Basic helpers have been extracted, now there's a pile more todo still
across the entire tree.
Reviewed-by: Daniel Thompson <daniel.thompson@linaro.org>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-57-daniel.vetter@ffwll.ch
This is now done in drm_prime.c
Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: Rob Herring <robh@kernel.org>
Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-47-daniel.vetter@ffwll.ch
If we use the gem fb helper as the prepare_fb hook, plus the
drm_prime.c import helpers now automatically setting obj->resv, we can
use the shared helpers to wait for fences instead of rolling our own.
Note that this relies on mtk setting drm_fb->obj, which is already
done in mtk_drm_framebuffer_init().
Aside: Probably can use the default commit_tail with this again, but I
didn't check for that.
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: CK Hu <ck.hu@mediatek.com>
Cc: Philipp Zabel <p.zabel@pengutronix.de>
Cc: Matthias Brugger <matthias.bgg@gmail.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-mediatek@lists.infradead.org
Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-45-daniel.vetter@ffwll.ch
The code in quota_getstate and quota_getstatev is strange; it
says the returned fs_quota_stat[v] structure has room for only
one type of time limits, so fills it in with the first enabled
quota, even though every quotactl command must have a type sent
in by the user.
Instead of just picking the first enabled quota, fill in the
reply with the timers for the quota type that was actually
requested.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Currently arm64 uses the default implementation of panic_smp_self_stop()
where the CPU runs in a cpu_relax() loop unable to receive IPIs anymore.
As a result, when two CPUs panic() simultaneously we get "SMP: failed to
stop secondary CPUs" warnings and extra delays before a reset, because
smp_send_stop() still tries to stop the other paniced CPU.
Provide an implementation of panic_smp_self_stop() that is identical to
the IPI CPU stop handler, so that the online status of stopped CPUs gets
properly updated.
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
The current code puts the stopped cpus in an 'yield' instruction loop.
Using a busy loop here is unnecessary, we can use the cpu_park_loop()
function here to do a wfi/wfe.
Signed-off-by: Jayachandran C <jnair@caviumnetworks.com>
Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
ZAC support added sense data requesting on error for both ZAC and ATA
devices. This seems to cause erratic error handling behaviors on some
SSDs where the device reports sense data availability and then
delivers the wrong content making EH take the wrong actions. The
failure mode was sporadic on a LITE-ON ssd and couldn't be reliably
reproduced.
There is no value in requesting sense data from non-ZAC ATA devices
while there's a significant risk of introducing EH misbehaviors which
are difficult to reproduce and fix. Let's do the sense data dancing
only for ZAC devices.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Tested-by: Masato Suzuki <masato.suzuki@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The memory regions specified by /memreserve/ are passed to
early_init_dt_reserve_memory_arch() with nomap=false, so it is
not suitable for reserving memory for Trusted Firmware-A etc.
Use the more robust /reserved-memory node with the no-map property
to prevent the kernel from mapping it.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
With commit d8e8fd0ebf ("mtd: rawnand: denali: decouple controller
and NAND chips"), the Denali NAND controller driver migrated to the
new controller/chip representation.
Update DT for it.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Consider, on one side, a bfq_queue Q that remains empty while in
service, and, on the other side, the pending I/O of bfq_queues that,
according to their timestamps, have to be served after Q. If an
uncontrolled amount of I/O from the latter bfq_queues were dispatched
while Q is waiting for its new I/O to arrive, then Q's bandwidth
guarantees would be violated. To prevent this, I/O dispatch is plugged
until Q receives new I/O (except for a properly controlled amount of
injected I/O). Unfortunately, preemption breaks I/O-dispatch plugging,
for the following reason.
Preemption is performed in two steps. First, Q is expired and
re-scheduled. Second, the new bfq_queue to serve is chosen. The first
step is needed by the second, as the second can be performed only
after Q's timestamps have been properly updated (done in the
expiration step), and Q has been re-queued for service. This
dependency is a consequence of the way how BFQ's scheduling algorithm
is currently implemented.
But Q is not re-scheduled at all in the first step, because Q is
empty. As a consequence, an uncontrolled amount of I/O may be
dispatched until Q becomes non empty again. This breaks Q's service
guarantees.
This commit addresses this issue by re-scheduling Q even if it is
empty. This in turn breaks the assumption that all scheduled queues
are non empty. Then a few extra checks are now needed.
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
BFQ enqueues the I/O coming from each process into a separate
bfq_queue, and serves bfq_queues one at a time. Each bfq_queue may be
served for at most timeout_sync milliseconds (default: 125 ms). This
service scheme is prone to the following inaccuracy.
While a bfq_queue Q1 is in service, some empty bfq_queue Q2 may
receive I/O, and, according to BFQ's scheduling policy, may become the
right bfq_queue to serve, in place of the currently in-service
bfq_queue. In this respect, postponing the service of Q2 to after the
service of Q1 finishes may delay the completion of Q2's I/O, compared
with an ideal service in which all non-empty bfq_queues are served in
parallel, and every non-empty bfq_queue is served at a rate
proportional to the bfq_queue's weight. This additional delay is equal
at most to the time Q1 may unjustly remain in service before switching
to Q2.
If Q1 and Q2 have the same weight, then this time is most likely
negligible compared with the completion time to be guaranteed to Q2's
I/O. In addition, first, one of the reasons why BFQ may want to serve
Q1 for a while is that this boosts throughput and, second, serving Q1
longer reduces BFQ's overhead. As a conclusion, it is usually better
not to preempt Q1 if both Q1 and Q2 have the same weight.
In contrast, as Q2's weight or priority becomes higher and higher
compared with that of Q1, the above delay becomes larger and larger,
compared with the I/O completion times that have to be guaranteed to
Q2 according to Q2's weight. So reducing this delay may be more
important than avoiding the costs of preempting Q1.
Accordingly, this commit preempts Q1 if Q2 has a higher weight or a
higher priority than Q1. Preemption causes Q1 to be re-scheduled, and
triggers a new choice of the next bfq_queue to serve. If Q2 really is
the next bfq_queue to serve, then Q2 will be set in service
immediately.
This change reduces the component of the I/O latency caused by the
above delay by about 80%. For example, on an (old) PLEXTOR PX-256M5
SSD, the maximum latency reported by fio drops from 15.1 to 3.2 ms for
a process doing sporadic random reads while another process is doing
continuous sequential reads.
Signed-off-by: Nicola Bottura <bottura.nicola95@gmail.com>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A bfq_queue Q may happen to be synchronized with another
bfq_queue Q2, i.e., the I/O of Q2 may need to be completed for Q to
receive new I/O. We call Q2 "waker queue".
If I/O plugging is being performed for Q, and Q is not receiving any
more I/O because of the above synchronization, then, thanks to BFQ's
injection mechanism, the waker queue is likely to get served before
the I/O-plugging timeout fires.
Unfortunately, this fact may not be sufficient to guarantee a high
throughput during the I/O plugging, because the inject limit for Q may
be too low to guarantee a lot of injected I/O. In addition, the
duration of the plugging, i.e., the time before Q finally receives new
I/O, may not be minimized, because the waker queue may happen to be
served only after other queues.
To address these issues, this commit introduces the explicit detection
of the waker queue, and the unconditional injection of a pending I/O
request of the waker queue on each invocation of
bfq_dispatch_request().
One may be concerned that this systematic injection of I/O from the
waker queue delays the service of Q's I/O. Fortunately, it doesn't. On
the contrary, next Q's I/O is brought forward dramatically, for it is
not blocked for milliseconds.
Reported-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Tested-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Until the base value for request service times gets finally computed
for a bfq_queue, the inject limit for that queue does depend on the
think-time state (short|long) of the queue. A timely update of the
think time then guarantees a quicker activation or deactivation of the
injection. Fortunately, the think time of a bfq_queue is updated in
the same code path as the inject limit; but after the inject limit.
This commits moves the update of the think time before the update of
the inject limit. For coherence, it moves the update of the seek time
too.
Reported-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Tested-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
I/O injection gets reduced if it increases the request service times
of the victim queue beyond a certain threshold. The threshold, in its
turn, is computed as a function of the base service time enjoyed by
the queue when it undergoes no injection.
As a consequence, for injection to work properly, the above base value
has to be accurate. In this respect, such a value may vary over
time. For example, it varies if the size or the spatial locality of
the I/O requests in the queue change. It is then important to update
this value whenever possible. This commit performs this update.
Reported-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Tested-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
One of the cases where the parameters for injection may be updated is
when there are no more in-flight I/O requests. The number of in-flight
requests is stored in the field bfqd->rq_in_driver of the descriptor
bfqd of the device. So, the controlled condition is
bfqd->rq_in_driver == 0.
Unfortunately, this is wrong because, the instruction that checks this
condition is in the code path that handles the completion of a
request, and, in particular, the instruction is executed before
bfqd->rq_in_driver is decremented in such a code path.
This commit fixes this issue by just replacing 0 with 1 in the
comparison.
Reported-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Tested-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Until the base value of the request service times gets finally
computed for a bfq_queue, the inject limit does depend on the
think-time state (short|long). The limit must be 0 or 1 if the think
time is deemed, respectively, as short or long. However, such a check
and possible limit update is performed only periodically, once per
second. So, to make the injection mechanism much more reactive, this
commit performs the update also every time the think-time state
changes.
In addition, in the following special case, this commit lets the
inject limit of a bfq_queue bfqq remain equal to 1 even if bfqq's
think time is short: bfqq's I/O is synchronized with that of some
other queue, i.e., bfqq may receive new I/O only after the I/O of the
other queue is completed. Keeping the inject limit to 1 allows the
blocking I/O to be served while bfqq is in service. And this is very
convenient both for bfqq and for the total throughput, as explained
in detail in the comments in bfq_update_has_short_ttime().
Reported-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Tested-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
With commit d8e8fd0ebf ("mtd: rawnand: denali: decouple controller
and NAND chips"), the Denali NAND controller driver migrated to the
new controller/chip representation.
Update DT for it.
In the new binding, the number of connected chips are described in
DT instead of run-time probed.
I added just one chip to the reference boards, where we do not know
if the on-board NAND device is a single chip or multiple chips.
If we added too many chips into DT, it would end up with the timeout
error in nand_scan_ident().
I changed all the pinctrl properties to use the single CS.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
cAVS modules do not require Type and Length header within the
set_module_params IPC. This is also true for Vendor modules. The
userspace (like tinymix) always appends this header to TLV controls
which are used for set_module_params. Simply assume this header is
always present in the payload and omit it from the IPC.
Signed-off-by: Kamil Lulko <kamilx.lulko@intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Before actual deletion, pipeline should enter RESET state. Currently,
pipe skips this checkpoint and goes straight to the finish line.
This is not the expected path by the FW, so correct it.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
rx_data and rx_bytes present for tx_wait_done are optional parameters.
If not provided, function should not attempt to copy received data.
This change fixes memcpy NULL pointer dereference issue occurring when
optional rx_data is NULL while received message size is non-zero.
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Since there are multiple IPCs being sent in a short span of time, there
is a possibility of more than one message being on the Rx list after
receiving response from firmware. In such cases, when the first
notification of interrupt from firmware is received, driver retrieves
the message from the Rx list but does not delete it from the list till
the next lock. In the meantime, when another interrupt is received from
the firmware, driver is reading the previous message again since the
previous message has not been removed from the list.
Signed-off-by: Gustaw Lewandowski <gustaw.lewandowski@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Host clears DONE bit to signal IPC target it has completed the
operation. Once this is done, IPC target i.e. DSP may proceed with the
next reply, filling registers with new portion of data.
Because of this, host should always read all registers prior to clearing
DONE and BUSY bits to ensure no desynchronization happens the time in
between clearing bits and reading message data (here, extension).
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
There are a few soc_tplg_dapm_widget_*_create functions with similar
content, but slightly different flow, unify their flow and make sure
that we go to error handler and free memory in case of failure.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Acked-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Provide helper functions and use them to free dtexts and dvalues in
topology. This is followup cleanup after related changes in this area as
suggested in:
https://mailman.alsa-project.org/pipermail/alsa-devel/2019-January/144761.html
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
When we unload Skylake driver we may end up calling
hdac_component_master_unbind(), it uses acomp->audio_ops, which we set
in hdmi_codec_probe(), so we need to set it to NULL in hdmi_codec_remove(),
otherwise we will dereference no longer existing pointer.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
When driver is probed, we iterate over NHLT and check if clk entries are
present. For each such entry we call register_skl_clk and keep the
result in data->clk[].
Currently data->clk is sparsely indexed using NHLT table iterator, while
when freeing we use number of registered entries. Let's just use
data->avail_clk_cnt as index, so it can be reset back in
unregister_src_clk.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
When we remove component we need to reverse things which were done on
init, this consists of topology cleanup, lists cleanup and releasing
firmware.
Currently cleanup handlers are put in wrong places or otherwise missing.
So add proper component cleanup function and perform cleanups in it.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Currently debugfs has no cleanup function. Add skl_debufs_exit function
so we can clean after ourselves properly.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
When we unload and reload machine driver, we shouldn't return that we
failed to initialize. This allows to reload machine driver, without
having to unload whole stack.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Change kzalloc to devm_kzalloc, so compr gets automatically freed when
it's no longer needed.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Currently on each driver reload internal counter is being increased. It
causes failure to enumerate driver devices, as they have hardcoded:
.codec_name = "ehdaudio0D2",
As there is currently no devices with multiple hda codecs and there is
currently no established way to reliably differentiate, between them,
always assign bus->idx = 0;
This fixes a problem when we unload and reload machine driver idx gets
incremented, so .codec_name would've needed to be set to "ehdaudio1D2"
after first reload and so on.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
If skl_probe_work() was not run driver ends up dereferencing NULL
pointer when operating on lists in skl_platform_unregister().
To fix this initialize lists in skl_create(). Also run
cancel_work_sync() before all cleanup functions, so we don't end up
unnecessarily running probe work.
Easily reproducible with:
while true; do modprobe snd_soc_skl; rmmod snd_soc_skl; done
(with the assumption that relevant drivers are added to blacklist on
system boot)
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
For BXT platforms, the recommended sequence to program the SDxFMT is to
first couple the stream, write the format and decouple again.
For all other platforms said sequence remains unchanged.
To prevent code duplication, IS_BXT (and consequently IS_CFL) macro is
relocated to hda_codec.h file so it can be accessed by SKL driver.
Signed-off-by: Paweł Harłoziński <pawel.harlozinski@intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
In case the requested gpio property is not found in the device tree, some
callers of gpiod_get_from_of_node() expect a return value of NULL, others
expect -ENOENT.
In particular devm_fwnode_get_index_gpiod_from_child() expects -ENOENT.
Currently it gets a NULL, which breaks the loop that tries all
gpio_suffixes. The result is that a gpio property is not found, even
though it is there.
This patch changes gpiod_get_from_of_node() to return -ENOENT instead
of NULL when the requested gpio property is not found in the device
tree. Additionally it modifies all calling functions to properly
evaluate the return value.
Another approach would be to leave the return value of
gpiod_get_from_of_node() as is and fix the bug in
devm_fwnode_get_index_gpiod_from_child(). Other callers would still need
to be reworked. The effort would be the same as with the chosen solution.
Signed-off-by: Georg Waibel <georg.waibel@sensor-technik.de>
Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
Clang warns when CONFIG_ACPI is unset:
drivers/pinctrl/qcom/pinctrl-sdm845.c:1320:5: warning: 'CONFIG_ACPI' is
not defined, evaluates to 0 [-Wundef]
#if CONFIG_ACPI
^
1 warning generated.
Use ifdef instead of if to resolve this.
Fixes: a229105d7a ("pinctrl: qcom: sdm845: Provide ACPI support")
Link: https://github.com/ClangBuiltLinux/linux/issues/569
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
readl_poll_timeout() calls usleep_range() to sleep between reads.
usleep_range() doesn't work efficiently for tiny values.
Raise the polling delay in qcom_qmp_phy_enable() to bring it in line
with the delay in qcom_qmp_phy_com_init().
Signed-off-by: Marc Gonzalez <marc.w.gonzalez@free.fr>
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
Locking is not needed for the phy_g12a_usb3_pcie_cr_bus_read/write() and
currently it causes the following BUG because of the usage of the
regmap_read_poll_timeout() running in spinlock_irq, configured by regmap fast_io.
Simply disable locking in the cr_regmap config since it's only used from the
PHY init callback function.
BUG: sleeping function called from invalid context at drivers/phy/amlogic/phy-meson-g12a-usb3-pcie.c:85
in_atomic(): 1, irqs_disabled(): 128, pid: 60, name: kworker/3:1
[snip]
Workqueue: events deferred_probe_work_func
Call trace:
dump_backtrace+0x0/0x190
show_stack+0x14/0x20
dump_stack+0x90/0xb4
___might_sleep+0xec/0x110
__might_sleep+0x50/0x88
phy_g12a_usb3_pcie_cr_bus_addr.isra.0+0x80/0x1a8
phy_g12a_usb3_pcie_cr_bus_read+0x34/0x1d8
_regmap_read+0x60/0xe0
_regmap_update_bits+0xc4/0x110
regmap_update_bits_base+0x60/0x90
phy_g12a_usb3_pcie_init+0xdc/0x210
phy_init+0x74/0xd0
dwc3_meson_g12a_probe+0x2cc/0x4d0
platform_drv_probe+0x50/0xa0
really_probe+0x20c/0x3b8
driver_probe_device+0x68/0x150
__device_attach_driver+0xa8/0x170
bus_for_each_drv+0x64/0xc8
__device_attach+0xd8/0x158
device_initial_probe+0x10/0x18
bus_probe_device+0x90/0x98
deferred_probe_work_func+0x94/0xe8
process_one_work+0x1e0/0x338
worker_thread+0x230/0x458
kthread+0x134/0x138
ret_from_fork+0x10/0x1c
Fixes: 36077e16c0 ("phy: amlogic: Add Amlogic G12A USB3 + PCIE Combo PHY Driver")
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Tested-by: Kevin Hilman <khilman@baylibre.com>
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
The actual layout for OCELOT_GPIO_ALT[01] when there are more than 32 pins
is interleaved, i.e. OCELOT_GPIO_ALT0[0], OCELOT_GPIO_ALT1[0],
OCELOT_GPIO_ALT0[1], OCELOT_GPIO_ALT1[1]. Introduce a new REG_ALT macro to
facilitate the register offset calculation and use it where necessary.
Fixes: da801ab56a pinctrl: ocelot: add MSCC Jaguar2 support
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
The third argument passed to REG is not the correct one and
ocelot_gpio_set_direction is not working for pins after 31. Fix that by
passing the pin number instead of the modulo 32 value.
Fixes: da801ab56a pinctrl: ocelot: add MSCC Jaguar2 support
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Some pin groups have park bits for multiple pins in one register.
Support this by turning the parked bit field into a parked bitmask
field. If no parked bits are supported, the bitmask can be 0.
Update the pingroup table on Tegra210, which is the only generation
where this is supported, with the parked bitmask.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Tested-by: Dmitry Osipenko <digetx@gmail.com>
Reviewed-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Rather than reuse the nvidia,tegra30-gpio compatible string
to find the GPIO controller on Tegra30, Tegra114, Tegra124 and
Tegra210, use the most specific compatible string for each
SoC generation for consistency.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Add a IIO driver that reports the angle between the lid and the base for
ChromeOS convertible device.
Tested on eve with ToT EC firmware.
Check driver is loaded and lid angle is correct.
Signed-off-by: Gwendal Grignou <gwendal@chromium.org>
Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
ARMv8.5 introduces the FRINT series of instructions for rounding floating
point numbers to integers. Provide a capability to userspace in order to
allow applications to determine if the system supports these instructions.
Signed-off-by: Mark Brown <broonie@kernel.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
The InfiniBand docs are plain text with no markups. So, all we needed to
do were to add the title markups and some markup sequences in order to
properly parse tables, lists and literal blocks.
At its new index.rst, let's add a :orphan: while this is not linked to the
main index.rst file, in order to avoid build warnings.
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
ARMv8.5 adds new instructions XAFLAG and AXFLAG to translate the
representation of the results of floating point comparisons between the
native ARM format and an alternative format used by some software. Add
a hwcap allowing userspace to determine if they are present, since we
referred to earlier CondM extensions as FLAGM call these extensions
FLAGM2.
Signed-off-by: Mark Brown <broonie@kernel.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
