active hweight a donating cgroup can
have is 2 so that it can donate 1 while keeping the other 1 for itself.
Earlier, we only donated from cgroups with sizable surpluses so this
condition was always true. However, with the precise donation algorithm
implemented, f1de2439ec ("blk-iocost: revamp donation amount
determination") made the donation amount calculation exact enabling even low
hweight cgroups to donate.
This means that in rare occasions, a cgroup with active hweight of 1 can
enter donation calculation triggering the following warning and then a
divide-by-zero oops.
WARNING: CPU: 4 PID: 0 at block/blk-iocost.c:1928 transfer_surpluses.cold+0x0/0x53 [884/94867]
...
RIP: 0010:transfer_surpluses.cold+0x0/0x53
Code: 92 ff 48 c7 c7 28 d1 ab b5 65 48 8b 34 25 00 ae 01 00 48 81 c6 90 06 00 00 e8 8b 3f fe ff 48 c7 c0 ea ff ff ff e9 95 ff 92 ff <0f> 0b 48 c7 c7 30 da ab b5 e8 71 3f fe ff 4c 89 e8 4d 85 ed 74 0
4
...
Call Trace:
<IRQ>
ioc_timer_fn+0x1043/0x1390
call_timer_fn+0xa1/0x2c0
__run_timers.part.0+0x1ec/0x2e0
run_timer_softirq+0x35/0x70
...
iocg: invalid donation weights in /a/b: active=1 donating=1 after=0
Fix it by excluding cgroups w/ active hweight < 2 from donating. Excluding
these extreme low hweight donations shouldn't affect work conservation in
any meaningful way.
Signed-off-by: Tejun Heo <tj@kernel.org>
Fixes: f1de2439ec ("blk-iocost: revamp donation amount determination")
Cc: stable@vger.kernel.org # v5.10+
Link: https://lore.kernel.org/r/Ybfh86iSvpWKxhVM@slm.duckdns.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
* for-next/perf-user-counter-access:
Documentation: arm64: Document PMU counters access from userspace
arm64: perf: Enable PMU counter userspace access for perf event
arm64: perf: Add userspace counter access disable switch
perf: Add a counter for number of user access events in context
x86: perf: Move RDPMC event flag to a common definition
* for-next/perf-smmu:
perf/smmuv3: Synthesize IIDR from CoreSight ID registers
perf/smmuv3: Add devicetree support
dt-bindings: Add Arm SMMUv3 PMCG binding
Some powers were changed during the jack insert detection
and clk's enable/disable in CCF.
If in parallel, the influence has a chance to detect
the wrong jack type, so add a lock.
Signed-off-by: Derek Fang <derek.fang@realtek.com>
Link: https://lore.kernel.org/r/20211214105033.471-1-derek.fang@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Merge series from Stephan Gerhold <stephan@gerhold.net>:
This series makes it possible to route audio through the combined
audio/modem DSP on MSM8916/APQ8016 devices instead of bypassing it using
the LPASS drivers. This is necessary to support certain functionality such
as voice call audio. See PATCH 4/5 for details.
Also, qcom,apq8016-sbc.txt is converted to DT schema by adding it to the
existing qcom,sm8250.yaml. The bindings are similar enough that it is easier
to share a single schema instead of duplicating everything into multiple ones.
Fix the following sparse warning: (new ones prefixed by >>)
>> sound/soc/mediatek/mt8192/mt8192-mt6359-rt1015-rt5682.c:370:33:
sparse: sparse: incorrect type in argument 3 (different base types)
sound/soc/mediatek/mt8192/mt8192-mt6359-rt1015-rt5682.c:370:33: sparse:
expected unsigned int to
sound/soc/mediatek/mt8192/mt8192-mt6359-rt1015-rt5682.c:370:33: sparse:
got restricted snd_pcm_format_t [usertype]
Correct discription of format, use S32_LE and S24_LE to distinguish the
different 32bit.
Signed-off-by: Jiaxin Yu <jiaxin.yu@mediatek.com>
Reported-by: kernel test robot <lkp@intel.com>
Reviewed-by: Tzung-Bi Shih <tzungbi@google.com>
Link: https://lore.kernel.org/r/20211209073224.21793-1-jiaxin.yu@mediatek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The only usage of sof_probe_compr_ops is to assign its address to the
cops field in the snd_soc_dai_driver struct (in
sound/soc/sof/intel/hda-dai.c). Make it const to allow the compiler to
put it in read-only memory.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Link: https://lore.kernel.org/r/20211204203220.54712-1-rikard.falkeborn@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In SOF OF device, pm_runtime_put_autosuspend() is not
matching any pm_runtime_get_sync().
This is imbalanced for PM runtime.
Also, for consistency we call pm_runtime_mark_last_busy()
before enabling PM runtime.
1. Remove pm_runtime_put_autosuspend() in probe_complete
2. Reorder PM runtime calls int probe_complete
Signed-off-by: Allen-KH Cheng <Allen-KH.Cheng@mediatek.com>
Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Link: https://lore.kernel.org/r/20211209200830.145005-1-daniel.baluta@oss.nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The FEAT_LSE atomic instructions include LD* instructions which return
the original value of a memory location can be used to directly
implement FETCH opertations. Each RETURN op is implemented as a copy of
the corresponding FETCH op with a trailing instruction to generate the
new value of the memory location. We only directly implement
*_fetch_add*(), for which we have a trailing `add` instruction.
As the compiler has no visibility of the `add`, this leads to less than
optimal code generation when consuming the result.
For example, the compiler cannot constant-fold the addition into later
operations, and currently GCC 11.1.0 will compile:
return __lse_atomic_sub_return(1, v) == 0;
As:
mov w1, #0xffffffff
ldaddal w1, w2, [x0]
add w1, w1, w2
cmp w1, #0x0
cset w0, eq // eq = none
ret
This patch improves this by replacing the `add` with C addition after
the inline assembly block, e.g.
ret += i;
This allows the compiler to manipulate `i`. This permits the compiler to
merge the `add` and `cmp` for the above, e.g.
mov w1, #0xffffffff
ldaddal w1, w1, [x0]
cmp w1, #0x1
cset w0, eq // eq = none
ret
With this change the assembly for each RETURN op is identical to the
corresponding FETCH op (including barriers and clobbers) so I've removed
the inline assembly and rewritten each RETURN op in terms of the
corresponding FETCH op, e.g.
| static inline void __lse_atomic_add_return(int i, atomic_t *v)
| {
| return __lse_atomic_fetch_add(i, v) + i
| }
The new construction does not adversely affect the common case, and
before and after this patch GCC 11.1.0 can compile:
__lse_atomic_add_return(i, v)
As:
ldaddal w0, w2, [x1]
add w0, w0, w2
... while having the freedom to do better elsewhere.
This is intended as an optimization and cleanup.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211210151410.2782645-6-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
We have overly conservative assembly constraints for the basic FEAT_LSE
atomic instructions, and using more accurate and permissive constraints
will allow for better code generation.
The FEAT_LSE basic atomic instructions have come in two forms:
LD{op}{order}{size} <Rs>, <Rt>, [<Rn>]
ST{op}{order}{size} <Rs>, [<Rn>]
The ST* forms are aliases of the LD* forms where:
ST{op}{order}{size} <Rs>, [<Rn>]
Is:
LD{op}{order}{size} <Rs>, XZR, [<Rn>]
For either form, both <Rs> and <Rn> are read but not written back to,
and <Rt> is written with the original value of the memory location.
Where (<Rt> == <Rs>) or (<Rt> == <Rn>), <Rt> is written *after* the
other register value(s) are consumed. There are no UNPREDICTABLE or
CONSTRAINED UNPREDICTABLE behaviours when any pair of <Rs>, <Rt>, or
<Rn> are the same register.
Our current inline assembly always uses <Rs> == <Rt>, treating this
register as both an input and an output (using a '+r' constraint). This
forces the compiler to do some unnecessary register shuffling and/or
redundant value generation.
For example, the compiler cannot reuse the <Rs> value, and currently GCC
11.1.0 will compile:
__lse_atomic_add(1, a);
__lse_atomic_add(1, b);
__lse_atomic_add(1, c);
As:
mov w3, #0x1
mov w4, w3
stadd w4, [x0]
mov w0, w3
stadd w0, [x1]
stadd w3, [x2]
We can improve this with more accurate constraints, separating <Rs> and
<Rt>, where <Rs> is an input-only register ('r'), and <Rt> is an
output-only value ('=r'). As <Rt> is written back after <Rs> is
consumed, it does not need to be earlyclobber ('=&r'), leaving the
compiler free to use the same register for both <Rs> and <Rt> where this
is desirable.
At the same time, the redundant 'r' constraint for `v` is removed, as
the `+Q` constraint is sufficient.
With this change, the above example becomes:
mov w3, #0x1
stadd w3, [x0]
stadd w3, [x1]
stadd w3, [x2]
I've made this change for the non-value-returning and FETCH ops. The
RETURN ops have a multi-instruction sequence for which we cannot use the
same constraints, and a subsequent patch will rewrite hte RETURN ops in
terms of the FETCH ops, relying on the ability for the compiler to reuse
the <Rs> value.
This is intended as an optimization.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211210151410.2782645-5-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
The FEAT_LSE atomic instructions include atomic bit-clear instructions
(`ldclr*` and `stclr*`) which can be used to directly implement ANDNOT
operations. Each AND op is implemented as a copy of the corresponding
ANDNOT op with a leading `mvn` instruction to apply a bitwise NOT to the
`i` argument.
As the compiler has no visibility of the `mvn`, this leads to less than
optimal code generation when generating `i` into a register. For
example, __lse_atomic_fetch_and(0xf, v) can be compiled to:
mov w1, #0xf
mvn w1, w1
ldclral w1, w1, [x2]
This patch improves this by replacing the `mvn` with NOT in C before the
inline assembly block, e.g.
i = ~i;
This allows the compiler to generate `i` into a register more optimally,
e.g.
mov w1, #0xfffffff0
ldclral w1, w1, [x2]
With this change the assembly for each AND op is identical to the
corresponding ANDNOT op (including barriers and clobbers), so I've
removed the inline assembly and rewritten each AND op in terms of the
corresponding ANDNOT op, e.g.
| static inline void __lse_atomic_and(int i, atomic_t *v)
| {
| return __lse_atomic_andnot(~i, v);
| }
This is intended as an optimization and cleanup.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211210151410.2782645-4-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>