adding:
- New secondary interrupt controller binding to support the wake-up
- Use the RSB bus instead of I2C for the PMIC on the H6
- HDMI support for the BananaPi M2-Zero
- New board: Topwise A721
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRcEzekXsqa64kGDp7j7w1vZxhRxQUCYGchiwAKCRDj7w1vZxhR
xW1/APsFH9bywTduCrsuJui1LdD3mVbhkN75tKfBw7Ce8rmloAD/U5dth8PKA4h6
yPCdc2k0XaK7ItcD3eveN2uwiNq89gU=
=Ktsz
-----END PGP SIGNATURE-----
Merge tag 'sunxi-dt-for-5.13-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into arm/dt
Our usual bunch of patches to support the Allwinner SoCs, this time
adding:
- New secondary interrupt controller binding to support the wake-up
- Use the RSB bus instead of I2C for the PMIC on the H6
- HDMI support for the BananaPi M2-Zero
- New board: Topwise A721
* tag 'sunxi-dt-for-5.13-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
ARM: dts: sun8i: h3: beelink-x2: Add power button
arm64: dts: allwinner: h6: Use RSB for AXP805 PMIC connection
ARM: dts: sunxi: h2-plus-bananapi-m2-zero: Add HDMI out
ARM: dts: sun4i: Add support for Topwise A721 tablet
dt-bindings: arm: Add Topwise A721
arm64: dts: allwinner: Move wakeup-capable IRQs to r_intc
arm64: dts: allwinner: Use the new r_intc binding
ARM: dts: sunxi: Move wakeup-capable IRQs to r_intc
ARM: dts: sunxi: h3/h5: Add r_intc node
ARM: dts: sunxi: Use the new r_intc binding
Link: https://lore.kernel.org/r/8a3e3271-bebe-4d27-a9e7-7b7a6311a38d.lettre@localhost
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
- Video IN (VIN) and Camera (CSI-2) support for the R-Car M3-W+ SoC,
- LED support for the Falcon development board,
- Preparatory display pipeline support for the R-Car V3U SoC,
- Miscellaneous fixes and improvements.
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQQ9qaHoIs/1I4cXmEiKwlD9ZEnxcAUCYGbVawAKCRCKwlD9ZEnx
cEgrAP9eZ3MAYaiATSSPB4KvErvRYc9dqMgHWfHX7Su+LpyPZAD/ViFlg1drXMNV
HNo3oZUWTChPek4zbe7EUcjpLhWzawk=
=fkRt
-----END PGP SIGNATURE-----
Merge tag 'renesas-arm-dt-for-v5.13-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-devel into arm/dt
Renesas ARM DT updates for v5.13 (take two)
- Video IN (VIN) and Camera (CSI-2) support for the R-Car M3-W+ SoC,
- LED support for the Falcon development board,
- Preparatory display pipeline support for the R-Car V3U SoC,
- Miscellaneous fixes and improvements.
* tag 'renesas-arm-dt-for-v5.13-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-devel:
arm64: dts: renesas: ulcb: Add cpu-supply property to a57_0 node
arm64: dts: renesas: salvator-common: Add cpu-supply property to a57_0 node
arm64: dts: renesas: r8a77950: Drop operating points above 1.5 GHz
arm64: dts: renesas: r8a779a0: Fix PMU interrupt
arm64: dts: renesas: r8a779a0: Add VSPD support
arm64: dts: renesas: r8a779a0: Add FCPVD support
arm64: dts: renesas: falcon-cpu: Add GP LEDs
arm64: dts: renesas: r8a77961: Add VIN and CSI-2 device nodes
ARM: dts: koelsch: Configure pull-up for SOFT_SW GPIO keys
arm64: dts: renesas: falcon: Move AVB0 to main DTS
arm64: dts: renesas: falcon: Move watchdog config to CPU board DTS
arm64: dts: renesas: falcon: Move console config to CPU board DTS
Link: https://lore.kernel.org/r/cover.1617359678.git.geert+renesas@glider.be
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Although every Beelink GS1 seems to have external 32768 Hz oscillator,
it works only on one from four tested. There are more reports of RTC
issues elsewhere, like Armbian forum.
One Beelink GS1 owner read RTC osc status register on Android which
shipped with the box. Reported value indicated problems with external
oscillator.
In order to fix RTC and related issues (HDMI-CEC and suspend/resume with
Crust) on all boards, switch to internal oscillator.
Fixes: 32507b8681 ("arm64: dts: allwinner: h6: Move ext. oscillator to board DTs")
Signed-off-by: Jernej Skrabec <jernej.skrabec@siol.net>
Tested-by: Clément Péron <peron.clem@gmail.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://lore.kernel.org/r/20210330184218.279738-1-jernej.skrabec@siol.net
Some drivers (phy, crypto, net) folders don't have sunxi in it but
allwinner. Add that keyword to match on it too.
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Acked-by: Chen-Yu Tsai <wens@csie.org>
Acked-by: Jernej Skrabec <jernej.skrabec@siol.net>
Link: https://lore.kernel.org/r/20210331130830.64182-2-maxime@cerno.tech
We've been struggling to get an LF-hosted mailing list for a while, but
now that lists.linux.dev is there we opted in.
Let's add it to MAINTAINERS.
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Acked-by: Chen-Yu Tsai <wens@csie.org>
Acked-by: Jernej Skrabec <jernej.skrabec@siol.net>
Link: https://lore.kernel.org/r/20210331130830.64182-1-maxime@cerno.tech
Commit 941432d007 ("arm64: dts: allwinner: Drop non-removable from
SoPine/LTS SD card") enabled the card detect GPIO for the SOPine module,
along the way with the Pine64-LTS, which share the same base .dtsi.
However while both boards indeed have a working CD GPIO on PF6, the
polarity is different: the SOPine modules uses a "push-pull" socket,
which has an active-high switch, while the Pine64-LTS use the more
traditional push-push socket and the common active-low switch.
Fix the polarity in the sopine.dtsi, and overwrite it in the LTS
board .dts, to make the SD card work again on systems using SOPine
modules.
Fixes: 941432d007 ("arm64: dts: allwinner: Drop non-removable from SoPine/LTS SD card")
Reported-by: Ashley <contact@victorianfox.com>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://lore.kernel.org/r/20210316144219.5973-1-andre.przywara@arm.com
The macros for the clock and reset indices for the RSB hardware block
were replaced with raw numbers when the RSB controller node was added.
This was done to avoid cross-tree dependencies.
Now that both the clk and DT changes have been merged, we can switch
back to using the macros.
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Add unit address to the ETB and ETM nodes.
It also allow us to get rid of the warnings:
../arch/arm/boot/dts/sama5d2.dtsi:43.6-57.4: Warning
(unit_address_vs_reg): /etb: node has a reg or ranges property, but no
unit name
../arch/arm/boot/dts/sama5d2.dtsi:59.6-73.4: Warning
(unit_address_vs_reg): /etm: node has a reg or ranges property, but no
unit name
when we compile with W=1.
Signed-off-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20210402170139.140595-1-nicolas.ferre@microchip.com
Having a button code and not a key code causes issues with libinput.
udev won't set ID_INPUT_KEY. If it is forced, then it causes a bug
within libinput.
Signed-off-by: Ludovic Desroches <ludovic.desroches@microchip.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20210402130227.21478-1-nicolas.ferre@microchip.com
We have several modifiers for plain pointers (%p, %px and %pK) and now
also the no_hash_pointers boot parameter. The documentation should help
to choose which variant to use. Importantly, we should discourage %px
in favor of %p (with the new boot parameter when debugging), and stress
that %pK should be only used for procfs and similar files, not dmesg
buffer. This patch clarifies the documentation in that regard.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Petr Mladek <pmladek@suse.com>
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20210225164639.27212-1-vbabka@suse.cz
This is the capacity in percentage, relative to design capacity.
Specifically, it is present in Apple Magic Mouse 2.
In contrast, usage 00850064 is also the capacity in percentage, but is
relative to full capacity. It is not mapped here because I don't have
such device.
Signed-off-by: John Chen <johnchen902@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
It is observed that the Magic Mouse 2 would not enter multi-touch mode
unless the mouse is connected before loading the module. It seems to be
a quirk specific to Magic Mouse 2
Retrying after 500ms fixes the problem for me. The delay can't be
reduced much further --- 300ms didn't work for me. Retrying immediately
after receiving an event didn't work either.
Signed-off-by: John Chen <johnchen902@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
It is observed that, with 3 button emulation, when middle button is
clicked, either the left button or right button is clicked as well. It
is caused by hidinput "correctly" acting on the event, oblivious to the
3 button emulation.
As raw_event has taken care of everything, no further processing is
needed. However, the only way to stop at raw_event is to return an error
(negative) value. Therefore, the processing is stopped at event instead.
[jkosina@suse.cz: fix comment style]
Signed-off-by: John Chen <johnchen902@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Instead of playing games with using irq_create_identity_mapping()
and irq_domain_associate(), drop the use of the former and only
use the latter, together with the allocation of the irq_desc
as needed.
It doesn't make the code less awful, but at least the intent
is clearer.
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Marc Zyngier <maz@kernel.org>
The Thinkpad X1 Tablet Thin keyboard's HID interface for the media-keys
and other special functions, is quite similar to the Thinkpad 10 ultrabook
keyboard's mouse/media-keys HID interface.
The only difference is that it needs a bit different key mappings.
Add support for the mute-LED and the non-standard media-keys on this
keyboard by re-using the tp10_ultrabook_kbd code combined with a new
lenovo_input_mapping_x1_tab_kbd() function.
Co-authored-by: Alexander Kobel <a-kobel@a-kobel.de>
Tested-by: Alexander Kobel <a-kobel@a-kobel.de>
Signed-off-by: Alexander Kobel <a-kobel@a-kobel.de>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Instead of looking for a hdev with a type of HID_TYPE_USBMOUSE find
the interface for the mute/mic-mute/fn-lock LEDs by checking for the
output-report which is used to set them.
This is a preparation patch for adding support for the LEDs on the
X1 tablet thin keyboard which uses the same output-report, but has
a separate (third) USB interface for the touchpad/mouse functionality.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
The mute and mic-mute LEDs should be automatically turned on/off based
on the audio-card's mixer settings.
Add the standardized default-trigger names for this, so that the alsa
code can turn the LEDs on/off as appropriate (on supported audio cards).
This brings the mute/mic-mute LED support inline with the thinkpad_acpi
support for the same LEDs in keyboards directly connected to the
laptop's embedded-controller.
Reviewed-by: Marek Behún <kabel@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Mapping the mic-mute button to KEY_MICMUTE is technically correct but
KEY_MICMUTE translates to a scancode of 256 (248 + 8) under X,
which does not fit in 8 bits, so it does not work.
Because of this userspace is expecting KEY_F20 instead,
theoretically KEY_MICMUTE should work under Wayland but even
there it does not work, because the desktop-environment is
listening only for KEY_F20 and not for KEY_MICMUTE.
Fixes: bc04b37ea0 ("HID: lenovo: Add ThinkPad 10 Ultrabook Keyboard support")
Reviewed-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
The LEDs can only by turned on/off, so max_brightness should be set to 1.
Without this the max_brightness sysfs-attribute will report 255 which is
wrong.
Reviewed-by: Marek Behún <kabel@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
The led_classdev already contains a cached value of the last set
brightness, the brightness_get callback is only meant for LED drivers
which can read back the actual / current brightness from the hardware.
Since lenovo_led_brightness_get() just returns the last set value
it does not add any functionality, so we can just remove it.
Reviewed-by: Marek Behún <kabel@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
The HID lenovo probe function only attaches drvdata to one of the
USB interfaces, but lenovo_event() will get called for all USB interfaces
to which hid-lenovo is bound.
This allows a malicious device to fake being a device handled by
hid-lenovo, which generates events for which lenovo_event() has
special handling (and thus dereferences hid_get_drvdata()) on another
interface triggering a NULL pointer exception.
Add a check for hid_get_drvdata() returning NULL, avoiding this
possible NULL pointer exception.
Fixes: bc04b37ea0 ("HID: lenovo: Add ThinkPad 10 Ultrabook Keyboard support")
Reviewed-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Fix the following issues with lenovo_led_set_tp10ubkbd() error handling:
1. On success hid_hw_raw_request() returns the number of bytes sent.
So we should check for (ret != 3) rather then for (ret != 0).
2. Actually propagate errors to the caller.
3. Since the LEDs are part of an USB keyboard-dock the mute LEDs can go
away at any time. Don't log an error when ret == -ENODEV and set the
LED_HW_PLUGGABLE flag to avoid errors getting logged when the USB gets
disconnected.
Fixes: bc04b37ea0 ("HID: lenovo: Add ThinkPad 10 Ultrabook Keyboard support")
Reviewed-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
The lenovo_led_brightness_set function may sleep, so we should have the
the led_class_dev's brightness_set_blocking callback point to it, rather
then the regular brightness_set callback.
When toggled through sysfs this is not a problem, but the brightness_set
callback may be called from atomic context when using LED-triggers.
Fixes: bc04b37ea0 ("HID: lenovo: Add ThinkPad 10 Ultrabook Keyboard support")
Reviewed-by: Marek Behún <kabel@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
The led_classdev already contains a cached value of the last set
brightness, the brightness_get callback is only meant for LED drivers
which can read back the actual / current brightness from the hardware.
Since elan_mute_led_get_brigtness() just returns the last set value
it does not add any functionality, so we can just remove it.
Reviewed-by: Marek Behún <kabel@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
The mute LED should be automatically turned on/off based
on the audio-card's mixer settings.
Add the standardized default-trigger name for this, so that the alsa
code can turn the LED on/off as appropriate (on supported audio cards).
Reviewed-by: Marek Behún <kabel@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Being integrated into an USB keyboard-dock the mute LED can go away
at any time, leading to the following errors:
[ 918.667671] elan 0003:04F3:0755.0002: Failed to set mute led brightness: -19
[ 918.667737] leds elan:red:mute: Setting an LED's brightness failed (-19)
Fix this by making the following changes:
1. Don't log an error from elan_mute_led_set_brigtness() when
ret == -ENODEV
2. Set the LED_HW_PLUGGABLE flag on the mute LED led_classdev
While at it also make sure that elan_mute_led_set_brigtness() returns
an error (-EIO) when ret != 3 but it is not an error (>= 0).
Reviewed-by: Marek Behún <kabel@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Per SMMUv3 spec, there is no Size and Addr field in the PREFETCH_CONFIG
command and they're not used by the driver. Remove them.
We can add them back if we're going to use PREFETCH_ADDR in the future.
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20210407084448.1838-1-yuzenghui@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
There are a few calls of usb_driver_claim_interface() but all of those
miss the proper error checks, as reported by Coverity. This patch
adds those missing checks.
Along with it, replace the magic pointer with -1 with a constant
USB_AUDIO_IFACE_UNUSED for better readability.
Reported-by: coverity-bot <keescook+coverity-bot@chromium.org>
Addresses-Coverity-ID: 1475943 ("Error handling issues")
Addresses-Coverity-ID: 1475944 ("Error handling issues")
Addresses-Coverity-ID: 1475945 ("Error handling issues")
Fixes: b1ce7ba619 ("ALSA: usb-audio: claim autodetected PCM interfaces all at once")
Fixes: e5779998bf ("ALSA: usb-audio: refactor code")
Link: https://lore.kernel.org/r/202104051059.FB7F3016@keescook
Link: https://lore.kernel.org/r/20210406113534.30455-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When a present pasid entry is disassembled, all kinds of pasid related
caches need to be flushed. But when a pasid entry is not being used
(PRESENT bit not set), we don't need to do this. Check the PRESENT bit
in intel_pasid_tear_down_entry() and avoid flushing caches if it's not
set.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20210320025415.641201-6-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When the Intel IOMMU is operating in the scalable mode, some information
from the root and context table may be used to tag entries in the PASID
cache. Software should invalidate the PASID-cache when changing root or
context table entries.
Suggested-by: Ashok Raj <ashok.raj@intel.com>
Fixes: 7373a8cc38 ("iommu/vt-d: Setup context and enable RID2PASID support")
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20210320025415.641201-4-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When the first level page table is used for IOVA translation, it only
supports Read-Only and Read-Write permissions. The Write-Only permission
is not supported as the PRESENT bit (implying Read permission) should
always set. When using second level, we still give separate permissions
that allows WriteOnly which seems inconsistent and awkward. We want to
have consistent behavior. After moving to 1st level, we don't want things
to work sometimes, and break if we use 2nd level for the same mappings.
Hence remove this configuration.
Suggested-by: Ashok Raj <ashok.raj@intel.com>
Fixes: b802d070a5 ("iommu/vt-d: Use iova over first level")
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20210320025415.641201-3-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The Address field of the Page Request Descriptor only keeps bit [63:12]
of the offending address. Convert it to a full address before reporting
it to device drivers.
Fixes: eb8d93ea3c ("iommu/vt-d: Report page request faults for guest SVA")
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20210320025415.641201-2-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
ce_add_elem() uses different return values to signal a result from
adding an element to the collector. Commit in Fixes: broke the case
where the element being added is not found in the array. Correct that.
[ bp: Rewrite commit message, add kernel-doc comments. ]
Fixes: de0e0624d8 ("RAS/CEC: Check count_threshold unconditionally")
Signed-off-by: William Roche <william.roche@oracle.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/1617722939-29670-1-git-send-email-william.roche@oracle.com
This clock must be always enabled to allow access to any registers in
fsys1 CMU. Until proper solution based on runtime PM is applied
(similar to what was done for Exynos5433), mark that clock as critical
so it won't be disabled.
It was observed on Samsung Galaxy S6 device (based on Exynos7420), where
UFS module is probed before pmic used to power that device.
In this case defer probe was happening and that clock was disabled by
UFS driver, causing whole boot to hang on next CMU access.
Fixes: 753195a749 ("clk: samsung: exynos7: Correct CMU_FSYS1 clocks names")
Signed-off-by: Paweł Chmiel <pawel.mikolaj.chmiel@gmail.com>
Acked-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/linux-clk/20201024154346.9589-1-pawel.mikolaj.chmiel@gmail.com
[s.nawrocki: Added comment in the code]
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
A few functions that were intentended for the perf events support are
currently declared in arch/x86/events/amd/iommu.h, which mens they are
not in scope for the actual function definition. Also amdkfd has started
using a few of them using externs in a .c file. End that misery by
moving the prototypes to the proper header.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210402143312.372386-5-hch@lst.de
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Remove exports for functions that are only used in the AMD IOMMU driver
itself, or the also always builtin perf events support.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210402143312.372386-4-hch@lst.de
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Use an explicit set_pgtable_quirks method instead that just passes
the actual quirk bitmask instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Li Yang <leoyang.li@nxp.com>
Link: https://lore.kernel.org/r/20210401155256.298656-20-hch@lst.de
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Instead make the global iommu_dma_strict paramete in iommu.c canonical by
exporting helpers to get and set it and use those directly in the drivers.
This make sure that the iommu.strict parameter also works for the AMD and
Intel IOMMU drivers on x86. As those default to lazy flushing a new
IOMMU_CMD_LINE_STRICT is used to turn the value into a tristate to
represent the default if not overriden by an explicit parameter.
[ported on top of the other iommu_attr changes and added a few small
missing bits]
Signed-off-by: Robin Murphy <robin.murphy@arm.com>.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210401155256.298656-19-hch@lst.de
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Use an explicit enable_nesting method instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Li Yang <leoyang.li@nxp.com>
Link: https://lore.kernel.org/r/20210401155256.298656-17-hch@lst.de
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The geometry information can be trivially queried from the iommu_domain
struture.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Li Yang <leoyang.li@nxp.com>
Link: https://lore.kernel.org/r/20210401155256.298656-16-hch@lst.de
Signed-off-by: Joerg Roedel <jroedel@suse.de>
DOMAIN_ATTR_PAGING is never used.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Li Yang <leoyang.li@nxp.com>
Link: https://lore.kernel.org/r/20210401155256.298656-15-hch@lst.de
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The win_addr and win_size parameters are always set to 0 and 1 << 36
respectively, so just hard code them.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210401155256.298656-14-hch@lst.de
Signed-off-by: Joerg Roedel <jroedel@suse.de>