| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
| |
Add the following machines into quirk,
Isnpiron 5420, Isnpiron 5520, Isnpiron 5720,
Isnpiron 7420, Isnpiron 7520, Isnpiron 7720
Signed-off-by: AceLan Kao <acelan.kao@canonical.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The patch 6ae3a0876185: "ACER: Add support for accelerometer sensor"
from Jun 1, 2012, leads to the following Smatch warning:
drivers/platform/x86/acer-wmi.c:1886 acer_wmi_accel_destroy()
error: don't call input_free_device() after input_unregister_device()
drivers/platform/x86/acer-wmi.c
1883 static void acer_wmi_accel_destroy(void)
1884 {
1885 input_unregister_device(acer_wmi_accel_dev);
1886 input_free_device(acer_wmi_accel_dev);
1887 }
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Matthew Garrett <mjg59@srcf.ucam.org>
Cc: Fengguang Wu <wfg@linux.intel.com>
Cc: joeyli <jlee@suse.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
drivers/platform/x86/acer-wmi.c:1836:18: sparse: Using plain integer as NULL pointer
drivers/platform/x86/acer-wmi.c:1836:
1833
1834 BUG_ON(!name || !ah);
1835
> 1836 handle = 0;
1837 status = acpi_get_devices(prop, acer_wmi_get_handle_cb,
1838 (void *)name, &handle);
1839
Reported-by: Fengguang Wu <wfg@linux.intel.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Matthew Garrett <mjg59@srcf.ucam.org>
Cc: Fengguang Wu <wfg@linux.intel.com>
Cc: joeyli <jlee@suse.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
There are some new video switch keys that used by newer machines.
0xA0 - SDSP HDMI only
0xA1 - SDSP LCD + HDMI
0xA2 - SDSP CRT + HDMI
0xA3 - SDSP TV + HDMI
But in Linux, there is no suitable userspace application to handle this,
so, mapping them all to KEY_SWITCHVIDEOMODE.
Signed-off-by: AceLan Kao <acelan.kao@canonical.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The BIOS of these machines will try to enable/disable wifi/bt in
their own sqeuence. It won't read the enable/disable parameter
in WMI command, but just iterates the wifi/bt's status described below
1st. enable wifi, enable bt
2nd. disable wifi, enable bt
3rd. enable wifi, disable bt
4th. disable wifi, disable bt
That will totally mess up the rfkill status, since we will try to read
wifi and bt's status and reset it again while booting up.
To avoid this, these machines should set the wapf value to 4,
that will let software totally control the wifi/bt's status and
BIOS will do nothing instead of sending out the 0x88(KEY_RFKILL) event
instead of 0x5e(wifi enable), 0x5f(wifi diable), 0x7d(bt enable), and
0x7e(bt disable) through WMI.
With this patch[1], it will handle the KEY_RFKILL event correctly and
will block/unblock wifi and bt together.
1. https://lkml.org/lkml/2012/5/21/75
Signed-off-by: AceLan Kao <acelan.kao@canonical.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Since ACPI devices ids were changed to use always upper-case letters, the ACPI
id of the extra keys (FNBT0000) was not maching the one defined in the driver
(FnBT0000), causing the extra keys not to work.
The patch replaces the driver id with the one reported by ACPI, fixing the
problem.
Signed-off-by: Miguel Gómez <magomez@igalia.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
| |
Classmate V4 laptop includes a new accelerometer that can't be handled by
previous driver. This patch adds a new driver to handle it.
[mjg: Fixed up the driver pm stuff]
Signed-off-by: Miguel Gómez <magomez@igalia.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
According to the ASUS WMI spec., to enable resume on lid open should
use the device ID(0x00120032), but it doesn't work indeed.
After discussing with ASUS' BIOS engineer, they say wake on lid open
doesn't have a uniq device ID(0x00120032) in the BIOS. It shares the same
device ID with deep S3(0x00120031), and the deep S3(resume on lid open)
is disable by default.
Adding this option in asus wmi sysfs
/sys/devices/platform/<platform>/lid_resume
so that userspace apps can enable/disable this feature by themselves.
Signed-off-by: AceLan Kao <acelan.kao@canonical.com>
Signed-off-by: Corentin Chary <corentin.chary@gmail.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
BugLink: https://bugs.launchpad.net/bugs/1000146
Some h/w that can adjust screen brightness through ACPI functions, but
can't turn on/off the backlight power correctly. So, we list those h/w in
quirks and try to turn on/off the backlight power through WMI.
Signed-off-by: AceLan Kao <acelan.kao@canonical.com>
Signed-off-by: Corentin Chary <corentin.chary@gmail.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
> Chassis Information
> Manufacturer: SAMSUNG ELECTRONICS CO., LTD.
> Type: Other
Type should be "Notebook", "Laptop", .. not "Other".
Signed-off-by: Corentin Chary <corentin.chary@gmail.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
On Samsung X360, the BIOS will set a flag (VDRV) if the generic
ACPI backlight device is used. This flag will definitively break
the backlight interface (even the vendor interface) untill next
reboot. It's why we should prevent video.ko from being used here
and we can't rely on a later call to acpi_video_unregister().
Signed-off-by: Corentin Chary <corentin.chary@gmail.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
| |
Signed-off-by: Corentin Chary <corentin.chary@gmail.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
| |
Instead of using directly acpi_video_unregister(), use
acpi_video_dmi_promote_vendor() (and make it call
acpi_video_unregister() if needed)
Signed-off-by: Corentin Chary <corentin.chary@gmail.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of adding a big blacklist in video_detect.c to set
ACPI_VIDEO_BACKLIGHT_DMI_VENDOR correctly, let external modules
promote or demote themselves when they know the generic video
module won't work.
Currently drivers where using acpi_video_unregister() directly
but:
- That didn't respect any acpi_backlight=[video|vendor] parameter
provided by the user.
- Any later call to acpi_video_register() would still re-load the
generic video module (and some gpu drivers are doing that).
This patch fix those two issues.
Signed-off-by: Corentin Chary <corentin.chary@gmail.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
| |
This device is present on Iconia Tab W500.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: joeyli <jlee@suse.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
According to responses from the BIOS team, ASUS_WMI_METHODID_DSTS2
(0x53545344) will be used as future DSTS ID. In addition, calling
asus_wmi_evaluate_method(ASUS_WMI_METHODID_DSTS2, 0, 0, NULL) returns
ASUS_WMI_UNSUPPORTED_METHOD in new ASUS laptop PCs. This patch fixes
no DSTS ID will be assigned in this case.
Signed-off-by: Alex Hung <alex.hung@canonical.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: stable@kernel.org
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio
Pull GPIO changes from Linus Walleij:
- New driver for AMD-8111 southbridge GPIOs
- New driver for Wolfson Micro Arizona devices
- Propagate device tree parse errors
- Probe deferral finalizations - all expected calls to GPIO will now
hopefully request deferral where apropriate
- Misc updates to TCA6424, WM8994, LPC32xx, PCF857x, Samsung MXC, OMAP
and PCA953X drivers.
Fix up gpio_idx conflicts in drivers/gpio/gpio-mxc.c
* tag 'gpio-for-v3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
gpio: of_get_named_gpio_flags() return -EPROBE_DEFER if GPIO not yet available
gpiolib: Defer failed gpio requests by default
MAINTAINERS: add entry OMAP GPIO driver
gpio/pca953x: increase variables size to support 24 bit of data
GPIO: PCA953X: Increase size of invert variable to support 24 bit
gpio/omap: move bank->dbck initialization to omap_gpio_mod_init()
gpio/mxc: use the edge_sel feature if available
gpio: propagate of_parse_phandle_with_args errors
gpio: samsung: add flags specifier to device-tree binding
gpiolib: Add support for Wolfson Microelectronics Arizona class devices
gpio: gpio-lpc32xx: Add gpio_to_irq mapping
gpio: pcf857x: share 8/16 bit access functions
gpio: LPC32xx: Driver cleanup
MAINTAINERS: Add Wolfson gpiolib drivers to the Wolfson entry
gpiolib: wm8994: Convert to devm_kzalloc()
gpiolib: wm8994: Use irq_domain mappings for gpios
gpio: add a driver for GPIO pins found on AMD-8111 south bridge chips
gpio/tca6424: merge I2C transactions, remove cast
gpio/of: fix a typo of comment message
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
of_get_named_gpio_flags() and of_get_named_gpio() return -EPROBE_DEFER if the
respective GPIO is not (yet) available. This is useful if driver's probe()
functions try to get a GPIO whose controller isn't probed yet. Thus, the driver
can be probed again later on.
The function still returns -EINVAL on other errors (parse error or node doesn't
exist). This way, the case of an optional/intentionally missing GPIO is handled
appropriately.
Signed-off-by: Roland Stigge <stigge@antcom.de>
Acked-by: Alexandre Pereira da Silva <aletes.xgr@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Since users must be explicitly provided with a GPIO number in order to
request one the overwhelmingly common case for failing to request will
be that the required GPIO driver has not yet registered and we should
therefore defer until it has registered.
In order to avoid having to code this logic in individual drivers have
gpio_request() return -EPROBE_DEFER when failing to look up the GPIO.
Drivers which don't want this behaviour can override it if they desire.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Since I've been maintaining this, making it official at the request of the
GPIO maintainers.
Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kevin Hilman <khilman@ti.com>
[corrected file path]
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Increase variable size from u16 to u32 to allocate 24 bit of data required for
the TCA6424 I/O expander device type.
Signed-off-by: Leed Aguilar <leed.aguilar@ti.com>
Signed-off-by: Chandrabhanu Mahapatra <cmahapatra@ti.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
TCA6424 is a low voltage 24 bit I2C and SMBus I/O expander of pca953x family
similar to its 16 bit predecessor TCA6416. It comes with three 8-bit active
Input, Output, Polarity Inversion and Configuration registers each. The polarity
of Input ports can be reversed by setting the appropiate bit in Polarity
Inversion registers.
The variables corresponding to Input, Output and Configuration registers have
already been updated to support 24 bit values. This patch thus updates the
invert variable of PCA953X platform data to support 24 bit.
Signed-off-by: Chandrabhanu Mahapatra <cmahapatra@ti.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Since the bank->dbck initialization in a one time operation there
is no need to keep this within gpio_debounce(). Therefore, moving
clk_get(bank->dbck) to omap_gpio_mod_init(). Since the value of
bank->dbck would be NULL at the beginning, this check has been
removed.
Signed-off-by: Tarun Kanti DebBarma <tarun.kanti@ti.com>
Reported-by: Paul Walmsley <paul@pwsan.com>
Reviewed-by: Jon Hunter <jon-hunter@ti.com>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Cousson, Benoit <b-cousson@ti.com>
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Some mxc processors have an edge_sel feature, which allows the IRQ to be
triggered by any edge.
This patch makes use of this feature if available, which skips mxc_flip_edge().
Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: Linus Walleij <linus.walleij@stericsson.com>
Acked-by: Sascha Hauer <kernel@pengutronix.de>
Cc: <linux-arm-kernel@lists.infradead.org>
Signed-off-by: Benoît Thébaudeau <benoit.thebaudeau@advansee.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Make of_get_named_gpio_flags propagate any error it receives from
of_parse_phandle_with_args instead of inconditionally returning -EINVAL.
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This adds a flags field to the gpio specifier for Samsung. I didn't
want to add yet another field in the already quite long specifier, so
I decided to compress it together with the Pull Up/Down settings instead.
This is needed to, for example, have a gpio-keys input that is active low.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The Arizona class devices provide some GPIOs for use in the system. This
driver provides support for these via gpiolib. Currently interrupts are
not supported, normally the GPIOs are outputs only.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
[Fold in WM5110 support patch]
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
This patch helps mapping with gpio_to_irq for the GPIOs that are irq enabled.
Signed-off-by: Roland Stigge <stigge@antcom.de>
Tested-by: Alexandre Pereira da Silva <aletes.xgr@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
This patch adds 8/16 bit write/read functions,
and shared gpio input/output/set/get functions.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Since LPC32xx is now switched over to devicetree based GPIO, the unused
lpc32xx_gpio_init() can be removed.
Further, the driver title changed since it referred to the wrong file.
Signed-off-by: Roland Stigge <stigge@antcom.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| | |
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| | |
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This has no practical impact at present since we don't support device tree
so any user must have set an irq_base but this will in future allow a
transition to device tree with minimal invasiveness.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Add a driver to use GPIO pins available on several AMD south bridges
(currently only AMD 8111 is supported).
Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| |\
| | |
| | |
| | | |
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This is a follow-up to ae79c1904 "[PATCH v2] Add support for TCA6424"
merged in the v3.5 merge window. It fixes comments made when the
patch was merged.
- Use 3 byte transfers instead of two separate transfers (2+1 byte)
- An unnecessary cast removed
Signed-off-by: Andreas Schallenberg <Andreas.Schallenberg@3alitytechnica.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
|
| | |
| | |
| | |
| | |
| | | |
Signed-off-by: Dong Aisheng <dong.aisheng@linaro.org>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile
Pull tile PCI build fixes from Chris Metcalf:
"This fixes a couple of issues with the pci_bus.subordinate removal
from pci-next as it affected the tile architecture. One commit does
the bombing for tilegx PCI (added during the merge window, so missed
the pci-next bombing) and the other commit undoes a buggy part of the
bombing for tilepro PCI."
* 'stable' of git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile:
tilepro pci: fix pci_bus.subordinate bad bombing from b918c62e
tilegx pci: fix semantic merge conflict with 3527ed81c
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The bombing to convert pci_bus.subordinate to busn_res.end accidentally
modified a "struct pci_dev" site, causing this file not to compile.
This commit reverts that code to use dev->subordinate again.
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Yinghai Lu removed pci_bus.subordinate in pci-next, which meant that
the tile-next changes to add tilegx PCI support don't build. This
was expected (seen in linux-next) and this one-line fix is along
the same lines as commit b918c62e for all other architectures.
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Recently, glibc made a change to suppress sign-conversion warnings in
FD_SET (glibc commit ceb9e56b3d1). This uncovered an issue with the
kernel's definition of __NFDBITS if applications #include
<linux/types.h> after including <sys/select.h>. A build failure would
be seen when passing the -Werror=sign-compare and -D_FORTIFY_SOURCE=2
flags to gcc.
It was suggested that the kernel should either match the glibc
definition of __NFDBITS or remove that entirely. The current in-kernel
uses of __NFDBITS can be replaced with BITS_PER_LONG, and there are no
uses of the related __FDELT and __FDMASK defines. Given that, we'll
continue the cleanup that was started with commit 8b3d1cda4f5f
("posix_types: Remove fd_set macros") and drop the remaining unused
macros.
Additionally, linux/time.h has similar macros defined that expand to
nothing so we'll remove those at the same time.
Reported-by: Jeff Law <law@redhat.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
CC: <stable@vger.kernel.org>
Signed-off-by: Josh Boyer <jwboyer@redhat.com>
[ .. and fix up whitespace as per akpm ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|\ \ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86/mm changes from Peter Anvin:
"The big change here is the patchset by Alex Shi to use INVLPG to flush
only the affected pages when we only need to flush a small page range.
It also removes the special INVALIDATE_TLB_VECTOR interrupts (32
vectors!) and replace it with an ordinary IPI function call."
Fix up trivial conflicts in arch/x86/include/asm/apic.h (added code next
to changed line)
* 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/tlb: Fix build warning and crash when building for !SMP
x86/tlb: do flush_tlb_kernel_range by 'invlpg'
x86/tlb: replace INVALIDATE_TLB_VECTOR by CALL_FUNCTION_VECTOR
x86/tlb: enable tlb flush range support for x86
mm/mmu_gather: enable tlb flush range in generic mmu_gather
x86/tlb: add tlb_flushall_shift knob into debugfs
x86/tlb: add tlb_flushall_shift for specific CPU
x86/tlb: fall back to flush all when meet a THP large page
x86/flush_tlb: try flush_tlb_single one by one in flush_tlb_range
x86/tlb_info: get last level TLB entry number of CPU
x86: Add read_mostly declaration/definition to variables from smp.h
x86: Define early read-mostly per-cpu macros
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The incompatible parameter of flush_tlb_mm_range cause build warning.
Fix it by correct parameter.
Ingo Molnar found that this could also cause a user space crash.
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reported-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Alex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1342747103-19765-1-git-send-email-alex.shi@intel.com
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This patch do flush_tlb_kernel_range by 'invlpg'. The performance pay
and gain was analyzed in previous patch
(x86/flush_tlb: try flush_tlb_single one by one in flush_tlb_range).
In the testing: http://lkml.org/lkml/2012/6/21/10
The pay is mostly covered by long kernel path, but the gain is still
quite clear, memory access in user APP can increase 30+% when kernel
execute this funtion.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1340845344-27557-10-git-send-email-alex.shi@intel.com
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
There are 32 INVALIDATE_TLB_VECTOR now in kernel. That is quite big
amount of vector in IDT. But it is still not enough, since modern x86
sever has more cpu number. That still causes heavy lock contention
in TLB flushing.
The patch using generic smp call function to replace it. That saved 32
vector number in IDT, and resolved the lock contention in TLB
flushing on large system.
In the NHM EX machine 4P * 8cores * HT = 64 CPUs, hackbench pthread
has 3% performance increase.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1340845344-27557-9-git-send-email-alex.shi@intel.com
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Not every tlb_flush execution moment is really need to evacuate all
TLB entries, like in munmap, just few 'invlpg' is better for whole
process performance, since it leaves most of TLB entries for later
accessing.
This patch also rewrite flush_tlb_range for 2 purposes:
1, split it out to get flush_blt_mm_range function.
2, clean up to reduce line breaking, thanks for Borislav's input.
My micro benchmark 'mummap' http://lkml.org/lkml/2012/5/17/59
show that the random memory access on other CPU has 0~50% speed up
on a 2P * 4cores * HT NHM EP while do 'munmap'.
Thanks Yongjie's testing on this patch:
-------------
I used Linux 3.4-RC6 w/ and w/o his patches as Xen dom0 and guest
kernel.
After running two benchmarks in Xen HVM guest, I found his patches
brought about 1%~3% performance gain in 'kernel build' and 'netperf'
testing, though the performance gain was not very stable in 'kernel
build' testing.
Some detailed testing results are below.
Testing Environment:
Hardware: Romley-EP platform
Xen version: latest upstream
Linux kernel: 3.4-RC6
Guest vCPU number: 8
NIC: Intel 82599 (10GB bandwidth)
In 'kernel build' testing in guest:
Command line | performance gain
make -j 4 | 3.81%
make -j 8 | 0.37%
make -j 16 | -0.52%
In 'netperf' testing, we tested TCP_STREAM with default socket size
16384 byte as large packet and 64 byte as small packet.
I used several clients to add networking pressure, then 'netperf' server
automatically generated several threads to response them.
I also used large-size packet and small-size packet in the testing.
Packet size | Thread number | performance gain
16384 bytes | 4 | 0.02%
16384 bytes | 8 | 2.21%
16384 bytes | 16 | 2.04%
64 bytes | 4 | 1.07%
64 bytes | 8 | 3.31%
64 bytes | 16 | 0.71%
Signed-off-by: Alex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1340845344-27557-8-git-send-email-alex.shi@intel.com
Tested-by: Ren, Yongjie <yongjie.ren@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This patch enabled the tlb flush range support in generic mmu layer.
Most of arch has self tlb flush range support, like ARM/IA64 etc.
X86 arch has no this support in hardware yet. But another instruction
'invlpg' can implement this function in some degree. So, enable this
feather in generic layer for x86 now. and maybe useful for other archs
in further.
Generic mmu_gather struct is protected by micro
HAVE_GENERIC_MMU_GATHER. Other archs that has flush range supported
own self mmu_gather struct. So, now this change is safe for them.
In future we may unify this struct and related functions on multiple
archs.
Thanks for Peter Zijlstra time and time reminder for multiple
architecture code safe!
Signed-off-by: Alex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1340845344-27557-7-git-send-email-alex.shi@intel.com
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
kernel will replace cr3 rewrite with invlpg when
tlb_flush_entries <= active_tlb_entries / 2^tlb_flushall_factor
if tlb_flushall_factor is -1, kernel won't do this replacement.
User can modify its value according to specific CPU/applications.
Thanks for Borislav providing the help message of
CONFIG_DEBUG_TLBFLUSH.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1340845344-27557-6-git-send-email-alex.shi@intel.com
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Testing show different CPU type(micro architectures and NUMA mode) has
different balance points between the TLB flush all and multiple invlpg.
And there also has cases the tlb flush change has no any help.
This patch give a interface to let x86 vendor developers have a chance
to set different shift for different CPU type.
like some machine in my hands, balance points is 16 entries on
Romely-EP; while it is at 8 entries on Bloomfield NHM-EP; and is 256 on
IVB mobile CPU. but on model 15 core2 Xeon using invlpg has nothing
help.
For untested machine, do a conservative optimization, same as NHM CPU.
Signed-off-by: Alex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1340845344-27557-5-git-send-email-alex.shi@intel.com
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
We don't need to flush large pages by PAGE_SIZE step, that just waste
time. and actually, large page don't need 'invlpg' optimizing according
to our micro benchmark. So, just flush whole TLB is enough for them.
The following result is tested on a 2CPU * 4cores * 2HT NHM EP machine,
with THP 'always' setting.
Multi-thread testing, '-t' paramter is thread number:
without this patch with this patch
./mprotect -t 1 14ns 13ns
./mprotect -t 2 13ns 13ns
./mprotect -t 4 12ns 11ns
./mprotect -t 8 14ns 10ns
./mprotect -t 16 28ns 28ns
./mprotect -t 32 54ns 52ns
./mprotect -t 128 200ns 200ns
Signed-off-by: Alex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1340845344-27557-4-git-send-email-alex.shi@intel.com
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
|