| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Remove intermediate scatter-gather table completely and
enable new DMA link API.
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Acked-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/f71638d50c9c79a462f2e0423501b1de77617656.1747747694.git.leon@kernel.org
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
Change the creation of mkey to be performed in multiple steps:
data allocation, DMA setup and actual call to HW to create that mkey.
In this new flow, the whole input to MKEY command is saved to eliminate
the need to keep array of pointers for DMA addresses for receive list
and in the future patches for send list too.
In addition to memory size reduce and elimination of unnecessary data
movements to set MKEY input, the code is prepared for future reuse.
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Acked-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/d4ad0384fbd1e23a607cbbe9e5756748f3a761d9.1747747694.git.leon@kernel.org
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
allocated_length is a multiple of page size and number of pages,
so let's change the functions to accept number of pages. This improves
code readability, simplifies buffer handling, and enables combining DMA
send/receive operations, as will be introduced in the next patches.
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Acked-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/76f39993d2ca0311b3bcfe56038a669d03926815.1747747694.git.leon@kernel.org
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
Merge two step DMA mapping API as basis for mlx5-vfio-pci uses.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
In situations where mapping/unmapping sequence can be controlled by
userspace, attempting to map over a region that has not yet been
unmapped is an error. But not something that should spam dmesg.
Now that there is a quirk, we can also drop the selftest_running
flag, and use the quirk instead for selftests.
Acked-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Link: https://lore.kernel.org/r/20250519175348.11924-6-robdclark@gmail.com
[will: Rename quirk to IO_PGTABLE_QUIRK_NO_WARN per Robin's suggestion]
Signed-off-by: Will Deacon <will@kernel.org>
|
|
The NIX_PARSE_S structure populated by hardware in the
NIX RX CQE has parsing information for the received packet.
A tracepoint to dump the all words of NIX_PARSE_S
is helpful in debugging packet parser.
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com>
Link: https://patch.msgid.link/1747331048-15347-1-git-send-email-sbhatta@marvell.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
After having factored out the provider part from mdio_bus.c, we can
make the mdio consumer / device layer a separate module. This also
allows to remove Kconfig symbol MDIO_DEVICE.
The module init / exit functions from mdio_bus.c no longer have to be
called from phy_device.c. The link order defined in
drivers/net/phy/Makefile ensures that init / exit functions are called
in the right order.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://patch.msgid.link/dba6b156-5748-44ce-b5e2-e8dc2fcee5a7@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
A few, quite rare, WMI attributes have names that are not compatible with
filenames, e.g. "Intel VT for Directed I/O (VT-d)".
For these cases the '/' gets replaced with '\' for display, but doesn't
get switched again when doing the WMI access.
Fix this by keeping the original attribute name and using that for sending
commands to the BIOS
Fixes: a40cd7ef22fb ("platform/x86: think-lmi: Add WMI interface support on Lenovo platforms")
Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Link: https://lore.kernel.org/r/20250520005027.3840705-1-mpearson-lenovo@squebb.ca
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
|
|
If user modifies the battery charge threshold an ACPI event is generated.
Confirmed with Lenovo FW team this is only generated on user event. As no
action is needed, ignore the event and prevent spurious kernel logs.
Reported-by: Derek Barbosa <debarbos@redhat.com>
Closes: https://lore.kernel.org/platform-driver-x86/7e9a1c47-5d9c-4978-af20-3949d53fb5dc@app.fastmail.com/T/#m5f5b9ae31d3fbf30d7d9a9d76c15fb3502dfd903
Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Armin Wolf <W_Armin@gmx.de>
Link: https://lore.kernel.org/r/20250517023348.2962591-1-mpearson-lenovo@squebb.ca
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
|
|
Merge series from Geert Uytterhoeven <geert+renesas@glider.be>:
This patch series (A) improves single transfer sizes in the MSIOF
driver, using two methods:
- By increasing the assumed FIFO sizes, impacting both PIO and DMA
transfers,
- By using two groups, impacting DMA transfers,
and (B) lets the recently-introduced MSIOF I2S drive reuse the SPI
driver's register definitions. All of this is covered with a thick
sauce of fixes for (harmless) bugs, cleanups, and refactorings.
Note that the driver uses the limitations as specified in the hardware
documentation. For discovering the actual FIFO sizes, I wrote some
crude test code that can be found at [2].
This is based on spi/for-next and sound-asoc/for-next, and has been
tested on a variery of R-Car SoCs.
[1] https://lore.kernel.org/cover.1746180072.git.geert+renesas@glider.be
[2] https://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers.git/log/?h=topic/msiof-fifo
|
|
Merge series from Mohammad Rafi Shaik <mohammad.rafi.shaik@oss.qualcomm.com>:
This patchset adds support for sound card on Qualcomm QCS9100 and
QCS9075 boards.
|
|
If either REGMAP_IRQ or REGMAP_MDIO are set then REGMAP is also set.
This then enables the selecting of IRQ_DOMAIN or MDIO_BUS from REGMAP
based on the above two symbols respectively. This makes it very easy
to end up with "circular dependencies".
Instead select the IRQ_DOMAIN or MDIO_BUS from the symbols that make
use of them. This is almost equivalent to before but makes it less
likely to end up with false circular dependency detections.
Signed-off-by: Andrew Davis <afd@ti.com>
Reported-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Closes: https://lore.kernel.org/r/bfe991fa-f54c-4d58-b2e0-34c4e4eb48f4@linaro.org/
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patch.msgid.link/20250516141722.13772-1-afd@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Add an useful info when failed to get irq/wakeirq due to -EPROBE_DEFER.
Before:
[ 15.737361] i2c 2-0050: deferred probe pending: (reason unknown)
After:
[ 15.816295] i2c 2-0050: deferred probe pending: tcpci: can't get irq
Signed-off-by: Xu Yang <xu.yang_2@nxp.com>
Reviewed-by: Carlos Song <carlos.song@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
|
|
Currently, it is possible that LPM is enabled while calling the set_lpm()
callback.
The current code performs a SET FEATURES command to disable DIPM if
policy < ATA_LPM_MED_POWER_WITH_DIPM, this means that it will currently
disable DIPM for policies:
ATA_LPM_UNKNOWN, ATA_LPM_MAX_POWER, ATA_LPM_MED_POWER
(but not for policy ATA_LPM_MED_POWER_WITH_DIPM).
The code called after calling the set_lpm() callback will later perform
a SET FEATURES command to enable DIPM, if
policy >= ATA_LPM_MED_POWER_WITH_DIPM.
As we can see DIPM will not be disabled before calling set_lpm() if the LPM
policy is: ATA_LPM_MED_POWER_WITH_DIPM, ATA_LPM_MIN_POWER_WITH_PARTIAL,
or ATA_LPM_MIN_POWER.
Make sure that we always disable DIPM before calling the set_lpm()
callback. This is because the set_lpm() callback is the function (for AHCI)
that sets the proper bits in PxSCTL.IPM, reflecting the support of the HBA.
PxSCTL.IPM controls the LPM states that the device is allowed to enter.
If the device tries to enter a state disabled by PxSCTL.IPM, the host will
NAK the transition.
If we do not disable DIPM before modifying PxSCTL.IPM, it is possible that
DIPM will try (and will be allowed to) enter a LPM state that the HBA does
not support (since we have not yet written PxSCTL.IPM, the HBA wasn't able
to NAK the transition).
While at it, remove the guard of host support for DIPM around the disabling
of DIPM. While it makes sense to take host support for DIPM into account
when enabling DIPM, it makes zero sense to take host support into account
when disabling DIPM.
If the host does not support DIPM, that is an even bigger reason why DIPM
should be disabled on the device side.
Signed-off-by: Niklas Cassel <cassel@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
|
|
Rename the no_dipm variable to host_has_dipm, by inverting the
expression, and and also having a clearer name.
No functional change.
Signed-off-by: Niklas Cassel <cassel@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
|
|
Rename the hipm and dipm variables to have a clearer name.
Also fold in the usage of no_dipm, as that is required in order to give
the dipm variable a more descriptive name.
No functional change.
Signed-off-by: Niklas Cassel <cassel@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
|
|
link->lpm_policy is initialized to ATA_LPM_UNKNOWN in ata_eh_reset().
ata_eh_set_lpm() is then only called if
link->lpm_policy != ap->target_lpm_policy (after reset)
and then only if link->lpm_policy > ATA_LPM_MAX_POWER (before
revalidation).
This means that ata_eh_set_lpm() is currently never called with
policy == ATA_LPM_UNKNOWN.
Add a WARN_ON_ONCE so that it is more obvious from reading the code that
this function is never called with policy == ATA_LPM_UNKNOWN.
Signed-off-by: Niklas Cassel <cassel@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
|
|
The comments describing which LPM policies that has DIPM enabled predates
the introduction of the LPM policies ATA_LPM_MIN_POWER_WITH_PARTIAL and
ATA_LPM_MED_POWER_WITH_DIPM. Update the DIPM comments to reflect reality.
Also remove the sentence that claims that "Order device and link
configurations such that the host always allows DIPM requests."
This comment is written before 24e0e61db3cb ("ata: libata: disallow
dev-initiated LPM transitions to unsupported states").
Even though the set_lpm() call is done before enabling DIPM, the host will
not always allow DIPM requests. For all LPM polcies where DIPM is enabled,
only DIPM requests to LPM states that are supported by the HBA will be
allowed.
Signed-off-by: Niklas Cassel <cassel@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
|
|
pci/iomap.c still contains warnings about those functions not behaving
in a managed manner if pcim_enable_device() was called. Since all hybrid
behavior that users could know about has been removed by now, those
explicit warnings are no longer necessary.
Remove the hybrid-devres usage warnings from the kernel-doc.
Signed-off-by: Philipp Stanner <phasta@kernel.org>
[kwilczynski: commit log]
Signed-off-by: Krzysztof Wilczyński <kwilczynski@kernel.org>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Link: https://lore.kernel.org/r/20250519112959.25487-8-phasta@kernel.org
|
|
When the demangling of the hybrid devres functions within PCI was
implemented, it was necessary to implement several PCI functions a
second time to avoid cyclic calls, since the hybrid functions in pci.c
call the managed functions in devres.c, which in turn can be directly
used outside of PCI and needed request infrastructure, too.
Therefore, __pcim_request_region_range(), __pci_release_region_range()
and wrappers around them were implemented.
The hybrid nature has recently been removed from all functions in pci.c.
Therefore, the functions in devres.c can now directly use their
counterparts in pci.c without causing a call-cycle.
Remove __pcim_request_region_range(), __pcim_request_region_range() and
the wrappers. Use the corresponding request functions from pci.c in
devres.c
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Signed-off-by: Krzysztof Wilczyński <kwilczynski@kernel.org>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Link: https://lore.kernel.org/r/20250519112959.25487-7-phasta@kernel.org
|
|
pcim_request_region_exclusive(), the only user in PCI devres that needed
exclusive region requests, has been removed.
All features related to exclusive requests can, therefore, be removed,
too. Remove them.
Signed-off-by: Philipp Stanner <phasta@kernel.org>
[kwilczynski: commit log]
Signed-off-by: Krzysztof Wilczyński <kwilczynski@kernel.org>
Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20250519112959.25487-6-phasta@kernel.org
|
|
We are already within another `#ifdef CONFIG_GPIOLIB_IRQCHIP` in
gpiochip_to_irq() so there's no need for another guard. Remove it.
Acked-by: Peng Fan <peng.fan@nxp.com>
Link: https://lore.kernel.org/r/20250519-gpio-irq-kconfig-fixes-v1-3-fe6ba1c6116d@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
|
|
This driver uses gpiochip_irq_reqres() and gpiochip_irq_relres() which
are only built with GPIOLIB_IRQCHIP=y. Add the missing Kconfig select.
Fixes: 7688a54d5b53 ("gpio: mpc8xxx: Make irq_chip immutable")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202505180309.1nosQMkI-lkp@intel.com/
Acked-by: Peng Fan <peng.fan@nxp.com>
Link: https://lore.kernel.org/r/20250519-gpio-irq-kconfig-fixes-v1-2-fe6ba1c6116d@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
|
|
This driver uses gpiochip_irq_reqres() and gpiochip_irq_relres() which
are only built with GPIOLIB_IRQCHIP=y. Add the missing Kconfig select.
Fixes: 20117cf426b6 ("gpio: pxa: Make irq_chip immutable")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202505181429.mzyIatOU-lkp@intel.com/
Acked-by: Peng Fan <peng.fan@nxp.com>
Link: https://lore.kernel.org/r/20250519-gpio-irq-kconfig-fixes-v1-1-fe6ba1c6116d@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
|
|
Currently, all SCMI devices with performance domains attempt to register
a cpufreq driver, even if their performance domains aren't used to
control the CPUs. The cpufreq framework only supports registering a
single driver, so only the first device will succeed. And if that device
isn't used for the CPUs, then cpufreq will scale the wrong domains.
To avoid this, return early from scmi_cpufreq_probe() if the probing
SCMI device isn't referenced by the CPU device phandles.
This keeps the existing assumption that all CPUs are controlled by a
single SCMI device.
Signed-off-by: Mike Tipton <quic_mdtipton@quicinc.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Cristian Marussi <cristian.marussi@arm.com>
Reviewed-by: Sudeep Holla <sudeep.holla@arm.com>
Tested-by: Cristian Marussi <cristian.marussi@arm.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
|
|
|
|
Introduce a Rust-based implementation of the cpufreq-dt driver, covering
most of the functionality provided by the existing C version. Some
features, such as retrieving platform data from `cpufreq-dt-platdev.c`,
are still pending.
The driver has been tested with QEMU, and frequency scaling works as
expected.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
|
|
According to the ISL28022 datasheet, bit15 of the current register is
representing -32768. Fix the calculation to properly handle this bit,
ensuring correct measurements for negative values.
Signed-off-by: Yikai Tsai <yikai.tsai.wiwynn@gmail.com>
Link: https://lore.kernel.org/r/20250519084055.3787-2-yikai.tsai.wiwynn@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
In the NVMe context, the term "shutdown" has a specific technical
meaning. To avoid confusion, this commit renames the nvme_mpath_
shutdown_disk function to nvme_mpath_remove_disk to better reflect
its purpose (i.e. removing the disk from the system). However,
nvme_mpath_remove_disk was already in use, and its functionality
is related to releasing or putting the head node disk. To resolve
this naming conflict and improve clarity, the existing nvme_mpath_
remove_disk function is also renamed to nvme_mpath_put_disk.
This renaming improves code readability and better aligns function
names with their actual roles.
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Currently, a multipath head disk node is not created for single-
ported NVMe adapters or private namespaces with non-unique NSID.
However, creating a head node in these cases can help transparently
handle transient PCIe link failures. Without a head node, features
like delayed removal cannot be leveraged, making it difficult to
tolerate such link failures. To address this, this commit introduces
nvme_core module parameter multipath_always_on.
When multipath_always_on is set to true, it forces the creation of a
multipath head node regardless NVMe disk or namespace type. So this
option allows the use of delayed removal of head node functionality
even for single-ported NVMe disks and private namespaces with a unique
NSID and thus helps transparently handle transient PCIe link failures.
By default multipath_always_on is set to false, thus preserving the
existing behavior. Setting it to true enables improved fault tolerance
in PCIe setups. Moreover, please note that enabling this option would
also implicitly enable nvme_core.multipath.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Currently, the multipath head node of an NVMe disk is removed
immediately as soon as all paths of the disk are removed. However,
this can cause issues in scenarios where:
- The disk hot-removal followed by re-addition.
- Transient PCIe link failures that trigger re-enumeration,
temporarily removing and then restoring the disk.
In these cases, removing the head node prematurely may lead to a head
disk node name change upon re-addition, requiring applications to
reopen their handles if they were performing I/O during the failure.
To address this, introduce a delayed removal mechanism of head disk
node. During transient failure, instead of immediate removal of head
disk node, the system waits for a configurable timeout, allowing the
disk to recover.
During transient disk failure, if application sends any IO then we
queue it instead of failing such IO immediately. If the disk comes back
online within the timeout, the queued IOs are resubmitted to the disk
ensuring seamless operation. In case disk couldn't recover from the
failure then queued IOs are failed to its completion and application
receives the error.
So this way, if disk comes back online within the configured period,
the head node remains unchanged, ensuring uninterrupted workloads
without requiring applications to reopen device handles.
A new sysfs attribute, named "delayed_removal_secs" is added under head
disk blkdev for user who wish to configure time for the delayed removal
of head disk node. The default value of this attribute is set to zero
second ensuring no behavior change unless explicitly configured.
Link: https://lore.kernel.org/linux-nvme/Y9oGTKCFlOscbPc2@infradead.org/
Link: https://lore.kernel.org/linux-nvme/Y+1aKcQgbskA2tra@kbusch-mbp.dhcp.thefacebook.com/
Suggested-by: Keith Busch <kbusch@kernel.org>
Suggested-by: Christoph Hellwig <hch@infradead.org>
[nilay: reworked based on the original idea/POC from Christoph and Keith]
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Redefine the max segments and max integrity limits based on the limiting
factors. This keeps exactly the same values for 4k PAGE_SIZE systems,
but increases the number of segments for larger page size as it properly
derives the scatterlist allocation based limit for them instead of
assuming a 4k PAGE_SIZE.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
|
|
This avoids open coding the variable size array arithmetics.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Leon Romanovsky <leon@kernel.org>
|
|
Open coding magic numbers in multiple places is never a good idea.
Signed-off-by: Leon Romanovsky <leon@kernel.org>
[hch: split from a larger patch]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com>
|
|
Add a separate flag to encode that the transfer is using the small
page sized pool, and use a normal 0..n count for the number of
descriptors.
Contains improvements and suggestions from Kanchan Joshi
<joshi.k@samsung.com> and Leon Romanovsky <leon@kernel.org>.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Reviewed-by: Leon Romanovsky <leon@kernel.org>
|
|
They are used for both PRPs and SGLs, and we use descriptor elsewhere
when referring to their allocations, so use that name here as well.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Leon Romanovsky <leon@kernel.org>
|
|
There is no real point in having a union of two pointer types here, just
use a void pointer as we mix and match types between the arms of the
union between the allocation and freeing side already.
Also rename the nr_allocations field to nr_descriptors to better describe
what it does.
Signed-off-by: Christoph Hellwig <hch@lst.de>
[leon: ported forward to include metadata SGL support]
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
|
|
Instead of keeping dedicated "bool aborted" variable, switch to a flags
flags that can be used for other flags as well.
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com>
|
|
No admin command defined in an NVMe specification supports metadata,
but to protect against vendor specific commands using metadata ensure
that we don't try to use SGLs for metadata on the admin queue, as NVMe
does not support SGLs on the admin queue for the PCI transport. Do
this by checking if the data transfer has been setup using SGLs as
that is required for using SGLs for metadata.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Leon Romanovsky <leon@kernel.org>
|
|
NVMe commands with over 8 KB of discontiguous data allocate PRP list
pages from the per-nvme_device dma_pool prp_page_pool or prp_small_pool.
Each call to dma_pool_alloc() and dma_pool_free() takes the per-dma_pool
spinlock. These device-global spinlocks are a significant source of
contention when many CPUs are submitting to the same NVMe devices. On a
workload issuing 32 KB reads from 16 CPUs (8 hypertwin pairs) across 2
NUMA nodes to 23 NVMe devices, we observed 2.4% of CPU time spent in
_raw_spin_lock_irqsave called from dma_pool_alloc and dma_pool_free.
Ideally, the dma_pools would be per-hctx to minimize contention. But
that could impose considerable resource costs in a system with many NVMe
devices and CPUs.
As a compromise, allocate per-NUMA-node PRP list DMA pools. Map each
nvme_queue to the set of DMA pools corresponding to its device and its
hctx's NUMA node. This reduces the _raw_spin_lock_irqsave overhead by
about half, to 1.2%. Preventing the sharing of PRP list pages across
NUMA nodes also makes them cheaper to initialize.
Link: https://lore.kernel.org/linux-nvme/CADUfDZqa=OOTtTTznXRDmBQo1WrFcDw1hBA7XwM7hzJ-hpckcA@mail.gmail.com/T/#u
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
nvme_init_hctx() and nvme_admin_init_hctx() are very similar. In
preparation for adding more logic, factor out a nvme_init_hctx-common()
helper.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
The lsrsp object is maintained by the LLDD. The lifetime of the lsrsp
object is implicit. Because there is no explicit cleanup/free call into
the LLDD, it is not safe to assume after xml_rsp_fails, that the lsrsp
is still valid. The LLDD could have freed the object already.
With the recent changes how fcloop tracks the resources, this is the
case. Thus don't access lsrsp after xml_rsp_fails.
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
The lifetime of the fcloop_lsreq is not tight to the lifetime of the
host or target port, thus there is no need anymore to synchronize the
cleanup path anymore.
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Add the missing fcloop_call_host_done calls so that the caller
frees resources when something goes wrong.
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Ensure that the tgtport is not going away as long portentry has a
pointer on it.
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
When nvmet_fc_unregister_targetport is called by the LLDD, it's not
possible to communicate with the host, thus all pending request will not
be process. Thus explicitly free them.
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
When the target port is gone, the lsrsp pointer is invalid. Thus don't
call the done function anymore instead just drop the response.
This happens when the target sends a disconnect association. After this
the target starts tearing down all resources and doesn't expect any
response.
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
fcloop depends on the host or the target to allocate the fcloop_lsreq
object. This means that the lifetime of the fcloop_lsreq is tied to
either the host or the target. Consequently, the host or the target must
cooperate during shutdown.
Unfortunately, this approach does not work well when the target forces a
shutdown, as there are dependencies that are difficult to resolve in a
clean way.
The simplest solution is to decouple the lifetime of the fcloop_lsreq
object by managing them directly within fcloop. Since this is not a
performance-critical path and only a small number of LS objects are used
during setup and cleanup, it does not significantly impact performance
to allocate them during normal operation.
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
The delete callback can be called either via the unregister function or
from the transport directly. Thus it is necessary ensure resources are
not freed multiple times.
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
The abort handling logic expects that the state and the fcpreq are only
accessed when holding the reqlock lock.
While at it, only handle the aborts in the abort handler.
Signed-off-by: Daniel Wagner <wagi@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
Leon Romanovsky
Alex Williamson
Rob Clark
Subbaraya Sundeep
Heiner Kallweit
Mark Pearson
Mark Brown
Andrew Davis
Xu Yang
Niklas Cassel
Philipp Stanner
Bartosz Golaszewski
Mike Tipton
Viresh Kumar
Yikai Tsai
Nilay Shroff
Christoph Hellwig
Leon Romanovsky
Caleb Sander Mateos
Daniel Wagner