Fixes the following W=1 kernel build warning(s):
drivers/soc/ti/knav_qmss_queue.c: In function ‘knav_setup_queue_pools’:
drivers/soc/ti/knav_qmss_queue.c:1310:6: warning: variable ‘ret’ set but not used [-Wunused-but-set-variable]
Cc: Santosh Shilimkar <ssantosh@kernel.org>
Cc: Sandeep Nair <sandeep_n@ti.com>
Cc: Cyril Chemparathy <cyril@ti.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
If a rstctrl reset bit is already deasserted, we can just bail out early
not wait for rstst to clear. Otherwise we can have deassert fail for
already deasserted resets.
Fixes: c5117a78dd ("soc: ti: omap-prm: poll for reset complete during de-assert")
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
pm_runtime_get_sync will increment pm usage counter even it
failed. Forgetting to pm_runtime_put_noidle will result in
reference leak in knav_queue_probe, so we should fix it.
Fixes: 41f93af900 ("soc: ti: add Keystone Navigator QMSS driver")
Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Since 5.10-rc1 i.MX is a devicetree-only platform and the existing
.id_table support in this driver was only useful for old non-devicetree
platforms.
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201116202403.29749-1-festevam@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Acked-by: Robert Jarzmik <robert.jarzmik@free.fr>
Cc: Daniel Mack <daniel@zonque.org>
Cc: Haojian Zhuang <haojian.zhuang@gmail.com>
Link: https://lore.kernel.org/r/20201027215252.25820-11-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Cc: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20201027215252.25820-10-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-9-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Acked-by: Zhou Wang <wangzhou1@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-8-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-7-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-6-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Acked-by: Jon Hunter <jonathanh@nvidia.com>
Cc: Laxman Dewangan <ldewangan@nvidia.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Link: https://lore.kernel.org/r/20201027215252.25820-5-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Cc: Green Wan <green.wan@sifive.com>
Link: https://lore.kernel.org/r/20201027215252.25820-4-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201027215252.25820-3-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-2-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The NAVSS UDMA will stuck if target IP module is disabled by PM while PSI-L
threads are paired UDMA<->IP and no further transfers is possible. This
could be the case for IPs J721E Main CPSW (cpsw9g).
Hence, to avoid such situation do PSI-L threads pairing only when UDMA
channel is going to be enabled as at this time DMA consumer module expected
to be active already.
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201030203000.4281-1-grygorii.strashko@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
When consumer requests channel power on the DMA controller device
and otherwise on the freeing channel resources.
Note, in some cases consumer acquires channel at the ->probe() stage and
releases it at the ->remove() stage. It will mean that DMA controller device
will be powered during all this time if there is no assist from hardware
to idle it. The above mentioned cases should be investigated separately
and individually.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lore.kernel.org/r/20201103183938.64752-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
In RING mode the ringacc does not access the ring memory. In this access
mode the ringacc coherency does not have meaning.
If the ring is configured in RING mode, then the ringacc itself will not
access to the ring memory. Only the requester (user) of the ring is going
to read/write to the memory.
Extend the ring configuration parameters with a device pointer to be used
for DMA API when the ring is configured in RING mode.
Extending the ring configuration struct will allow per ring selection of
device to be used for allocation, thus allowing per ring coherency.
To avoid regression, fall back to use the ringacc dev in case the alloc_dev
is not provided.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
The ringacc driver has been converted to use the new set_cfg function to
configure the ring, the old config ops can be removed.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Switch to the new set_cfg to configure the ring.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
The sysfw ring configuration message has been extended to include virtid
and asel value for the ring.
Add the ASEL_VALID to TI_SCI_MSG_VALUE_RM_ALL_NO_ORDER as it is required
for DMA rings.
Instead of extending the current .config() ops - which would need same
patch change in the ringacc driver - add ti_sci_msg_rm_ring_cfg struct and
a new ops using it to configure the ring.
This will allow easy update path in case new members are added for the ring
configuration.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
The ring_get_cfg (0x1111 message) is not used and it is not supported by
sysfw for a long time.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Sysfw added 'extended_ch_type' to the tx_ch_cfg_req message which should be
used when BCDMA block copy channels are configured:
extended_ch_type = 0 : the channel is split tx channel (tchan)
extended_ch_type = 1 : the channel is block copy channel (bchan)
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Allocate MSI entries for both first and second range if they are valid
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Sysfw added support for a second range in the resource range API to be able
to describe complex allocations mainly for DMA channels.
Update the ti_sci part to consider the second range as well.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Use the ti_sci_resource_desc directly and update it's start and num members
directly instead of requiring individual parameters for them.
This will allow easy extension of the RM parameters without changing API.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
The system controller's resource manager have support for configuring the
TDTYPE of TCHAN_CFG register on j721e.
With this parameter the teardown completion can be controlled:
TDTYPE == 0: Return without waiting for peer to complete the teardown
TDTYPE == 1: Wait for peer to complete the teardown
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Tero Kristo <t-kristo@ti.com>
Tested-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Add the sysfs attribute bits in ABI/stable for shared wq support.
Signed-off-by: Jing Lin <jing.lin@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Link: https://lore.kernel.org/r/160382008649.3911367.10851752182908509837.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add code to "complete" a descriptor when the descriptor or its completion
address hit a fault error when SVA mode is being used. This error can be
triggered due to bad programming by the user. A lock is introduced in order
to protect the descriptor completion lists since the fault handler will run
from the system work queue after being scheduled in the interrupt handler.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Link: https://lore.kernel.org/r/160382008092.3911367.12766483427643278985.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add shared workqueue support that includes the support of Shared Virtual
memory (SVM) or in similar terms On Demand Paging (ODP). The shared
workqueue uses the enqcmds command in kernel and will respond with retry if
the workqueue is full. Shared workqueue only works when there is PASID
support from the IOMMU.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Link: https://lore.kernel.org/r/160382007499.3911367.26043087963708134.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The xor_hw_desc local variable is assigned but never read:
drivers/dma/ppc4xx/adma.c: In function ‘ppc440spe_desc_set_src_mult’:
drivers/dma/ppc4xx/adma.c:562:17: warning: variable ‘xor_hw_desc’ set but not used [-Wunused-but-set-variable]
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201019155756.21445-2-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The ppc440spe_adma_chan_list file-scope variable is not used outside of
the unit so it can be made static.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201019155756.21445-1-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The sama7g5 version of the XDMAC supports priority configuration and
outstanding capabilities.
Add defines for the specific registers for this configuration, together
with recommended settings.
However the settings are very different if the XDMAC is a mem2mem or a
per2mem controller.
Thus, we need to differentiate according to device tree property.
Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Link: https://lore.kernel.org/r/20201016093918.290137-1-eugen.hristev@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
SAMA7G5 SoC uses a slightly different variant of the AT_XDMAC.
Added support by a new compatible and a layout struct that copes
to the specific version considering the compatible string.
Only the differences in register map are present in the layout struct.
I reworked the register access for this part that has the differences.
Also the Source/Destination Interface bits are no longer valid for this
variant of the XDMAC. Thus, the layout also has a bool for specifying
whether these bits are required or not.
Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Link: https://lore.kernel.org/r/20201016093850.290053-1-eugen.hristev@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The PERID in the CC register for mem2mem operations must match an unused
PERID.
The PERID field is 7 bits, but the selected value is 0x3f.
On later products we can have more reserved PERIDs for actual peripherals,
thus this needs to be increased to maximum size.
Changing the value to 0x7f, which is the maximum for 7 bits field.
Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Reviewed-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Link: https://lore.kernel.org/r/20201016093725.289880-1-eugen.hristev@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Make use of the new struct_size() helper instead of the offsetof() idiom.
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20201008141828.GA20325@embeddedor
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Use a more generic form for __section that requires quotes to avoid
complications with clang and gcc differences.
Remove the quote operator # from compiler_attributes.h __section macro.
Convert all unquoted __section(foo) uses to quoted __section("foo").
Also convert __attribute__((section("foo"))) uses to __section("foo")
even if the __attribute__ has multiple list entry forms.
Conversion done using the script at:
https://lore.kernel.org/lkml/75393e5ddc272dc7403de74d645e6c6e0f4e70eb.camel@perches.com/2-convert_section.pl
Signed-off-by: Joe Perches <joe@perches.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@gooogle.com>
Reviewed-by: Miguel Ojeda <ojeda@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
tid_addr is not a "pointer to (pointer to int in userspace)"; it is in
fact a "pointer to (pointer to int in userspace) in userspace". So
sparse rightfully complains about passing a kernel pointer to
put_user().
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 453431a549 ("mm, treewide: rename kzfree() to
kfree_sensitive()") renamed kzfree() to kfree_sensitive(),
but it left a compatibility definition of kzfree() to avoid
being too disruptive.
Since then a few more instances of kzfree() have slipped in.
Just get rid of them and remove the compatibility definition
once and for all.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
