Changes to hrtimer mode (potentially made by __hrtimer_init_sleeper on
PREEMPT_RT) are not visible to hrtimer_start_range_ns, thus not
accounted for by hrtimer_start_expires call paths. In particular,
__wait_event_hrtimeout suffers from this problem as we have, for
example:
fs/aio.c::read_events
wait_event_interruptible_hrtimeout
__wait_event_hrtimeout
hrtimer_init_sleeper_on_stack <- this might "mode |= HRTIMER_MODE_HARD"
on RT if task runs at RT/DL priority
hrtimer_start_range_ns
WARN_ON_ONCE(!(mode & HRTIMER_MODE_HARD) ^ !timer->is_hard)
fires since the latter doesn't see the change of mode done by
init_sleeper
Fix it by making __wait_event_hrtimeout call hrtimer_sleeper_start_expires,
which is aware of the special RT/DL case, instead of hrtimer_start_range_ns.
Reported-by: Bruno Goncalves <bgoncalv@redhat.com> Signed-off-by: Juri Lelli <juri.lelli@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Daniel Bristot de Oliveira <bristot@kernel.org> Reviewed-by: Valentin Schneider <vschneid@redhat.com> Link: https://lore.kernel.org/r/20220627095051.42470-1-juri.lelli@redhat.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Here the kworker is waiting on msdos_sb_info::s_lock which is held by
tar which is in turn waiting for a buffer which is locked waiting to be
flushed, but this operation is plugged in the kworker.
The lock is a normal struct mutex, so tsk_is_pi_blocked() will always
return false on !RT and thus the behaviour changes for RT.
It seems that the intent here is to skip blk_flush_plug() in the case
where a non-preemptible lock (such as a spinlock) has been converted to
a rtmutex on RT, which is the case covered by the SM_RTLOCK_WAIT
schedule flag. But sched_submit_work() is only called from schedule()
which is never called in this scenario, so the check can simply be
deleted.
Looking at the history of the -rt patchset, in fact this change was
present from v5.9.1-rt20 until being dropped in v5.13-rt1 as it was part
of a larger patch [1] most of which was replaced by commit 40b43958ad73
("sched/core: Rework the __schedule() preempt argument").
As described in [1]:
The schedule process must distinguish between blocking on a regular
sleeping lock (rwsem and mutex) and a RT-only sleeping lock (spinlock
and rwlock):
- rwsem and mutex must flush block requests (blk_schedule_flush_plug())
even if blocked on a lock. This can not deadlock because this also
happens for non-RT.
There should be a warning if the scheduling point is within a RCU read
section.
- spinlock and rwlock must not flush block requests. This will deadlock
if the callback attempts to acquire a lock which is already acquired.
Similarly to being preempted, there should be no warning if the
scheduling point is within a RCU read section.
and with the tsk_is_pi_blocked() in the scheduler path, we hit the first
issue.
The generic IPI code depends on the IRQ affinity mask being allocated
and initialized. This will not be the case if SMP is disabled. Fix up
the remaining driver that selected GENERIC_IRQ_IPI in a non-SMP config.
Function irq_chip::irq_request_resources() is reported as optional
in the declaration of struct irq_chip.
If the parent irq_chip does not implement it, we should ignore it
and return.
The time spent in select_idle_cpu() is visible to netperf and might have a negative
impact.
[Symptom analysis]
The patch [1] from Mel Gorman has been applied to track the efficiency
of select_idle_sibling. Copy the indicators here:
SIS Search Efficiency(se_eff%):
A ratio expressed as a percentage of runqueues scanned versus
idle CPUs found. A 100% efficiency indicates that the target,
prev or recent CPU of a task was idle at wakeup. The lower the
efficiency, the more runqueues were scanned before an idle CPU
was found.
SIS Domain Search Efficiency(dom_eff%):
Similar, except only for the slower SIS
patch.
SIS Fast Success Rate(fast_rate%):
Percentage of SIS that used target, prev or
recent CPUs.
SIS Success rate(success_rate%):
Percentage of scans that found an idle CPU.
The test is based on Aubrey's schedtests tool, including netperf, hackbench,
schbench and tbench.
According to the test above, if the system becomes busy, the
SIS Search Efficiency(se_eff%) drops significantly. Although some
benchmarks would finally find an idle CPU(success_rate% = 100%), it is
doubtful whether it is worth it to search the whole LLC domain.
[Proposal]
It would be ideal to have a crystal ball to answer this question:
How many CPUs must a wakeup path walk down, before it can find an idle
CPU? Many potential metrics could be used to predict the number.
One candidate is the sum of util_avg in this LLC domain. The benefit
of choosing util_avg is that it is a metric of accumulated historic
activity, which seems to be smoother than instantaneous metrics
(such as rq->nr_running). Besides, choosing the sum of util_avg
would help predict the load of the LLC domain more precisely, because
SIS_PROP uses one CPU's idle time to estimate the total LLC domain idle
time.
In summary, the lower the util_avg is, the more select_idle_cpu()
should scan for idle CPU, and vice versa. When the sum of util_avg
in this LLC domain hits 85% or above, the scan stops. The reason to
choose 85% as the threshold is that this is the imbalance_pct(117)
when a LLC sched group is overloaded.
Introduce the quadratic function:
y = SCHED_CAPACITY_SCALE - p * x^2
and y'= y / SCHED_CAPACITY_SCALE
x is the ratio of sum_util compared to the CPU capacity:
x = sum_util / (llc_weight * SCHED_CAPACITY_SCALE)
y' is the ratio of CPUs to be scanned in the LLC domain,
and the number of CPUs to scan is calculated by:
nr_scan = llc_weight * y'
Choosing quadratic function is because:
[1] Compared to the linear function, it scans more aggressively when the
sum_util is low.
[2] Compared to the exponential function, it is easier to calculate.
[3] It seems that there is no accurate mapping between the sum of util_avg
and the number of CPUs to be scanned. Use heuristic scan for now.
For a platform with 112 CPUs per LLC, the number of CPUs to scan is:
sum_util% 0 5 15 25 35 45 55 65 75 85 86 ...
scan_nr 112 111 108 102 93 81 65 47 25 1 0 ...
For a platform with 16 CPUs per LLC, the number of CPUs to scan is:
sum_util% 0 5 15 25 35 45 55 65 75 85 86 ...
scan_nr 16 15 15 14 13 11 9 6 3 0 0 ...
Furthermore, to minimize the overhead of calculating the metrics in
select_idle_cpu(), borrow the statistics from periodic load balance.
As mentioned by Abel, on a platform with 112 CPUs per LLC, the
sum_util calculated by periodic load balance after 112 ms would
decay to about 0.5 * 0.5 * 0.5 * 0.7 = 8.75%, thus bringing a delay
in reflecting the latest utilization. But it is a trade-off.
Checking the util_avg in newidle load balance would be more frequent,
but it brings overhead - multiple CPUs write/read the per-LLC shared
variable and introduces cache contention. Tim also mentioned that,
it is allowed to be non-optimal in terms of scheduling for the
short-term variations, but if there is a long-term trend in the load
behavior, the scheduler can adjust for that.
When SIS_UTIL is enabled, the select_idle_cpu() uses the nr_scan
calculated by SIS_UTIL instead of the one from SIS_PROP. As Peter and
Mel suggested, SIS_UTIL should be enabled by default.
This patch is based on the util_avg, which is very sensitive to the
CPU frequency invariance. There is an issue that, when the max frequency
has been clamp, the util_avg would decay insanely fast when
the CPU is idle. Commit 8e933d6e2080 ("cpufreq: intel_pstate: Handle no_turbo
in frequency invariance") could be used to mitigate this symptom, by adjusting
the arch_max_freq_ratio when turbo is disabled. But this issue is still
not thoroughly fixed, because the current code is unaware of the user-specified
max CPU frequency.
[Test result]
netperf and tbench were launched with 25% 50% 75% 100% 125% 150%
175% 200% of CPU number respectively. Hackbench and schbench were launched
by 1, 2 ,4, 8 groups. Each test lasts for 100 seconds and repeats 3 times.
The following is the benchmark result comparison between
baseline:vanilla v5.19-rc1 and compare:patched kernel. Positive compare%
indicates better performance.
There is -87.9% less CPU scans after patched, which indicates lower overhead.
Besides, with this patch applied, there is -13% less rq lock contention
in perf-profile.calltrace.cycles-pp._raw_spin_lock.raw_spin_rq_lock_nested
.try_to_wake_up.default_wake_function.woken_wake_function.
This might help explain the performance improvement - Because this patch allows
the waking task to remain on the previous CPU, rather than grabbing other CPUs'
lock.
Each hackbench test is a:
hackbench -g $job --process/threads --pipe/sockets -l 1000000 -s 100
hackbench.throughput
=========
case load baseline(std%) compare%( std%)
process-pipe 1 group 1.00 ( 1.29) +0.57 ( 0.47)
process-pipe 2 groups 1.00 ( 0.27) +0.77 ( 0.81)
process-pipe 4 groups 1.00 ( 0.26) +1.17 ( 0.02)
process-pipe 8 groups 1.00 ( 0.15) -4.79 ( 0.02)
process-sockets 1 group 1.00 ( 0.63) -0.92 ( 0.13)
process-sockets 2 groups 1.00 ( 0.03) -0.83 ( 0.14)
process-sockets 4 groups 1.00 ( 0.40) +5.20 ( 0.26)
process-sockets 8 groups 1.00 ( 0.04) +3.52 ( 0.03)
threads-pipe 1 group 1.00 ( 1.28) +0.07 ( 0.14)
threads-pipe 2 groups 1.00 ( 0.22) -0.49 ( 0.74)
threads-pipe 4 groups 1.00 ( 0.05) +1.88 ( 0.13)
threads-pipe 8 groups 1.00 ( 0.09) -4.90 ( 0.06)
threads-sockets 1 group 1.00 ( 0.25) -0.70 ( 0.53)
threads-sockets 2 groups 1.00 ( 0.10) -0.63 ( 0.26)
threads-sockets 4 groups 1.00 ( 0.19) +11.92 ( 0.24)
threads-sockets 8 groups 1.00 ( 0.08) +4.31 ( 0.11)
Each schbench test is a:
schbench -m $job -t 28 -r 100 -s 30000 -c 30000
schbench.latency_90%_us
========
case load baseline(std%) compare%( std%)
normal 1 mthread 1.00 ( 31.22) -7.36 ( 20.25)*
normal 2 mthreads 1.00 ( 2.45) -0.48 ( 1.79)
normal 4 mthreads 1.00 ( 1.69) +0.45 ( 0.64)
normal 8 mthreads 1.00 ( 5.47) +9.81 ( 14.28)
*Consider the Standard Deviation, this -7.36% regression might not be valid.
Also, a OLTP workload with a commercial RDBMS has been tested, and there
is no significant change.
There were concerns that unbalanced tasks among CPUs would cause problems.
For example, suppose the LLC domain is composed of 8 CPUs, and 7 tasks are
bound to CPU0~CPU6, while CPU7 is idle:
Since the util_avg ratio is 87.5%( = 7/8 ), which is higher than 85%,
select_idle_cpu() will not scan, thus CPU7 is undetected during scan.
But according to Mel, it is unlikely the CPU7 will be idle all the time
because CPU7 could pull some tasks via CPU_NEWLY_IDLE.
lkp(kernel test robot) has reported a regression on stress-ng.sock on a
very busy system. According to the sched_debug statistics, it might be caused
by SIS_UTIL terminates the scan and chooses a previous CPU earlier, and this
might introduce more context switch, especially involuntary preemption, which
impacts a busy stress-ng. This regression has shown that, not all benchmarks
in every scenario benefit from idle CPU scan limit, and it needs further
investigation.
Besides, there is slight regression in hackbench's 16 groups case when the
LLC domain has 16 CPUs. Prateek mentioned that we should scan aggressively
in an LLC domain with 16 CPUs. Because the cost to search for an idle one
among 16 CPUs is negligible. The current patch aims to propose a generic
solution and only considers the util_avg. Something like the below could
be applied on top of the current patch to fulfill the requirement:
For LLC domain with 16 CPUs, the nr_scan will be expanded to 2 times large.
The smaller the CPU number this LLC domain has, the larger nr_scan will be
expanded. This needs further investigation.
There is also ongoing work[2] from Abel to filter out the busy CPUs during
wakeup, to further speed up the idle CPU scan. And it could be a following-up
optimization on top of this change.
Suggested-by: Tim Chen <tim.c.chen@intel.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Chen Yu <yu.c.chen@intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Tested-by: Yicong Yang <yangyicong@hisilicon.com> Tested-by: Mohini Narkhede <mohini.narkhede@intel.com> Tested-by: K Prateek Nayak <kprateek.nayak@amd.com> Link: https://lore.kernel.org/r/20220612163428.849378-1-yu.c.chen@intel.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Add checks verifying number of inodes stored in the superblock matches
the number computed from number of inodes per group. Also verify we have
at least one block worth of inodes per group. This prevents crashes on
corrupted filesystems.
Reported-by: syzbot+d273f7d7f58afd93be48@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
Pages mapped in user-space with PROT_MTE have the allocation tags either
zeroed or copied/restored to some user values. In order for the kernel
to access such pages via page_address(), resetting the tag in
page->flags was necessary. This tag resetting was deferred to
set_pte_at() -> mte_sync_page_tags() but it can race with another CPU
reading the flags (via page_to_virt()):
Since now the post_alloc_hook() function resets the page->flags tag when
unpoisoning is skipped for user pages (including the __GFP_ZEROTAGS
case), revert the arm64 commit calling page_kasan_tag_reset().
To fix this issue, keep the table->data as &insn->current_mode and
use container_of() to retrieve the insn pointer. Another mutex is
used to protect against the current_mode update but not for retrieving
insn_emulation as table->data is no longer changing.
Enable tracing of the execve*() system calls with the
syscalls:sys_exit_execve tracepoint by removing the call to
forget_syscall() when starting a new thread and preserving the value of
regs->syscallno across exec.
On most architectures, IRQ flag tracing is disabled in NMI context, and
architectures need to define and select TRACE_IRQFLAGS_NMI_SUPPORT in
order to enable this.
Commit:
460e7f8d66abdf4b ("lockdep: Prepare for NMI IRQ state tracking")
Permitted IRQ flag tracing in NMI context, allowing lockdep to work in
NMI context where an architecture had suitable entry logic. At the time,
most architectures did not have such suitable entry logic, and this broke
lockdep on such architectures. Thus, this was partially disabled in
commit:
166add0845e612f3 ("locking/lockdep: Fix TRACE_IRQFLAGS vs. NMIs")
... with architectures needing to select TRACE_IRQFLAGS_NMI_SUPPORT to
enable IRQ flag tracing in NMI context.
Currently TRACE_IRQFLAGS_NMI_SUPPORT is defined under
arch/x86/Kconfig.debug. Move it to arch/Kconfig so architectures can
select it without having to provide their own definition.
Since the regular TRACE_IRQFLAGS_SUPPORT is selected by
arch/x86/Kconfig, the select of TRACE_IRQFLAGS_NMI_SUPPORT is moved
there too.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Will Deacon <will@kernel.org> Link: https://lore.kernel.org/r/20220511131733.4074499-2-mark.rutland@arm.com Signed-off-by: Will Deacon <will@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When kernel is booted with idle=nomwait do not use MWAIT as the
default idle state.
If the user boots the kernel with idle=nomwait, it is a clear
direction to not use mwait as the default idle state.
However, the current code does not take this into consideration
while selecting the default idle state on x86.
Fix it by checking for the idle=nomwait boot option in
prefer_mwait_c1_over_halt().
Also update the documentation around idle=nomwait appropriately.
If a process is killed or otherwise exits while having active network
connections and many threads waiting on epoll_wait, the threads will all
be woken immediately, but not removed from ep->wq. Then when network
traffic scans ep->wq in wake_up, every wakeup attempt will fail, and will
not remove the entries from the list.
This means that the cost of the wakeup attempt is far higher than usual,
does not decrease, and this also competes with the dying threads trying to
actually make progress and remove themselves from the wq.
Handle this by removing visited epoll wq entries unconditionally, rather
than only when the wakeup succeeds - the structure of ep_poll means that
the only potential loss is the timed_out->eavail heuristic, which now can
race and result in a redundant ep_send_events attempt. (But only when
incoming data and a timeout actually race, not on every timeout)
Shakeel added:
: We are seeing this issue in production with real workloads and it has
: caused hard lockups. Particularly network heavy workloads with a lot
: of threads in epoll_wait() can easily trigger this issue if they get
: killed (oom-killed in our case).
Link: https://lkml.kernel.org/r/xm26fsjotqda.fsf@google.com Signed-off-by: Ben Segall <bsegall@google.com> Tested-by: Shakeel Butt <shakeelb@google.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Shakeel Butt <shakeelb@google.com> Cc: Eric Dumazet <edumazet@google.com> Cc: Roman Penyaev <rpenyaev@suse.de> Cc: Jason Baron <jbaron@akamai.com> Cc: Khazhismel Kumykov <khazhy@google.com> Cc: Heiher <r@hev.cc> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In nf_tables_updtable, if nf_tables_table_enable returns an error,
nft_trans_destroy is called to free the transaction object.
nft_trans_destroy() calls list_del(), but the transaction was never
placed on a list -- the list head is all zeroes, this results in
a null dereference:
BUG: KASAN: null-ptr-deref in nft_trans_destroy+0x26/0x59
Call Trace:
nft_trans_destroy+0x26/0x59
nf_tables_newtable+0x4bc/0x9bc
[..]
Its sane to assume that nft_trans_destroy() can be called
on the transaction object returned by nft_trans_alloc(), so
make sure the list head is initialised.
Fixes: 24afa2cc4521 ("netfilter: nf_tables: use new transaction infrastructure to handle table") Reported-by: mingi cho <mgcho.minic@gmail.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When doing lookups for rules on the same batch by using its ID, a rule from
a different chain can be used. If a rule is added to a chain but tries to
be positioned next to a rule from a different chain, it will be linked to
chain2, but the use counter on chain1 would be the one to be incremented.
When looking for rules by ID, use the chain that was used for the lookup by
name. The chain used in the context copied to the transaction needs to
match that same chain. That way, struct nft_rule does not need to get
enlarged with another member.
Fixes: 923354fe5352 ("netfilter: nf_tables: add NFTA_RULE_ID attribute") Fixes: 3c095cb5288d ("netfilter: nf_tables: Support RULE_ID reference in new rule") Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com> Cc: <stable@vger.kernel.org> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When doing lookups for chains on the same batch by using its ID, a chain
from a different table can be used. If a rule is added to a table but
refers to a chain in a different table, it will be linked to the chain in
table2, but would have expressions referring to objects in table1.
Then, when table1 is removed, the rule will not be removed as its linked to
a chain in table2. When expressions in the rule are processed or removed,
that will lead to a use-after-free.
When looking for chains by ID, use the table that was used for the lookup
by name, and only return chains belonging to that same table.
Fixes: 40929d8e631b ("netfilter: nf_tables: add NFTA_RULE_CHAIN_ID attribute") Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com> Cc: <stable@vger.kernel.org> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When doing lookups for sets on the same batch by using its ID, a set from a
different table can be used.
Then, when the table is removed, a reference to the set may be kept after
the set is freed, leading to a potential use-after-free.
When looking for sets by ID, use the table that was used for the lookup by
name, and only return sets belonging to that same table.
This fixes CVE-2022-2586, also reported as ZDI-CAN-17470.
Reported-by: Team Orca of Sea Security (@seasecresponse) Fixes: 8406eef378c0 ("netfilter: nf_tables: use new transaction infrastructure to handle sets") Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com> Cc: <stable@vger.kernel.org> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For High-Speed Transfers the prepare_one_trb function is calculating the
multiplier setting for the trb based on the length parameter of the trb
currently prepared. This assumption is wrong. For trbs with a sg list,
the length of the actual request has to be taken instead.
Fixes: 152d9411f321 ("usb: dwc3: gadget: Correct ISOC DATA PIDs for short packets") Cc: stable <stable@kernel.org> Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de> Link: https://lore.kernel.org/r/20220704141812.1532306-3-m.grzeschik@pengutronix.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The function __dwc3_prepare_one_trb has many parameters. Since it is
only used in dwc3_prepare_one_trb there is no point in keeping the
function. We merge both functions and get rid of the big list of
parameters.
Fixes: 152d9411f321 ("usb: dwc3: gadget: Correct ISOC DATA PIDs for short packets") Cc: stable <stable@kernel.org> Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de> Link: https://lore.kernel.org/r/20220704141812.1532306-2-m.grzeschik@pengutronix.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Usb core introduce the mechanism of giveback of URB in tasklet context to
reduce hardware interrupt handling time. On some test situation(such as
FIO with 4KB block size), when tasklet callback function called to
giveback URB, interrupt handler add URB node to the bh->head list also.
If check bh->head list again after finish all URB giveback of local_list,
then it may introduce a "dynamic balance" between giveback URB and add URB
to bh->head list. This tasklet callback function may not exit for a long
time, which will cause other tasklet function calls to be delayed. Some
real-time applications(such as KB and Mouse) will see noticeable lag.
In order to prevent the tasklet function from occupying the cpu for a long
time at a time, new URBS will not be added to the local_list even though
the bh->head list is not empty. But also need to ensure the left URB
giveback to be processed in time, so add a member high_prio for structure
giveback_urb_bh to prioritize tasklet and schelule this tasklet again if
bh->head list is not empty.
At the same time, we are able to prioritize tasklet through structure
member high_prio. So, replace the local high_prio_bh variable with this
structure member in usb_hcd_giveback_urb.
Fixes: 9f6c41f2859f ("USB: HCD: support giveback of URB in tasklet context") Cc: stable <stable@kernel.org> Reviewed-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Weitao Wang <WeitaoWang-oc@zhaoxin.com> Link: https://lore.kernel.org/r/20220726074918.5114-1-WeitaoWang-oc@zhaoxin.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
coresight devices track their connections (output connections) and
hold a reference to the fwnode. When a device goes away, we walk through
the devices on the coresight bus and make sure that the references
are dropped. This happens both ways:
a) For all output connections from the device, drop the reference to
the target device via coresight_release_platform_data()
b) Iterate over all the devices on the coresight bus and drop the
reference to fwnode if *this* device is the target of the output
connection, via coresight_remove_conns()->coresight_remove_match().
However, the coresight_remove_match() doesn't clear the fwnode field,
after dropping the reference, this causes use-after-free and
additional refcount drops on the fwnode.
e.g., if we have two devices, A and B, with a connection, A -> B.
If we remove B first, B would clear the reference on B, from A
via coresight_remove_match(). But when A is removed, it still has
a connection with fwnode still pointing to B. Thus it tries to drops
the reference in coresight_release_platform_data(), raising the bells
like :
When CONFIG_CPUMASK_OFFSTACK and CONFIG_DEBUG_PER_CPU_MAPS is selected,
cpu_max_bits_warn() generates a runtime warning similar as below while
we show /proc/cpuinfo. Fix this by using nr_cpu_ids (the runtime limit)
instead of NR_CPUS to iterate CPUs.
On a bare-metal Power8 system that doesn't have an "ibm,power-rng", a
malicious QEMU and guest that ignore the absence of the
KVM_CAP_PPC_HWRNG flag, and calls H_RANDOM anyway, will dereference a
NULL pointer.
In practice all Power8 machines have an "ibm,power-rng", but let's not
rely on that, add a NULL check and early return in
powernv_get_random_real_mode().
Fixes: 1746d4c9c4c7 ("KVM: PPC: Book3S HV: Add fast real-mode H_RANDOM implementation.") Cc: stable@vger.kernel.org # v4.1+ Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20220727143219.2684192-1-mpe@ellerman.id.au Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On FSL_BOOK3E, _PAGE_RW is defined with two bits, one for user and one
for supervisor. As soon as one of the two bits is set, the page has
to be display as RW. But the way it is implemented today requires both
bits to be set in order to display it as RW.
Instead of display RW when _PAGE_RW bits are set and R otherwise,
reverse the logic and display R when _PAGE_RW bits are all 0 and
RW otherwise.
This change has no impact on other platforms as _PAGE_RW is a single
bit on all of them.
By default old pre-3.0 Freescale PCIe controllers reports invalid PCI Class
Code 0x0b20 for PCIe Root Port. It can be seen by lspci -b output on P2020
board which has this pre-3.0 controller:
$ lspci -bvnn
00:00.0 Power PC [0b20]: Freescale Semiconductor Inc P2020E [1957:0070] (rev 21)
!!! Invalid class 0b20 for header type 01
Capabilities: [4c] Express Root Port (Slot-), MSI 00
Fix this issue by programming correct PCI Class Code 0x0604 for PCIe Root
Port to the Freescale specific PCIe register 0x474.
Fix activated by U-Boot stay active also after booting Linux kernel.
But boards which use older U-Boot version without that fix are affected and
still require this fix.
So implement this class code fix also in kernel fsl_pci.c driver.
test_bit(), as any other bitmap op, takes `unsigned long *` as a
second argument (pointer to the actual bitmap), as any bitmap
itself is an array of unsigned longs. However, the ia64_get_irr()
code passes a ref to `u64` as a second argument.
This works with the ia64 bitops implementation due to that they
have `void *` as the second argument and then cast it later on.
This works with the bitmap API itself due to that `unsigned long`
has the same size on ia64 as `u64` (`unsigned long long`), but
from the compiler PoV those two are different.
Define @irr as `unsigned long` to fix that. That implies no
functional changes. Has been hidden for 16 years!
The three bugs are here:
__func__, s3a_buf->s3a_data->exp_id);
__func__, md_buf->metadata->exp_id);
__func__, dis_buf->dis_data->exp_id);
The list iterator 's3a_buf/md_buf/dis_buf' will point to a bogus
position containing HEAD if the list is empty or no element is found.
This case must be checked before any use of the iterator, otherwise
it will lead to a invalid memory access.
To fix this bug, add an check. Use a new variable '*_iter' as the
list iterator, while use the old variable '*_buf' as a dedicated
pointer to point to the found element.
Add function mb_cache_entry_delete_or_get() to delete mbcache entry if
it is unused and also add a function to wait for entry to become unused
- mb_cache_entry_wait_unused(). We do not share code between the two
deleting function as one of them will go away soon.
CC: stable@vger.kernel.org Fixes: 75980e9575cf ("ext4: convert to mbcache2") Signed-off-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20220712105436.32204-2-jack@suse.cz Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Do not reclaim entries that are currently used by somebody from a
shrinker. Firstly, these entries are likely useful. Secondly, we will
need to keep such entries to protect pending increment of xattr block
refcount.
CC: stable@vger.kernel.org Fixes: 75980e9575cf ("ext4: convert to mbcache2") Signed-off-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20220712105436.32204-1-jack@suse.cz Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There's a KASAN warning in raid10_remove_disk when running the lvm
test lvconvert-raid-reshape.sh. We fix this warning by verifying that the
value "number" is valid.
BUG: KASAN: slab-out-of-bounds in raid10_remove_disk+0x61/0x2a0 [raid10]
Read of size 8 at addr ffff889108f3d300 by task mdX_raid10/124682
Last potentially related work creation:
kasan_save_stack+0x1e/0x40
__kasan_record_aux_stack+0x9e/0xc0
kvfree_call_rcu+0x84/0x480
timerfd_release+0x82/0x140
L __fput+0xfa/0x400
task_work_run+0x80/0xc0
exit_to_user_mode_prepare+0x155/0x160
syscall_exit_to_user_mode+0x12/0x40
do_syscall_64+0x42/0xc0
entry_SYSCALL_64_after_hwframe+0x46/0xb0
Second to last potentially related work creation:
kasan_save_stack+0x1e/0x40
__kasan_record_aux_stack+0x9e/0xc0
kvfree_call_rcu+0x84/0x480
timerfd_release+0x82/0x140
__fput+0xfa/0x400
task_work_run+0x80/0xc0
exit_to_user_mode_prepare+0x155/0x160
syscall_exit_to_user_mode+0x12/0x40
do_syscall_64+0x42/0xc0
entry_SYSCALL_64_after_hwframe+0x46/0xb0
The buggy address belongs to the object at ffff889108f3d200
which belongs to the cache kmalloc-256 of size 256
The buggy address is located 0 bytes to the right of
256-byte region [ffff889108f3d200, ffff889108f3d300)
Memory state around the buggy address: ffff889108f3d200: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ffff889108f3d280: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>ffff889108f3d300: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
^ ffff889108f3d380: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff889108f3d400: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Cc: stable@vger.kernel.org Signed-off-by: Song Liu <song@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When we ran the lvm test "shell/integrity-blocksize-3.sh" on a kernel with
kasan, we got failure in write_page.
The reason for the failure is that md_bitmap_destroy is called before
destroying the thread and the thread may be waiting in the function
write_page for the bio to complete. When the thread finishes waiting, it
executes "if (test_bit(BITMAP_WRITE_ERROR, &bitmap->flags))", which
triggers the kasan warning.
Note that the commit 18f37975fcd8 that caused this bug claims that it is
neede for md-cluster, you should check md-cluster and possibly find
another bugfix for it.
BUG: KASAN: use-after-free in write_page+0x18d/0x680 [md_mod]
Read of size 8 at addr ffff889162030c78 by task mdX_raid1/5539
Overlayfs may fail to complete updates when a filesystem lacks
fileattr/xattr syscall support and responds with an ENOSYS error code,
resulting in an unexpected "Function not implemented" error.
This bug may occur with FUSE filesystems, such as davfs2.
# when "some-file" exists in the lowerdir, this fails with "Function
# not implemented", with dmesg showing "overlayfs: failed to retrieve
# lower fileattr (/some-file, err=-38)"
touch /test/mnt/some-file
The underlying cause of this regresion is actually in FUSE, which fails to
translate the ENOSYS error code returned by userspace filesystem (which
means that the ioctl operation is not supported) to ENOTTY.
While requesting a new mailbox command, driver does not write any data to
unused registers. Initialize the unused register value to zero while
requesting a new mailbox command to prevent stale entry access by firmware.
When a SCSI device is removed while in active use, currently sg will
immediately return -ENODEV on any attempt to wait for active commands that
were sent before the removal. This is problematic for commands that use
SG_FLAG_DIRECT_IO since the data buffer may still be in use by the kernel
when userspace frees or reuses it after getting ENODEV, leading to
corrupted userspace memory (in the case of READ-type commands) or corrupted
data being sent to the device (in the case of WRITE-type commands). This
has been seen in practice when logging out of a iscsi_tcp session, where
the iSCSI driver may still be processing commands after the device has been
marked for removal.
Change the policy to allow userspace to wait for active sg commands even
when the device is being removed. Return -ENODEV only when there are no
more responses to read.
Link: https://lore.kernel.org/r/5ebea46f-fe83-2d0b-233d-d0dcb362dd0a@cybernetics.com Cc: <stable@vger.kernel.org> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Tony Battersby <tonyb@cybernetics.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The driver use the non-managed form of the register function in
isl29028_remove(). To keep the release order as mirroring the ordering
in probe, the driver should use non-managed form in probe, too.
iio_format_avail_range() should print range as follow [min, step, max], so
the function was previously calling iio_format_list() with length = 3,
length variable refers to the array size of values not the number of
elements. In case of non IIO_VAL_INT values each element has integer part
and decimal part. With length = 3 this would cause premature end of loop
and result in printing only one element.
UML generally does not provide access to special CPU instructions like
RDRAND, and execution tends to be rather deterministic, with no real
hardware interrupts, making good randomness really very hard, if not
all together impossible. Not only is this a security eyebrow raiser, but
it's also quite annoying when trying to do various pieces of UML-based
automation that takes a long time to boot, if ever.
Fix this by trivially calling getrandom() in the host and using that
seed as "bootloader randomness", which initializes the rng immediately
at UML boot.
The old behavior can be restored the same way as on any other arch, by
way of CONFIG_TRUST_BOOTLOADER_RANDOMNESS=n or
random.trust_bootloader=0. So seen from that perspective, this just
makes UML act like other archs, which is positive in its own right.
Additionally, wire up arch_get_random_{int,long}() in the same way, so
that reseeds can also make use of the host RNG, controllable by
CONFIG_TRUST_CPU_RANDOMNESS and random.trust_cpu, per usual.
Cc: stable@vger.kernel.org Acked-by: Johannes Berg <johannes@sipsolutions.net> Acked-By: Anton Ivanov <anton.ivanov@cambridgegreys.com> Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 3f4b2c9bc008 ("um: fix and optimize xor select template for CONFIG64 and timetravel mode")
caused a build regression when CONFIG_XOR_BLOCKS and CONFIG_UML_TIME_TRAVEL_SUPPORT
are selected.
Fix it by removing the straying parenthesis.
Cc: stable@vger.kernel.org Fixes: 3f4b2c9bc008 ("um: fix and optimize xor select template for CONFIG64 and timetravel mode") Signed-off-by: Benjamin Beichler <benjamin.beichler@uni-rostock.de>
[rw: Added commit message] Signed-off-by: Richard Weinberger <richard@nod.at> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In current implementation the Arasan NAND driver is updating the
system clock(i.e., anand->clk) in accordance to the timing modes
(i.e., SDR or NVDDR). But as per the Arasan NAND controller spec the
flash clock or the NAND bus clock(i.e., nfc->bus_clk), need to be
updated instead. This patch keeps the system clock unchanged and updates
the NAND bus clock as per the timing modes.
According to the Arasan NAND controller spec, the flash clock rate for SDR
must be <= 100 MHz, while for NV-DDR it must be the same as the rate of the
CLK line for the mode. The driver previously always set 100 MHz for NV-DDR,
which would result in incorrect behavior for NV-DDR modes 0-4.
The appropriate clock rate can be calculated from the NV-DDR timing
parameters as 1/tCK, or for rates measured in picoseconds,
10^12 / nand_nvddr_timings->tCK_min.
Then even if we can only mount it RO, we will still cause metadata
update for log replay:
BTRFS info (device dm-1): flagging fs with big metadata feature
BTRFS info (device dm-1): using free space tree
BTRFS info (device dm-1): has skinny extents
BTRFS info (device dm-1): start tree-log replay
This is definitely against RO compact flag requirement.
[CAUSE]
RO compact flag only forces us to do RO mount, but we will still do log
replay for plain RO mount.
Thus this will result us to do log replay and update metadata.
This can be very problematic for new RO compat flag, for example older
kernel can not understand v2 cache, and if we allow metadata update on
RO mount and invalidate/corrupt v2 cache.
[FIX]
Just reject the mount unless rescue=nologreplay is provided:
BTRFS error (device dm-1): cannot replay dirty log with unsupport optional features (0x40000000), try rescue=nologreplay instead
We don't want to set rescue=nologreply directly, as this would make the
end user to read the old data, and cause confusion.
Since the such case is really rare, we're mostly fine to just reject the
mount with an error message, which also includes the proper workaround.
CC: stable@vger.kernel.org #4.9+ Signed-off-by: Qu Wenruo <wqu@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Syzbot found a Use After Free bug in compute_effective_progs().
The reproducer creates a number of BPF links, and causes a fault
injected alloc to fail, while calling bpf_link_detach on them.
Link detach triggers the link to be freed by bpf_link_free(),
which calls __cgroup_bpf_detach() and update_effective_progs().
If the memory allocation in this function fails, the function restores
the pointer to the bpf_cgroup_link on the cgroup list, but the memory
gets freed just after it returns. After this, every subsequent call to
update_effective_progs() causes this already deallocated pointer to be
dereferenced in prog_list_length(), and triggers KASAN UAF error.
To fix this issue don't preserve the pointer to the prog or link in the
list, but remove it and replace it with a dummy prog without shrinking
the table. The subsequent call to __cgroup_bpf_detach() or
__cgroup_bpf_detach() will correct it.
Alex Deucher [Tue, 19 Jul 2022 18:56:59 +0000 (14:56 -0400)]
drm/amdgpu: fix check in fbdev init
The new vkms virtual display code is atomic so there is
no need to call drm_helper_disable_unused_functions()
when it is enabled. Doing so can result in a segfault.
When the driver switched from the old virtual display code
to the new atomic virtual display code, it was missed that
we enable virtual display unconditionally under SR-IOV
so the checks here missed that case. Add the missing
check for SR-IOV.
There is no equivalent of this patch for Linus' tree
because the relevant code no longer exists. This patch
is only relevant to kernels 5.15 and 5.16.
When pinning a buffer, we should check to see if there are any
additional restrictions imposed by bo->preferred_domains. This will
prevent the BO from being moved to an invalid domain when pinning.
For example, this can happen if the user requests to create a BO in GTT
domain for display scanout. amdgpu_dm will allow pinning to either VRAM
or GTT domains, since DCN can scanout from either or. However, in
amdgpu_bo_pin_restricted(), pinning to VRAM is preferred if there is
adequate carveout. This can lead to pinning to VRAM despite the user
requesting GTT placement for the BO.
v2: Allow the kernel to override the domain, which can happen when
exporting a BO to a V4L camera (for example).
Signed-off-by: Leo Li <sunpeng.li@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It looks like that when we moved nouveau over to using drm_dp_aux_init()
and registering it's aux bus during late connector registration, we totally
forgot to fix the failure codepath in nouveau_connector_create() - as it
still seems to assume that drm_dp_aux_init() can fail (it can't).
So, let's fix that and also add a missing check to ensure that we've
properly allocated nv_connector->aux.name while we're at it.
Signed-off-by: Lyude Paul <lyude@redhat.com> Reviewed-by: David Airlie <airlied@linux.ie> Fixes: 765333894d8c ("drm/nouveau/kms/nv50-: Move AUX adapter reg to connector late register/early unregister") Cc: <stable@vger.kernel.org> # v5.14+ Link: https://patchwork.freedesktop.org/patch/msgid/20220526204313.656473-1-lyude@redhat.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
While trying to fix another issue, it occurred to me that I don't actually
think there is any situation where we want pm_runtime_put() in nouveau to
be synchronous. In fact, this kind of just seems like it would cause
issues where we may unexpectedly block a thread we don't expect to be
blocked.
So, let's only use pm_runtime_put_autosuspend().
Changes since v1:
* Use pm_runtime_put_autosuspend(), not pm_runtime_put()
Signed-off-by: Lyude Paul <lyude@redhat.com> Reviewed-by: David Airlie <airlied@linux.ie> Fixes: 41cef002e760 ("drm/nouveau: Intercept ACPI_VIDEO_NOTIFY_PROBE") Cc: Hans de Goede <hdegoede@redhat.com> Cc: <stable@vger.kernel.org> # v4.10+ Link: https://patchwork.freedesktop.org/patch/msgid/20220714174234.949259-3-lyude@redhat.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix a number of compile errors by including the correct header
files. Examples are shown below.
../drivers/gpu/drm/hyperv/hyperv_drm_modeset.c: In function 'hyperv_blit_to_vram_rect':
../drivers/gpu/drm/hyperv/hyperv_drm_modeset.c:25:48: error: invalid use of undefined type 'struct drm_framebuffer'
25 | struct hyperv_drm_device *hv = to_hv(fb->dev);
| ^~
../drivers/gpu/drm/hyperv/hyperv_drm_modeset.c: In function 'hyperv_connector_get_modes':
../drivers/gpu/drm/hyperv/hyperv_drm_modeset.c:59:17: error: implicit declaration of function 'drm_add_modes_noedid' [-Werror=implicit-function-declaration]
59 | count = drm_add_modes_noedid(connector,
| ^~~~~~~~~~~~~~~~~~~~
../drivers/gpu/drm/hyperv/hyperv_drm_modeset.c:62:9: error: implicit declaration of function 'drm_set_preferred_mode'; did you mean 'drm_mm_reserve_node'? [-Werror=implicit-function-declaration]
62 | drm_set_preferred_mode(connector, hv->preferred_width,
| ^~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de> Fixes: aebd89482ab4 ("drm/hyperv: Add DRM driver for hyperv synthetic video device") Fixes: 6f5734c32c2c ("drm: Drop drm_framebuffer.h from drm_crtc.h") Fixes: b6a276d7208a ("drm: Drop drm_edid.h from drm_crtc.h") Cc: Deepak Rawat <drawat.floss@gmail.com> Cc: Thomas Zimmermann <tzimmermann@suse.de> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Maxime Ripard <mripard@kernel.org> Cc: linux-hyperv@vger.kernel.org Cc: dri-devel@lists.freedesktop.org Cc: <stable@vger.kernel.org> # v5.14+ Acked-by: Maxime Ripard <maxime@cerno.tech> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20220622083413.12573-1-tzimmermann@suse.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The dmas property is used to hold the dmaengine channel used for audio
output.
Older device trees were missing that property, so if it's not there we
disable the audio output entirely.
However, some overlays have set an empty value to that property, mostly
to workaround the fact that overlays cannot remove a property. Let's add
a test for that case and if it's empty, let's disable it as well.
The vmapping of dma-buf may succeed, but DRM SHMEM rejects the IOMEM
mapping, and thus, drm_gem_shmem_vmap_locked() should unvmap the IOMEM
before erroring out.
Cc: stable@vger.kernel.org Fixes: a78fa6cee23f ("drm/gem: Use struct dma_buf_map in GEM vmap ops and convert GEM backends") Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20220630200058.1883506-2-dmitry.osipenko@collabora.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use ww_acquire_fini() in the error code paths. Otherwise lockdep
thinks that lock is held when lock's memory is freed after the
drm_gem_lock_reservations() error. The ww_acquire_context needs to be
annotated as "released", which fixes the noisy "WARNING: held lock freed!"
splat of VirtIO-GPU driver with CONFIG_DEBUG_MUTEXES=y and enabled lockdep.
Cc: stable@vger.kernel.org Fixes: 0d8f01908cc6e ("drm: Add helpers for locking an array of BO reservations.") Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20220630200405.1883897-2-dmitry.osipenko@collabora.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The 12/24hr flag in the RX-8035 can be found in the hour register,
instead of the CTRL1 on the RX-8025. This was overlooked when
support for the RX-8035 was added, and was causing read errors when
the hour register 'overflowed'.
To deal with the relevant register not always being visible in
the relevant functions, determine the 12/24 mode at startup and
store it in the driver state.
Modules always live before the kernel, MODULES_END is fixed but
MODULES_VADDR isn't fixed, it depends on the kernel size.
Let's add it to virtual kernel memory layout dump.
As MODULES is only defined for CONFIG_64BIT, so we dump it when
CONFIG_64BIT=y.
Current task of executing crash kexec will be schedule out when panic is
triggered by RCU Stall, as it needs to wait rcu completion. It lead to
inability to enter the crash system.
The implementation of machine_crash_shutdown() is non-standard for RISC-V
according to other Arch's implementation(eg, x86, arm64), we need to send
IPI to stop secondary harts.
When use 'echo c > /proc/sysrq-trigger' to trigger kdump, riscv_crash_save_regs()
will be called to save regs for vmcore, we found "epc" value 00ffffffa5537400
is not a valid kernel virtual address, but is a user virtual address. Other
regs(eg, ra, sp, gp...) are correct kernel virtual address.
Actually 0x00ffffffb0dd9400 is the user mode PC of 'PID: 113 Comm: sh', which
is saved in the task's stack.
Use __smp_processor_id() to avoid check the preemption context when
CONFIG_DEBUG_PREEMPT enabled, as we will enter crash kernel and no
return.
Without the patch,
[ 103.781044] sysrq: Trigger a crash
[ 103.784625] Kernel panic - not syncing: sysrq triggered crash
[ 103.837634] CPU1: off
[ 103.889668] CPU2: off
[ 103.933479] CPU3: off
[ 103.939424] Starting crashdump kernel...
[ 103.943442] BUG: using smp_processor_id() in preemptible [00000000] code: sh/346
[ 103.950884] caller is debug_smp_processor_id+0x1c/0x26
[ 103.956051] CPU: 0 PID: 346 Comm: sh Kdump: loaded Not tainted 5.10.113-00002-gce03f03bf4ec-dirty #149
[ 103.965355] Call Trace:
[ 103.967805] [<ffffffe00020372a>] walk_stackframe+0x0/0xa2
[ 103.973206] [<ffffffe000bcf1f4>] show_stack+0x32/0x3e
[ 103.978258] [<ffffffe000bd382a>] dump_stack_lvl+0x72/0x8e
[ 103.983655] [<ffffffe000bd385a>] dump_stack+0x14/0x1c
[ 103.988705] [<ffffffe000bdc8fe>] check_preemption_disabled+0x9e/0xaa
[ 103.995057] [<ffffffe000bdc926>] debug_smp_processor_id+0x1c/0x26
[ 104.001150] [<ffffffe000206c64>] machine_kexec+0x22/0xd0
[ 104.006463] [<ffffffe000291a7e>] __crash_kexec+0x6a/0xa4
[ 104.011774] [<ffffffe000bcf3fa>] panic+0xfc/0x2b0
[ 104.016480] [<ffffffe000656ca4>] sysrq_reset_seq_param_set+0x0/0x70
[ 104.022745] [<ffffffe000657310>] __handle_sysrq+0x8c/0x154
[ 104.028229] [<ffffffe0006577e8>] write_sysrq_trigger+0x5a/0x6a
[ 104.034061] [<ffffffe0003d90e0>] proc_reg_write+0x58/0xd4
[ 104.039459] [<ffffffe00036cff4>] vfs_write+0x7e/0x254
[ 104.044509] [<ffffffe00036d2f6>] ksys_write+0x58/0xbe
[ 104.049558] [<ffffffe00036d36a>] sys_write+0xe/0x16
[ 104.054434] [<ffffffe000201b9a>] ret_from_syscall+0x0/0x2
[ 104.067863] Will call new kernel at ecc00000 from hart id 0
[ 104.074939] FDT image at fc5ee000
[ 104.079523] Bye...
With the patch we can got clear output,
[ 67.740553] sysrq: Trigger a crash
[ 67.744166] Kernel panic - not syncing: sysrq triggered crash
[ 67.809123] CPU1: off
[ 67.865210] CPU2: off
[ 67.909075] CPU3: off
[ 67.919123] Starting crashdump kernel...
[ 67.924900] Will call new kernel at ecc00000 from hart id 0
[ 67.932045] FDT image at fc5ee000
[ 67.935560] Bye...
Fixes: fa994d4a9a44 ("riscv: use hart id instead of cpu id on machine_kexec") Reviewed-by: Guo Ren <guoren@kernel.org> Reviewed-by: Heiko Stuebner <heiko@sntech.de> Reviewed-by: Atish Patra <atishp@rivosinc.com> Signed-off-by: Xianting Tian <xianting.tian@linux.alibaba.com> Link: https://lore.kernel.org/r/20220811074150.3020189-2-xianting.tian@linux.alibaba.com Cc: stable@vger.kernel.org Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix device tree schema validation error messages for the SiFive
Unmatched: ' cache-sets:0:0: 1024 was expected'.
The existing bindings allow for just 1024 cache-sets but the fu740 on
Unmatched the has 2048 cache-sets. The ISA itself permits any arbitrary
power of two, however this is not supported by dt-schema. The RTL for
the IP, to which the number of cache-sets is a tunable parameter, has
been released publicly so speculatively adding a small number of
"reasonable" values seems unwise also.
Instead, as the binding only supports two distinct controllers: add 2048
and explicitly lock it to the fu740's l2 cache while limiting 1024 to
the l2 cache on the fu540.
Fixes: 64ed198bacfc ("dt-bindings: riscv: Update l2 cache DT documentation to add support for SiFive FU740") Reported-by: Atul Khare <atulkhare@rivosinc.com> Signed-off-by: Conor Dooley <conor.dooley@microchip.com> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20220803185359.942928-1-mail@conchuod.ie Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In riscv the process of uprobe going to clear spie before exec
the origin insn,and set spie after that.But When access the page
which origin insn has been placed a page fault may happen and
irq was disabled in arch_uprobe_pre_xol function,It cause a WARN
as follows.
There is no need to clear/set spie in arch_uprobe_pre/post/abort_xol.
We can just remove it.
For all syscalls in 32-bit compat mode on 64-bit kernels the upper
32-bits of the 64-bit registers are zeroed out, so a negative 32-bit
signed value will show up as positive 64-bit signed value.
This behaviour breaks the io_pgetevents_time64() syscall which expects
signed 64-bit values for the "min_nr" and "nr" parameters.
Fix this by switching to the compat_sys_io_pgetevents_time64() syscall,
which uses "compat_long_t" types for those parameters.
Fix the output of /proc/iomem to show the real hardware device name
including the pa_pathname, e.g. "Merlin 160 Core Centronics [8:16:0]".
Up to now only the pa_pathname ("[8:16.0]") was shown.
Some code paths cannot guarantee the inode have any dentry alias. So
WARN_ON() all !dentry may flood the kernel logs.
For example, when an overlayfs inode is watched by inotifywait (1), and
someone is trying to read the /proc/$(pidof inotifywait)/fdinfo/INOTIFY_FD,
at that time if the dentry has been reclaimed by kernel (such as
echo 2 > /proc/sys/vm/drop_caches), there will be a WARN_ON(). The
printed call stack would be like:
For some sev ioctl interfaces, input may be passed that is less than or
equal to SEV_FW_BLOB_MAX_SIZE, but larger than the data that PSP
firmware returns. In this case, kmalloc will allocate memory that is the
size of the input rather than the size of the data. Since PSP firmware
doesn't fully overwrite the buffer, the sev ioctl interfaces with the
issue may return uninitialized slab memory.
Currently, all of the ioctl interfaces in the ccp driver are safe, but
to prevent future problems, change all ioctl interfaces that allocate
memory with kmalloc to use kzalloc and memset the data buffer to zero
in sev_ioctl_do_platform_status.
Fixes: c9a292daea55 ("crypto: ccp: Use the stack and common buffer for status commands") Fixes: 98298cc8e7162 ("crypto: ccp: Implement SEV_PEK_CSR ioctl command") Fixes: 83c3fc3c72095 ("crypto: ccp: Implement SEV_PDH_CERT_EXPORT ioctl command") Fixes: 7d79c4402d1d9 ("crypto: ccp - introduce SEV_GET_ID2 command") Cc: stable@vger.kernel.org Reported-by: Andy Nguyen <theflow@google.com> Suggested-by: David Rientjes <rientjes@google.com> Suggested-by: Peter Gonda <pgonda@google.com> Signed-off-by: John Allen <john.allen@amd.com> Reviewed-by: Peter Gonda <pgonda@google.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unlike other copying operations on ITER_PIPE, copy_mc_to_iter() can
result in a short copy. In that case we need to trim the unused
buffers, as well as the length of partially filled one - it's not
enough to set ->head, ->iov_offset and ->count to reflect how
much had we copied. Not hard to fix, fortunately...
I'd put a helper (pipe_discard_from(pipe, head)) into pipe_fs_i.h,
rather than iov_iter.c - it has nothing to do with iov_iter and
having it will allow us to avoid an ugly kludge in fs/splice.c.
We could put it into lib/iov_iter.c for now and move it later,
but I don't see the point going that way...
Cc: stable@kernel.org # 4.19+ Fixes: b796fdde4bdf "lib/iov_iter: Fix pipe handling in _copy_to_iter_mcsafe()" Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Christian Brauner (Microsoft) <brauner@kernel.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
usbnet uses the work usbnet_deferred_kevent() to perform tasks which may
sleep. On disconnect, completion of the work was originally awaited in
->ndo_stop(). But in 2003, that was moved to ->disconnect() by historic
commit "[PATCH] USB: usbnet, prevent exotic rtnl deadlock":
The change was made because back then, the kernel's workqueue
implementation did not allow waiting for a single work. One had to wait
for completion of *all* work by calling flush_scheduled_work(), and that
could deadlock when waiting for usbnet_deferred_kevent() with rtnl_mutex
held in ->ndo_stop().
The commit solved one problem but created another: It causes a
use-after-free in USB Ethernet drivers aqc111.c, asix_devices.c,
ax88179_178a.c, ch9200.c and smsc75xx.c:
* If the drivers receive a link change interrupt immediately before
disconnect, they raise EVENT_LINK_RESET in their (non-sleepable)
->status() callback and schedule usbnet_deferred_kevent().
* usbnet_deferred_kevent() invokes the driver's ->link_reset() callback,
which calls netif_carrier_{on,off}().
* That in turn schedules the work linkwatch_event().
Because usbnet_deferred_kevent() is awaited after unregister_netdev(),
netif_carrier_{on,off}() may operate on an unregistered netdev and
linkwatch_event() may run after free_netdev(), causing a use-after-free.
In 2010, usbnet was changed to only wait for a single instance of
usbnet_deferred_kevent() instead of *all* work by commit b489eba7fc5c
("drivers/net: don't use flush_scheduled_work()").
Unfortunately the commit neglected to move the wait back to
->ndo_stop(). Rectify that omission at long last.
There is no need to directly skip over to the SCROLL_REDRAW case while
the logo is still shown.
When using DRM, this change has no effect because the code will reach
the SCROLL_REDRAW case immediately anyway.
But if you run an accelerated fbdev driver and have
FRAMEBUFFER_CONSOLE_LEGACY_ACCELERATION enabled, console scrolling is
slowed down by factors so that it feels as if you use a 9600 baud
terminal.
So, drop those unnecessary checks and speed up fbdev console
acceleration during bootup.
The user may use the fbcon=vc:<n1>-<n2> option to tell fbcon to take
over the given range (n1...n2) of consoles. The value for n1 and n2
needs to be a positive number and up to (MAX_NR_CONSOLES - 1).
The given values were not fully checked against those boundaries yet.
To fix the issue, convert first_fb_vc and last_fb_vc to unsigned
integers and check them against the upper boundary, and make sure that
first_fb_vc is smaller than last_fb_vc.
If cooling_device_stats_setup() fails to create the stats object, it
must clear the last slot in cooling_device_attr_groups that was
initially empty (so as to make it possible to add stats attributes to
the cooling device attribute groups).
Failing to do so may cause the stats attributes to be created by
mistake for a device that doesn't have a stats object, because the
slot in question might be populated previously during the registration
of another cooling device.
Fixes: 9db8901ea0af ("thermal: Add cooling device's statistics in sysfs") Reported-by: Di Shen <di.shen@unisoc.com> Tested-by: Di Shen <di.shen@unisoc.com> Cc: 4.17+ <stable@vger.kernel.org> # 4.17+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
All creation paths except for O_TMPFILE handle umask in the vfs directly
if the filesystem doesn't support or enable POSIX ACLs. If the filesystem
does then umask handling is deferred until posix_acl_create().
Because, O_TMPFILE misses umask handling in the vfs it will not honor
umask settings. Fix this by adding the missing umask handling.
Link: https://lore.kernel.org/r/1657779088-2242-2-git-send-email-xuyang2018.jy@fujitsu.com Fixes: ef81791ba709 ("[O_TMPFILE] it's still short a few helpers, but infrastructure should be OK now...") Cc: <stable@vger.kernel.org> # 4.19+ Reported-by: Christian Brauner (Microsoft) <brauner@kernel.org> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Reviewed-and-Tested-by: Jeff Layton <jlayton@kernel.org> Acked-by: Christian Brauner (Microsoft) <brauner@kernel.org> Signed-off-by: Yang Xu <xuyang2018.jy@fujitsu.com> Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If something manages to set the maximum file size to MAX_OFFSET+1, this
can cause the xfs and ext4 filesystems at least to become corrupt.
Ordinarily, the kernel protects against userspace trying this by
checking the value early in the truncate() and ftruncate() system calls
calls - but there are at least two places that this check is bypassed:
(1) Cachefiles will round up the EOF of the backing file to DIO block
size so as to allow DIO on the final block - but this might push
the offset negative. It then calls notify_change(), but this
inadvertently bypasses the checking. This can be triggered if
someone puts an 8EiB-1 file on a server for someone else to try and
access by, say, nfs.
(2) ksmbd doesn't check the value it is given in set_end_of_file_info()
and then calls vfs_truncate() directly - which also bypasses the
check.
In both cases, it is potentially possible for a network filesystem to
cause a disk filesystem to be corrupted: cachefiles in the client's
cache filesystem; ksmbd in the server's filesystem.
nfsd is okay as it checks the value, but we can then remove this check
too.
Fix this by adding a check to inode_newsize_ok(), as called from
setattr_prepare(), thereby catching the issue as filesystems set up to
perform the truncate with minimal opportunity for bypassing the new
check.
Fixes: eed34cba8a1f ("cachefiles: Implement backing file wrangling") Fixes: 9be54045676d ("cifsd: add file operations") Signed-off-by: David Howells <dhowells@redhat.com> Reported-by: Jeff Layton <jlayton@kernel.org> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Namjae Jeon <linkinjeon@kernel.org> Cc: stable@kernel.org Acked-by: Alexander Viro <viro@zeniv.linux.org.uk>
cc: Steve French <sfrench@samba.org>
cc: Hyunchul Lee <hyc.lee@gmail.com>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Dave Wysochanski <dwysocha@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is another Asus K42JZ model with the PCI SSID 1043:1313
that requires the quirk ALC269VB_FIXUP_ASUS_MIC_NO_PRESENCE.
Add the corresponding entry to the quirk table.
The 12,1 model requires the same configuration as the 12,2 model
to enable headphones but has a different codec SSID. Adds
12,1 SSID for matching quirk.
There is another LENOVO 20149 (Type1Sku0) Notebook model with
CX20590, the device PCI SSID is 17aa:3977, which headphones are
not responding, that requires the quirk CXT_PINCFG_LENOVO_NOTEBOOK.
Add the corresponding entry to the quirk table.
Set pm_power_off to NULL like on all other architectures, check if it
is set in machine_halt() and machine_power_off() and fallback to
default_power_off if no other power driver got registered.
This brings riscv architecture inline with all other architectures,
and allows to reuse exiting power drivers unmodified.
Kernels without legacy SBI v0.1 extensions (CONFIG_RISCV_SBI_V01 is
not set), do not set pm_power_off to sbi_shutdown(). There is no
support for SBI v0.3 system reset extension either. This prevents
using gpio_poweroff on SiFive HiFive Unmatched.
Tested on SiFive HiFive unmatched, with a dtb specifying gpio-poweroff
node and kernel complied without CONFIG_RISCV_SBI_V01.
Commit a4317b8a36c4 ("KVM: x86: Fix recording of guest steal time
/ preempted status", 2021-11-11) open coded the previous call to
kvm_map_gfn, but in doing so it dropped the comparison between the cached
guest physical address and the one in the MSR. This cause an incorrect
cache hit if the guest modifies the steal time address while the memslots
remain the same. This can happen with kexec, in which case the steal
time data is written at the address used by the old kernel instead of
the old one.
While at it, rename the variable from gfn to gpa since it is a plain
physical address and not a right-shifted one.
Reported-by: Dave Young <ruyang@redhat.com> Reported-by: Xiaoying Yan <yiyan@redhat.com> Analyzed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Cc: David Woodhouse <dwmw@amazon.co.uk> Cc: stable@vger.kernel.org Fixes: a4317b8a36c4 ("KVM: x86: Fix recording of guest steal time / preempted status") Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit a4317b8a36c4 ("KVM: x86: Fix recording of guest steal time
/ preempted status", 2021-11-11) open coded the previous call to
kvm_map_gfn, but in doing so it dropped the comparison between the cached
guest physical address and the one in the MSR. This cause an incorrect
cache hit if the guest modifies the steal time address while the memslots
remain the same. This can happen with kexec, in which case the preempted
bit is written at the address used by the old kernel instead of
the old one.
Cc: David Woodhouse <dwmw@amazon.co.uk> Cc: stable@vger.kernel.org Fixes: a4317b8a36c4 ("KVM: x86: Fix recording of guest steal time / preempted status") Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mark kvm_mmu_x86_module_init() with __init, the entire reason it exists
is to initialize variables when kvm.ko is loaded, i.e. it must never be
called after module initialization.
Fixes: 335aa133e6a5 ("KVM: x86/mmu: Resolve nx_huge_pages when kvm.ko is loaded") Cc: stable@vger.kernel.org Reviewed-by: Kai Huang <kai.huang@intel.com> Tested-by: Michael Roth <michael.roth@amd.com> Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220803224957.1285926-2-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Windows 10/11 guests with Hyper-V role (WSL2) enabled are observed to
hang upon boot or shortly after when a non-default TSC frequency was
set for L1. The issue is observed on a host where TSC scaling is
supported. The problem appears to be that Windows doesn't use TSC
scaling for its guests, even when the feature is advertised, and KVM
filters SECONDARY_EXEC_TSC_SCALING out when creating L2 controls from
L1's VMCS. This leads to L2 running with the default frequency (matching
host's) while L1 is running with an altered one.
Keep SECONDARY_EXEC_TSC_SCALING in secondary exec controls for L2 when
it was set for L1. TSC_MULTIPLIER is already correctly computed and
written by prepare_vmcs02().
When injecting a #GP on LLDT/LTR due to a non-canonical LDT/TSS base, set
the error code to the selector. Intel SDM's says nothing about the #GP,
but AMD's APM explicitly states that both LLDT and LTR set the error code
to the selector, not zero.
Note, a non-canonical memory operand on LLDT/LTR does generate a #GP(0),
but the KVM code in question is specific to the base from the descriptor.
Fixes: 2268660cef03 ("KVM: x86: Emulator ignores LDTR/TR extended base on LLDT/LTR") Cc: stable@vger.kernel.org Signed-off-by: Sean Christopherson <seanjc@google.com> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Link: https://lore.kernel.org/r/20220711232750.1092012-3-seanjc@google.com Signed-off-by: Sean Christopherson <seanjc@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Wait to mark the TSS as busy during LTR emulation until after all fault
checks for the LTR have passed. Specifically, don't mark the TSS busy if
the new TSS base is non-canonical.
Opportunistically drop the one-off !seg_desc.PRESENT check for TR as the
only reason for the early check was to avoid marking a !PRESENT TSS as
busy, i.e. the common !PRESENT is now done before setting the busy bit.
Inject a #UD if L1 attempts VMXON with a CR0 or CR4 that is disallowed
per the associated nested VMX MSRs' fixed0/1 settings. KVM cannot rely
on hardware to perform the checks, even for the few checks that have
higher priority than VM-Exit, as (a) KVM may have forced CR0/CR4 bits in
hardware while running the guest, (b) there may incompatible CR0/CR4 bits
that have lower priority than VM-Exit, e.g. CR0.NE, and (c) userspace may
have further restricted the allowed CR0/CR4 values by manipulating the
guest's nested VMX MSRs.
Note, despite a very strong desire to throw shade at Jim, commit 0025faf79e9a ("kvm: nVMX: Remove superfluous VMX instruction fault checks")
is not to blame for the buggy behavior (though the comment...). That
commit only removed the CR0.PE, EFLAGS.VM, and COMPATIBILITY mode checks
(though it did erroneously drop the CPL check, but that has already been
remedied). KVM may force CR0.PE=1, but will do so only when also
forcing EFLAGS.VM=1 to emulate Real Mode, i.e. hardware will still #UD.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216033 Fixes: 1eb91cc48533 ("KVM: nVMX: Implement VMXON and VMXOFF") Reported-by: Eric Li <ercli@ucdavis.edu> Cc: stable@vger.kernel.org Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220607213604.3346000-4-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Check that the guest (L2) and host (L1) CR4 values that would be loaded
by nested VM-Enter and VM-Exit respectively are valid with respect to
KVM's (L0 host) allowed CR4 bits. Failure to check KVM reserved bits
would allow L1 to load an illegal CR4 (or trigger hardware VM-Fail or
failed VM-Entry) by massaging guest CPUID to allow features that are not
supported by KVM. Amusingly, KVM itself is an accomplice in its doom, as
KVM adjusts L1's MSR_IA32_VMX_CR4_FIXED1 to allow L1 to enable bits for
L2 based on L1's CPUID model.
Note, although nested_{guest,host}_cr4_valid() are _currently_ used if
and only if the vCPU is post-VMXON (nested.vmxon == true), that may not
be true in the future, e.g. emulating VMXON has a bug where it doesn't
check the allowed/required CR0/CR4 bits.
Cc: stable@vger.kernel.org Fixes: 3ce5a61d0f22 ("KVM: nVMX: fix checks on CR{0,4} during virtual VMX operation") Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220607213604.3346000-3-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Restrict the nVMX MSRs based on KVM's config, not based on the guest's
current config. Using the guest's config to audit the new config
prevents userspace from restoring the original config (KVM's config) if
at any point in the past the guest's config was restricted in any way.
Fixes: 651c35318378 ("KVM: nVMX: support restore of VMX capability MSRs") Cc: stable@vger.kernel.org Cc: David Matlack <dmatlack@google.com> Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220607213604.3346000-6-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Split the common x86 parts of kvm_is_valid_cr4(), i.e. the reserved bits
checks, into a separate helper, __kvm_is_valid_cr4(), and export only the
inner helper to vendor code in order to prevent nested VMX from calling
back into vmx_is_valid_cr4() via kvm_is_valid_cr4().
On SVM, this is a nop as SVM doesn't place any additional restrictions on
CR4.
On VMX, this is also currently a nop, but only because nested VMX is
missing checks on reserved CR4 bits for nested VM-Enter. That bug will
be fixed in a future patch, and could simply use kvm_is_valid_cr4() as-is,
but nVMX has _another_ bug where VMXON emulation doesn't enforce VMX's
restrictions on CR0/CR4. The cleanest and most intuitive way to fix the
VMXON bug is to use nested_host_cr{0,4}_valid(). If the CR4 variant
routes through kvm_is_valid_cr4(), using nested_host_cr4_valid() won't do
the right thing for the VMXON case as vmx_is_valid_cr4() enforces VMX's
restrictions if and only if the vCPU is post-VMXON.
Cc: stable@vger.kernel.org Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220607213604.3346000-2-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / kernel.git / log