PSI accounts stalls for each cgroup separately and aggregates it
at each level of the hierarchy. This may cause non-negligible overhead
for some workloads when under deep level of the hierarchy.
commit 3958e2d0c34e ("cgroup: make per-cgroup pressure stall tracking configurable")
make PSI to skip per-cgroup stall accounting, only account system-wide
to avoid this each level overhead.
But for our use case, we also want leaf cgroup PSI stats accounted for
userspace adjustment on that cgroup, apart from only system-wide adjustment.
So this patch introduce a per-cgroup PSI accounting disable/re-enable
interface "cgroup.pressure", which is a read-write single value file that
allowed values are "0" and "1", the defaults is "1" so per-cgroup
PSI stats is enabled by default.
Implementation details:
It should be relatively straight-forward to disable and re-enable
state aggregation, time tracking, averaging on a per-cgroup level,
if we can live with losing history from while it was disabled.
I.e. the avgs will restart from 0, total= will have gaps.
But it's hard or complex to stop/restart groupc->tasks[] updates,
which is not implemented in this patch. So we always update
groupc->tasks[] and PSI_ONCPU bit in psi_group_change() even when
the cgroup PSI stats is disabled.
Suggested-by: Johannes Weiner <hannes@cmpxchg.org> Suggested-by: Tejun Heo <tj@kernel.org> Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20220907090332.2078-1-zhouchengming@bytedance.com
Chengming Zhou [Thu, 25 Aug 2022 16:41:10 +0000 (00:41 +0800)]
sched/psi: Cache parent psi_group to speed up group iteration
We use iterate_groups() to iterate each level psi_group to update
PSI stats, which is a very hot path.
In current code, iterate_groups() have to use multiple branches and
cgroup_parent() to get parent psi_group for each level, which is not
very efficient.
This patch cache parent psi_group in struct psi_group, only need to get
psi_group of task itself first, then just use group->parent to iterate.
Chengming Zhou [Thu, 25 Aug 2022 16:41:08 +0000 (00:41 +0800)]
sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure
Now PSI already tracked workload pressure stall information for
CPU, memory and IO. Apart from these, IRQ/SOFTIRQ could have
obvious impact on some workload productivity, such as web service
workload.
When CONFIG_IRQ_TIME_ACCOUNTING, we can get IRQ/SOFTIRQ delta time
from update_rq_clock_task(), in which we can record that delta
to CPU curr task's cgroups as PSI_IRQ_FULL status.
Note we don't use PSI_IRQ_SOME since IRQ/SOFTIRQ always happen in
the current task on the CPU, make nothing productive could run
even if it were runnable, so we only use PSI_IRQ_FULL.
Johannes Weiner [Thu, 25 Aug 2022 16:41:07 +0000 (00:41 +0800)]
sched/psi: Remove NR_ONCPU task accounting
We put all fields updated by the scheduler in the first cacheline of
struct psi_group_cpu for performance.
Since we want add another PSI_IRQ_FULL to track IRQ/SOFTIRQ pressure,
we need to reclaim space first. This patch remove NR_ONCPU task accounting
in struct psi_group_cpu, use one bit in state_mask to track instead.
Chengming Zhou [Thu, 25 Aug 2022 16:41:06 +0000 (00:41 +0800)]
sched/psi: Optimize task switch inside shared cgroups again
Way back when PSI_MEM_FULL was accounted from the timer tick, task
switching could simply iterate next and prev to the common ancestor to
update TSK_ONCPU and be done.
Then memstall ticks were replaced with checking curr->in_memstall
directly in psi_group_change(). That meant that now if the task switch
was between a memstall and a !memstall task, we had to iterate through
the common ancestors at least ONCE to fix up their state_masks.
We added the identical_state filter to make sure the common ancestor
elimination was skipped in that case. It seems that was always a
little too eager, because it caused us to walk the common ancestors
*twice* instead of the required once: the iteration for next could
have stopped at the common ancestor; prev could have updated TSK_ONCPU
up to the common ancestor, then finish to the root without changing
any flags, just to get the new curr->in_memstall into the state_masks.
This patch recognizes this and makes it so that we walk to the root
exactly once if state_mask needs updating, which is simply catching up
on a missed optimization that could have been done in commit 7fae6c8171d2
("psi: Use ONCPU state tracking machinery to detect reclaim") directly.
Apart from this, it's also necessary for the next patch "sched/psi: remove
NR_ONCPU task accounting". Suppose we walk the common ancestors twice:
We previously used tasks[NR_ONCPU] to record TSK_ONCPU, tasks[NR_ONCPU]++
in (1) then tasks[NR_ONCPU]-- in (2), so tasks[NR_ONCPU] still be correct.
The next patch change to use one bit in state mask to record TSK_ONCPU,
PSI_ONCPU bit will be set in (1), but then be cleared in (2), which cause
the psi_group_cpu has task running on CPU but without PSI_ONCPU bit set!
With this patch, we will never walk the common ancestors twice, so won't
have above problem.
Chengming Zhou [Thu, 25 Aug 2022 16:41:05 +0000 (00:41 +0800)]
sched/psi: Move private helpers to sched/stats.h
This patch move psi_task_change/psi_task_switch declarations out of
PSI public header, since they are only needed for implementing the
PSI stats tracking in sched/stats.h
psi_task_switch is obvious, psi_task_change can't be public helper
since it doesn't check psi_disabled static key. And there is no
any user now, so put it in sched/stats.h too.
Chengming Zhou [Thu, 25 Aug 2022 16:41:03 +0000 (00:41 +0800)]
sched/psi: Don't create cgroup PSI files when psi_disabled
commit 3958e2d0c34e ("cgroup: make per-cgroup pressure stall tracking configurable")
make PSI can be configured to skip per-cgroup stall accounting. And
doesn't expose PSI files in cgroup hierarchy.
Chengming Zhou [Thu, 25 Aug 2022 16:41:02 +0000 (00:41 +0800)]
sched/psi: Fix periodic aggregation shut off
We don't want to wake periodic aggregation work back up if the
task change is the aggregation worker itself going to sleep, or
we'll ping-pong forever.
Previously, we would use psi_task_change() in psi_dequeue() when
task going to sleep, so this check was put in psi_task_change().
But commit 4117cebf1a9f ("psi: Optimize task switch inside shared cgroups")
defer task sleep handling to psi_task_switch(), won't go through
psi_task_change() anymore.
So this patch move this check to psi_task_switch().
Merge tag 'powerpc-6.0-4' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
- Fix handling of PCI domains in /proc on 32-bit systems using the
recently added support for numbering buses from zero for each domain.
- A fix and a revert for some changes to use READ/WRITE_ONCE() which
caused problems with KASAN enabled due to sanitisation calls being
introduced in low-level paths that can't cope with it.
- Fix build errors on 32-bit caused by the syscall table being
misaligned sometimes.
- Two fixes to get IBM Cell native machines booting again, which had
bit-rotted while my QS22 was temporarily out of action.
- Fix the papr_scm driver to not assume the order of events returned by
the hypervisor is stable, and a related compile fix.
Thanks to Aneesh Kumar K.V, Christophe Leroy, Jordan Niethe, Kajol Jain,
Masahiro Yamada, Nathan Chancellor, Pali Rohár, Vaibhav Jain, and Zhouyi
Zhou.
* tag 'powerpc-6.0-4' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/papr_scm: Ensure rc is always initialized in papr_scm_pmu_register()
Revert "powerpc/irq: Don't open code irq_soft_mask helpers"
powerpc: Fix hard_irq_disable() with sanitizer
powerpc/rtas: Fix RTAS MSR[HV] handling for Cell
Revert "powerpc: Remove unused FW_FEATURE_NATIVE references"
powerpc: align syscall table for ppc32
powerpc/pci: Enable PCI domains in /proc when PCI bus numbers are not unique
powerpc/papr_scm: Fix nvdimm event mappings
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull kvm fixes from Paolo Bonzini:
"s390:
- PCI interpretation compile fixes
RISC-V:
- fix unused variable warnings in vcpu_timer.c
- move extern sbi_ext declarations to a header
x86:
- check validity of argument to KVM_SET_MP_STATE
- use guest's global_ctrl to completely disable guest PEBS
- fix a memory leak on memory allocation failure
- mask off unsupported and unknown bits of IA32_ARCH_CAPABILITIES
- fix build failure with Clang integrated assembler
- fix MSR interception
- always flush TLBs when enabling dirty logging"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: x86: check validity of argument to KVM_SET_MP_STATE
perf/x86/core: Completely disable guest PEBS via guest's global_ctrl
KVM: x86: fix memoryleak in kvm_arch_vcpu_create()
KVM: x86: Mask off unsupported and unknown bits of IA32_ARCH_CAPABILITIES
KVM: s390: pci: Hook to access KVM lowlevel from VFIO
riscv: kvm: move extern sbi_ext declarations to a header
riscv: kvm: vcpu_timer: fix unused variable warnings
KVM: selftests: Fix ambiguous mov in KVM_ASM_SAFE()
KVM: selftests: Fix KVM_EXCEPTION_MAGIC build with Clang
KVM: VMX: Heed the 'msr' argument in msr_write_intercepted()
kvm: x86: mmu: Always flush TLBs when enabling dirty logging
kvm: x86: mmu: Drop the need_remote_flush() function
Nick Desaulniers [Thu, 1 Sep 2022 17:59:13 +0000 (10:59 -0700)]
Makefile.extrawarn: re-enable -Wformat for clang; take 2
-Wformat was recently re-enabled for builds with clang, then quickly
re-disabled, due to concerns stemming from the frequency of default
argument promotion related warning instances.
commit 258fafcd0683 ("Makefile.extrawarn: re-enable -Wformat for clang")
commit 21f9c8a13bb2 ("Revert "Makefile.extrawarn: re-enable -Wformat for clang"")
ISO WG14 has ratified N2562 to address default argument promotion
explicitly for printf, as part of the upcoming ISO C2X standard.
The behavior of clang was changed in clang-16 to not warn for the cited
cases in all language modes.
Add a version check, so that users of clang-16 now get the full effect
of -Wformat. For older clang versions, re-enable flags under the
-Wformat group that way users still get some useful checks related to
format strings, without noisy default argument promotion warnings. I
intentionally omitted -Wformat-y2k and -Wformat-security from being
re-enabled, which are also part of -Wformat in clang-16.
Merge tag 'gpio-fixes-for-v6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux
Pull gpio fixes from Bartosz Golaszewski:
"A a set of fixes from the GPIO subsystem.
Most are small driver fixes except the realtek-otto driver patch which
is pretty big but addresses a significant flaw that can cause the CPU
to stay infinitely busy on uncleared ISR on some platforms.
Summary:
- MAINTAINERS update
- fix resource leaks in gpio-mockup and gpio-pxa
- add missing locking in gpio-pca953x
- use 32-bit I/O in gpio-realtek-otto
- make irq_chip structures immutable in four more drivers"
* tag 'gpio-fixes-for-v6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
gpio: ws16c48: Make irq_chip immutable
gpio: 104-idio-16: Make irq_chip immutable
gpio: 104-idi-48: Make irq_chip immutable
gpio: 104-dio-48e: Make irq_chip immutable
gpio: realtek-otto: switch to 32-bit I/O
gpio: pca953x: Add mutex_lock for regcache sync in PM
gpio: mockup: remove gpio debugfs when remove device
gpio: pxa: use devres for the clock struct
MAINTAINERS: rectify entry for XILINX GPIO DRIVER
Merge tag 'for-linus-6.0-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip
Pull xen fixes from Juergen Gross:
- a minor fix for the Xen grant driver
- a small series fixing a recently introduced problem in the Xen
blkfront/blkback drivers with negotiation of feature usage
* tag 'for-linus-6.0-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
xen/grants: prevent integer overflow in gnttab_dma_alloc_pages()
xen-blkfront: Cache feature_persistent value before advertisement
xen-blkfront: Advertise feature-persistent as user requested
xen-blkback: Advertise feature-persistent as user requested
Merge tag 'loongarch-fixes-6.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson
Pull LoongArch fixes from Huacai Chen:
"Fix several build errors or warnings, cleanup some code, and adjust
arch_do_signal_or_restart() to adapt generic entry"
* tag 'loongarch-fixes-6.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson:
LoongArch: mm: Remove the unneeded result variable
LoongArch: Fix arch_remove_memory() undefined build error
LoongArch: Fix section mismatch due to acpi_os_ioremap()
LoongArch: Improve dump_tlb() output messages
LoongArch: Adjust arch_do_signal_or_restart() to adapt generic entry
LoongArch: Avoid orphan input sections
Merge tag 'input-for-v6.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Pull input fixes from Dmitry Torokhov:
- GT1158 ID added to Goodix touchscreen driver
- Boeder Force Feedback Wheel USB added to iforce joystick driver
- fixup for iforce driver to avoid hangups
- fix autoloading of rk805-pwrkey driver.
* tag 'input-for-v6.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: iforce - add support for Boeder Force Feedback Wheel
Input: iforce - wake up after clearing IFORCE_XMIT_RUNNING flag
Input: goodix - add compatible string for GT1158
MAINTAINERS: add include/dt-bindings/input to INPUT DRIVERS
Input: rk805-pwrkey - fix module autoloading
Input: goodix - add support for GT1158
dt-bindings: input: touchscreen: add compatible string for Goodix GT1158
Merge tag 'tty-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Pull tty/serial driver fixes from Greg KH:
"Here are some small tty/serial/vt driver fixes for 6.0-rc4 that
resolve a number of reported issues:
- n_gsm fixups for previous changes that caused problems
- much-reported serdev crash fix that showed up in 6.0-rc1
- vt font selection bugfix
- kerneldoc build warning fixes
- other tiny serial core fixes
All of these have been in linux-next for a while with no reported
problems"
* tag 'tty-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
tty: n_gsm: avoid call of sleeping functions from atomic context
tty: n_gsm: replace kicktimer with delayed_work
tty: n_gsm: initialize more members at gsm_alloc_mux()
tty: n_gsm: add sanity check for gsm->receive in gsm_receive_buf()
tty: serial: atmel: Preserve previous USART mode if RS485 disabled
tty: serial: lpuart: disable flow control while waiting for the transmit engine to complete
tty: Fix lookahead_buf crash with serdev
serial: fsl_lpuart: RS485 RTS polariy is inverse
vt: Clear selection before changing the font
serial: document start_rx member at struct uart_ops
Merge tag 'staging-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
Pull staging driver fixes from Greg KH:
"Here are three small staging driver fixes for 6.0-rc4 that resolve
some reported problems and add some a device id:
- new device id for r8188eu driver
- use-after-free bugfixes for the rtl8712 driver
- fix up firmware dependency problem for the r8188eu driver
All of these have been in linux-next for a while with no reported
problems"
* tag 'staging-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
staging: rtl8712: fix use after free bugs
staging: r8188eu: Add Rosewill USB-N150 Nano to device tables
staging: r8188eu: add firmware dependency
Merge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk fixes from Stephen Boyd:
"Here's a collection of primarily clk driver fixes, with a couple fixes
to the core framework.
We had to revert out a commit that affected boot on some devices that
have the CLK_OPS_PARENT_ENABLE flag set. It isn't critical to have
that fix so we'll try again next time.
Driver side fixes include:
- Plug an OF-node refcount bug in the TI clk driver
- Fix the error handling in the raspberry pi firmware get_rate so
that errors don't look like valid frequencies
- Avoid going out of bounds in the raspberry pi driver too if the
video firmware returns something we're not expecting"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
Revert "clk: core: Honor CLK_OPS_PARENT_ENABLE for clk gate ops"
clk: bcm: rpi: Show clock id limit in error case
clk: bcm: rpi: Add missing newline
clk: bcm: rpi: Prevent out-of-bounds access
clk: bcm: rpi: Fix error handling of raspberrypi_fw_get_rate
clk: core: Fix runtime PM sequence in clk_core_unprepare()
clk: core: Honor CLK_OPS_PARENT_ENABLE for clk gate ops
clk: ti: Fix missing of_node_get() ti_find_clock_provider()
Merge tag 'hwmon-for-v6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging
Pull hwmon fixes from Guenter Roeck:
- Fix out of bounds access in gpio-fan driver
- Fix VOUT margin caching in PMBus core
- Avoid error message after -EPROBE_DEFER from devm_regulator_register()
* tag 'hwmon-for-v6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
hwmon: (gpio-fan) Fix array out of bounds access
hwmon: (pmbus) Fix vout margin caching
hwmon: (pmbus) Use dev_err_probe() to filter -EPROBE_DEFER error messages
Steven Price [Fri, 2 Sep 2022 11:26:12 +0000 (12:26 +0100)]
mm: pagewalk: Fix race between unmap and page walker
The mmap lock protects the page walker from changes to the page tables
during the walk. However a read lock is insufficient to protect those
areas which don't have a VMA as munmap() detaches the VMAs before
downgrading to a read lock and actually tearing down PTEs/page tables.
For users of walk_page_range() the solution is to simply call pte_hole()
immediately without checking the actual page tables when a VMA is not
present. We now never call __walk_page_range() without a valid vma.
For walk_page_range_novma() the locking requirements are tightened to
require the mmap write lock to be taken, and then walking the pgd
directly with 'no_vma' set.
This in turn means that all page walkers either have a valid vma, or
it's that special 'novma' case for page table debugging. As a result,
all the odd '(!walk->vma && !walk->no_vma)' tests can be removed.
Fixes: dd2283f2605e ("mm: mmap: zap pages with read mmap_sem in munmap") Reported-by: Jann Horn <jannh@google.com> Signed-off-by: Steven Price <steven.price@arm.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com> Cc: Konstantin Khlebnikov <koct9i@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
LoongArch: Fix section mismatch due to acpi_os_ioremap()
Now acpi_os_ioremap() is marked with __init because it calls memblock_
is_memory() which is also marked with __init in the !ARCH_KEEP_MEMBLOCK
case. However, acpi_os_ioremap() is called by ordinary functions such
as acpi_os_{read, write}_memory() and causes section mismatch warnings:
Fix these warnings by selecting ARCH_KEEP_MEMBLOCK unconditionally and
removing the __init modifier of acpi_os_ioremap(). This can also give a
chance to track "memory" and "reserved" memblocks after early boot.
Huacai Chen [Wed, 31 Aug 2022 03:19:27 +0000 (11:19 +0800)]
LoongArch: Adjust arch_do_signal_or_restart() to adapt generic entry
Commit 8ba62d37949e248c69 ("task_work: Call tracehook_notify_signal from
get_signal on all architectures") adjust arch_do_signal_or_restart() for
all architectures. LoongArch hasn't been upstream yet at that time and
can be still built successfully without adjustment because this function
has a weak version with the correct prototype. It is obviously that we
should convert LoongArch to use new API, otherwise some signal handlings
will be lost.
Ard Biesheuvel [Wed, 24 Aug 2022 15:31:10 +0000 (17:31 +0200)]
LoongArch: Avoid orphan input sections
Ensure that all input sections are listed explicitly in the linker
script, and issue a warning otherwise. This ensures that the binary
image matches the PE/COFF and other image metadata exactly, which is
important for things like code signing.
Merge tag 'block-6.0-2022-09-02' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
- NVMe pull request via Christoph:
- error handling fix for the new auth code (Hannes Reinecke)
- fix unhandled tcp states in nvmet_tcp_state_change (Maurizio
Lombardi)
- add NVME_QUIRK_BOGUS_NID for Lexar NM610 (Shyamin Ayesh)
- Add documentation for the ublk driver merged in this merge window
(Ming)
* tag 'block-6.0-2022-09-02' of git://git.kernel.dk/linux-block:
Documentation: document ublk
nvmet-tcp: fix unhandled tcp states in nvmet_tcp_state_change()
nvmet-auth: add missing goto in nvmet_setup_auth()
nvme-pci: add NVME_QUIRK_BOGUS_NID for Lexar NM610
Merge tag 'io_uring-6.0-2022-09-02' of git://git.kernel.dk/linux-block
Pull io_uring fixes from Jens Axboe:
- A single fix for over-eager retries for networking (Pavel)
- Revert the notification slot support for zerocopy sends.
It turns out that even after more than a year or development and
testing, there's not full agreement on whether just using plain
ordered notifications is Good Enough to avoid the complexity of using
the notifications slots. Because of that, we decided that it's best
left to a future final decision.
We can always bring back this feature, but we can't really change it
or remove it once we've released 6.0 with it enabled. The reverts
leave the usual CQE notifications as the primary interface for
knowing when data was sent, and when it was acked. (Pavel)
* tag 'io_uring-6.0-2022-09-02' of git://git.kernel.dk/linux-block:
selftests/net: return back io_uring zc send tests
io_uring/net: simplify zerocopy send user API
io_uring/notif: remove notif registration
Revert "io_uring: rename IORING_OP_FILES_UPDATE"
Revert "io_uring: add zc notification flush requests"
selftests/net: temporarily disable io_uring zc test
io_uring/net: fix overexcessive retries
Merge tag '6.0-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Pull cifs fixes from Steve French:
"Five fixes, all also marked for stable:
- fixes for collapse range and insert range (also fixes xfstest
generic/031)
- memory leak fix"
* tag '6.0-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
cifs: fix small mempool leak in SMB2_negotiate()
smb3: use filemap_write_and_wait_range instead of filemap_write_and_wait
smb3: fix temporary data corruption in insert range
smb3: fix temporary data corruption in collapse range
smb3: Move the flush out of smb2_copychunk_range() into its callers
Merge tag 'landlock-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/mic/linux
Pull landlock fix from Mickaël Salaün:
"This fixes a mis-handling of the LANDLOCK_ACCESS_FS_REFER right when
multiple rulesets/domains are stacked.
The expected behaviour was that an additional ruleset can only
restrict the set of permitted operations, but in this particular case,
it was potentially possible to re-gain the LANDLOCK_ACCESS_FS_REFER
right"
* tag 'landlock-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/mic/linux:
landlock: Fix file reparenting without explicit LANDLOCK_ACCESS_FS_REFER
Merge tag 'drm-fixes-2022-09-02' of git://anongit.freedesktop.org/drm/drm
Pull drm fixes from Dave Airlie:
"Regular fixes pull. One core dma-buf fix, then two weeks of i915
fixes, a lot of amdgpu fixes mostly for new IP, and a bunch of msm
fixes, mostly modesetting ones.
Nothing seems too bad at this point.
dma-buf/dma-resv:
- Fence-handling fix
i915:
- GVT fixes including fix for a CommetLake regression in mmio table
and misc doc and typo fixes
- Fix CCS handling
- Fix for guc requests after reset
- Display DSI related fixes
- Display backlight related fixes
- Fix for a null pointer dereference
- HDMI related quirk for ECS Liva Q2 with GLK graphics
- Skip wm/ddb readout for disabled pipes
amdgpu:
- FRU error message fix
- MES 11 updates
- DCN 3.2.x fixes
- DCN 3.1.4 fixes
- Fix possible use after free in CS IOCTL
- SMU 13.0.x fixes
- Fix iolink reporting on devices with direct connections to CPU
- GFX10 tap delay firmware fixes
msm:
- Fix for inconsistent indenting in msm_dsi_dphy_timing_calc_v3().
- Fix to make eDP the first connector in the connected list.
- Fix to populate intf_cfg correctly before calling reset_intf_cfg().
- Specify the correct number of DSI regulators for SDM660.
- Specify the correct number of DSI regulators for MSM8996.
- Fix for removing DP_RECOVERED_CLOCK_OUT_EN bit for tps4 link training
- Fix probe-deferral crash in gpu devfreq
- Fix gpu debugfs deadlock"
* tag 'drm-fixes-2022-09-02' of git://anongit.freedesktop.org/drm/drm: (51 commits)
drm/amd/amdgpu: skip ucode loading if ucode_size == 0
drm/amdgpu: only init tap_delay ucode when it's included in ucode binary
drm/amd/display: Fix black flash when switching from ODM2to1 to ODMBypass
drm/amd/display: Fix check for stream and plane
drm/amd/display: Re-initialize viewport after pipe merge
drm/amd/display: Use correct plane for CAB cursor size allocation
drm/amdgpu: ensure no PCIe peer access for CPU XGMI iolinks
drm/amd/pm: bump SMU 13.0.0 driver_if header version
drm/amd/pm: use vbios carried pptable for all SMU13.0.7 SKUs
drm/amd/pm: use vbios carried pptable for those supported SKUs
drm/amd/display: fix wrong register access
drm/amd/display: use actual cursor size instead of max for CAB allocation
drm/amd/display: disable display fresh from MALL on an edge case for DCN321
drm/amd/display: Fix CAB cursor size allocation for DCN32/321
drm/amd/display: Missing HPO instance added
drm/amd/display: set dig fifo read start level to 7 before dig fifo reset
drm/amdgpu: Fix use-after-free in amdgpu_cs_ioctl
drm/amd/display: Fix OTG H timing reset for dcn314
drm/amd/display: Fix DCN32 DPSTREAMCLK_CNTL programming
drm/amdgpu: Update mes_v11_api_def.h
...
Merge tag 'char-misc-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char/misc driver fixes from Greg KH:
"Here are some small char/misc and other driver fixes for 6.0-rc4.
Included in here are:
- binder fixes for previous fixes, and a few more fixes uncovered by
them.
- iio driver fixes
- soundwire driver fixes
- fastrpc driver fixes for memory corruption on some hardware
- peci driver fix
- mhi driver fix
All of these have been in linux-next with no reported problems"
* tag 'char-misc-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
binder: fix alloc->vma_vm_mm null-ptr dereference
misc: fastrpc: increase maximum session count
misc: fastrpc: fix memory corruption on open
misc: fastrpc: fix memory corruption on probe
soundwire: qcom: fix device status array range
bus: mhi: host: Fix up null pointer access in mhi_irq_handler
soundwire: qcom: remove duplicate reset control get
iio: light: cm32181: make cm32181_pm_ops static
iio: ad7292: Prevent regulator double disable
dt-bindings: iio: gyroscope: bosch,bmg160: correct number of pins
iio: adc: mcp3911: use correct formula for AD conversion
iio: adc: mcp3911: correct "microchip,device-addr" property
Revert "binder_alloc: Add missing mmap_lock calls when using the VMA"
binder_alloc: Add missing mmap_lock calls when using the VMA
binder: fix UAF of ref->proc caused by race condition
iio: light: cm3605: Fix an error handling path in cm3605_probe()
iio: adc: mcp3911: make use of the sign bit
peci: cpu: Fix use-after-free in adev_release()
peci: aspeed: fix error check return value of platform_get_irq()
Merge tag 'usb-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB/Thunderbolt driver fixes from Greg KH:
"Here are a lot of small USB and Thunderbolt driver fixes for 6.0-rc4
for reported problems. Included in here are:
- new usb-serial driver ids
- dwc3 driver bugfixes for reported problems with 6.0-rc1
- new device quirks, and reverts of some quirks that were incorrect
- gadget driver bugfixes for reported problems
- USB host controller bugfixes (xhci and others)
- other small USB fixes, details in the shortlog
- small thunderbolt driver fixes
All of these have been in linux-next with no reported issues"
* tag 'usb-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (51 commits)
Revert "usb: gadget: udc-xilinx: replace memcpy with memcpy_toio"
usb: storage: Add ASUS <0x0b05:0x1932> to IGNORE_UAS
USB: serial: ch341: fix disabled rx timer on older devices
USB: serial: ch341: fix lost character on LCR updates
USB: serial: cp210x: add Decagon UCA device id
Revert "usb: add quirks for Lenovo OneLink+ Dock"
usb: cdns3: fix issue with rearming ISO OUT endpoint
usb: cdns3: fix incorrect handling TRB_SMM flag for ISOC transfer
usb: gadget: mass_storage: Fix cdrom data transfers on MAC-OS
media: mceusb: Use new usb_control_msg_*() routines
USB: core: Prevent nested device-reset calls
USB: gadget: Fix obscure lockdep violation for udc_mutex
usb: dwc2: fix wrong order of phy_power_on and phy_init
usb: gadget: udc-xilinx: replace memcpy with memcpy_toio
usb: typec: Remove retimers properly
usb: dwc3: disable USB core PHY management
usb: add quirks for Lenovo OneLink+ Dock
USB: serial: option: add support for Cinterion MV32-WA/WB RmNet mode
USB: serial: ftdi_sio: add Omron CS1W-CIF31 device id
USB: serial: option: add Quectel EM060K modem
...
Merge tag 'platform-drivers-x86-v6.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
Pull x86 platform driver fixes from Hans de Goede:
"Various small fixes and hardware-id additions"
* tag 'platform-drivers-x86-v6.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
platform/x86: p2sb: Fix UAF when caller uses resource name
platform/x86: asus-wmi: Increase FAN_CURVE_BUF_LEN to 32
platform/mellanox: Remove redundant 'NULL' check
platform/mellanox: Remove unnecessary code
platform/mellanox: mlxreg-lc: Fix locking issue
platform/mellanox: mlxreg-lc: Fix coverity warning
platform/x86: acer-wmi: Acer Aspire One AOD270/Packard Bell Dot keymap fixes
platform/x86: thinkpad_acpi: Explicitly set to balanced mode on startup
platform/x86: asus-wmi: Fix the name of the mic-mute LED classdev
platform/surface: aggregator_registry: Add HID devices for sensors and UCSI client to SP8
platform/surface: aggregator_registry: Rename HID device nodes based on new findings
platform/surface: aggregator_registry: Rename HID device nodes based on their function
platform/surface: aggregator_registry: Add support for Surface Laptop Go 2
platform/x86: x86-android-tablets: Fix broken touchscreen on Chuwi Hi8 with Windows BIOS
platform/x86: pmc_atom: Fix SLP_TYPx bitfield mask
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"It's a lot smaller than last week, with the star of the show being a
couple of fixes to head.S addressing a boot regression introduced by
the recent overhaul of that code in non-default configurations (i.e.
KASLR disabled).
The first of those two resolves the issue reported (and bisected) by
Mikulus in the wait_on_bit() thread.
Summary:
- Fix two boot issues caused by the recent head.S rework when !KASLR
- Fix calculation of crashkernel memory reservation
- Fix bogus error check in PMU IRQ probing code"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: mm: Reserve enough pages for the initial ID map
perf/arm_pmu_platform: fix tests for platform_get_irq() failure
arm64: head: Ignore bogus KASLR displacement on non-relocatable kernels
arm64/kexec: Fix missing extra range for crashkres_low.
Mickaël Salaün [Wed, 31 Aug 2022 20:38:40 +0000 (22:38 +0200)]
landlock: Fix file reparenting without explicit LANDLOCK_ACCESS_FS_REFER
This change fixes a mis-handling of the LANDLOCK_ACCESS_FS_REFER right
when multiple rulesets/domains are stacked. The expected behaviour was
that an additional ruleset can only restrict the set of permitted
operations, but in this particular case, it was potentially possible to
re-gain the LANDLOCK_ACCESS_FS_REFER right.
With the introduction of LANDLOCK_ACCESS_FS_REFER, we added the first
globally denied-by-default access right. Indeed, this lifted an initial
Landlock limitation to rename and link files, which was initially always
denied when the source or the destination were different directories.
This led to an inconsistent backward compatibility behavior which was
only taken into account if no domain layer were using the new
LANDLOCK_ACCESS_FS_REFER right. However, when restricting a thread with
a new ruleset handling LANDLOCK_ACCESS_FS_REFER, all inherited parent
rulesets/layers not explicitly handling LANDLOCK_ACCESS_FS_REFER would
behave as if they were handling this access right and with all their
rules allowing it. This means that renaming and linking files could
became allowed by these parent layers, but all the other required
accesses must also be granted: all layers must allow file removal or
creation, and renaming and linking operations cannot lead to privilege
escalation according to the Landlock policy. See detailed explanation
in commit b91c3e4ea756 ("landlock: Add support for file reparenting with
LANDLOCK_ACCESS_FS_REFER").
To say it another way, this bug may lift the renaming and linking
limitations of the initial Landlock version, and a same ruleset can
enforce different restrictions depending on previous or next enforced
ruleset (i.e. inconsistent behavior). The LANDLOCK_ACCESS_FS_REFER right
cannot give access to data not already allowed, but this doesn't follow
the contract of the first Landlock ABI. This fix puts back the
limitation for sandboxes that didn't opt-in for this additional right.
For instance, if a first ruleset allows LANDLOCK_ACCESS_FS_MAKE_REG on
/dst and LANDLOCK_ACCESS_FS_REMOVE_FILE on /src, renaming /src/file to
/dst/file is denied. However, without this fix, stacking a new ruleset
which allows LANDLOCK_ACCESS_FS_REFER on / would now permit the
sandboxed thread to rename /src/file to /dst/file .
This change fixes the (absolute) rule access rights, which now always
forbid LANDLOCK_ACCESS_FS_REFER except when it is explicitly allowed
when creating a rule.
Making all domain handle LANDLOCK_ACCESS_FS_REFER was an initial
approach but there is two downsides:
* it makes the code more complex because we still want to check that a
rule allowing LANDLOCK_ACCESS_FS_REFER is legitimate according to the
ruleset's handled access rights (i.e. ABI v1 != ABI v2);
* it would not allow to identify if the user created a ruleset
explicitly handling LANDLOCK_ACCESS_FS_REFER or not, which will be an
issue to audit Landlock.
Instead, this change adds an ACCESS_INITIALLY_DENIED list of
denied-by-default rights, which (only) contains
LANDLOCK_ACCESS_FS_REFER. All domains are treated as if they are also
handling this list, but without modifying their fs_access_masks field.
A side effect is that the errno code returned by rename(2) or link(2)
*may* be changed from EXDEV to EACCES according to the enforced
restrictions. Indeed, we now have the mechanic to identify if an access
is denied because of a required right (e.g. LANDLOCK_ACCESS_FS_MAKE_REG,
LANDLOCK_ACCESS_FS_REMOVE_FILE) or if it is denied because of missing
LANDLOCK_ACCESS_FS_REFER rights. This may result in different errno
codes than for the initial Landlock version, but this approach is more
consistent and better for rename/link compatibility reasons, and it
wasn't possible before (hence no backport to ABI v1). The
layout1.rename_file test reflects this change.
Add 4 layout1.refer_denied_by_default* test suites to check that the
behavior of a ruleset not handling LANDLOCK_ACCESS_FS_REFER (ABI v1) is
unchanged even if another layer handles LANDLOCK_ACCESS_FS_REFER (i.e.
ABI v1 precedence). Make sure rule's absolute access rights are correct
by testing with and without a matching path. Add test_rename() and
test_exchange() helpers.
Extend layout1.inval tests to check that a denied-by-default access
right is not necessarily part of a domain's handled access rights.
Test coverage for security/landlock is 95.3% of 599 lines according to
gcc/gcov-11.
Fixes: b91c3e4ea756 ("landlock: Add support for file reparenting with LANDLOCK_ACCESS_FS_REFER") Reviewed-by: Paul Moore <paul@paul-moore.com> Reviewed-by: Günther Noack <gnoack3000@gmail.com> Link: https://lore.kernel.org/r/20220831203840.1370732-1-mic@digikod.net Cc: stable@vger.kernel.org
[mic: Constify and slightly simplify test helpers] Signed-off-by: Mickaël Salaün <mic@digikod.net>
Dan Carpenter [Thu, 1 Sep 2022 15:35:20 +0000 (18:35 +0300)]
xen/grants: prevent integer overflow in gnttab_dma_alloc_pages()
The change from kcalloc() to kvmalloc() means that arg->nr_pages
might now be large enough that the "args->nr_pages << PAGE_SHIFT" can
result in an integer overflow.
Fixes: b3f7931f5c61 ("xen/gntdev: switch from kcalloc() to kvcalloc()") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Juergen Gross <jgross@suse.com> Link: https://lore.kernel.org/r/YxDROJqu/RPvR0bi@kili Signed-off-by: Juergen Gross <jgross@suse.com>
SeongJae Park [Wed, 31 Aug 2022 16:58:24 +0000 (16:58 +0000)]
xen-blkfront: Cache feature_persistent value before advertisement
Xen blkfront advertises its support of the persistent grants feature
when it first setting up and when resuming in 'talk_to_blkback()'.
Then, blkback reads the advertised value when it connects with blkfront
and decides if it will use the persistent grants feature or not, and
advertises its decision to blkfront. Blkfront reads the blkback's
decision and it also makes the decision for the use of the feature.
Commit 402c43ea6b34 ("xen-blkfront: Apply 'feature_persistent' parameter
when connect"), however, made the blkfront's read of the parameter for
disabling the advertisement, namely 'feature_persistent', to be done
when it negotiate, not when advertise. Therefore blkfront advertises
without reading the parameter. As the field for caching the parameter
value is zero-initialized, it always advertises as the feature is
disabled, so that the persistent grants feature becomes always disabled.
This commit fixes the issue by making the blkfront does parmeter caching
just before the advertisement.
Fixes: 402c43ea6b34 ("xen-blkfront: Apply 'feature_persistent' parameter when connect") Cc: <stable@vger.kernel.org> # 5.10.x Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com> Signed-off-by: SeongJae Park <sj@kernel.org> Tested-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com> Reviewed-by: Juergen Gross <jgross@suse.com> Link: https://lore.kernel.org/r/20220831165824.94815-4-sj@kernel.org Signed-off-by: Juergen Gross <jgross@suse.com>
SeongJae Park [Wed, 31 Aug 2022 16:58:23 +0000 (16:58 +0000)]
xen-blkfront: Advertise feature-persistent as user requested
The advertisement of the persistent grants feature (writing
'feature-persistent' to xenbus) should mean not the decision for using
the feature but only the availability of the feature. However, commit 74a852479c68 ("xen-blkfront: add a parameter for disabling of persistent
grants") made a field of blkfront, which was a place for saving only the
negotiation result, to be used for yet another purpose: caching of the
'feature_persistent' parameter value. As a result, the advertisement,
which should follow only the parameter value, becomes inconsistent.
This commit fixes the misuse of the semantic by making blkfront saves
the parameter value in a separate place and advertises the support based
on only the saved value.
Fixes: 74a852479c68 ("xen-blkfront: add a parameter for disabling of persistent grants") Cc: <stable@vger.kernel.org> # 5.10.x Suggested-by: Juergen Gross <jgross@suse.com> Signed-off-by: SeongJae Park <sj@kernel.org> Tested-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com> Reviewed-by: Juergen Gross <jgross@suse.com> Link: https://lore.kernel.org/r/20220831165824.94815-3-sj@kernel.org Signed-off-by: Juergen Gross <jgross@suse.com>
SeongJae Park [Wed, 31 Aug 2022 16:58:22 +0000 (16:58 +0000)]
xen-blkback: Advertise feature-persistent as user requested
The advertisement of the persistent grants feature (writing
'feature-persistent' to xenbus) should mean not the decision for using
the feature but only the availability of the feature. However, commit aac8a70db24b ("xen-blkback: add a parameter for disabling of persistent
grants") made a field of blkback, which was a place for saving only the
negotiation result, to be used for yet another purpose: caching of the
'feature_persistent' parameter value. As a result, the advertisement,
which should follow only the parameter value, becomes inconsistent.
This commit fixes the misuse of the semantic by making blkback saves the
parameter value in a separate place and advertises the support based on
only the saved value.
Fixes: aac8a70db24b ("xen-blkback: add a parameter for disabling of persistent grants") Cc: <stable@vger.kernel.org> # 5.10.x Suggested-by: Juergen Gross <jgross@suse.com> Signed-off-by: SeongJae Park <sj@kernel.org> Tested-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com> Reviewed-by: Juergen Gross <jgross@suse.com> Link: https://lore.kernel.org/r/20220831165824.94815-2-sj@kernel.org Signed-off-by: Juergen Gross <jgross@suse.com>
Nathan Chancellor [Tue, 30 Aug 2022 15:12:56 +0000 (08:12 -0700)]
powerpc/papr_scm: Ensure rc is always initialized in papr_scm_pmu_register()
Clang warns:
arch/powerpc/platforms/pseries/papr_scm.c:492:6: warning: variable 'rc' is used uninitialized whenever 'if' condition is true [-Wsometimes-uninitialized]
if (!p->stat_buffer_len)
^~~~~~~~~~~~~~~~~~~
arch/powerpc/platforms/pseries/papr_scm.c:523:64: note: uninitialized use occurs here
dev_info(&p->pdev->dev, "nvdimm pmu didn't register rc=%d\n", rc);
^~
include/linux/dev_printk.h:150:67: note: expanded from macro 'dev_info'
dev_printk_index_wrap(_dev_info, KERN_INFO, dev, dev_fmt(fmt), ##__VA_ARGS__)
^~~~~~~~~~~
include/linux/dev_printk.h:110:23: note: expanded from macro 'dev_printk_index_wrap'
_p_func(dev, fmt, ##__VA_ARGS__); \
^~~~~~~~~~~
arch/powerpc/platforms/pseries/papr_scm.c:492:2: note: remove the 'if' if its condition is always false
if (!p->stat_buffer_len)
^~~~~~~~~~~~~~~~~~~~~~~~
arch/powerpc/platforms/pseries/papr_scm.c:484:8: note: initialize the variable 'rc' to silence this warning
int rc, nodeid;
^
= 0
1 warning generated.
The call to papr_scm_pmu_check_events() was eliminated but a return code
was not added to the if statement. Add the same return code from
papr_scm_pmu_check_events() for this condition so there is no more
warning.
If there is a context switch before "stb r9,2354(r31)", r31 may
not equal to r13, in such case, irq soft mask will not work.
The usual solution of marking the code ineligible for instrumentation
forces the code out-of-line, which we would prefer to avoid. Christophe
proposed a partial revert, but Nick raised some concerns with that. So
for now do a full revert.
Merge tag 'soundwire-6.0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire into char-misc-linus
Vinod writes:
"soundwire fixes for v6.0
This contains two fixes to qcom sdw driver which resolve duplicate reset
control get and second one fixes device array indices."
* tag 'soundwire-6.0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire:
soundwire: qcom: fix device status array range
soundwire: qcom: remove duplicate reset control get
Dave Airlie [Fri, 2 Sep 2022 01:26:29 +0000 (11:26 +1000)]
Merge tag 'drm-intel-fixes-2022-09-01' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes
- Fix for a null pointer dereference (Lukasz)
- HDMI related quirk for ECS Liva Q2 with GLK graphics (Diego)
- Skip wm/ddb readout for disabled pipes (Ville)
Paolo Bonzini [Thu, 11 Aug 2022 16:41:25 +0000 (12:41 -0400)]
KVM: x86: check validity of argument to KVM_SET_MP_STATE
An invalid argument to KVM_SET_MP_STATE has no effect other than making the
vCPU fail to run at the next KVM_RUN. Since it is extremely unlikely that
any userspace is relying on it, fail with -EINVAL just like for other
architectures.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Like Xu [Wed, 31 Aug 2022 03:35:24 +0000 (11:35 +0800)]
perf/x86/core: Completely disable guest PEBS via guest's global_ctrl
When a guest PEBS counter is cross-mapped by a host counter, software
will remove the corresponding bit in the arr[global_ctrl].guest and
expect hardware to perform a change of state "from enable to disable"
via the msr_slot[] switch during the vmx transaction.
The real world is that if user adjust the counter overflow value small
enough, it still opens a tiny race window for the previously PEBS-enabled
counter to write cross-mapped PEBS records into the guest's PEBS buffer,
when arr[global_ctrl].guest has been prioritised (switch_msr_special stuff)
to switch into the enabled state, while the arr[pebs_enable].guest has not.
Close this window by clearing invalid bits in the arr[global_ctrl].guest.
Cc: linux-perf-users@vger.kernel.org Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sean Christopherson <seanjc@google.com> Fixes: 854250329c02 ("KVM: x86/pmu: Disable guest PEBS temporarily in two rare situations") Signed-off-by: Like Xu <likexu@tencent.com>
Message-Id: <20220831033524.58561-1-likexu@tencent.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Miaohe Lin [Thu, 1 Sep 2022 12:23:00 +0000 (20:23 +0800)]
KVM: x86: fix memoryleak in kvm_arch_vcpu_create()
When allocating memory for mci_ctl2_banks fails, KVM doesn't release
mce_banks leading to memoryleak. Fix this issue by calling kfree()
for it when kcalloc() fails.
Fixes: 281b52780b57 ("KVM: x86: Add emulation for MSR_IA32_MCx_CTL2 MSRs.") Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Message-Id: <20220901122300.22298-1-linmiaohe@huawei.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Jim Mattson [Tue, 30 Aug 2022 17:49:47 +0000 (10:49 -0700)]
KVM: x86: Mask off unsupported and unknown bits of IA32_ARCH_CAPABILITIES
KVM should not claim to virtualize unknown IA32_ARCH_CAPABILITIES
bits. When kvm_get_arch_capabilities() was originally written, there
were only a few bits defined in this MSR, and KVM could virtualize all
of them. However, over the years, several bits have been defined that
KVM cannot just blindly pass through to the guest without additional
work (such as virtualizing an MSR promised by the
IA32_ARCH_CAPABILITES feature bit).
Define a mask of supported IA32_ARCH_CAPABILITIES bits, and mask off
any other bits that are set in the hardware MSR.
Cc: Paolo Bonzini <pbonzini@redhat.com> Fixes: 5b76a3cff011 ("KVM: VMX: Tell the nested hypervisor to skip L1D flush on vmentry") Signed-off-by: Jim Mattson <jmattson@google.com> Reviewed-by: Vipin Sharma <vipinsh@google.com> Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Message-Id: <20220830174947.2182144-1-jmattson@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Dave Airlie [Thu, 1 Sep 2022 19:58:52 +0000 (05:58 +1000)]
Merge tag 'drm-msm-fixes-2022-08-27' of https://gitlab.freedesktop.org/drm/msm into drm-fixes
Fixes for v6.0
- Fix for inconsistent indenting in function msm_dsi_dphy_timing_calc_v3.
This fixes a smatch warning reported by kbot
- Fix to make eDP the first connector in the connected list. This was
mainly done to address a screen corruption issue we were seeing on
sc7280 boards which have eDP as the primary display. The corruption
itself is from usermode but we decided to fix it this way because
things work correct with the primary display as the first one for
usermode
- Fix to populate intf_cfg correctly before calling reset_intf_cfg().
Without this, the display pipeline is not torn down correctly for
writeback
- Specify the correct number of DSI regulators for SDM660. It should
have been 1 but 2 was mentioned
- Specify the correct number of DSI regulators for MSM8996. It should
have been 3 but 2 was mentioned
- Fix for removing DP_RECOVERED_CLOCK_OUT_EN bit for tps4 link training
for DP. This was causing link training failures and hence no display
for a specific DP to HDMI cable on chromebooks
- Fix probe-deferral crash in gpu devfreq
- Fix gpu debugfs deadlock
Dave Airlie [Thu, 1 Sep 2022 19:56:25 +0000 (05:56 +1000)]
Merge tag 'amd-drm-fixes-6.0-2022-08-31' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes
amd-drm-fixes-6.0-2022-08-31:
amdgpu:
- FRU error message fix
- MES 11 updates
- DCN 3.2.x fixes
- DCN 3.1.4 fixes
- Fix possible use after free in CS IOCTL
- SMU 13.0.x fixes
- Fix iolink reporting on devices with direct connections to CPU
- GFX10 tap delay firmware fixes
- tcp: make global challenge ack rate limitation per net-ns and
default disabled
- tg3: fix potential hang-up on system reboot
- mac802154: fix reception for no-daddr packets
Misc:
- r8152: add PID for the lenovo onelink+ dock"
* tag 'net-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (56 commits)
net/smc: Remove redundant refcount increase
Revert "sch_cake: Return __NET_XMIT_STOLEN when consuming enqueued skb"
tcp: make global challenge ack rate limitation per net-ns and default disabled
tcp: annotate data-race around challenge_timestamp
net: dsa: hellcreek: Print warning only once
ip: fix triggering of 'icmp redirect'
sch_cake: Return __NET_XMIT_STOLEN when consuming enqueued skb
selftests: net: sort .gitignore file
Documentation: networking: correct possessive "its"
kcm: fix strp_init() order and cleanup
mlxbf_gige: compute MDIO period based on i1clk
ethernet: rocker: fix sleep in atomic context bug in neigh_timer_handler
net: lan966x: improve error handle in lan966x_fdma_rx_get_frame()
nfp: fix the access to management firmware hanging
net: phy: micrel: Make the GPIO to be non-exclusive
net: virtio_net: fix notification coalescing comments
net/sched: fix netdevice reference leaks in attach_default_qdiscs()
net: sched: tbf: don't call qdisc_put() while holding tree lock
net: Use u64_stats_fetch_begin_irq() for stats fetch.
net: dsa: xrs700x: Use irqsave variant for u64 stats update
...
If the gfp flag used for the memory allocation already has __GFP_ZERO,
then there is no need to explicitly clear the "struct devres_node". It is
already zeroed.
This saves a few cycles when using devm_zalloc() and co.
In the case of devres_alloc() (which calls __devres_alloc_node()), the
compiler could remove the test and the memset() because it should be able
to see that the __GFP_ZERO flag is set.
So this would make the code both faster and smaller.
Wolfram Sang [Thu, 18 Aug 2022 20:59:56 +0000 (22:59 +0200)]
driver_core: move from strlcpy with unused retval to strscpy
Follow the advice of the below link and prefer 'strscpy' in this
subsystem. Conversion is 1:1 because the return value is not used.
Generated by a coccinelle script.
Merge tag 'slab-for-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab
Pull slab fix from Vlastimil Babka:
- A fix from Waiman Long to avoid a theoretical deadlock reported by
lockdep.
* tag 'slab-for-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab:
mm/slab_common: Deleting kobject in kmem_cache_destroy() without holding slab_mutex/cpu_hotplug_lock
Tejun Heo [Sun, 28 Aug 2022 05:04:40 +0000 (19:04 -1000)]
cgroup: Implement cgroup_file_show()
Add cgroup_file_show() which allows toggling visibility of a cgroup file
using the new kernfs_show(). This will be used to hide psi interface files
on cgroups where it's disabled.
Tejun Heo [Sun, 28 Aug 2022 05:04:39 +0000 (19:04 -1000)]
kernfs: Implement kernfs_show()
Currently, kernfs nodes can be created hidden and activated later by calling
kernfs_activate() to allow creation of multiple nodes to succeed or fail as
a unit. This is an one-way one-time-only transition. This patch introduces
kernfs_show() which can toggle visibility dynamically.
As the currently proposed use - toggling the cgroup pressure files - only
requires operating on leaf nodes, for the sake of simplicity, restrict it as
such for now.
Hiding uses the same mechanism as deactivation and likewise guarantees that
there are no in-flight operations on completion. KERNFS_ACTIVATED and
KERNFS_HIDDEN are used to manage the interactions between activations and
show/hide operations. A node is visible iff both activated & !hidden.
Tejun Heo [Sun, 28 Aug 2022 05:04:38 +0000 (19:04 -1000)]
kernfs: Factor out kernfs_activate_one()
Factor out kernfs_activate_one() from kernfs_activate() and reorder
operations so that KERNFS_ACTIVATED now simply indicates whether activation
was attempted on the node ignoring whether activation took place. As the
flag doesn't have a reader, the refactoring and reordering shouldn't cause
any behavior difference.
Tejun Heo [Sun, 28 Aug 2022 05:04:37 +0000 (19:04 -1000)]
kernfs: Add KERNFS_REMOVING flags
KERNFS_ACTIVATED tracks whether a given node has ever been activated. As a
node was only deactivated on removal, this was used for
1. Drain optimization (removed by the previous patch).
2. To hide !activated nodes
3. To avoid double activations
4. Reject adding children to a node being removed
5. Skip activaing a node which is being removed.
We want to decouple deactivation from removal so that nodes can be
deactivated and hidden dynamically, which makes KERNFS_ACTIVATED useless for
all of the above purposes.
#1 is already gone. #2 and #3 can instead test whether the node is currently
active. A new flag KERNFS_REMOVING is added to explicitly mark nodes which
are being removed for #4 and #5.
While this leaves KERNFS_ACTIVATED with no users, leave it be as it will be
used in a following patch.
Tejun Heo [Sun, 28 Aug 2022 05:04:36 +0000 (19:04 -1000)]
kernfs: Improve kernfs_drain() and always call on removal
__kernfs_remove() was skipping draining based on KERNFS_ACTIVATED - whether
the node has ever been activated since creation. Instead, update it to
always call kernfs_drain() which now drains or skips based on the precise
drain conditions. This ensures that the nodes will be deactivated and
drained regardless of their states.
This doesn't make meaningful difference now but will enable deactivating and
draining nodes dynamically by making removals safe when racing those
operations.
While at it, drop / update comments.
v2: Fix the inverted test on kernfs_should_drain_open_files() noted by
Chengming. This was fixed by the next unrelated patch in the previous
posting.
Tejun Heo [Sun, 28 Aug 2022 05:04:35 +0000 (19:04 -1000)]
kernfs: Skip kernfs_drain_open_files() more aggressively
Track the number of mmapped files and files that need to be released and
skip kernfs_drain_open_file() if both are zero, which are the precise
conditions which require draining open_files. The early exit test is
factored into kernfs_should_drain_open_files() which is now tested by
kernfs_drain_open_files()'s caller - kernfs_drain().
This isn't a meaningful optimization on its own but will enable future
stand-alone kernfs_deactivate() implementation.
v2: Chengming noticed that on->nr_to_release was leaking after ->open()
failure. Fix it by telling kernfs_unlink_open_file() that it's called
from the ->open() fail path and should dec the counter. Use kzalloc() to
allocate kernfs_open_node so that the tracking fields are correctly
initialized.
Tejun Heo [Sun, 28 Aug 2022 05:04:32 +0000 (19:04 -1000)]
kernfs: Simply by replacing kernfs_deref_open_node() with of_on()
kernfs_node->attr.open is an RCU pointer to kernfs_open_node. However, RCU
dereference is currently only used in kernfs_notify(). Everywhere else,
either we're holding the lock which protects it or know that the
kernfs_open_node is pinned becaused we have a pointer to a kernfs_open_file
which is hanging off of it.
kernfs_deref_open_node() is used for the latter case - accessing
kernfs_open_node from kernfs_open_file. The lifetime and visibility rules
are simple and clear here. To someone who can access a kernfs_open_file, its
kernfs_open_node is pinned and visible through of->kn->attr.open.
Replace kernfs_deref_open_node() which simpler of_on(). The former takes
both @kn and @of and RCU deref @kn->attr.open while sanity checking with
@of. The latter takes @of and uses protected deref on of->kn->attr.open.
As the return value can't be NULL, remove the error handling in the callers
too.
Merge tag 'sound-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"Just handful changes at this time. The only major change is the
regression fix about the x86 WC-page buffer allocation.
The rest are trivial data-race fixes for ALSA sequencer core, the
possible out-of-bounds access fixes in the new ALSA control hash code,
and a few device-specific workarounds and fixes"
* tag 'sound-6.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: usb-audio: Add quirk for LH Labs Geek Out HD Audio 1V5
ALSA: hda/realtek: Add speaker AMP init for Samsung laptops with ALC298
ALSA: control: Re-order bounds checking in get_ctl_id_hash()
ALSA: control: Fix an out-of-bounds bug in get_ctl_id_hash()
ALSA: hda: intel-nhlt: Correct the handling of fmt_config flexible array
ALSA: seq: Fix data-race at module auto-loading
ALSA: seq: oss: Fix data-race for max_midi_devs access
ALSA: memalloc: Revive x86-specific WC page allocations again
Andy Shevchenko [Thu, 1 Sep 2022 11:34:06 +0000 (14:34 +0300)]
platform/x86: p2sb: Fix UAF when caller uses resource name
We have to copy only selected fields from the original resource.
Because a PCI device will be removed immediately after getting
its resources, we may not use any allocated data, hence we may
not copy any pointers.
Consider the following scenario:
1/ a caller of p2sb_bar() gets the resource;
2/ the resource has been copied by platform_device_add_data()
in order to create a platform device;
3/ the platform device creation will call for the device driver's
->probe() as soon as a match found;
4/ the ->probe() takes given resources (see 2/) and tries to
access one of its field, i.e. 'name', in the
__devm_ioremap_resource() to create a pretty looking output;
5/ but the 'name' is a dangling pointer because p2sb_bar()
removed a PCI device, which 'name' had been copied to
the caller's memory.
6/ UAF (Use-After-Free) as a result.
Kudos to Mika for the initial analisys of the issue.
Luke D. Jones [Sun, 28 Aug 2022 07:46:38 +0000 (19:46 +1200)]
platform/x86: asus-wmi: Increase FAN_CURVE_BUF_LEN to 32
Fix for TUF laptops returning with an -ENOSPC on calling
asus_wmi_evaluate_method_buf() when fetching default curves. The TUF method
requires at least 32 bytes space.
This also moves and changes the pr_debug() in fan_curve_check_present() to
pr_warn() in fan_curve_get_factory_default() so that there is at least some
indication in logs of why it fails.
Russ Weight [Wed, 31 Aug 2022 00:25:18 +0000 (17:25 -0700)]
firmware_loader: Fix memory leak in firmware upload
In the case of firmware-upload, an instance of struct fw_upload is
allocated in firmware_upload_register(). This data needs to be freed
in fw_dev_release(). Create a new fw_upload_free() function in
sysfs_upload.c to handle the firmware-upload specific memory frees
and incorporate the missing kfree call for the fw_upload structure.
Russ Weight [Mon, 29 Aug 2022 17:45:57 +0000 (10:45 -0700)]
firmware_loader: Fix use-after-free during unregister
In the following code within firmware_upload_unregister(), the call to
device_unregister() could result in the dev_release function freeing the
fw_upload_priv structure before it is dereferenced for the call to
module_put(). This bug was found by the kernel test robot using
CONFIG_KASAN while running the firmware selftests.
Pavel Begunkov [Thu, 1 Sep 2022 10:54:04 +0000 (11:54 +0100)]
io_uring/net: simplify zerocopy send user API
Following user feedback, this patch simplifies zerocopy send API. One of
the main complaints is that the current API is difficult with the
userspace managing notification slots, and then send retries with error
handling make it even worse.
Instead of keeping notification slots change it to the per-request
notifications model, which posts both completion and notification CQEs
for each request when any data has been sent, and only one CQE if it
fails. All notification CQEs will have IORING_CQE_F_NOTIF set and
IORING_CQE_F_MORE in completion CQEs indicates whether to wait a
notification or not.
IOSQE_CQE_SKIP_SUCCESS is disallowed with zerocopy sends for now.
This is less flexible, but greatly simplifies the user API and also the
kernel implementation. We reuse notif helpers in this patch, but in the
future there won't be need for keeping two requests.
Carlos Llamas [Mon, 29 Aug 2022 20:12:48 +0000 (20:12 +0000)]
binder: fix alloc->vma_vm_mm null-ptr dereference
Syzbot reported a couple issues introduced by commit 44e602b4e52f
("binder_alloc: add missing mmap_lock calls when using the VMA"), in
which we attempt to acquire the mmap_lock when alloc->vma_vm_mm has not
been initialized yet.
This can happen if a binder_proc receives a transaction without having
previously called mmap() to setup the binder_proc->alloc space in [1].
Also, a similar issue occurs via binder_alloc_print_pages() when we try
to dump the debugfs binder stats file in [2].
Fix these issues by setting up alloc->vma_vm_mm pointer during open()
and caching directly from current->mm. This guarantees we have a valid
reference to take the mmap_lock during scenarios described above.
Johan Hovold [Mon, 29 Aug 2022 08:05:30 +0000 (10:05 +0200)]
misc: fastrpc: fix memory corruption on open
The probe session-duplication overflow check incremented the session
count also when there were no more available sessions so that memory
beyond the fixed-size slab-allocated session array could be corrupted in
fastrpc_session_alloc() on open().
Johan Hovold [Mon, 29 Aug 2022 08:05:29 +0000 (10:05 +0200)]
misc: fastrpc: fix memory corruption on probe
Add the missing sanity check on the probed-session count to avoid
corrupting memory beyond the fixed-size slab-allocated session array
when there are more than FASTRPC_MAX_SESSIONS sessions defined in the
devicetree.
Merge tag 'nvme-6.0-2022-09-01' of git://git.infradead.org/nvme into block-6.0
Pull NVMe fixes from Christoph:
"nvme fixes for Linux 6.0
- error handling fix for the new auth code (Hannes Reinecke)
- fix unhandled tcp states in nvmet_tcp_state_change (Maurizio Lombardi)
- add NVME_QUIRK_BOGUS_NID for Lexar NM610 (Shyamin Ayesh)"
* tag 'nvme-6.0-2022-09-01' of git://git.infradead.org/nvme:
nvmet-tcp: fix unhandled tcp states in nvmet_tcp_state_change()
nvmet-auth: add missing goto in nvmet_setup_auth()
nvme-pci: add NVME_QUIRK_BOGUS_NID for Lexar NM610
Hu Xiaoying [Thu, 1 Sep 2022 04:57:37 +0000 (12:57 +0800)]
usb: storage: Add ASUS <0x0b05:0x1932> to IGNORE_UAS
USB external storage device(0x0b05:1932), use gnome-disk-utility tools
to test usb write < 30MB/s.
if does not to load module of uas for this device, can increase the
write speed from 20MB/s to >40MB/s.
Suggested-by: Matthias Kaehlcke <mka@chromium.org> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Hu Xiaoying <huxiaoying@kylinos.cn> Link: https://lore.kernel.org/r/20220901045737.3438046-1-huxiaoying@kylinos.cn Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / users / dwmw2 / linux.git / log