S0ix blocker counter is available in PWRM space. Add support to
read and show S0ix blocker counter value through debugfs.
Signed-off-by: Xi Pardee <xi.pardee@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240426002752.2504282-3-xi.pardee@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Update Lunar Lake signal status map. This status map has been updated
since the map was merged. This patch updates the signal status map to
the lastest version.
Signed-off-by: Xi Pardee <xi.pardee@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240426002752.2504282-2-xi.pardee@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Convert the incoming page to a folio and use it throughout, saving
several calls to compound_head(). Also use 'pos' for file position
rather than the ambiguous 'offset' and convert 'length' to type size_t
in case we get some truly ridiculous sized folios in the future. This
function should now be large-folio safe, but I may have missed
something.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Add missing MODULE_DESCRIPTION() to ISST modules.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-11-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Add dev_fmt for formatting log messages.
No functional impact is expected.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-10-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Use in_range() macro to simplify range check.
No functional impact is expected.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-9-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
A partitioned system has two different PCI VSEC devices per package.
A non-partitioned device has only one PCI VSEC device per package.
The current implementation only supports non partitioned systems.
Each partition maps a set of power domains. Other than reading from
different MMIO regions, there is no change in the SST functionality.
The scope of SST control is still per power domain. Hence user space
does not need to be aware of existence of partitions.
With partitions, existing per package information defined using struct
tpmi_sst_struct is enhanced to store information for both partitions. A
mapping function map_partition_power_domain_id() is introduced, which
maps to correct partition and index. This mapping function is called
in get_instance() and isst_if_clos_assoc(), before indexing into
tpmi_sst_struct->power_domain_info[].
The TPMI core platform info provides partition id and compute die ID
mask for each partition. Use this information to order power domains,
so that compute dies are presented before IO dies to match hardware
defined compute die ID for each CPU.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Reviewed-by: Zhang Rui <rui.zhang@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-8-srinivas.pandruvada@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Instead of long lines for assignment to tpmi_sst->power_domain_info, use
a local variable pd_info and assign later. Also move the assignment
of number of resources after the assignment of pd_info.
No functional change is expected.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-7-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Define a local variable for &auxdev->dev and use to shorten length of
lines. No functional change is done.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-6-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
TPMI information header added additional fields in version 2. Some of the
reserved fields in version 1 are used to define new fields.
Parse new fields and export as part of platform data. These fields include:
- PCI segment ID
- Partition ID of the package: If a package is represented by more than
one PCI device, then partition ID along with cdie_mask, describes the
scope. For example to update get/set properties for a compute die, one
of the PCI MMIO region is selected from the partition ID.
- cdie_mask: Mask of all compute dies in this partition.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Zhang Rui <rui.zhang@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-5-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Align comments in kernel-doc for the struct intel_tpmi_plat_info.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-4-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Check the major version from TPMI information header and fail to load
driver if the version is not supported.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240423204619.3946901-3-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
When tpmi_process_info() returns error, fail to load the driver.
This can happen if call to ioremap() returns error.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Cc: stable@vger.kernel.org # v6.3+
Link: https://lore.kernel.org/r/20240423204619.3946901-2-srinivas.pandruvada@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
baseboard, Protonic MECSBC, Wolfvision PF5.
The panthor driver for Mali Valhall GPUs landed, so a number of boards
enable their gpu (Cool Pi, Theobroma-Systems boards, QuartzPro64,
Rock5b, EVB1)
Also the USBDP phy driver landed, allowing the usb3 dual-role controllers
to be used on EVB1, Rock 5A and 5B, Indiedroid-Nova, Theobroma-Systems
Tiger and Jaguar.
A lot new peripherals for the Khadas Edge 2 (rtc, uart, sfc, adc, ir,
usb, pcie, tf-card, pmic); PCIe3 support on Jaguar, audio support for
the rk3308 and cache descriptions for rk356x and rk3328.
Corrected model names for boards from Radxa, Pine64, Powkiddy, Anberic
and general more dt cleanups.
-----BEGIN PGP SIGNATURE-----
iQFEBAABCAAuFiEE7v+35S2Q1vLNA3Lx86Z5yZzRHYEFAmYqxdYQHGhlaWtvQHNu
dGVjaC5kZQAKCRDzpnnJnNEdgUkyB/wOV4fCk53ryrNAY40Gh+/WVfOfO+FnEQvx
p57s90ZkkGHsK5CHcO+9MFxHM1Sjl4QbgqmYhnAuuH9jEJmlMRZw+SyDnF39hzaE
hTBpiY0rQe4IVXxfTy4Khph2cv6ba2kfTa2SOYaCq6+K5RoGfMICVjDGoZVpbMlm
KA8ewaZ0fq1gj9cwQ5tzfTQVKl6yeRbLbjKbRRjz0LoOOz3r2WjmWMA4s9l/67uD
/yvNO4BMYONx1YifNRxfVuxmsbsF9l9jEHojRTdPGRhRGT0PQHY9/X8jh2DPvAQt
0krD20rYZmCfICiR80NIw/KgBb7gmmzIcQlYOw73CesenEHBV5Mu
=1NvS
-----END PGP SIGNATURE-----
gpgsig -----BEGIN PGP SIGNATURE-----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=gZv5
-----END PGP SIGNATURE-----
Merge tag 'v6.10-rockchip-dts64-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into soc/dt
New boards: ArmSom Sige7, GameForce Chi,Forlinx FET3588-C with OK3588-C
baseboard, Protonic MECSBC, Wolfvision PF5.
The panthor driver for Mali Valhall GPUs landed, so a number of boards
enable their gpu (Cool Pi, Theobroma-Systems boards, QuartzPro64,
Rock5b, EVB1)
Also the USBDP phy driver landed, allowing the usb3 dual-role controllers
to be used on EVB1, Rock 5A and 5B, Indiedroid-Nova, Theobroma-Systems
Tiger and Jaguar.
A lot new peripherals for the Khadas Edge 2 (rtc, uart, sfc, adc, ir,
usb, pcie, tf-card, pmic); PCIe3 support on Jaguar, audio support for
the rk3308 and cache descriptions for rk356x and rk3328.
Corrected model names for boards from Radxa, Pine64, Powkiddy, Anberic
and general more dt cleanups.
* tag 'v6.10-rockchip-dts64-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip: (64 commits)
arm64: dts: rockchip: add dual-role usb3 hosts to rk3588 Tiger-Haikou
arm64: dts: rockchip: add usb-id extcon on rk3588 tiger
arm64: dts: rockchip: fix comment for upper usb3 port
arm64: dts: rockchip: fix pcie-refclk frequency on rk3588 tiger
arm64: dts: rockchip: correct gpio_pwrctrl1 typos on rk3588(s) boards
arm64: dts: rockchip: Correct the model names for Pine64 boards
dt-bindings: arm: rockchip: Correct the descriptions for Pine64 boards
arm64: dts: rockchip: Add ArmSom Sige7 board
dt-bindings: arm: rockchip: Add ArmSoM Sige7
dt-bindings: vendor-prefixes: add ArmSoM
arm64: dts: rockchip: add PCIe3 support on rk3588-jaguar
arm64: dts: rockchip: move uart2 pinmux to dtsi on rk3588-tiger
arm64: dts: rockchip: Add USB-C Support for rk3588s-indiedroid-nova
arm64: dts: rockchip: correct the model name for Radxa ROCK 3A
dt-bindings: arm: rockchip: correct the model name for Radxa ROCK 3A
arm64: dts: rockchip: Correct the model names for Radxa ROCK 5 boards
dt-bindings: arm: rockchip: Correct the descriptions for Radxa boards
arm64: dts: rockchip: add lower USB3 port to rock-5b
arm64: dts: rockchip: add upper USB3 port to rock-5a
arm64: dts: rockchip: add USB3 to rk3588-evb1
...
Link: https://lore.kernel.org/r/15361932.O9o76ZdvQC@phil
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
The modpost script is not happy
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/platform/x86/classmate-laptop.o
because there is a missing module description.
Add it to the module.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Link: https://lore.kernel.org/r/20240423161108.2636958-1-andriy.shevchenko@linux.intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Add a new driver for the custom fast charging protocol found on Lenovo Yoga
Tablet 2 1380F / 1380L models.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240422131649.46002-1-hdegoede@redhat.com
Function gfs2_freeze_unlock() is always called with &sdp->sd_freeze_gh
as its argument, so clean up the code by passing in sdp instead.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Add a new driver for the MSI WMI Platform interface. The underlying
ACPI WMI interface supports many features, but so far only reading
of fan speed sensors is implemented.
The driver was reverse-engineered based on a user request to the
lm-sensors project, see the github issue for details.
The ACPI WMI interface used by this driver seems to use the same
embedded controller interface as the msi-ec driver, but supports
automatic discovery of supported machines without relying on a
DMI whitelist.
The driver was tested by the user who created the github issue.
Closes: https://github.com/lm-sensors/lm-sensors/issues/475
Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Link: https://lore.kernel.org/r/20240421191145.3189-1-W_Armin@gmx.de
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
This patch elaborates on the use of PC by expanding the PC acronym,
explaining the units, and the relative position to which the offset
applies.
Signed-off-by: Dave Thaler <dthaler1968@googlemail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20240426231126.5130-1-dthaler1968@gmail.com
acpi_send_ev is a bool and everywhere else true/false is used
to set it. Replace the one instance using 0 with false.
Suggested-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240429093446.130322-1-hdegoede@redhat.com
The hotkey combination Fn + G can be used to disable the trackpoint
doubletap feature on Windows. Add matching functionality for Linux.
Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Vishnu Sankar <vishnuocv@gmail.com>
Link: https://lore.kernel.org/r/20240417173124.9953-4-mpearson-lenovo@squebb.ca
[hdegoede@redhat.com: Adjust for switch to sparse-keymap keymaps]
[hdegoede@redhat.com: Do not log unknown event msg for doubletap when disabled]
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-25-hdegoede@redhat.com
New Lenovo platforms are adding the FN+N key to generate system debug
details that support can use for collecting important details on any
customer cases for Windows.
Add the infrastructure so we can do the same on Linux by sending
a KEY_VENDOR keycode to userspace.
Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Nitin Joshi <njoshi1@lenovo.com>
Link: https://lore.kernel.org/r/20240417173124.9953-3-mpearson-lenovo@squebb.ca
[hdegoede@redhat.com: Adjust for switch to sparse-keymap keymaps]
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-24-hdegoede@redhat.com
Lenovo trackpoints are adding the ability to generate a doubletap event.
This handles the doubletap event and sends the KEY_PROG4 event to
userspace. Despite the driver itself not using KEY_PROG1 - KEY_PROG3 this
still uses KEY_PROG4 because of some keys being remapped to KEY_PROG1 -
KEY_PROG3 by default by the upstream udev hwdb containing:
evdev:name:ThinkPad Extra Buttons:dmi:bvn*:bvr*:bd*:svnLENOVO*:pn*:*
...
KEYBOARD_KEY_17=prog1
KEYBOARD_KEY_1a=f20 # Microphone mute button
KEYBOARD_KEY_45=bookmarks
KEYBOARD_KEY_46=prog2 # Fn + PrtSc, on Windows: Snipping tool
KEYBOARD_KEY_4a=prog3 # Fn + Right shift, on Windows: No idea
Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Vishnu Sankar <vishnuocv@gmail.com>
Link: https://lore.kernel.org/r/20240417173124.9953-2-mpearson-lenovo@squebb.ca
[hdegoede@redhat.com: Adjust for switch to sparse-keymap keymaps]
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-23-hdegoede@redhat.com
The X1 carbon 2014 / 2nd gen's adaptive keyboard top row's "Home" mode,
which is 1 of the 2 modes Linux supports, has clipping-tool and cloud
buttons which so far are not mapped.
I assume these were left as KEY_RESERVED because no suitable KEY_FOO codes
were available when support was added.
In the mean time we have gotten KEY_SELECTIVE_SCREENSHOT and this has been
used for the clipping-tool function under Fn + PrtSc on more traditional
ThinkPad keyboards already.
Finding a KEY_FOO code for the cloud key is harder looking at the symbol
it seems to refer to cloud-storage which made me think of file syncing,
or file transfer which has let me to pick KEY_XFER for this.
Note this is based on looking at a picture of the adaptive top row
in Home mode and has not been tested on an actual adaptive keyboard.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-21-hdegoede@redhat.com
Switch the hotkey keymap handling over to the sparse-keymap helpers,
there should be no functional changes from this.
Note all the mappings to KEY_UNKNOWN are removed since that is the default
behavior of sparse_keymap_report_event() for unknown scancodes.
Also drop the big comment about making changes to the keymaps since
the contents of that comment are mostly obsolete.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-20-hdegoede@redhat.com
The input core already filters out EV_KEY events for KEY_RESERVED,
remove the check for this in tpacpi_input_send_key() and rely
on the input core filtering instead.
Also change tpacpi_input_send_key() to only report the scancode
once instead of reporting it on both press and release. Together
these 2 changes make tpacpi_input_send_key() behave the same as
sparse_keymap_report_event().
The goal of this patch is to have a separate commit with
the slightly different behavior from sparse_keymap_report_event()
before switching over to using the sparse-keymap helpers.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-19-hdegoede@redhat.com
Change the default keymap to report the correct keycodes for the volume and
brightness keys. Reporting key events for these is already filtered out by
the hotkey_reserved_mask which masks these keys out of hotkey_user_mask at
initialization time, so there is no need to also map them to KEY_RESERVED.
This avoids users, who want these to be reported, having to also remap
the keycodes on top of overriding hotkey_user_mask to report these
and Linux userspace has already been overriding the KEY_RESERVED mappings
with the correct keycodes through udev/hwdb/60-keyboard.hwdb for years now.
Also drop hotkey_unmap() it was only used to dynamically map the brightness
keys to KEY_RESERVED and after removing that it has no remaining users.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-18-hdegoede@redhat.com
Change the hotkey_reserved_mask initialization to hardcode the list
of reserved keys. There are only a few reserved keys and the code to
iterate over the keymap will be removed when moving to sparse-keymaps.
Note only the 32 original hotkeys are affected by the hotkey_*_mask values:
if (i < sizeof(hotkey_reserved_mask)*8)
hotkey_reserved_mask |= 1 << i;
The (i < sizeof(hotkey_reserved_mask)*8) condition translates to (i < 32)
so this code only ever set bits in hotkey_reserved_mask for the 32 original
hotkeys. Therefor this patch does not set any bits in hotkey_reserved_mask
for the KEY_RESERVED mappings for the adaptive keyboard scancodes.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-17-hdegoede@redhat.com
Petr Machata says:
====================
mlxsw: Improve events processing performance
Amit Cohen writes:
Spectrum ASICs only support a single interrupt, it means that all the
events are handled by one IRQ (interrupt request) handler.
Currently, we schedule a tasklet to handle events in EQ, then we also use
tasklet for CQ, SDQ and RDQ. Tasklet runs in softIRQ (software IRQ)
context, and will be run on the same CPU which scheduled it. It means that
today we have one CPU which handles all the packets (both network packets
and EMADs) from hardware.
The existing implementation is not efficient and can be improved.
Measuring latency of EMADs in the driver (without the time in FW) shows
that latency is increased by factor of 28 (x28) when network traffic is
handled by the driver.
Measuring throughput in CPU shows that CPU can handle ~35% less packets
of specific flow when corrupted packets are also handled by the driver.
There are cases that these values even worse, we measure decrease of ~44%
packet rate.
This can be improved if network packet and EMADs will be handled in
parallel by several CPUs, and more than that, if different types of traffic
will be handled in parallel. We can achieve this using NAPI.
This set converts the driver to process completions from hardware via NAPI.
The idea is to add NAPI instance per CQ (which is mapped 1:1 to SDQ/RDQ),
which means that each DQ can be handled separately. we have DQ for EMADs
and DQs for each trap group (like LLDP, BGP, L3 drops, etc..). See more
details in commit messages.
An additional improvement which is done as part of this set is related to
doorbells' ring. The idea is to handle small chunks of Rx packets (which
is also recommended using NAPI) and ring doorbells once per chunk. This
reduces the access to hardware which is expensive (time wise) and might
take time because of memory barriers.
With this set we can see better performance.
To summerize:
EMADs latency:
+------------------------------------------------------------------------+
| | Before this set | Now |
|------------------|---------------------------|-------------------------|
| Increased factor | x28 | x1.5 |
+------------------------------------------------------------------------+
Note that we can see even measurements that show better latency when
traffic is handled by the driver.
Throughput:
+------------------------------------------------------------------------+
| | Before this set | Now |
|-------------|----------------------------|-----------------------------|
| Reduced | 35% | 6% |
| packet rate | | |
+------------------------------------------------------------------------+
Additional improvements are planned - use page pool for buffer allocations
and avoid cache miss of each SKB using napi_build_skb().
Patch set overview:
Patches #1-#2 improve access to hardware by reducing dorbells' rings
Patch #3-#4 are preaparations for NAPI usage
Patch #5 converts the driver to use NAPI
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Spectrum ASICs only support a single interrupt, that means that all the
events are handled by one IRQ (interrupt request) handler. Once an
interrupt is received, we schedule tasklet to handle events from EQ and
then schedule tasklets to handle completions from CQs. Tasklet runs in
softIRQ (software IRQ) context, and will be run on the same CPU which
scheduled it. That means that today we use only one CPU to handle all the
packets (both network packets and EMADs) from hardware.
This can be improved using NAPI. The idea is to use NAPI instance per
CQ, which is mapped 1:1 to DQ (RDQ or SDQ). NAPI poll method can be run
in kernel thread, so then the driver will be able to handle WQEs in several
CPUs. Convert the existing code to use NAPI APIs.
Add NAPI instance as part of 'struct mlxsw_pci_queue' and initialize it
as part of CQs initialization. Set the appropriate poll method and dummy
net device, according to queue number, similar to tasklet setup. For CQs
which are used for completions of RDQ, use Rx poll method and
'napi_dev_rx', which is set as 'threaded'. It means that Rx poll method
will run in kernel context, so several RDQs will be handled in parallel.
For CQs which are used for completions of SDQ, use Tx poll method and
'napi_dev_tx', this method will run in softIRQ context, as it is
recommended in NAPI documentation, as Tx packets' processing is short task.
Convert mlxsw_pci_cq_{rx,tx}_tasklet() to poll methods. Handle 'budget'
argument - ignore it in Tx poll method, as it is recommended to not limit
Tx processing. For Rx processing, handle up to 'budget' completions.
Return 'work_done' which is the amount of completions that were handled.
Handle the following cases:
1. After processing 'budget' completions, the driver still has work to do:
Return work-done = budget. In that case, the NAPI instance will be
polled again (without the need to be rescheduled). Do not re-arm the
queue, as NAPI will handle the reschedule, so we do not have to involve
hardware to send an additional interrupt for the completions that should
be processed.
2. Event processing has been completed:
Call napi_complete_done() to mark NAPI processing as completed, which
means that the poll method will not be rescheduled. Re-arm the queue,
as all completions were handled.
In case that poll method handled exactly 'budget' completions, return
work-done = budget -1, to distinguish from the case that driver still
has completions to handle. Otherwise, return the amount of completions
that were handled.
Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The next patch will set the driver to use NAPI for event processing. Then
tasklet mechanism will be used only for EQ. Reorganize 'mlxsw_pci_queue'
to hold EQ and CQ attributes in a union. For now, add tasklet for both EQ
and CQ. This will be changed in the next patch, as 'tasklet_struct' will be
replaced with NAPI instance.
Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlxsw will use NAPI for event processing in a next patch. As preparation,
add two dummy net devices and initialize them.
NAPI instance should be attached to net device. Usually each queue is used
by a single net device in network drivers, so the mapping between net
device to NAPI instance is intuitive. In our case, Rx queues are not per
port, they are per trap-group. Tx queues are mapped to net devices, but we
do not have a separate queue for each local port, several ports share the
same queue.
Use init_dummy_netdev() to initialize dummy net devices for NAPI.
To run NAPI poll method in a kernel thread, the net device which NAPI
instance is attached to should be marked as 'threaded'. It is
recommended to handle Tx packets in softIRQ context, as usually this is
a short task - just free the Tx packet which has been transmitted.
Rx packets handling is more complicated task, so drivers can use a
dedicated kernel thread to process them. It allows processing packets from
different Rx queues in parallel. We would like to handle only Rx packets in
kernel threads, which means that we will use two dummy net devices
(one for Rx and one for Tx). Set only one of them with 'threaded' as it
will be used for Rx processing. Do not fail in case that setting 'threaded'
fails, as it is better to use regular softIRQ NAPI rather than preventing
the driver from loading.
Note that the net devices are initialized with init_dummy_netdev(), so
they are not registered, which means that they will not be visible to user.
It will not be possible to change 'threaded' configuration from user
space, but it is reasonable in our case, as there is no another
configuration which makes sense, considering that user has no influence
on the usage of each queue.
Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, for each CQE in CQ, we ring CQ doorbell, then handle RDQ and
ring RDQ doorbell. Finally we ring CQ arm doorbell - once per CQ tasklet.
The idea of ringing CQ doorbell before RDQ doorbell, is to be sure that
when we post new WQE (after RDQ is handled), there is an available CQE.
This was done because of a hardware bug as part of
commit c9ebea04cb ("mlxsw: pci: Ring CQ's doorbell before RDQ's").
There is no real reason to ring RDQ and CQ doorbells for each completion,
it is better to handle several completions and reduce number of ringings,
as access to hardware is expensive (time wise) and might take time because
of memory barriers.
A previous patch changed CQ tasklet to handle up to 64 Rx packets. With
this limitation, we can ring CQ and RDQ doorbells once per CQ tasklet.
The counters of the doorbells are increased by the amount of packets
that we handled, then the device will know for which completion to send
an additional event.
To avoid reordering CQ and RDQ doorbells' ring, let the tasklet to ring
also RDQ doorbell, mlxsw_pci_cqe_rdq_handle() handles the counter but
does not ring the doorbell.
Note that with this change there is no need to copy the CQE, as we ring CQ
doorbell only after Rx packet processing (which uses the CQE) is done.
Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can get many completions in one interrupt. Currently, the CQ tasklet
handles up to half queue size completions, and then arms the hardware to
generate additional events, which means that in case that there were
additional completions that we did not handle, we will get immediately an
additional interrupt to handle the rest.
The decision to handle up to half of the queue size is arbitrary and was
determined in 2015, when mlxsw driver was added to the kernel. One
additional fact that should be taken into account is that while WQEs
from RDQ are handled, the CPU that handles the tasklet is dedicated for
this task, which means that we might hold the CPU for a long time.
Handle WQEs in smaller chucks, then arm CQ doorbell to notify the hardware
to send additional notifications. Set the chunk size to 64 as this number
is recommended using NAPI and the driver will use NAPI in a next patch.
Note that for now we use ARM doorbell to retrigger CQ tasklet, but with
NAPI it will be more efficient as software will reschedule the poll
method and we will not involve hardware for that.
Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Do not send ACPI netlink events for unknown hotkeys, to avoid userspace
starting to rely on them. Instead these should be added to the keymap to
send evdev events.
This should not cause a behavior change for existing laptop models since
all currently known 0x1xxx events have a mapping.
In hindsight the ACPI netlink events should have been suppressed for
the adaptive keyboard and extended hotkeys events too. But the kernel has
been sending ACPI netlink events for those for a long time now, so we
cannot just stop sending them without potentially causing issues for
existing users who may depend on these.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-16-hdegoede@redhat.com
All callers of tpacpi_input_send_key() first call tpacpi_driver_event(),
move the tpacpi_driver_event() inside tpacpi_input_send_key() to avoid
code duplication.
For the original hotkey codes 0x1001 - 0x1020 tpacpi_driver_event() never
returns true. So the added "return true;" inside tpacpi_input_send_key()
never happens when called from tpacpi_hotkey_send_key() so behavior does
not change.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-15-hdegoede@redhat.com
Move the mapping of hkey events to scancodes to tpacpi_input_send_key(),
this results in a nice cleanup and prepares things for adding sparse-keymap
support.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-14-hdegoede@redhat.com
Both are only 1 / 2 lines and both only have 1 caller fold the contents
into tpacpi_hotkey_send_key() which is their single caller.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-13-hdegoede@redhat.com
Call tpacpi_driver_event() at the top of hotkey_notify_hotkey() for all
(orig / adaptive / extended) hotkey types, rather then having the orig
code path call tpacpi_input_send_key_masked() which calls it through
hotkey_driver_event() and having the adaptive / extended helpers call
it separately.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-12-hdegoede@redhat.com
Move hotkey_user_mask check to tpacpi_input_send_key(), this is
a preparation patch for further refactoring.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-11-hdegoede@redhat.com
Move the special handling (send_acpi_ev = false, hotkey_source_mask check)
for original hotkeys out of the switch-case in hotkey_notify_hotkey().
This is a preparation patch for further refactoring.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-10-hdegoede@redhat.com
Factor out the adaptive kbd non hotkey event handling into
adaptive_keyboard_change_row() and adaptive_keyboard_s_quickview_row()
helpers and move the handling of TP_HKEY_EV_DFR_CHANGE_ROW and
TP_HKEY_EV_DFR_S_QUICKVIEW_ROW to tpacpi_driver_event().
This groups all the handling of hotkey events which do not emit
a key press event together in tpacpi_driver_event().
This also drops the returning of false as known-event value when
adaptive_keyboard_get_mode() / adaptive_keyboard_set_mode() fail.
These functions already log an error on failure, returning false just
leads to an extra messgae being logged about the hkey event being
unknown, which is wrong as the event is not unknown.
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-9-hdegoede@redhat.com
tpacpi_driver_event() already only responds to hkey events which it knows
about. Make it return a bool and return true when it has handled the event.
This avoids the need to list TP_HKEY_EV_foo values to which it responds
both in its caller and in the function itself.
Instead callers can now call it unconditionally and check the return value.
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240424122834.19801-8-hdegoede@redhat.com
