Pull "Dts related changes for omaps for v4.16 merge window" from Tony
Lindgren:
These changes are mostly improvments for various devices.
Note that these are based on my earlier fixes branch
omap-for-v4.15/fixes-dt to avoid a pointless merge conflict
between a fix and removal. The summary of changes is:
- Fix audio codec reset pin for am335x-pepper and n900, this
has been always broken and won't get fixed until the related
driver changes are also merged in for v4.16, so not urgent
- Fix tps65917 powerhold property for dra76-evm
- Changes to logicpd boards to remove MTD partition information
and to add support for omap35xx variants by setting up common
dts files for the logicpd boards
- Disable dra7 USB metastability workaround, this won't do
anything until the related driver changes are also merged
into v4.16, so not urgent
* tag 'omap-for-v4.16/dt-signed' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
ARM: dts: Move move WiFi bindings to logicpd-torpedo-37xx-devkit
ARM: dts: Add minimal support for LogicPD OMAP35xx Torpedo devkit
ARM: dts: Move most of logicpd-torpedo-37xx-devkit to logicpd-torpedo-baseboard
ARM: dts: Add minimal support for LogicPD OMAP35xx SOM-LV devkit
ARM: dts: Move most of logicpd-som-lv-37xx-devkit.dts to logicpd-som-lv-baseboard.dtsi
ARM: dts: dra7: Add missing hdmi audio DMA channel information
ARM: dts: dra7: Disable USB metastability workaround for USB2
ARM: dts: omap3: logicpd kits: Remove partition information
ARM: dts: DRA76-EVM: Set powerhold property for tps65917
ARM: dts: omap3-n900: Fix the audio CODEC's reset pin
ARM: dts: am335x-pepper: Fix the audio CODEC's reset pin
Pull "Clock related dts changes for omaps for v4.16 merge window" from
Tony Lindgren:
This branch contains a series of dts changes from Tero Kristo to
start using clkctrl clocks.
Note that this branch is based on a merge of omap-for-v4.16/soc-signed
and an immutable commit from Tero Kristo fe7020e64f ("clk: ti: omap4:
clkctrl data fixes for opt-clocks") that is also in clk-next.
* tag 'omap-for-v4.16/dt-clk-signed' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap: (57 commits)
ARM: dts: dm816x: add clkctrl nodes
ARM: dts: dm814x: add clkctrl nodes
ARM: dts: am43xx: add clkctrl nodes
ARM: dts: am33xx: add clkctrl nodes
ARM: dts: dra7: add clkctrl nodes
ARM: dts: omap5: add clkctrl nodes
ARM: dts: omap4: add clkctrl nodes
ARM: dts: dm816x: add bus functionality to base PRCM node
ARM: dts: am43xx: add bus functionality to base PRCM node
ARM: dts: am33xx: add bus functionality to base PRCM node
ARM: dts: dra7: add bus functionality to base PRCM nodes
ARM: dts: omap4: add bus functionality to base PRCM nodes
ARM: dts: omap5: add bus functionality to base PRCM nodes
ARM: dts: dm816x: add fck under timers1/2
ARM: dts: dm814x: add fck under timers1/2
ARM: dts: dra7: add fck under timer1
ARM: dts: am43xx: add fck under timers1/2
ARM: dts: am33xx: add fck under timers1/2
ARM: dts: omap4: add fck under timer1
ARM: dts: omap5: add fck under timer1
...
Pull "Amlogic 64-bit DT updates for v4.16" from Kevin Hilman
- meson-gx: add VPU power domain support
- odroid-c2: add HDMI and CEC nodes
- misc cleanups
* tag 'amlogic-dt64' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/khilman/linux-amlogic:
ARM64: dts: meson-gxm: fix q200 interrupt number
ARM64: dts: meson-gxm: add the PHY interrupt line on Khadas VIM2
ARM64: dts: meson: add comments with the GPIO for the PHY interrupts
ARM64: dts: amlogic: use generic bus node names
ARM64: dts: meson: drop "sana" clock from SAR ADC
ARM64: dts: odroid-c2: Add HDMI and CEC Nodes
ARM64: dts: meson-gx: grow reset controller memory zone
ARM64: dts: meson-gx: Add HDMI_5V regulator on selected boards
ARM64: dts: meson-gx: add VPU power domain
Pull "Amlogic 32-bit DT changes for v4.16" from Kevin Hilman:
- meson8: GPIO IRQ support
- switch to stable UART bindings w/correct clock
- add more L2 cache settings
- drop unused ADC clock
* tag 'amlogic-dt' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/khilman/linux-amlogic:
ARM: meson: enable MESON_IRQ_GPIO also for MACH_MESON8
ARM: dts: meson8: enable the GPIO interrupt controller
ARM: dts: meson8b: use stable UART bindings with correct gate clock
ARM: dts: meson8: use stable UART bindings with correct gate clock
ARM: dts: meson: drop "sana" clock from SAR ADC
ARM: dts: meson8: add more L2 cache settings
ARM: dts: meson8b: add more L2 cache settings
Pull "Renesas ARM64 Based SoC DT Updates for v4.16" from Simon Horman:
* Use r8a77970 (V3M) CPG core clock and SYSC power domain macros
These may be used in place of numeric constants now that they
are present in Linus's tree.
* Add r8a77970 (V3M) Starter Kit board support
This includes basic support to bring up the board with a serial
console and EtherAVB support
* Add IPMMU nodes and connections to on-chip devices
on r8a7795 (H3), r8a7796 (M3-W), r8a77970 (V3M) and r8a77995 (D3) SoCs
Simon Horman says "With these patches applied a white list enabled IPMMU
driver may be used to check silicon revision and then enable IPMMU in the
known working cases."
* Enable DMA for SCIF2 on r8a77995 (D2) SoC
* Increase the number of GPIO bank 1 ports to 29 on r8a7795 (H3) SoC
This adds support for the GP-1-28 port pin of the r8a7795 (H3) ES2.0 SoC
* Add support for CAN to r8a77995 (D3) SoC
Ulrich Hecht says "This is a by-the-datasheet implementation, with the
datasheet missing some bits, namely the pin map. I filled in the gaps...
by deducing the information from pin numbers already in the PFC driver,
so careful scrutiny is advised."
* Add support for SDHI to r8a77995 (D3) SoC
* Add SoC name to file header of r8a7795 (H3) and r8a7796 (M3-W)
Salvator-X and Salvator-XS board files
Geert Uytterhoeven says "With the proliferation of Salvator-X and
Salvator-XS boards carrying different R-Car Gen3 SoCs variants, several
DTS files ended up having the same file headers.
Add the SoC names to the file headers to avoid confusion."
* Add device note for ROHM BD9571MWV PMIC to
r8a7795 (H3) and r8a7796 (M3-W) Salvator-X and Salvator-XS boards.
Geert Uytterhoeven says "This was based on the example in the DT binding
documentation, but using IRQ0 instead of a GPIO interrupt, as that
matches the schematics, and because INTC-EX is a simpler block."
* Enable USB2.0 channel 0 on r8a77970 (V3M) ULCB Kingfisher board
Vladimir Barinov says "The dedicated USB0_PWEN pin is used to control
CN13 VBUS source from U43 power supply. MAX3355 can also provide VBUS,
hence it should be disabled via OTG_OFFVBUSn node coming from gpio
expander TCA9539. Set MAX3355 enabled using OTG_EXTLPn node to be able
to read OTG ID of CN13."
* Add support for r8a7795 (M3-W) Salvator-XS board
Geert Uytterhoeven says "This patch series adds support for the version
of the Salvator-XS development board equipped with an R-Car M3-W SiP.
The DT was based on work for the Salvator-X and -XS boards with M3-W
resp. H3 SiPs."
* Add watchdog timer support to r8a77970 (V3M) eagle board
Geert Uytterhoven says "This allows to use the watchdog timer to reset
the board, until PSCI is enhanced to include such functionality."
* Use Use R-Car SDHI Gen3 fallback on r8a7795 (H3) and r8a7796 (M3-W) SoCs
* Set driver type for MMC on r8a7795 (H3) and r8a7796 (M3-W) Salvator-X and
Salvator-XS boards.
Wolfram Sang says "These boards are known to have eMMC issues with the
default driver type. Specify a working one."
* tag 'renesas-arm64-dt-for-v4.16' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/horms/renesas: (54 commits)
arm64: dts: renesas: r8a77970: use SYSC power domain macros
arm64: dts: renesas: r8a77970: use CPG core clock macros
arm64: dts: renesas: v3msk: add EtherAVB support
arm64: dts: renesas: initial V3MSK board device tree
arm64: dts: renesas: r8a77995: Connect Ethernet-AVB to IPMMU-RT
arm64: dts: renesas: r8a77995: Add IPMMU device nodes
arm64: dts: renesas: r8a77970: Enable IPMMU-DS1, RT and MM
arm64: dts: renesas: r8a77970: Connect Ethernet-AVB to IPMMU-RT
arm64: dts: renesas: r8a77970: Tie SYS-DMAC to IPMMU-DS1
arm64: dts: renesas: r8a77970: Add IPMMU device nodes
arm64: dts: renesas: r8a77995: add DMA for SCIF2
arm64: dts: renesas: r8a77970: sort includes
arm64: dts: renesas: r8a7795: Increase the number of GPIO bank 1 ports to 29
arm64: dts: renesas: r8a77995: Add CAN FD support
arm64: dts: renesas: r8a77995: Add CAN support
arm64: dts: renesas: r8a77995: Add CAN external clock support
arm64: dts: renesas: r8a7795-salvator-xs: Add SoC name to file header
arm64: dts: renesas: r8a7796-salvator-x: Add SoC name to file header
arm64: dts: renesas: r8a7795-salvator-x: Add SoC name to file header
arm64: dts: renesas: r8a7795-es1-salvator-x: Add SoC name to file header
...
Pull "Renesas ARM Based SoC DT Updates for v4.16" from Simon Horman:
* Convert to named i2c-gpio bindings
Geert Uytterhoeven says "Commits 7d29f509d2 ("dt-bindings: i2c:
i2c-gpio: Add support for named gpios") and 05c7477885 ("i2c: gpio:
Add support for named gpios in DT") introduced named i2c-gpio DT
bindings, and deprecated the more error-prone unnamed variant.
This patch series switches all Renesas boards to the new bindings, and
adds the missing GPIO_OPEN_DRAIN I/O flags, which were implicitly
assumed before..."
... Note that after this series is applied, the i2c-gpio buses are no
longer detected when booting new DTBs on old (v4.14 and older) kernels,
which should not be an issue. Booting old DTBs on new kernels is not
affected."
* Update DTS for CMT DT binding rework
Geert Uytterhoeven says "This patch series updates the CMT device nodes
in the various Renesas DTS files sh_cmt clocksource driver for the recent
DT binding rework that was merged in v4.14-rc1 and v4.15-rc1..."
* Add SMP support to r8a7794 (R-Car E2) SoC
Sergei Shtylyov says "Add the device tree node for the Advanced Power
Management Unit (APMU). Use the "enable-method" prop to point out that
the APMU should be used for the SMP support."
* Correct primary compatible value for eeprom
on r7s72100 (RZ/A1H) genmai and r8a7791 (R-Car M2-W) koelsh boards
Geert Uytterhoeven says "The Renesas part numbers of the two-wire serial
interface EEPROMs do not follow the 24Cxx pattern, but the R1EX24xxx
pattern.
Hence change the primary compatible values to the appropriate variant of
"renesas,r1ex24xxx", like is already done on Gose.""
* Move cec_clock to root node on r8a7791 (R-Car M2-W) koelsh board
r8a7791 (R-Car M2-W) koelsh board
* Use R-Car SDHI and Ether Gen1 and 2 fallback compat strings
Use recently posted R-Car SDHI and Ether Gen 1 and 2 fallback
compat strings in the DT of Renesas ARM based SoCs.
* Add IIC cores to dtsi of r8a7745 (RZ/G1E) SoC
* Rework DT architecture for r8a7745 (RZ/G1E) iW-RainboW-G22D development
platform and add serial support.
Fabrizio Castro says "... define a new DT architecture for the
iW-RainboW-G22D SODIMM Development Platform to include the configuration
with the HDMI daughter board and to define the serial interfaces."
* Add USB function support to
r8a7745 (RZ/G1E) iW-RainboW-G22D development platform
* Add PCIEC and ttySC3 support to r8a7743 (RZ/G1M) iW-RainboW-G20M-Qseven SoM
* Add VIN support to r8a7743 (RZ/G1M) and r8a7745 (RZ/G1E) SoCs
* Add CAN and HDMI support to r8a7743 (RZ/G1M) iW-RainboW-G20D-Qseven and
r8a7745 (RZ/G1E) iW-RainboW-G22D development platforms
* tag 'renesas-dt-for-v4.16' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/horms/renesas: (61 commits)
ARM: dts: alt: Convert to named i2c-gpio bindings
ARM: dts: koelsch: Convert to named i2c-gpio bindings
ARM: dts: lager: Convert to named i2c-gpio bindings
ARM: dts: armadillo800eva: Convert to named i2c-gpio bindings
ARM: dts: sh73a0: Remove CMT renesas,channels-mask
ARM: dts: r8a7794: Remove CMT renesas,channels-mask
ARM: dts: r8a7793: Remove CMT renesas,channels-mask
ARM: dts: r8a7791: Remove CMT renesas,channels-mask
ARM: dts: r8a7790: Remove CMT renesas,channels-mask
ARM: dts: r8a7740: Remove CMT renesas,channels-mask
ARM: dts: r8a73a4: Remove CMT renesas,channels-mask
ARM: dts: r8a7794: Update CMT compat strings
ARM: dts: r8a7793: Update CMT compat strings
ARM: dts: r8a7791: Update CMT compat strings
ARM: dts: r8a7790: Update CMT compat strings
ARM: dts: r8a73a4: Update CMT compat string
ARM: dts: r8a7794: Add SMP support
ARM: dts: genmai: Correct primary compatible value for eeprom
ARM: dts: koelsch: Correct primary compatible value for eeprom
ARM: dts: r8a7745: add VIN dt support
...
Pull "Renesas ARM Based SoC DT Bindings Updates for v4.16" from Simon Horman:
* Document V3MSK board bindings
These are the bindings for the R-Car V3M Starter Kit
* Document M3-W-based Salvator-XS board bingigns
Geert Uytterhoeven says "The Renesas Salvator-XS (Salvator-X 2nd version)
development board can be equipped with either an R-Car H3 ES2.0 or M3-W
ES1.x SiP, which are pin-compatible."
* tag 'renesas-dt-bindings-for-v4.16' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/horms/renesas:
arm64: renesas: document V3MSK board bindings
ARM: shmobile: Document Renesas M3-W-based Salvator-XS board DT bindings
ARCH_S3C24XX is a very exceptional platform that some DT files in
arch/arm/boot/dts/, but does not select USE_OF.
All the other platforms with DT files correctly select USE_OF
directly or indirectly (Most of them are either ARCH_MULTIPLATFORM
or ARM_SINGLE_ARMV7M).
With ARCH_S3C24XX fixed, "ifeq ($(CONFIG_OF),y)" in DT Makefile
can be deleted.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
The at24 driver allows to register I2C EEPROM chips using different vendor
and devices, but the I2C subsystem does not take the vendor into account
when matching using the I2C table since it only has device entries.
But when matching using an OF table, both the vendor and device has to be
taken into account so the driver defines only a set of compatible strings
using the "atmel" vendor as a generic fallback for compatible I2C devices.
So add this generic fallback to the device node compatible string to make
the device to match the driver using the OF device ID table.
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
The at24 driver allows to register I2C EEPROM chips using different vendor
and devices, but the I2C subsystem does not take the vendor into account
when matching using the I2C table since it only has device entries.
But when matching using an OF table, both the vendor and device has to be
taken into account so the driver defines only a set of compatible strings
using the "atmel" vendor as a generic fallback for compatible I2C devices.
So add this generic fallback to the device node compatible string to make
the device to match the driver using the OF device ID table.
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
The at24 driver allows to register I2C EEPROM chips using different vendor
and devices, but the I2C subsystem does not take the vendor into account
when matching using the I2C table since it only has device entries.
But when matching using an OF table, both the vendor and device has to be
taken into account so the driver defines only a set of compatible strings
using the "atmel" vendor as a generic fallback for compatible I2C devices.
So add this generic fallback to the device node compatible string to make
the device to match the driver using the OF device ID table.
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
This reverts commit 04e35f4495.
SELinux runs with secureexec for all non-"noatsecure" domain transitions,
which means lots of processes end up hitting the stack hard-limit change
that was introduced in order to fix a race with prlimit(). That race fix
will need to be redesigned.
Reported-by: Laura Abbott <labbott@redhat.com>
Reported-by: Tomáš Trnka <trnka@scm.com>
Cc: stable@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull Page Table Isolation (PTI) v4.14 backporting base tree from Ingo Molnar:
"This tree contains the v4.14 PTI backport preparatory tree, which
consists of four merges of upstream trees and 7 cherry-picked commits,
which the upcoming PTI work depends on"
NOTE! The resulting tree is exactly the same as the original base tree
(ie the diff between this commit and its immediate first parent is
empty).
The only reason for this merge is literally to have a common point for
the actual PTI changes so that the commits can be shared in both the
4.15 and 4.14 trees.
* 'WIP.x86-pti.base-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm/kasan: Don't use vmemmap_populate() to initialize shadow
locking/barriers: Convert users of lockless_dereference() to READ_ONCE()
locking/barriers: Add implicit smp_read_barrier_depends() to READ_ONCE()
bpf: fix build issues on um due to mising bpf_perf_event.h
perf/x86: Enable free running PEBS for REGS_USER/INTR
x86: Make X86_BUG_FXSAVE_LEAK detectable in CPUID on AMD
x86/cpufeature: Add User-Mode Instruction Prevention definitions
Pull Page Table Isolation (PTI) preparatory tree from Ingo Molnar:
"This does a rename to free up linux/pti.h to be used by the upcoming
page table isolation feature"
* 'WIP.x86-pti.base.prep-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
drivers/misc/intel/pti: Rename the header file to free up the namespace
Pull timer fix from Thomas Gleixner:
"A single bugfix which prevents arbitrary sigev_notify values in
posix-timers"
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
posix-timer: Properly check sigevent->sigev_notify
Pull dmaengine fixes from Vinod Koul:
"This time consisting of fixes in a bunch of drivers and the dmatest
module:
- Fix for disable clk on error path in fsl-edma driver
- Disable clk fail fix in jz4740 driver
- Fix long pending bug in dmatest driver for dangling pointer
- Fix potential NULL pointer dereference in at_hdmac driver
- Error handling path in ioat driver"
* tag 'dmaengine-fix-4.15-rc4' of git://git.infradead.org/users/vkoul/slave-dma:
dmaengine: fsl-edma: disable clks on all error paths
dmaengine: jz4740: disable/unprepare clk if probe fails
dmaengine: dmatest: move callback wait queue to thread context
dmaengine: at_hdmac: fix potential NULL pointer dereference in atc_prep_dma_interleaved
dmaengine: ioat: Fix error handling path
With CONFIG_MTD=m and CONFIG_CRAMFS=y, we now get a link failure:
fs/cramfs/inode.o: In function `cramfs_mount': inode.c:(.text+0x220): undefined reference to `mount_mtd'
fs/cramfs/inode.o: In function `cramfs_mtd_fill_super':
inode.c:(.text+0x6d8): undefined reference to `mtd_point'
inode.c:(.text+0xae4): undefined reference to `mtd_unpoint'
This adds a more specific Kconfig dependency to avoid the broken
configuration.
Alternatively we could make CRAMFS itself depend on "MTD || !MTD" with a
similar result.
Fixes: 99c18ce580 ("cramfs: direct memory access support")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Pitre <nico@linaro.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs fixes from Al Viro:
"The alloc_super() one is a regression in this merge window, lazytime
thing is older..."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
VFS: Handle lazytime in do_mount()
alloc_super(): do ->s_umount initialization earlier
Pull ext4 fixes from Ted Ts'o:
"Fix a regression which caused us to fail to interpret symlinks in very
ancient ext3 file system images.
Also fix two xfstests failures, one of which could cause an OOPS, plus
an additional bug fix caught by fuzz testing"
* tag 'ext4_for_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
ext4: fix crash when a directory's i_size is too small
ext4: add missing error check in __ext4_new_inode()
ext4: fix fdatasync(2) after fallocate(2) operation
ext4: support fast symlinks from ext3 file systems
[ Note, this is a Git cherry-pick of the following commit:
a23f06f06d ("bpf: fix build issues on um due to mising bpf_perf_event.h")
... for easier x86 PTI code testing and back-porting. ]
Since c895f6f703 ("bpf: correct broken uapi for
BPF_PROG_TYPE_PERF_EVENT program type") um (uml) won't build
on i386 or x86_64:
[...]
CC init/main.o
In file included from ../include/linux/perf_event.h:18:0,
from ../include/linux/trace_events.h:10,
from ../include/trace/syscall.h:7,
from ../include/linux/syscalls.h:82,
from ../init/main.c:20:
../include/uapi/linux/bpf_perf_event.h:11:32: fatal error:
asm/bpf_perf_event.h: No such file or directory #include
<asm/bpf_perf_event.h>
[...]
Lets add missing bpf_perf_event.h also to um arch. This seems
to be the only one still missing.
Fixes: c895f6f703 ("bpf: correct broken uapi for BPF_PROG_TYPE_PERF_EVENT program type")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Suggested-by: Richard Weinberger <richard@sigma-star.at>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Richard Weinberger <richard@sigma-star.at>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[ Note, this is a Git cherry-pick of the following commit:
a47ba4d77e ("perf/x86: Enable free running PEBS for REGS_USER/INTR")
... for easier x86 PTI code testing and back-porting. ]
Currently free running PEBS is disabled when user or interrupt
registers are requested. Most of the registers are actually
available in the PEBS record and can be supported.
So we just need to check for the supported registers and then
allow it: it is all except for the segment register.
For user registers this only works when the counter is limited
to ring 3 only, so this also needs to be checked.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20170831214630.21892-1-andi@firstfloor.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[ Note, this is a Git cherry-pick of the following commit:
2b67799bdf25 ("x86: Make X86_BUG_FXSAVE_LEAK detectable in CPUID on AMD")
... for easier x86 PTI code testing and back-porting. ]
The latest AMD AMD64 Architecture Programmer's Manual
adds a CPUID feature XSaveErPtr (CPUID_Fn80000008_EBX[2]).
If this feature is set, the FXSAVE, XSAVE, FXSAVEOPT, XSAVEC, XSAVES
/ FXRSTOR, XRSTOR, XRSTORS always save/restore error pointers,
thus making the X86_BUG_FXSAVE_LEAK workaround obsolete on such CPUs.
Signed-Off-By: Rudolf Marek <r.marek@assembler.cz>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Borislav Petkov <bp@suse.de>
Tested-by: Borislav Petkov <bp@suse.de>
Cc: Andy Lutomirski <luto@amacapital.net>
Link: https://lkml.kernel.org/r/bdcebe90-62c5-1f05-083c-eba7f08b2540@assembler.cz
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull rdma fixes from Jason Gunthorpe:
"More fixes from testing done on the rc kernel, including more SELinux
testing. Looking forward, lockdep found regression today in ipoib
which is still being fixed.
Summary:
- Fix for SELinux on the umad SMI path. Some old hardware does not
fill the PKey properly exposing another bug in the newer SELinux
code.
- Check the input port as we can exceed array bounds from this user
supplied value
- Users are unable to use the hash field support as they want due to
incorrect checks on the field restrictions, correct that so the
feature works as intended
- User triggerable oops in the NETLINK_RDMA handler
- cxgb4 driver fix for a bad interaction with CQ flushing in iser
caused by patches in this merge window, and bad CQ flushing during
normal close.
- Unbalanced memalloc_noio in ipoib in an error path"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
IB/ipoib: Restore MM behavior in case of tx_ring allocation failure
iw_cxgb4: only insert drain cqes if wq is flushed
iw_cxgb4: only clear the ARMED bit if a notification is needed
RDMA/netlink: Fix general protection fault
IB/mlx4: Fix RSS hash fields restrictions
IB/core: Don't enforce PKey security on SMI MADs
IB/core: Bound check alternate path port number
Pull i2c fixes from Wolfram Sang:
"Two bugfixes for the AT24 I2C eeprom driver and some minor corrections
for I2C bus drivers"
* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: piix4: Fix port number check on release
i2c: stm32: Fix copyrights
i2c-cht-wc: constify platform_device_id
eeprom: at24: change nvmem stride to 1
eeprom: at24: fix I2C device selection for runtime PM
Pull NFS client fixes from Anna Schumaker:
"This has two stable bugfixes, one to fix a BUG_ON() when
nfs_commit_inode() is called with no outstanding commit requests and
another to fix a race in the SUNRPC receive codepath.
Additionally, there are also fixes for an NFS client deadlock and an
xprtrdma performance regression.
Summary:
Stable bugfixes:
- NFS: Avoid a BUG_ON() in nfs_commit_inode() by not waiting for a
commit in the case that there were no commit requests.
- SUNRPC: Fix a race in the receive code path
Other fixes:
- NFS: Fix a deadlock in nfs client initialization
- xprtrdma: Fix a performance regression for small IOs"
* tag 'nfs-for-4.15-3' of git://git.linux-nfs.org/projects/anna/linux-nfs:
SUNRPC: Fix a race in the receive code path
nfs: don't wait on commit in nfs_commit_inode() if there were no commit requests
xprtrdma: Spread reply processing over more CPUs
nfs: fix a deadlock in nfs client initialization
This reverts commits 5c9d2d5c26, c7da82b894, and e7fe7b5cae.
We'll probably need to revisit this, but basically we should not
complicate the get_user_pages_fast() case, and checking the actual page
table protection key bits will require more care anyway, since the
protection keys depend on the exact state of the VM in question.
Particularly when doing a "remote" page lookup (ie in somebody elses VM,
not your own), you need to be much more careful than this was. Dave
Hansen says:
"So, the underlying bug here is that we now a get_user_pages_remote()
and then go ahead and do the p*_access_permitted() checks against the
current PKRU. This was introduced recently with the addition of the
new p??_access_permitted() calls.
We have checks in the VMA path for the "remote" gups and we avoid
consulting PKRU for them. This got missed in the pkeys selftests
because I did a ptrace read, but not a *write*. I also didn't
explicitly test it against something where a COW needed to be done"
It's also not entirely clear that it makes sense to check the protection
key bits at this level at all. But one possible eventual solution is to
make the get_user_pages_fast() case just abort if it sees protection key
bits set, which makes us fall back to the regular get_user_pages() case,
which then has a vma and can do the check there if we want to.
We'll see.
Somewhat related to this all: what we _do_ want to do some day is to
check the PAGE_USER bit - it should obviously always be set for user
pages, but it would be a good check to have back. Because we have no
generic way to test for it, we lost it as part of moving over from the
architecture-specific x86 GUP implementation to the generic one in
commit e585513b76 ("x86/mm/gup: Switch GUP to the generic
get_user_page_fast() implementation").
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: "Jérôme Glisse" <jglisse@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull networking fixes from David Miller:
1) Clamp timeouts to INT_MAX in conntrack, from Jay Elliot.
2) Fix broken UAPI for BPF_PROG_TYPE_PERF_EVENT, from Hendrik
Brueckner.
3) Fix locking in ieee80211_sta_tear_down_BA_sessions, from Johannes
Berg.
4) Add missing barriers to ptr_ring, from Michael S. Tsirkin.
5) Don't advertise gigabit in sh_eth when not available, from Thomas
Petazzoni.
6) Check network namespace when delivering to netlink taps, from Kevin
Cernekee.
7) Kill a race in raw_sendmsg(), from Mohamed Ghannam.
8) Use correct address in TCP md5 lookups when replying to an incoming
segment, from Christoph Paasch.
9) Add schedule points to BPF map alloc/free, from Eric Dumazet.
10) Don't allow silly mtu values to be used in ipv4/ipv6 multicast, also
from Eric Dumazet.
11) Fix SKB leak in tipc, from Jon Maloy.
12) Disable MAC learning on OVS ports of mlxsw, from Yuval Mintz.
13) SKB leak fix in skB_complete_tx_timestamp(), from Willem de Bruijn.
14) Add some new qmi_wwan device IDs, from Daniele Palmas.
15) Fix static key imbalance in ingress qdisc, from Jiri Pirko.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (76 commits)
net: qcom/emac: Reduce timeout for mdio read/write
net: sched: fix static key imbalance in case of ingress/clsact_init error
net: sched: fix clsact init error path
ip_gre: fix wrong return value of erspan_rcv
net: usb: qmi_wwan: add Telit ME910 PID 0x1101 support
pkt_sched: Remove TC_RED_OFFLOADED from uapi
net: sched: Move to new offload indication in RED
net: sched: Add TCA_HW_OFFLOAD
net: aquantia: Increment driver version
net: aquantia: Fix typo in ethtool statistics names
net: aquantia: Update hw counters on hw init
net: aquantia: Improve link state and statistics check interval callback
net: aquantia: Fill in multicast counter in ndev stats from hardware
net: aquantia: Fill ndev stat couters from hardware
net: aquantia: Extend stat counters to 64bit values
net: aquantia: Fix hardware DMA stream overload on large MRRS
net: aquantia: Fix actual speed capabilities reporting
sock: free skb in skb_complete_tx_timestamp on error
s390/qeth: update takeover IPs after configuration change
s390/qeth: lock IP table while applying takeover changes
...
Pull USB fixes from Greg KH:
"Here are some USB fixes for 4.15-rc4.
There is the usual handful gadget/dwc2/dwc3 fixes as always, for
reported issues. But the most important things in here is the core fix
from Alan Stern to resolve a nasty security bug (my first attempt is
reverted, Alan's was much cleaner), as well as a number of usbip fixes
from Shuah Khan to resolve those reported security issues.
All of these have been in linux-next with no reported issues"
* tag 'usb-4.15-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
USB: core: prevent malicious bNumInterfaces overflow
Revert "USB: core: only clean up what we allocated"
USB: core: only clean up what we allocated
Revert "usb: gadget: allow to enable legacy drivers without USB_ETH"
usb: gadget: webcam: fix V4L2 Kconfig dependency
usb: dwc2: Fix TxFIFOn sizes and total TxFIFO size issues
usb: dwc3: gadget: Fix PCM1 for ISOC EP with ep->mult less than 3
usb: dwc3: of-simple: set dev_pm_ops
usb: dwc3: of-simple: fix missing clk_disable_unprepare
usb: dwc3: gadget: Wait longer for controller to end command processing
usb: xhci: fix TDS for MTK xHCI1.1
xhci: Don't add a virt_dev to the devs array before it's fully allocated
usbip: fix stub_send_ret_submit() vulnerability to null transfer_buffer
usbip: prevent vhci_hcd driver from leaking a socket pointer address
usbip: fix stub_rx: harden CMD_SUBMIT path to handle malicious input
usbip: fix stub_rx: get_pipe() to validate endpoint number
tools/usbip: fixes potential (minor) "buffer overflow" (detected on recent gcc with -Werror)
USB: uas and storage: Add US_FL_BROKEN_FUA for another JMicron JMS567 ID
usb: musb: da8xx: fix babble condition handling
Pull staging fixes from Greg KH:
"Here are some small staging driver fixes for 4.15-rc4.
One patch for the ccree driver to prevent an unitialized value from
being returned to a caller, and the other fixes a logic error in the
pi433 driver"
* tag 'staging-4.15-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
staging: pi433: Fixes issue with bit shift in rf69_get_modulation
staging: ccree: Uninitialized return in ssi_ahash_import()
Pull virtio regression fixes from Michael Tsirkin:
"Fixes two issues in the latest kernel"
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
virtio_mmio: fix devm cleanup
ptr_ring: fix up after recent ptr_ring changes
Pull device mapper fixes from Mike Snitzer:
- fix a particularly nasty DM core bug in a 4.15 refcount_t conversion.
- fix various targets to dm_register_target after module __init
resources created; otherwise racing lvm2 commands could result in a
NULL pointer during initialization of associated DM kernel module.
- fix regression in bio-based DM multipath queue_if_no_path handling.
- fix DM bufio's shrinker to reclaim more than one buffer per scan.
* tag 'for-4.15/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm bufio: fix shrinker scans when (nr_to_scan < retain_target)
dm mpath: fix bio-based multipath queue_if_no_path handling
dm: fix various targets to dm_register_target after module __init resources created
dm table: fix regression from improper dm_dev_internal.count refcount_t conversion
Pull SCSI fixes from James Bottomley:
"The most important one is the bfa fix because it's easy to oops the
kernel with this driver (this includes the commit that corrects the
compiler warning in the original), a regression in the new timespec
conversion in aacraid and a regression in the Fibre Channel ELS
handling patch.
The other three are a theoretical problem with termination in the
vendor/host matching code and a use after free in lpfc.
The additional patches are a fix for an I/O hang in the mq code under
certain circumstances and a rare oops in some debugging code"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: core: Fix a scsi_show_rq() NULL pointer dereference
scsi: MAINTAINERS: change FCoE list to linux-scsi
scsi: libsas: fix length error in sas_smp_handler()
scsi: bfa: fix type conversion warning
scsi: core: run queue if SCSI device queue isn't ready and queue is idle
scsi: scsi_devinfo: cleanly zero-pad devinfo strings
scsi: scsi_devinfo: handle non-terminated strings
scsi: bfa: fix access to bfad_im_port_s
scsi: aacraid: address UBSAN warning regression
scsi: libfc: fix ELS request handling
scsi: lpfc: Use after free in lpfc_rq_buf_free()
Pull MMC fixes from Ulf Hansson:
"A couple of MMC fixes:
- fix use of uninitialized drv_typ variable
- apply NO_CMD23 quirk to some specific SD cards to make them work"
* tag 'mmc-v4.15-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:
mmc: core: apply NO_CMD23 quirk to some specific cards
mmc: core: properly init drv_type
Pull ceph fix from Ilya Dryomov:
"CephFS inode trimming fix from Zheng, marked for stable"
* tag 'ceph-for-4.15-rc4' of git://github.com/ceph/ceph-client:
ceph: drop negative child dentries before try pruning inode's alias
Pull overlayfs fixes from Miklos Szeredi:
- fix incomplete syncing of filesystem
- fix regression in readdir on ovl over 9p
- only follow redirects when needed
- misc fixes and cleanups
* 'overlayfs-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
ovl: fix overlay: warning prefix
ovl: Use PTR_ERR_OR_ZERO()
ovl: Sync upper dirty data when syncing overlayfs
ovl: update ctx->pos on impure dir iteration
ovl: Pass ovl_get_nlink() parameters in right order
ovl: don't follow redirects if redirect_dir=off
Currently mdio read/write takes around ~115us as the timeout
between status check is set to 100us.
By reducing the timeout to 1us mdio read/write takes ~15us to
complete. This improves the link up event response.
Signed-off-by: Hemanth Puranik <hpuranik@codeaurora.org>
Acked-by: Timur Tabi <timur@codeaurora.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull arm64 fixes from Will Deacon:
"There are some significant fixes in here for FP state corruption,
hardware access/dirty PTE corruption and an erratum workaround for the
Falkor CPU.
I'm hoping that things finally settle down now, but never say never...
Summary:
- Fix FPSIMD context switch regression introduced in -rc2
- Fix ABI break with SVE CPUID register reporting
- Fix use of uninitialised variable
- Fixes to hardware access/dirty management and sanity checking
- CPU erratum workaround for Falkor CPUs
- Fix reporting of writeable+executable mappings
- Fix signal reporting for RAS errors"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: fpsimd: Fix copying of FP state from signal frame into task struct
arm64/sve: Report SVE to userspace via CPUID only if supported
arm64: fix CONFIG_DEBUG_WX address reporting
arm64: fault: avoid send SIGBUS two times
arm64: hw_breakpoint: Use linux/uaccess.h instead of asm/uaccess.h
arm64: Add software workaround for Falkor erratum 1041
arm64: Define cputype macros for Falkor CPU
arm64: mm: Fix false positives in set_pte_at access/dirty race detection
arm64: mm: Fix pte_mkclean, pte_mkdirty semantics
arm64: Initialise high_memory global variable earlier
