| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Now that the IAVF driver simply uses dev_alloc_page() + free_page() with
no custom recycling logics, it can easily be switched to using Page
Pool / libeth API instead.
This allows to removing the whole dancing around headroom, HW buffer
size, and page order. All DMA-for-device is now done in the PP core,
for-CPU -- in the libeth helper.
Use skb_mark_for_recycle() to bring back the recycling and restore the
performance. Speaking of performance: on par with the baseline and
faster with the PP optimization series applied. But the memory usage for
1500b MTU is now almost 2x lower (x86_64) thanks to allocating a page
every second descriptor.
Signed-off-by: Alexander Lobakin <[email protected]>
Signed-off-by: Tony Nguyen <[email protected]>
|
|
Add a couple intuitive helpers to hide Rx buffer implementation details
in the library and not multiplicate it between drivers. The settings are
sorta optimized for 100G+ NICs, but nothing really HW-specific here.
Use the new page_pool_dev_alloc() to dynamically switch between
split-page and full-page modes depending on MTU, page size, required
headroom etc. For example, on x86_64 with the default driver settings
each page is shared between 2 buffers. Turning on XDP (not in this
series) -> increasing headroom requirement pushes truesize out of 2048
boundary, leading to that each buffer starts getting a full page.
The "ceiling" limit is %PAGE_SIZE, as only order-0 pages are used to
avoid compound overhead. For the above architecture, this means maximum
linear frame size of 3712 w/o XDP.
Not that &libeth_buf_queue is not a complete queue/ring structure for
now, rather a shim, but eventually the libeth-enabled drivers will move
to it, with iavf being the first one.
Signed-off-by: Alexander Lobakin <[email protected]>
Signed-off-by: Tony Nguyen <[email protected]>
|
|
Each driver is responsible for syncing buffers written by HW for CPU
before accessing them. Almost each PP-enabled driver uses the same
pattern, which could be shorthanded into a static inline to make driver
code a little bit more compact.
Introduce a simple helper which performs DMA synchronization for the
size passed from the driver. It can be used even when the pool doesn't
manage DMA-syncs-for-device, just make sure the page has a correct DMA
address set via page_pool_set_dma_addr().
Signed-off-by: Alexander Lobakin <[email protected]>
Reviewed-by: Ilias Apalodimas <[email protected]>
Signed-off-by: Tony Nguyen <[email protected]>
|
|
There are several functions taking pointers to data they don't modify.
This includes statistics fetching, page and page_pool parameters, etc.
Constify the pointers, so that call sites will be able to pass const
pointers as well.
No functional changes, no visible changes in functions sizes.
Reviewed-by: Ilias Apalodimas <[email protected]>
Signed-off-by: Alexander Lobakin <[email protected]>
Signed-off-by: Tony Nguyen <[email protected]>
|
|
Add NUMA-aware counterparts for kvmalloc_array() and kvcalloc() to be
able to flexibly allocate arrays for a particular node.
Rewrite kvmalloc_array() to kvmalloc_array_node(NUMA_NO_NODE) call.
Acked-by: Vlastimil Babka <[email protected]>
Reviewed-by: Przemek Kitszel <[email protected]>
Signed-off-by: Alexander Lobakin <[email protected]>
Signed-off-by: Tony Nguyen <[email protected]>
|
|
Not a secret there's a ton of code duplication between two and more Intel
ethernet modules.
Before introducing new changes, which would need to be copied over again,
start decoupling the already existing duplicate functionality into a new
module, which will be shared between several Intel Ethernet drivers.
Add the lookup table which converts 8/10-bit hardware packet type into
a parsed bitfield structure for easy checking packet format parameters,
such as payload level, IP version, etc. This is currently used by i40e,
ice and iavf and it's all the same in all three drivers.
The only difference introduced in this implementation is that instead of
defining a 256 (or 1024 in case of ice) element array, add unlikely()
condition to limit the input to 154 (current maximum non-reserved packet
type). There's no reason to waste 600 (or even 3600) bytes only to not
hurt very unlikely exception packets.
The hash computation function now takes payload level directly as a
pkt_hash_type. There's a couple cases when non-IP ptypes are marked as
L3 payload and in the previous versions their hash level would be 2, not
3. But skb_set_hash() only sees difference between L4 and non-L4, thus
this won't change anything at all.
The module is behind the hidden Kconfig symbol, which the drivers will
select when needed. The exports are behind 'LIBIE' namespace to limit
the scope of the functions.
Not that non-HW-specific symbols will live in yet another module,
libeth. This is done to easily distinguish pretty generic code ready
for reusing by any other vendor and/or for moving the layer up from
the code useful in Intel's 1-100G drivers only.
Signed-off-by: Alexander Lobakin <[email protected]>
Signed-off-by: Tony Nguyen <[email protected]>
|
|
Introduce two new BPF kfuncs, bpf_preempt_disable and
bpf_preempt_enable. These kfuncs allow disabling preemption in BPF
programs. Nesting is allowed, since the intended use cases includes
building native BPF spin locks without kernel helper involvement. Apart
from that, this can be used to per-CPU data structures for cases where
programs (or userspace) may preempt one or the other. Currently, while
per-CPU access is stable, whether it will be consistent is not
guaranteed, as only migration is disabled for BPF programs.
Global functions are disallowed from being called, but support for them
will be added as a follow up not just preempt kfuncs, but rcu_read_lock
kfuncs as well. Static subprog calls are permitted. Sleepable helpers
and kfuncs are disallowed in non-preemptible regions.
Signed-off-by: Kumar Kartikeya Dwivedi <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alexei Starovoitov <[email protected]>
|
|
Another ambiguous use of strncpy() is to copy from strings that may not
be NUL-terminated. These cases depend on having the destination buffer
be explicitly larger than the source buffer's maximum size, having
the size of the copy exactly match the source buffer's maximum size,
and for the destination buffer to get explicitly NUL terminated.
This usually happens when parsing protocols or hardware character arrays
that are not guaranteed to be NUL-terminated. The code pattern is
effectively this:
char dest[sizeof(src) + 1];
strncpy(dest, src, sizeof(src));
dest[sizeof(dest) - 1] = '\0';
In practice it usually looks like:
struct from_hardware {
...
char name[HW_NAME_SIZE] __nonstring;
...
};
struct from_hardware *p = ...;
char name[HW_NAME_SIZE + 1];
strncpy(name, p->name, HW_NAME_SIZE);
name[NW_NAME_SIZE] = '\0';
This cannot be replaced with:
strscpy(name, p->name, sizeof(name));
because p->name is smaller and not NUL-terminated, so FORTIFY will
trigger when strnlen(p->name, sizeof(name)) is used. And it cannot be
replaced with:
strscpy(name, p->name, sizeof(p->name));
because then "name" may contain a 1 character early truncation of
p->name.
Provide an unambiguous interface for converting a maybe not-NUL-terminated
string to a NUL-terminated string, with compile-time buffer size checking
so that it can never fail at runtime: memtostr() and memtostr_pad(). Also
add KUnit tests for both.
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Kees Cook <[email protected]>
|
|
./include/linux/coresight.h: linux/amba/bus.h is included more than once.
Reported-by: Abaci Robot <[email protected]>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=8869
Signed-off-by: Jiapeng Chong <[email protected]>
Reviewed-by: James Clark <[email protected]>
Signed-off-by: Suzuki K Poulose <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
It is impossible to use init_dummy_netdev together with alloc_netdev()
as the 'setup' argument.
This is because alloc_netdev() initializes some fields in the net_device
structure, and later init_dummy_netdev() memzero them all. This causes
some problems as reported here:
https://lore.kernel.org/all/[email protected]/
Split the init_dummy_netdev() function in two. Create a new function called
init_dummy_netdev_core() that does not memzero the net_device structure.
Then have init_dummy_netdev() memzero-ing and calling
init_dummy_netdev_core(), keeping the old behaviour.
init_dummy_netdev_core() is the new function that could be called as an
argument for alloc_netdev().
Also, create a helper to allocate and initialize dummy net devices,
leveraging init_dummy_netdev_core() as the setup argument. This function
basically simplify the allocation of dummy devices, by allocating and
initializing it. Freeing the device continue to be done through
free_netdev()
Suggested-by: Jakub Kicinski <[email protected]>
Signed-off-by: Breno Leitao <[email protected]>
Reviewed-by: Ido Schimmel <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
arch_update_hw_pressure()
Now that cpufreq provides a pressure value to the scheduler, rename
arch_update_thermal_pressure into HW pressure to reflect that it returns
a pressure applied by HW (i.e. with a high frequency change) and not
always related to thermal mitigation but also generated by max current
limitation as an example. Such high frequency signal needs filtering to be
smoothed and provide an value that reflects the average available capacity
into the scheduler time scale.
Signed-off-by: Vincent Guittot <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Tested-by: Lukasz Luba <[email protected]>
Reviewed-by: Qais Yousef <[email protected]>
Reviewed-by: Lukasz Luba <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
Provide to the scheduler a feedback about the temporary max available
capacity. Unlike arch_update_thermal_pressure(), this doesn't need to be
filtered as the pressure will happen for dozens of ms or more.
Signed-off-by: Vincent Guittot <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Tested-by: Lukasz Luba <[email protected]>
Reviewed-by: Qais Yousef <[email protected]>
Reviewed-by: Lukasz Luba <[email protected]>
Reviewed-by: Dhruva Gole <[email protected]>
Acked-by: Rafael J. Wysocki <[email protected]>
Acked-by: Viresh Kumar <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
Move the SYSCTL_TABLE_TYPE_{DEFAULT,PERMANENTLY_EMPTY} enums from
ctl_table to ctl_table_header.
Removing the mutable member is necessary to constify static instances
of struct ctl_table.
Move the initialization of the sysctl_mount_point type into
init_header() where all the other header fields are also initialized.
As a side-effect the memory usage of the sysctl core is reduced.
Each ctl_table_header instance can manage multiple ctl_table instances
and is only allocated when the table is actually registered.
This saves 8 bytes of memory per ctl_table on 64bit, 4 due to the enum
field itself and 4 due to padding.
Signed-off-by: Thomas Weißschuh <[email protected]>
Signed-off-by: Joel Granados <[email protected]>
|
|
The permissions callback should not modify the ctl_table. Enforce this
expectation via the typesystem. This is a step to put "struct ctl_table"
into .rodata throughout the kernel.
The patch was created with the following coccinelle script:
@@
identifier func, head, ctl;
@@
int func(
struct ctl_table_header *head,
- struct ctl_table *ctl)
+ const struct ctl_table *ctl)
{ ... }
(insert_entry() from fs/proc/proc_sysctl.c is a false-positive)
No additional occurrences of '.permissions =' were found after a
tree-wide search for places missed by the conccinelle script.
Reviewed-by: Joel Granados <[email protected]>
Signed-off-by: Thomas Weißschuh <[email protected]>
Signed-off-by: Joel Granados <[email protected]>
|
|
Remove the 'table' argument from set_ownership as it is never used. This
change is a step towards putting "struct ctl_table" into .rodata and
eventually having sysctl core only use "const struct ctl_table".
The patch was created with the following coccinelle script:
@@
identifier func, head, table, uid, gid;
@@
void func(
struct ctl_table_header *head,
- struct ctl_table *table,
kuid_t *uid, kgid_t *gid)
{ ... }
No additional occurrences of 'set_ownership' were found after doing a
tree-wide search.
Reviewed-by: Joel Granados <[email protected]>
Signed-off-by: Thomas Weißschuh <[email protected]>
Signed-off-by: Joel Granados <[email protected]>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into char-misc-next
Jonathan writes:
IIO: 1st set of new device support, features and cleanup for 6.10
The Analog Device team (Paul Cercueil and Nuno Sa) have been working on
improving high speed device handling. They have had some support in their
own tree for many years, so it is great to see them bring it to upstream.
Some of that is seen here, with the first output device using the
IIO dmaengine infrastructure and a new DAC backend FPGA IP driver.
This makes use of a new set of interfaces to allow backend and
front end driver communication in a fashion that in theory at least
allows for a single driver for a given ADC / DAC independent of
the IP to which is being used to deal with the data bus and DMA aspects
of working with these devices. It is early days for this new
generic way of handling split devices, but as it's kernel internals only
we can merrily change anything about it as a wider diversity of devices
show up and we get a better feel for what works.
Alongside the usual set of new drivers and features we have
the automatic cleanup of fwnode_handle_put() which didn't quite make
it in last cycle. The equivalent DT version was merged by Rob Herring
via the DT binding tree and one patch using that in IIO can also be
found in this pull request. Rob has been making extensive use of that
infrastructure in the DT core which is good to see and provides more
evidence this basic approach is useful.
In some cases, the IIO driver was converted over from DT only to
using the generic firmware description handling of property.h
including using the new macros. The general preference for IIO
is to use this more generic handling where possible - a bunch of other
drivers have been converted this cycle as well.
New device support
==================
adi,ad7173
- New driver supporting AD7172-2, AD7172-4 AD7173-9, AD7175-2, AD7175-8,
AD7176-2 and AD7177-2 ADCs.
- Follow up fix for an accidental use of logic not instead of bitwise.
adi,ad7944
- New driver supporting AD7944, AD7985 and AD7986 pin compatible ADCs.
- Later patch added use of new spi_optimize_message() to reduce overheads
of setting up a reused message.
- Additional changes later in series reduced code duplication.
adi,ad9739a RF DAC
- New driver for this 14-bit 2.5 GSPS DAC via an LVDS interface.
adi,axi-dac
- Support for this FPGA IP used to send data to high performance DACs over
an interface such as JESD204B/C or parallel interfaces. Used in
conjunction with a DAC driver. The initial user is the ad9739a.
The dmaengine-buffer needed various changes to make it bidirectional.
avago,apds9306
- New driver for this ambient light sensor.
- Fix much later in this pull for an off by 1 error.
New device IDs
==============
For these at most an ID and a instance of chip specific data was needed.
Always nice to see manufacturers sticking to an existing software interface
for new parts.
allwinner,sun20i
- Add support for h616.
invensense,mpu6050
- Add support for ICM42688
maxim,max30102
- Add compatible for MAX30101
ti,dac5571
- Add compatible for DAC081C081
General
=======
fwnode_handle
- Support for cleanup.h based __free(fwnode_handle)
- Loop macro using this for looping over child nodes without needing to
call fwnode_handle_put() in ever early exit from the loop.
- Used in:
* adi,ad3552r
* adi,ad4130
* adi,ad5770r
* adi,ad74413r
* adi,ad7173
* adi,adfm2000
* linear,ltc2688
* linear,ltc2983
* maxim,max11410
* microchip,pac1934
* qcom,spmi-adc
* renesas,rz2gl
* st,ab8500
* st,stm32 (Fix for failure to set return value precedes this patch,
providing an example of why enabling direct returns makes bugs
less likely)
- Conversions to fwnode also using the cleanup logic
* adi,ad7124
* adi,ad7292
* freescale,fsl-imx25-gcq
- Other conversions to fwnode where the new cleanup handling isn't useful
* adi,ad7192
* avia,hx711
* freescale,mma8452
* nxp,fxls8962af
* st,spear
* ti,twl4030
Features
========
adi,adxl345
- Support SPI_3WIRE mode.
adi,ad9944
- Support 3-wire mode, note this isn't normal 3-wire SPI (unlike the
adxl345 change above), but rather a wiring scheme where the SPI
chip select is used to trigger conversions rather than using a
separate pin.
- Add some device specific documentation, mostly around the various wiring
schemes.
invensense,mpu6050
- Add Wake on Motion support as an IIO event and as a wake-up source.
linear,ltc2983
- Add vdd-supply.
ti,hdc3020
- Add power management using trigger on demand mode and adding suspend and
resume handling.
- Use reset GPIO if available.
Cleanup and fixes
================
iio core
- Use the various autocleanup and lock guards from cleanup.h to simplify
the IIO core.
- Don't set the pointer used for iio_priv() if it is zero sized as that
points beyond the end of the allocation. No driver actually uses it
in that case but good to clean this up.
various drivers
- Drop unnecessary casts of other pointer types to void *
docs
- Add missing ABI entry for in_temp_input.
adi,adx345
- General cleanup prior to adding spi-3wire mode.
adi,axi-adc
- Be more flexible and allow minor version changes as these are expected
to be backwards compatible.
avago,apds9300/9600
- Merge near identical bindings. The drivers are quite different, but
the bindings can be shared. The apds9306 binding introduced in this
series uses this shared binding doc as well.
- Add missing vdd-supply
- Update binding to use IRQ_TYPE_LEVEL_LOW instead of 8.
bosch,bmp280
- Organize headers
freescale,fxl-imx25-gcq
- Use devm_ for remaining probe() time setup allowing dropping
of specific error handling and remove() functions.
infineon,dps310
- Fix handling of negative temperatures
- Bring style of other similar calls inline with the form needed
for temperatures
- Ensure error handling of regmap calls is consistent within the driver.
- Simplify scale reading logic.
invensense,mpu6050
- Flip logic in binding to exclude devices without i2c-gate instead
of opting in. The list is expected to be much shorter as all recent
devices support this feature.
honeywell,hsc030pa
- Use spi_read() instead of opening coding.
renesas,rcar
- Use device_for_each_child_of_node_scoped() to remove need to manually
release. Left over from series the rest of which went in during 6.9.
st,ab8500
- Fix naming of function parameters in kernel-doc
* tag 'iio-for-6.10a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio: (108 commits)
iio: adc: ti-ads131e08: Use device_for_each_child_node_scoped() to simplify error paths.
iio: adc: adi-axi-adc: only error out in major version mismatch
iio: dac: support the ad9739a RF DAC
iio: dac: add support for AXI DAC IP core
iio: backend: add new functionality
dt-bindings: iio: dac: add docs for AD9739A
dt-bindings: iio: dac: add docs for AXI DAC IP
iio: buffer-dmaengine: Enable write support
iio: buffer-dmaengine: Support specifying buffer direction
iio: buffer-dma: Enable buffer write support
iio: buffer-dma: Rename iio_dma_buffer_data_available()
iio: buffer-dma: add iio_dmaengine_buffer_setup()
iio: pressure: dps310: simplify scale factor reading
iio: pressure: dps310: consistently check return value of `regmap_read`
iio: pressure: dps310: introduce consistent error handling
iio: pressure: dps310: support negative temperature values
dt-bindings: iio: adc: Add GPADC for Allwinner H616
iio: dac: ad5755: make use of of_device_id table
iio: imu: inv_icm42600: add support of ICM-42688-P
dt-bindings: iio: imu: add icm42688 inside inv_icm42600
...
|
|
To support sleepable async callbacks, we need to tell push_async_cb()
whether the cb is sleepable or not.
The verifier now detects that we are in bpf_wq_set_callback_impl and
can allow a sleepable callback to happen.
Signed-off-by: Benjamin Tissoires <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alexei Starovoitov <[email protected]>
|
|
Jonathan Heathcote reported a regression caused by blamed commit
on aarch64 architecture.
x86 happens to have irq-safe __this_cpu_add_return()
and __this_cpu_sub(), but this is not generic.
I think my confusion came from "struct sock" argument,
because these helpers are called with a locked socket.
But the memory accounting is per-proto (and per-cpu after
the blamed commit). We might cleanup these helpers later
to directly accept a "struct proto *proto" argument.
Switch to this_cpu_add_return() and this_cpu_xchg()
operations, and get rid of preempt_disable()/preempt_enable() pairs.
Fast path becomes a bit faster as a result :)
Many thanks to Jonathan Heathcote for his awesome report and
investigations.
Fixes: 3cd3399dd7a8 ("net: implement per-cpu reserves for memory_allocated")
Reported-by: Jonathan Heathcote <[email protected]>
Closes: https://lore.kernel.org/netdev/VI1PR01MB42407D7947B2EA448F1E04EFD10D2@VI1PR01MB4240.eurprd01.prod.exchangelabs.com/
Signed-off-by: Eric Dumazet <[email protected]>
Acked-by: Soheil Hassas Yeganeh <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>
|
|
Currently bpf_wq_cancel_and_free() is just a placeholder as there is
no memory allocation for bpf_wq just yet.
Again, duplication of the bpf_timer approach
Signed-off-by: Benjamin Tissoires <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alexei Starovoitov <[email protected]>
|
|
Mostly a copy/paste from the bpf_timer API, without the initialization
and free, as they will be done in a separate patch.
Signed-off-by: Benjamin Tissoires <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alexei Starovoitov <[email protected]>
|
|
The devm_regulator_get_enable_optional() should be a 'call and forget'
API, meaning, when it is used to enable the regulators, the API does not
provide a handle to do any further control of the regulators. It gives
no real benefit to return an error from the stub if CONFIG_REGULATOR is
not set.
On the contrary, returning an error is causing problems to drivers when
hardware is such it works out just fine with no regulator control.
Returning an error forces drivers to specifically handle the case where
CONFIG_REGULATOR is not set, making the mere existence of the stub
questionalble.
Change the stub implementation for the
devm_regulator_get_enable_optional() to return Ok so drivers do not
separately handle the case where the CONFIG_REGULATOR is not set.
Signed-off-by: Matti Vaittinen <[email protected]>
Fixes: da279e6965b3 ("regulator: Add devm helpers for get and enable")
Reviewed-by: Guenter Roeck <[email protected]>
Link: https://lore.kernel.org/r/ZiedtOE00Zozd3XO@fedora
Signed-off-by: Mark Brown <[email protected]>
|
|
https://gitlab.freedesktop.org/drm/xe/kernel into drm-next
UAPI Changes:
- Remove unused flags (Francois Dugast)
- Extend uAPI to query HuC micro-controler firmware version (Francois Dugast)
- drm/xe/uapi: Define topology types as indexes rather than masks
(Francois Dugast)
- drm/xe/uapi: Restore flags VM_BIND_FLAG_READONLY and VM_BIND_FLAG_IMMEDIATE
(Francois Dugast)
- devcoredump updates. Some touching the output format.
(José Roberto de Souza, Matthew Brost)
- drm/xe/hwmon: Add infra to support card power and energy attributes
- Improve LRC, HWSP and HWCTX error capture. (Maarten Lankhorst)
- drm/xe/uapi: Add IP version and stepping to GT list query (Matt roper)
- Invalidate userptr VMA on page pin fault (Matthew Brost)
- Improve xe_bo_move tracepoint (Priyanka Danamudi)
- Align fence output format in ftrace log
Cross-driver Changes:
- drm/i915/hwmon: Get rid of devm (Ashutosh Dixit)
(Acked-by: Rodrigo Vivi <[email protected]>)
- drm/i915/display: convert inner wakeref get towards get_if_in_use
(SOB Rodrigo Vivi)
- drm/i915: Convert intel_runtime_pm_get_noresume towards raw wakeref
(Committer, SOB Jani Nikula)
Driver Changes:
- Fix for unneeded CCS metadata allocation (Akshata Jahagirdar)
- Fix for fix multicast support for Xe_LP platforms (Andrzej Hajda)
- A couple of build fixes (Arnd Bergmann)
- Fix register definition (Ashutosh Dixit)
- Add BMG mocs table (Balasubramani Vivekanandan)
- Replace sprintf() across driver (Bommu Krishnaiah)
- Add an xe2 workaround (Bommu Krishnaiah)
- Makefile fix (Dafna Hirschfeld)
- force_wake_get error value check (Daniele Ceraolo Spurio)
- Handle GSCCS ER interrupt (Daniele Ceraolo Spurio)
- GSC Workaround (Daniele Ceraolo Spurio)
- Build error fix (Dawei Li)
- drm/xe/gt: Add L3 bank mask to GT topology (Francois Dugast)
- Implement xe2- and GuC workarounds (Gustavo Sousa, Haridhar Kalvala,
Himal rasad Ghimiray, John Harrison, Matt Roper, Radhakrishna Sripada,
Vinay Belgaumkar, Badal Nilawar)
- xe2hpg compression (Himal Ghimiray Prasad)
- Error code cleanups and fixes (Himal Prasad Ghimiray)
- struct xe_device cleanup (Jani Nikula)
- Avoid validating bos when only requesting an exec dma-fence
(José Roberto de Souza)
- Remove debug message from migrate_clear (José Roberto de Souza)
- Nuke EXEC_QUEUE_FLAG_PERSISTENT leftover internal flag (José Roberto de Souza)
- Mark dpt and related vma as uncached (Juha-Pekka Heikkila)
- Hwmon updates (Karthik Poosa)
- KConfig fix when ACPI_WMI selcted (Lu Yao)
- Update intel_uncore_read*() return types (Luca Coelho)
- Mocs updates (Lucas De Marchi, Matt Roper)
- Drop dynamic load-balancing workaround (Lucas De Marchi)
- Fix a PVC workaround (Lucas De Marchi)
- Group live kunit tests into a single module (Lucas De Marchi)
- Various code cleanups (Lucas De Marchi)
- Fix a ggtt init error patch and move ggtt invalidate out of ggtt lock
(Maarten Lankhorst)
- Fix a bo leak (Marten Lankhorst)
- Add LRC parsing for more GPU instructions (Matt Roper)
- Add various definitions for hardware and IP (Matt Roper)
- Define all possible engines in media IP descriptors (Matt Roper)
- Various cleanups, asserts and code fixes (Matthew Auld)
- Various cleanups and code fixes (Matthew Brost)
- Increase VM_BIND number of per-ioctl Ops (Matthew Brost, Paulo Zanoni)
- Don't support execlists in xe_gt_tlb_invalidation layer (Matthew Brost)
- Handle timing out of already signaled jobs gracefully (Matthew Brost)
- Pipeline evict / restore of pinned BOs during suspend / resume (Matthew Brost)
- Do not grab forcewakes when issuing GGTT TLB invalidation via GuC
(Matthew Brost)
- Drop ggtt invalidate from display code (Matthew Brost)
- drm/xe: Add XE_BO_GGTT_INVALIDATE flag (Matthew Brost)
- Add debug messages for MMU notifier and VMA invalidate (Matthew Brost)
- Use ordered wq for preempt fence waiting (Matthew Brost)
- Initial development for SR-IOV support including some refactoring
(Michal Wajdeczko)
- Various GuC- and GT- related cleanups and fixes (Michal Wajdeczko)
- Move userptr over to start using hmm_range_fault (Oak Zeng)
- Add new PCI IDs to DG2 platform (Ravi Kumar Vodapalli)
- Pcode - and VRAM initialization check update (Riana Tauro)
- Large PM update including i915 display patches, and a fix for one of those.
(Rodrigo Vivi)
- Introduce performance tuning changes for Xe2_HPG (Shekhar Chauhan)
- GSC / HDCP updates (Suraj Kandpal)
- Minor code cleanup (Tejas Upadhyay)
- Rework / fix rebind TLB flushing and move rebind into the drm_exec locking loop
(Thomas Hellström)
- Backmerge (Thomas Hellström)
- GuC updates and fixes (Vinay Belgaumkar, Zhanjun Dong)
Signed-off-by: Dave Airlie <[email protected]>
# -----BEGIN PGP SIGNATURE-----
#
# iHUEABYKAB0WIQRskUM7w1oG5rx2IZO4FpNVCsYGvwUCZiestQAKCRC4FpNVCsYG
# v8dLAQCDFUR7R5rwSdfqzNy+Djg+9ZgmtzVEfHZ+rI2lTReaCwEAhWeK7UooIMV0
# vGsSdsqGsJQm4VLRzE6H1yemCCQOBgM=
# =HouD
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 23 Apr 2024 22:42:29 AEST
# gpg: using EDDSA key 6C91433BC35A06E6BC762193B81693550AC606BF
# gpg: Can't check signature: No public key
# Conflicts:
# drivers/gpu/drm/xe/xe_device_types.h
# drivers/gpu/drm/xe/xe_vm.c
# drivers/gpu/drm/xe/xe_vm_types.h
From: Thomas Hellstrom <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/Zievlb1wvqDg1ovi@fedora
|
|
MSI functions for allocation and free can be directly used by
the device drivers without any wrapper provided by bus drivers.
So export these MSI functions.
Also, add a wrapper API to allocate MSIs providing only the
number of interrupts rather than range for simpler driver usage.
Signed-off-by: Nipun Gupta <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alex Williamson <[email protected]>
|
|
A zone write plug BIO work function blk_zone_wplug_bio_work() calls
submit_bio_noacct_nocheck() to execute the next unplugged BIO. This
function may block. So executing zone plugs BIO works using the block
layer global kblockd workqueue can potentially lead to preformance or
latency issues as the number of concurrent work for a workqueue is
limited to WQ_DFL_ACTIVE (256).
1) For a system with a large number of zoned disks, issuing write
requests to otherwise unused zones may be delayed wiating for a work
thread to become available.
2) Requeue operations which use kblockd but are independent of zone
write plugging may alsoi end up being delayed.
To avoid these potential performance issues, create a workqueue per
zoned device to execute zone plugs BIO work. The workqueue max active
parameter is set to the maximum number of zone write plugs allocated
with the zone write plug mempool. This limit is equal to the maximum
number of open zones of the disk and defaults to 128 for disks that do
not have a limit on the number of open zones.
Fixes: dd291d77cc90 ("block: Introduce zone write plugging")
Signed-off-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
|
|
We need the usb/thunderbolt fixes in here as well.
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
We want the kernfs fixes in here as well.
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
We want the tty fixes in here as well, and it resolves a merge conflict
in:
drivers/tty/serial/serial_core.c
as well.
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
We need the char/misc fixes in here as well to work off of.
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
MT8188
Add LVTS thermal controller definition for MT8188.
Signed-off-by: Nicolas Pitre <[email protected]>
Reviewed-by: Rob Herring <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
MT8186
Add LVTS thermal controller definition for MT8186.
Signed-off-by: Nicolas Pitre <[email protected]>
Acked-by: Krzysztof Kozlowski <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
Add cmdq_pkt_acquire_event() function to support CMDQ user making
an instruction for acquiring event.
CMDQ users can use cmdq_pkt_acquire_event() as `mutex_lock`
and cmdq_pkt_clear_event() as `mutex_unlock` to protect the global
resource modified instructions between them.
cmdq_pkt_acquire_event() would wait for event to be cleared.
After event is cleared by cmdq_pkt_clear_event() in other GCE threads,
cmdq_pkt_acquire_event() would set event and keep executing next
instruction. So the mutex would work like this:
cmdq_pkt_acquire_event() /* mutex lock */
/* critical secton instructions that modified global resource */
cmdq_pkt_clear_event() /* mutex unlock */
Prevent the critical section instructions from being affected by other
GCE threads.
Signed-off-by: Jason-JH.Lin <[email protected]>
Reviewed-by: AngeloGioacchino Del Regno <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
Add cmdq_pkt_poll_addr function to support CMDQ user making
an instruction for polling a specific address of hardware rigster
to check the value with or without mask.
POLL is a legacy operation in GCE, so it does not support SPR and
CMDQ_CODE_LOGIC. To support polling the register address which doesn't
have the subsys id, CMDQ users need to make an instruction with GPR and
CMDQ_CODE_MASK operation to move the register address to be poll into GPR.
Then users can make an POLL instruction with GPR to poll the register
address assigned in previous instruction.
Signed-off-by: Jason-JH.Lin <[email protected]>
Reviewed-by: AngeloGioacchino Del Regno <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
Add cmdq_pkt_mem_move() function to support CMDQ user making
an instruction for moving a value from a source address to a
destination address.
Signed-off-by: Jason-JH.Lin <[email protected]>
Reviewed-by: AngeloGioacchino Del Regno <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
Add specific purpose register definitions for GCE, so CMDQ users can
use them as a buffer to store data.
Signed-off-by: Jason-JH.Lin <[email protected]>
Reviewed-by: AngeloGioacchino Del Regno <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
cmdq_pkt_create() and cmdq_pkt_destroy() is not suitable for
client drivers so each client driver has implement its own
function. This refinement would pass struct cmdq_pkt pointer into
cmdq_pkt_create(). In addition, client driver has the struct
cmdq_client information, so it's not necessary to store this
information in struct cmdq_pkt. After this refinement, client
drivers could use these helper funciton instead of implementing
its own version.
Signed-off-by: Chun-Kuang Hu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
cmdq_pkt_flush_async() is not used by all client drivers (MediaTek
drm driver and MediaTek mdp3 driver), so remove it.
Signed-off-by: Chun-Kuang Hu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
cmdq_pkt_eoc() append eoc command to CMDQ packet. eoc command
would ask GCE to generate IRQ. It's usually appended to the end
of packet to notify all command in the packet is done.
Reviewed-by: AngeloGioacchino Del Regno <[email protected]>
Signed-off-by: Chun-Kuang Hu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
cmdq_pkt_jump_rel() append relative jump command to the packet.
Relative jump change PC to the target address with offset from
current PC.
Signed-off-by: Chun-Kuang Hu <[email protected]>
Reviewed-by: AngeloGioacchino Del Regno <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
In order to distinguish absolute jump and relative jump,
cmdq_pkt_jump() append absolute jump command, so rename it to
cmdq_pkt_jump_abs().
Signed-off-by: Chun-Kuang Hu <[email protected]>
Reviewed-by: AngeloGioacchino Del Regno <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
[Angelo: Added temporary wrapper to avoid build breakage]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
In original design, cmdq_pkt_jump() call cmdq_get_shift_pa() every
time to get shift_pa. But the shift_pa is constant value for each
SoC, so client driver just need to call cmdq_get_shift_pa() once
and pass shift_pa to cmdq_pkt_jump() to prevent frequent function
call.
Signed-off-by: Chun-Kuang Hu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: AngeloGioacchino Del Regno <[email protected]>
|
|
The current implementation of the fpga region assumes that the low-level
module registers a driver for the parent device and uses its owner pointer
to take the module's refcount. This approach is problematic since it can
lead to a null pointer dereference while attempting to get the region
during programming if the parent device does not have a driver.
To address this problem, add a module owner pointer to the fpga_region
struct and use it to take the module's refcount. Modify the functions for
registering a region to take an additional owner module parameter and
rename them to avoid conflicts. Use the old function names for helper
macros that automatically set the module that registers the region as the
owner. This ensures compatibility with existing low-level control modules
and reduces the chances of registering a region without setting the owner.
Also, update the documentation to keep it consistent with the new interface
for registering an fpga region.
Fixes: 0fa20cdfcc1f ("fpga: fpga-region: device tree control for FPGA")
Suggested-by: Greg Kroah-Hartman <[email protected]>
Suggested-by: Xu Yilun <[email protected]>
Reviewed-by: Russ Weight <[email protected]>
Signed-off-by: Marco Pagani <[email protected]>
Acked-by: Xu Yilun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Xu Yilun <[email protected]>
|
|
Maíra needs a backmerge to apply v3d patches, and Danilo for some
nouveau patches.
Signed-off-by: Maxime Ripard <[email protected]>
|
|
Commit a72bbec70da2 ("crash: hotplug support for kexec_load()")
introduced a new kexec flag, `KEXEC_UPDATE_ELFCOREHDR`. Kexec tool uses
this flag to indicate to the kernel that it is safe to modify the
elfcorehdr of the kdump image loaded using the kexec_load system call.
However, it is possible that architectures may need to update kexec
segments other then elfcorehdr. For example, FDT (Flatten Device Tree)
on PowerPC. Introducing a new kexec flag for every new kexec segment
may not be a good solution. Hence, a generic kexec flag bit,
`KEXEC_CRASH_HOTPLUG_SUPPORT`, is introduced to share the CPU/Memory
hotplug support intent between the kexec tool and the kernel for the
kexec_load system call.
Now we have two kexec flags that enables crash hotplug support for
kexec_load system call. First is KEXEC_UPDATE_ELFCOREHDR (only used in
x86), and second is KEXEC_CRASH_HOTPLUG_SUPPORT (for all architectures).
To simplify the process of finding and reporting the crash hotplug
support the following changes are introduced.
1. Define arch specific function to process the kexec flags and
determine crash hotplug support
2. Rename the @update_elfcorehdr member of struct kimage to
@hotplug_support and populate it for both kexec_load and
kexec_file_load syscalls, because architecture can update more than
one kexec segment
3. Let generic function crash_check_hotplug_support report hotplug
support for loaded kdump image based on value of @hotplug_support
To bring the x86 crash hotplug support in line with the above points,
the following changes have been made:
- Introduce the arch_crash_hotplug_support function to process kexec
flags and determine crash hotplug support
- Remove the arch_crash_hotplug_[cpu|memory]_support functions
Signed-off-by: Sourabh Jain <[email protected]>
Acked-by: Baoquan He <[email protected]>
Acked-by: Hari Bathini <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://msgid.link/[email protected]
|
|
In the event of memory hotplug or online/offline events, the crash
memory hotplug notifier `crash_memhp_notifier()` receives a
`memory_notify` object but doesn't forward that object to the
generic and architecture-specific crash hotplug handler.
The `memory_notify` object contains the starting PFN (Page Frame Number)
and the number of pages in the hot-removed memory. This information is
necessary for architectures like PowerPC to update/recreate the kdump
image, specifically `elfcorehdr`.
So update the function signature of `crash_handle_hotplug_event()` and
`arch_crash_handle_hotplug_event()` to accept the `memory_notify` object
as an argument from crash memory hotplug notifier.
Since no such object is available in the case of CPU hotplug event, the
crash CPU hotplug notifier `crash_cpuhp_online()` passes NULL to the
crash hotplug handler.
Signed-off-by: Sourabh Jain <[email protected]>
Acked-by: Baoquan He <[email protected]>
Acked-by: Hari Bathini <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Update header inclusions to follow IWYU (Include What You Use)
principle.
Signed-off-by: Andy Shevchenko <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
The devm_regulator_get_enable() should be a 'call and forget' API,
meaning, when it is used to enable the regulators, the API does not
provide a handle to do any further control of the regulators. It gives
no real benefit to return an error from the stub if CONFIG_REGULATOR is
not set.
On the contrary, returning and error is causing problems to drivers when
hardware is such it works out just fine with no regulator control.
Returning an error forces drivers to specifically handle the case where
CONFIG_REGULATOR is not set, making the mere existence of the stub
questionalble. Furthermore, the stub of the regulator_enable() seems to
be returning Ok.
Change the stub implementation for the devm_regulator_get_enable() to
return Ok so drivers do not separately handle the case where the
CONFIG_REGULATOR is not set.
Signed-off-by: Matti Vaittinen <[email protected]>
Reported-by: Aleksander Mazur <[email protected]>
Suggested-by: Guenter Roeck <[email protected]>
Fixes: da279e6965b3 ("regulator: Add devm helpers for get and enable")
Reviewed-by: Guenter Roeck <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kuba/linux into for-6.10/io_uring
Merge net changes required for the upcoming send zerocopy improvements.
* 'for-uring-ubufops' of git://git.kernel.org/pub/scm/linux/kernel/git/kuba/linux:
net: add callback for setting a ubuf_info to skb
net: extend ubuf_info callback to ops structure
Signed-off-by: Jens Axboe <[email protected]>
|
|
Pavel Begunkov says:
====================
implement io_uring notification (ubuf_info) stacking (net part)
To have per request buffer notifications each zerocopy io_uring send
request allocates a new ubuf_info. However, as an skb can carry only
one uarg, it may force the stack to create many small skbs hurting
performance in many ways.
The patchset implements notification, i.e. an io_uring's ubuf_info
extension, stacking. It attempts to link ubuf_info's into a list,
allowing to have multiple of them per skb.
liburing/examples/send-zerocopy shows up 6 times performance improvement
for TCP with 4KB bytes per send, and levels it with MSG_ZEROCOPY. Without
the patchset it requires much larger sends to utilise all potential.
bytes | before | after (Kqps)
1200 | 195 | 1023
4000 | 193 | 1386
8000 | 154 | 1058
====================
Link: https://lore.kernel.org/all/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>
|
|
Update header inclusions to follow IWYU (Include What You Use)
principle.
Signed-off-by: Andy Shevchenko <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: William Breathitt Gray <[email protected]>
|
|
At the moment an skb can only have one ubuf_info associated with it,
which might be a performance problem for zerocopy sends in cases like
TCP via io_uring. Add a callback for assigning ubuf_info to skb, this
way we will implement smarter assignment later like linking ubuf_info
together.
Note, it's an optional callback, which should be compatible with
skb_zcopy_set(), that's because the net stack might potentially decide
to clone an skb and take another reference to ubuf_info whenever it
wishes. Also, a correct implementation should always be able to bind to
an skb without prior ubuf_info, otherwise we could end up in a situation
when the send would not be able to progress.
Reviewed-by: Jens Axboe <[email protected]>
Reviewed-by: David Ahern <[email protected]>
Signed-off-by: Pavel Begunkov <[email protected]>
Reviewed-by: Willem de Bruijn <[email protected]>
Link: https://lore.kernel.org/all/b7918aadffeb787c84c9e72e34c729dc04f3a45d.1713369317.git.asml.silence@gmail.com
Signed-off-by: Jakub Kicinski <[email protected]>
|