There is a deadlock in rtw_joinbss_event_prehandle(), which is shown
below:
(Thread 1) | (Thread 2)
| _set_timer()
rtw_joinbss_event_prehandle()| mod_timer()
spin_lock_bh() //(1) | (wait a time)
... | _rtw_join_timeout_handler()
del_timer_sync() | spin_lock_bh() //(2)
(wait timer to stop) | ...
We hold pmlmepriv->lock in position (1) of thread 1 and
use del_timer_sync() to wait timer to stop, but timer handler
also need pmlmepriv->lock in position (2) of thread 2.
As a result, rtw_joinbss_event_prehandle() will block forever.
This patch extracts del_timer_sync() from the protection of
spin_lock_bh(), which could let timer handler to obtain
the needed lock. What`s more, we change spin_lock_bh() to
spin_lock_irq() in _rtw_join_timeout_handler() in order to
prevent deadlock.
Signed-off-by: Duoming Zhou <duoming@zju.edu.cn>
Link: https://lore.kernel.org/r/20220409064953.67420-1-duoming@zju.edu.cn
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is a deadlock in rtw_surveydone_event_callback(),
which is shown below:
(Thread 1) | (Thread 2)
| _set_timer()
rtw_surveydone_event_callback()| mod_timer()
spin_lock_bh() //(1) | (wait a time)
... | rtw_scan_timeout_handler()
del_timer_sync() | spin_lock_bh() //(2)
(wait timer to stop) | ...
We hold pmlmepriv->lock in position (1) of thread 1 and use
del_timer_sync() to wait timer to stop, but timer handler
also need pmlmepriv->lock in position (2) of thread 2.
As a result, rtw_surveydone_event_callback() will block forever.
This patch extracts del_timer_sync() from the protection of
spin_lock_bh(), which could let timer handler to obtain
the needed lock. What`s more, we change spin_lock_bh() in
rtw_scan_timeout_handler() to spin_lock_irq(). Otherwise,
spin_lock_bh() will also cause deadlock() in timer handler.
Signed-off-by: Duoming Zhou <duoming@zju.edu.cn>
Link: https://lore.kernel.org/r/20220409061836.60529-1-duoming@zju.edu.cn
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Adhere to linux coding style.
Reported by checkpatch:
spaces preferred around that '{operator}'.
Signed-off-by: Mahak Gupta <mahak_g@cs.iitr.ac.in>
Link: https://lore.kernel.org/r/20220413054517.6343-1-mahak_g@cs.iitr.ac.in
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove the optional remove callback for the soc_codec_dev_gbaudio
structure. The only place it is referenced is
snd_soc_component_remove() which is only called if the sound_component
pointer is non-null. The null function pointers here can be optionally
ommitted. When a sound component is registered this way, the remove
callback is optional. We can safely remove the whole gbcodec_remove
function, which used to be an empty function with a void return type.
Suggested-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jaehee Park <jhpark1013@gmail.com>
Link: https://lore.kernel.org/r/12037ae2502ad7d0311bcdf2178c3d2156293236.1649824370.git.jhpark1013@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Else is not necessary after return and break statements, hence remove
it.
Reported by checkpatch:
WARNING: else is not generally useful after a break or return
Signed-off-by: Mahak Gupta <mahak_g@cs.iitr.ac.in>
Link: https://lore.kernel.org/r/20220413052759.4859-1-mahak_g@cs.iitr.ac.in
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove compat ALSA card, which has overlapping functionality with the
two other cards described by the driver (HDMI and headphones)
This handles TODO item "Revisit multi-cards options and PCM route mixer
control".
Move the S/PDIF device that was part of the compat ALSA card to the HDMI
card.
Only enable headphones card by default, because HDMI breaks when using
both vc4 and bcm2835-audio with HDMI card enabled.
Signed-off-by: Adrien Thierry <athierry@redhat.com>
Link: https://lore.kernel.org/r/20220408150359.26661-1-athierry@redhat.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead of "oneOf:" choices, use "allOf:" and "if:" to define clocks,
clock-names, resets, and reset-names that can be taken by the compatible
string.
The order of clock-names and reset-names doesn't change here.
Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/1648637715-19262-6-git-send-email-hayashi.kunihiko@socionext.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
There is no clock-names and reset-names for Pro5 SoC, that should have two
properties, "gio" and "link" like usb3-ssphy.
And according to the existing PXs2 devicetree, the clock-names for PXs2 SoC
should have "link" and "phy", and minItems of clocks should be 2.
Fixes: 134ab2845a ("dt-bindings: phy: Convert UniPhier USB3-PHY conroller to json-schema")
Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/1648637715-19262-5-git-send-email-hayashi.kunihiko@socionext.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The usb-device doesn't need "vbus-supply" property, so the property should
be removed from required.
Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/1648637715-19262-3-git-send-email-hayashi.kunihiko@socionext.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
UniPhier USB2 PHY controller can apply vbus-supply.
Add "vbus-supply" property to fix the following warning.
uniphier-pro4-ace.dtb: usb-glue: phy@2: 'vbus-supply' does not match any of the regexes: 'pinctrl-[0-9]+'
From schema: Documentation/devicetree/bindings/phy/socionext,uniphier-usb2-phy.yaml
Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/1648637715-19262-2-git-send-email-hayashi.kunihiko@socionext.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
DCTL is short for D-MAC control that V1 chip uses this H2C to configure
security CAM. Implement the callers in next patch.
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20220413010804.8941-3-pkshih@realtek.com
In order to support new chip that has capability of 160M, we need new
format to fill new information, so add a new V1 ID for newer use. Since
most fields are the same, fill fields according to the function ID of chip.
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20220413010804.8941-2-pkshih@realtek.com
The recent change for memory allocator replaced the SG-buffer handling
helper for x86 with the standard non-contiguous page handler. This
works for most cases, but there is a corner case I obviously
overlooked, namely, the fallback of non-contiguous handler without
IOMMU. When the system runs without IOMMU, the core handler tries to
use the continuous pages with a single SGL entry. It works nicely for
most cases, but when the system memory gets fragmented, the large
allocation may fail frequently.
Ideally the non-contig handler could deal with the proper SG pages,
it's cumbersome to extend for now. As a workaround, here we add new
types for (minimalistic) SG allocations, instead, so that the
allocator falls back to those types automatically when the allocation
with the standard API failed.
BTW, one better (but pretty minor) improvement from the previous
SG-buffer code is that this provides the proper mmap support without
the PCM's page fault handling.
Fixes: 2c95b92ecd ("ALSA: memalloc: Unify x86 SG-buffer handling (take#3)")
BugLink: https://gitlab.freedesktop.org/pipewire/pipewire/-/issues/2272
BugLink: https://bugzilla.suse.com/show_bug.cgi?id=1198248
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20220413054808.7547-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Adds the secure capability OPTEE_FFA_SEC_CAP_ARG_OFFSET to indicate that
OP-TEE with FF-A can support an argument struct at a non-zero offset into
a passed shared memory object.
Signed-off-by: Jens Wiklander <jens.wiklander@linaro.org>
Adds OPTEE_SMC_CALL_WITH_RPC_ARG and OPTEE_SMC_CALL_WITH_REGD_ARG where
the struct optee_msg_arg to be used for RPC is appended in the memory
following the normal argument struct optee_msg_arg. This is an
optimization to avoid caching the RPC argument struct while still
maintaining similar performance as if it was cached.
OPTEE_SMC_CALL_WITH_REGD_ARG optimized one step further by using a
registered shared memory object instead. It's in other aspects identical
to OPTEE_SMC_CALL_WITH_RPC_ARG.
The presence of OPTEE_SMC_CALL_WITH_RPC_ARG and
OPTEE_SMC_CALL_WITH_REGD_ARG is indicated by the new
OPTEE_SMC_SEC_CAP_RPC_ARG bit returned by
OPTEE_SMC_EXCHANGE_CAPABILITIES. OPTEE_SMC_EXCHANGE_CAPABILITIES also
reports the number of arguments that the RPC argument struct must have
room for.
OPTEE_SMC_CALL_WITH_RPC_ARG and OPTEE_SMC_CALL_WITH_ARG can be used
interleaved with difference that when OPTEE_SMC_CALL_WITH_RPC_ARG is
used the RPC argument struct to be used is the one appended to the
normal argument struct. The same is true for
OPTEE_SMC_CALL_WITH_REGD_ARG.
Reviewed-by: Sumit Garg <sumit.garg@linaro.org>
Signed-off-by: Jens Wiklander <jens.wiklander@linaro.org>
Renames the field rpc_arg_count in struct optee to rpc_param_count.
Function parameter names and local variables are also renamed to match.
Reviewed-by: Sumit Garg <sumit.garg@linaro.org>
Signed-off-by: Jens Wiklander <jens.wiklander@linaro.org>
The local variables 'T' and 'UT' are needlessly declared as *unsigned*
*long* -- the corresponding parameters of ata_timing_compute() are both
declared as *int*. While fixing up those declarations, also make the
'via_clock' and 'T' variables *const* as they are never re-assigned
after initialization -- the object code should remain the same as gcc
previously used copy propagation anyway...
Found by Linux Verification Center (linuxtesting.org) with the SVACE static
analysis tool.
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
sil680_sel{dev|reg}() return a PCI config space address but needlessly
use the *unsigned long* type for that, whereas the PCI config space
accessors take *int* for the address parameter. Switch these functions
to returning *int*, updating the local variables at their call sites.
Get rid of the 'base' local variables in these functions, while at it...
Found by Linux Verification Center (linuxtesting.org) with the SVACE static
analysis tool.
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Add devicetree support for SDHCI controller found in Qualcomm SDX65
platform. The SDHCI controller is based on the MSM SDHCI v5 IP.
Signed-off-by: Rohit Agarwal <quic_rohiagar@quicinc.com>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1649670615-21268-4-git-send-email-quic_rohiagar@quicinc.com
Align RPM requests node with DT schema by using hyphen instead of
underscore.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
[bjorn: Fixed up qcom-{apq8074,msm8974}-*.dts to match the qcom-msm8974.dtsi]
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20220401201035.189106-8-krzysztof.kozlowski@linaro.org
After adding DT schema for the SPI controller, the Qualcomm GENI Serial
Engine QUP Wrapper Controller bindings can reference it directly for
full schema validation.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Kuldeep Singh <singh.kuldeep87k@gmail.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20220404064017.68634-2-krzysztof.kozlowski@linaro.org
The RPM clock controller receive input clock ("xo"). It is modelled on
only one chip - MSM8953.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20220401201035.189106-11-krzysztof.kozlowski@linaro.org
Convert the Qualcomm RPM Clock Controller bindings to DT schema and
include it in parent's schema (SMD RPM).
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20220401201035.189106-10-krzysztof.kozlowski@linaro.org
Convert the Qualcomm Shared Memory Driver bindings to DT Schema.
Changes against original bindings: enforce only specific names of child
nodes, instead of any names.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20220401201035.189106-9-krzysztof.kozlowski@linaro.org
Remove regulator-always-on property where not necessary and mark regulators
that are not supposed to be voted active on boot with regulator-boot-on.
While at it, reorder the load properties to make it look more decent.
Reorder PMICs to fix a probe defer caused by messy dependencies and Linux's
inability to handle them (at least for now).
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20220319174645.340379-12-konrad.dybcio@somainline.org
Make sure the necessary clocks are kept on after clk_cleanup (until MDSS
is properly handled by its own driver) and touch up the fb address to
prevent some weird shifting. It's still not perfect, but at least the
kernel log doesn't start a third deep into your screen..
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
[bjorn: Folded in change of framebuffer base address, from Konrad]
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20220319174645.340379-10-konrad.dybcio@somainline.org
