In case a guest isn't consuming incoming network traffic as fast as it
is coming in, xen-netback is buffering network packages in unlimited
numbers today. This can result in host OOM situations.
Commit f48da8b14d04ca8 ("xen-netback: fix unlimited guest Rx internal
queue and carrier flapping") meant to introduce a mechanism to limit
the amount of buffered data by stopping the Tx queue when reaching the
data limit, but this doesn't work for cases like UDP.
When hitting the limit don't queue further SKBs, but drop them instead.
In order to be able to tell Rx packages have been dropped increment the
rx_dropped statistics counter in this case.
It should be noted that the old solution to continue queueing SKBs had
the additional problem of an overflow of the 32-bit rx_queue_len value
would result in intermittent Tx queue enabling.
Commit 1d5d48523900a4b ("xen-netback: require fewer guest Rx slots when
not using GSO") introduced a security problem in netback, as an
interface would only be regarded to be stalled if no slot is available
in the rx queue ring page. In case the SKB at the head of the queued
requests will need more than one rx slot and only one slot is free the
stall detection logic will never trigger, as the test for that is only
looking for at least one slot to be free.
Fix that by testing for the needed number of slots instead of only one
slot being available.
In order to not have to take the rx queue lock that often, store the
number of needed slots in the queue data. As all SKB dequeue operations
happen in the rx queue kernel thread this is safe, as long as the
number of needed slots is accessed via READ/WRITE_ONCE() only and
updates are always done with the rx queue lock held.
Add a small helper for obtaining the number of free slots.
This is part of XSA-392
Fixes: 1d5d48523900a4b ("xen-netback: require fewer guest Rx slots when not using GSO") Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Xen console driver is still vulnerable for an attack via excessive
number of events sent by the backend. Fix that by using a lateeoi event
channel.
For the normal domU initial console this requires the introduction of
bind_evtchn_to_irq_lateeoi() as there is no xenbus device available
at the time the event channel is bound to the irq.
As the decision whether an interrupt was spurious or not requires to
test for bytes having been read from the backend, move sending the
event into the if statement, as sending an event without having found
any bytes to be read is making no sense at all.
The Xen netfront driver is still vulnerable for an attack via excessive
number of events sent by the backend. Fix that by using lateeoi event
channels.
For being able to detect the case of no rx responses being added while
the carrier is down a new lock is needed in order to update and test
rsp_cons and the number of seen unconsumed responses atomically.
The Xen blkfront driver is still vulnerable for an attack via excessive
number of events sent by the backend. Fix that by using lateeoi event
channels.
This patch causes a Tx only workload to go to sleep even when it does
not have to, leading to misserable performance in skb mode. It fixed
one rare problem but created a much worse one, so this need to be
reverted while I try to craft a proper solution to the original
problem.
Fixes: bd0687c18e63 ("xsk: Do not sleep in poll() when need_wakeup set") Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20211217145646.26449-1-magnus.karlsson@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix drivers/bus/ti-sysc.c:2494:13: error: variable 'error' set but not
used introduced by commit 9d881361206e ("bus: ti-sysc: Add quirk handling
for reinit on context lost").
Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit marks accesses to the rcu_state.n_force_qs. These data
races are hard to make happen, but syzkaller was equal to the task.
Reported-by: syzbot+e08a83a1940ec3846cd5@syzkaller.appspotmail.com Acked-by: Marco Elver <elver@google.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In resp_mode_select() sanity check the block descriptor len to avoid UAF.
BUG: KASAN: use-after-free in resp_mode_select+0xa4c/0xb40 drivers/scsi/scsi_debug.c:2509
Read of size 1 at addr ffff888026670f50 by task scsicmd/15032
Change min_t() to use type "u32" instead of type "int" to avoid stack out
of bounds. With min_t() type "int" the values get sign extended and the
larger value gets used causing stack out of bounds.
BUG: KASAN: stack-out-of-bounds in memcpy include/linux/fortify-string.h:191 [inline]
BUG: KASAN: stack-out-of-bounds in sg_copy_buffer+0x1de/0x240 lib/scatterlist.c:976
Read of size 127 at addr ffff888072607128 by task syz-executor.7/18707
If the size arg to kcalloc() is zero, it returns ZERO_SIZE_PTR. Because of
that, for a following NULL pointer check to work on the returned pointer,
kcalloc() must not be called with the size arg equal to zero. Return early
without error before the kcalloc() call if size arg is zero.
BUG: KASAN: null-ptr-deref in memcpy include/linux/fortify-string.h:191 [inline]
BUG: KASAN: null-ptr-deref in sg_copy_buffer+0x138/0x240 lib/scatterlist.c:974
Write of size 4 at addr 0000000000000010 by task syz-executor.1/22789
Syzbot triggered the following warning in ovl_workdir_create() ->
ovl_create_real():
if (!err && WARN_ON(!newdentry->d_inode)) {
The reason is that the cgroup2 filesystem returns from mkdir without
instantiating the new dentry.
Weird filesystems such as this will be rejected by overlayfs at a later
stage during setup, but to prevent such a warning, call ovl_mkdir_real()
directly from ovl_workdir_create() and reject this case early.
Since mxl111sf_* devices call mxl111sf_ctrl_msg() in ->frontend_attach()
internally we need to initialize state->msg_lock before
frontend_attach(). To achieve it, ->probe() call added to all mxl111sf_*
devices, which will simply initiaize mutex.
Reported-and-tested-by: syzbot+5ca0bf339f13c4243001@syzkaller.appspotmail.com Fixes: 8572211842af ("[media] mxl111sf: convert to new DVB USB") Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Do not sleep in poll() when the need_wakeup flag is set. When this
flag is set, the application needs to explicitly wake up the driver
with a syscall (poll, recvmsg, sendmsg, etc.) to guarantee that Rx
and/or Tx processing will be processed promptly. But the current code
in poll(), sleeps first then wakes up the driver. This means that no
driver processing will occur (baring any interrupts) until the timeout
has expired.
Fix this by checking the need_wakeup flag first and if set, wake the
driver and return to the application. Only if need_wakeup is not set
should the process sleep if there is a timeout set in the poll() call.
Fixes: 77cd0d7b3f25 ("xsk: add support for need_wakeup flag in AF_XDP rings") Reported-by: Keith Wiles <keith.wiles@intel.com> Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Link: https://lore.kernel.org/bpf/20211214102607.7677-1-magnus.karlsson@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A new warning in clang points out a few places in this driver where a
bitwise OR is being used with boolean types:
drivers/input/touchscreen.c:81:17: warning: use of bitwise '|' with boolean operands [-Wbitwise-instead-of-logical]
data_present = touchscreen_get_prop_u32(dev, "touchscreen-min-x",
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This use of a bitwise OR is intentional, as bitwise operations do not
short circuit, which allows all the calls to touchscreen_get_prop_u32()
to happen so that the last parameter is initialized while coalescing the
results of the calls to make a decision after they are all evaluated.
To make this clearer to the compiler, use the '|=' operator to assign
the result of each touchscreen_get_prop_u32() call to data_present,
which keeps the meaning of the code the same but makes it obvious that
every one of these calls is expected to happen.
Signed-off-by: Nathan Chancellor <nathan@kernel.org> Reported-by: Nick Desaulniers <ndesaulniers@google.com> Reviewed-by: Nick Desaulniers <ndesaulniers@google.com> Link: https://lore.kernel.org/r/20211014205757.3474635-1-nathan@kernel.org Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Anders Roxell <anders.roxell@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Even after commit e1d7ba873555 ("time: Always make sure wall_to_monotonic
isn't positive") it is still possible to make wall_to_monotonic positive
by running the following code:
The reason is that the second parameter of timespec64_compare(), ts_delta,
may be unnormalized because the delta is calculated with an open coded
substraction which causes the comparison of tv_sec to yield the wrong
result:
Commit fab8a02b73eb ("serial: 8250_fintek: Enable high speed mode on Fintek F81866")
introduced support to use high baudrate with Fintek SuperIO UARTs. It'll
change clocksources when the UART probed.
But when user add kernel parameter "console=ttyS0,115200 console=tty0" to make
the UART as console output, the console will output garbled text after the
following kernel message.
The issue is occurs in following step:
probe_setup_port() -> fintek_8250_goto_highspeed()
It change clocksource from 115200 to 921600 with wrong time, it should change
clocksource in set_termios() not in probed. The following 3 patches are
implemented change clocksource in fintek_8250_set_termios().
Due to the high baud rate had implemented above 3 patches and the patch
Commit fab8a02b73eb ("serial: 8250_fintek: Enable high speed mode on Fintek F81866")
is bugged, So this patch will remove it.
Fixes: fab8a02b73eb ("serial: 8250_fintek: Enable high speed mode on Fintek F81866") Signed-off-by: Ji-Ze Hong (Peter Hong) <hpeter+linux_kernel@gmail.com> Link: https://lore.kernel.org/r/20211215075835.2072-1-hpeter+linux_kernel@gmail.com Cc: stable <stable@vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The donation calculation logic assumes that the donor has non-zero
after-donation hweight, so the lowest active hweight a donating cgroup can
have is 2 so that it can donate 1 while keeping the other 1 for itself.
Earlier, we only donated from cgroups with sizable surpluses so this
condition was always true. However, with the precise donation algorithm
implemented, f1de2439ec43 ("blk-iocost: revamp donation amount
determination") made the donation amount calculation exact enabling even low
hweight cgroups to donate.
This means that in rare occasions, a cgroup with active hweight of 1 can
enter donation calculation triggering the following warning and then a
divide-by-zero oops.
Fix it by excluding cgroups w/ active hweight < 2 from donating. Excluding
these extreme low hweight donations shouldn't affect work conservation in
any meaningful way.
When creating a subvolume, at create_subvol(), we allocate an anonymous
device and later call btrfs_get_new_fs_root(), which in turn just calls
btrfs_get_root_ref(). There we call btrfs_init_fs_root() which assigns
the anonymous device to the root, but if after that call there's an error,
when we jump to 'fail' label, we call btrfs_put_root(), which frees the
anonymous device and then returns an error that is propagated back to
create_subvol(). Than create_subvol() frees the anonymous device again.
When this happens, if the anonymous device was not reallocated after
the first time it was freed with btrfs_put_root(), we get a kernel
message like the following:
(...)
[13950.282466] BTRFS: error (device dm-0) in create_subvol:663: errno=-5 IO failure
[13950.283027] ida_free called for id=65 which is not allocated.
[13950.285974] BTRFS info (device dm-0): forced readonly
(...)
If the anonymous device gets reallocated by another btrfs filesystem
or any other kernel subsystem, then bad things can happen.
So fix this by setting the root's anonymous device to 0 at
btrfs_get_root_ref(), before we call btrfs_put_root(), if an error
happened.
Fixes: 2dfb1e43f57dd3 ("btrfs: preallocate anon block device at first phase of snapshot creation") CC: stable@vger.kernel.org # 5.10+ Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: Filipe Manana <fdmanana@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>
Line 1169 (#3) allocates a memory chunk for victim_name by kmalloc(),
but when the function returns in line 1184 (#4) victim_name allocated
by line 1169 (#3) is not freed, which will lead to a memory leak.
There is a similar snippet of code in this function as allocating a memory
chunk for victim_name in line 1104 (#1) as well as releasing the memory
in line 1116 (#2).
We should kfree() victim_name when the return value of backref_in_log()
is less than zero and before the function returns in line 1184 (#4).
When generalising GPIO support and adding support for CP2102N, the GPIO
registration for some CP2105 devices accidentally broke. Specifically,
when all the pins of a port are in "modem" mode, and thus unavailable
for GPIO use, the GPIO chip would now be registered without having
initialised the number of GPIO lines. This would in turn be rejected by
gpiolib and some errors messages would be printed (but importantly probe
would still succeed).
Fix this by initialising the number of GPIO lines before registering the
GPIO chip.
Note that as for the other device types, and as when all CP2105 pins are
muxed for LED function, the GPIO chip is registered also when no pins
are available for GPIO use.
Masking all unused MSI-X entries is done to ensure that a crash kernel
starts from a clean slate, which correponds to the reset state of the
device as defined in the PCI-E specificion 3.0 and later:
Vector Control for MSI-X Table Entries
--------------------------------------
"00: Mask bit: When this bit is set, the function is prohibited from
sending a message using this MSI-X Table entry.
...
This bit’s state after reset is 1 (entry is masked)."
A Marvell NVME device fails to deliver MSI interrupts after trying to
enable MSI-X interrupts due to that masking. It seems to take the MSI-X
mask bits into account even when MSI-X is disabled.
While not specification compliant, this can be cured by moving the masking
into the success path, so that the MSI-X table entries stay in device reset
state when the MSI-X setup fails.
[ tglx: Move it into the success path, add comment and amend changelog ]
Fixes: aa8092c1d1f1 ("PCI/MSI: Mask all unused MSI-X entries") Signed-off-by: Stefan Roese <sr@denx.de> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: linux-pci@vger.kernel.org Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: Michal Simek <michal.simek@xilinx.com> Cc: Marek Vasut <marex@denx.de> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20211210161025.3287927-1-sr@denx.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
PCI_MSIX_FLAGS_MASKALL is set in the MSI-X control register at MSI-X
interrupt setup time. It's cleared on success, but the error handling path
only clears the PCI_MSIX_FLAGS_ENABLE bit.
That's incorrect as the reset state of the PCI_MSIX_FLAGS_MASKALL bit is
zero. That can be observed via lspci:
When activate_stm_id_vb_detection is enabled, ID and Vbus detection relies
on sensing comparators. This detection needs time to stabilize.
A delay was already applied in dwc2_resume() when reactivating the
detection, but it wasn't done in dwc2_probe().
This patch adds delay after enabling STM ID/VBUS detection. Then, ID state
is good when initializing gadget and host, and avoid to get a wrong
Connector ID Status Change interrupt.
Fixes: a415083a11cc ("usb: dwc2: add support for STM32MP15 SoCs USB OTG HS and FS") Cc: stable <stable@vger.kernel.org> Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com> Signed-off-by: Amelie Delaunay <amelie.delaunay@foss.st.com> Link: https://lore.kernel.org/r/20211207124510.268841-1-amelie.delaunay@foss.st.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
syzbot is reporting that an unprivileged user who logged in from tty
console can crash the system using a reproducer shown below [1], for
n_hdlc_tty_wakeup() is synchronously calling n_hdlc_send_frames().
int main(int argc, char *argv[])
{
const int disc = 0xd;
ioctl(1, TIOCSETD, &disc);
while (1) {
ioctl(1, TCXONC, 0);
write(1, "", 1);
ioctl(1, TCXONC, 1); /* Kernel panic - not syncing: scheduling while atomic */
}
}
----------
Linus suspected that "struct tty_ldisc"->ops->write_wakeup() must not
sleep, and Jiri confirmed it from include/linux/tty_ldisc.h. Thus, defer
n_hdlc_send_frames() from n_hdlc_tty_wakeup() to a WQ context like
net/nfc/nci/uart.c does.
Link: https://syzkaller.appspot.com/bug?extid=5f47a8cea6a12b77a876 Reported-by: syzbot <syzbot+5f47a8cea6a12b77a876@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Analyzed-by: Fabio M. De Francesco <fmdefrancesco@gmail.com> Suggested-by: Linus Torvalds <torvalds@linux-foundation.org> Confirmed-by: Jiri Slaby <jirislaby@kernel.org> Reviewed-by: Fabio M. De Francesco <fmdefrancesco@gmail.com> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Link: https://lore.kernel.org/r/40de8b7e-a3be-4486-4e33-1b1d1da452f8@i-love.sakura.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The ability to write to MSR_IA32_PERF_CAPABILITIES from the host should
not depend on guest visible CPUID entries, even if just to allow
creating/restoring guest MSRs and CPUIDs in any sequence.
Fixes: 27461da31089 ("KVM: x86/pmu: Support full width counting") Suggested-by: Sean Christopherson <seanjc@google.com> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20211216165213.338923-3-vkuznets@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This change causes boot lockups when using "arlyprintk=xdbc" because
ktime can not be used at this point in time in the boot process. Also,
it is not needed for very small delays like this.
Szymon rightly pointed out that the previous check for the endpoint
direction in bRequestType was not looking at only the bit involved, but
rather the whole value. Normally this is ok, but for some request
types, bits other than bit 8 could be set and the check for the endpoint
length could not stall correctly.
When CONFIG_FSL_PMC is set to n, no value is assigned to cpu_up_prepare
in the mpc85xx_pm_ops structure. As a result, oops is triggered in
smp_85xx_start_cpu().
The libbpf CI reported occasional failure in btf_skc_cls_ingress:
test_syncookie:FAIL:Unexpected syncookie states gen_cookie:80326634 recv_cookie:0
bpf prog error at line 97
"error at line 97" means the bpf prog cannot find the listening socket
when the final ack is received. It then skipped processing
the syncookie in the final ack which then led to "recv_cookie:0".
The problem is the userspace program did not do accept() and went
ahead to close(listen_fd) before the kernel (and the bpf prog) had
a chance to process the final ack.
The fix is to add accept() call so that the userspace will wait for
the kernel to finish processing the final ack first before close()-ing
everything.
Fixes: 9a856cae2217 ("bpf: selftest: Add test_btf_skc_cls_ingress") Reported-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20211216191630.466151-1-kafai@fb.com Signed-off-by: Sasha Levin <sashal@kernel.org>
The descriptor list is a shared resource across all of the transmit queues, and
the locking mechanism used today only protects concurrency across a given
transmit queue between the transmit and reclaiming. This creates an opportunity
for the SYSTEMPORT hardware to work on corrupted descriptors if we have
multiple producers at once which is the case when using multiple transmit
queues.
This was particularly noticeable when using multiple flows/transmit queues and
it showed up in interesting ways in that UDP packets would get a correct UDP
header checksum being calculated over an incorrect packet length. Similarly TCP
packets would get an equally correct checksum computed by the hardware over an
incorrect packet length.
The SYSTEMPORT hardware maintains an internal descriptor list that it re-arranges
when the driver produces a new descriptor anytime it writes to the
WRITE_PORT_{HI,LO} registers, there is however some delay in the hardware to
re-organize its descriptors and it is possible that concurrent TX queues
eventually break this internal allocation scheme to the point where the
length/status part of the descriptor gets used for an incorrect data buffer.
The fix is to impose a global serialization for all TX queues in the short
section where we are writing to the WRITE_PORT_{HI,LO} registers which solves
the corruption even with multiple concurrent TX queues being used.
In nginx/wrk benchmark, there's a hung problem with high probability
on case likes that: (client will last several minutes to exit)
server: smc_run nginx
client: smc_run wrk -c 10000 -t 1 http://server
Client hangs with the following backtrace:
0 [ffffa7ce8Of3bbf8] __schedule at ffffffff9f9eOd5f
1 [ffffa7ce8Of3bc88] schedule at ffffffff9f9eløe6
2 [ffffa7ce8Of3bcaO] schedule_timeout at ffffffff9f9e3f3c
3 [ffffa7ce8Of3bd2O] wait_for_common at ffffffff9f9el9de
4 [ffffa7ce8Of3bd8O] __flush_work at ffffffff9fOfeOl3
5 [ffffa7ce8øf3bdfO] smc_release at ffffffffcO697d24 [smc]
6 [ffffa7ce8Of3be2O] __sock_release at ffffffff9f8O2e2d
7 [ffffa7ce8Of3be4ø] sock_close at ffffffff9f8ø2ebl
8 [ffffa7ce8øf3be48] __fput at ffffffff9f334f93
9 [ffffa7ce8Of3be78] task_work_run at ffffffff9flOlff5
10 [ffffa7ce8Of3beaO] do_exit at ffffffff9fOe5Ol2
11 [ffffa7ce8Of3bflO] do_group_exit at ffffffff9fOe592a
12 [ffffa7ce8Of3bf38] __x64_sys_exit_group at ffffffff9fOe5994
13 [ffffa7ce8Of3bf4O] do_syscall_64 at ffffffff9f9d4373
14 [ffffa7ce8Of3bfsO] entry_SYSCALL_64_after_hwframe at ffffffff9fa0007c
This issue dues to flush_work(), which is used to wait for
smc_connect_work() to finish in smc_release(). Once lots of
smc_connect_work() was pending or all executing work dangling,
smc_release() has to block until one worker comes to free, which
is equivalent to wait another smc_connnect_work() to finish.
In order to fix this, There are two changes:
1. For those idle smc_connect_work(), cancel it from the workqueue; for
executing smc_connect_work(), waiting for it to finish. For that
purpose, replace flush_work() with cancel_work_sync().
2. Since smc_connect() hold a reference for passive closing, if
smc_connect_work() has been cancelled, release the reference.
Fixes: 24ac3a08e658 ("net/smc: rebuild nonblocking connect") Reported-by: Tony Lu <tonylu@linux.alibaba.com> Tested-by: Dust Li <dust.li@linux.alibaba.com> Reviewed-by: Dust Li <dust.li@linux.alibaba.com> Reviewed-by: Tony Lu <tonylu@linux.alibaba.com> Signed-off-by: D. Wythe <alibuda@linux.alibaba.com> Acked-by: Karsten Graul <kgraul@linux.ibm.com> Link: https://lore.kernel.org/r/1639571361-101128-1-git-send-email-alibuda@linux.alibaba.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When printing netdev features %pNF already takes care of the 0x prefix,
remove the explicit one.
Fixes: 6413139dfc64 ("skbuff: increase verbosity when dumping skb data") Signed-off-by: Gal Pressman <gal@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Packet sockets may switch ring versions. Avoid misinterpreting state
between versions, whose fields share a union. rx_owner_map is only
allocated with a packet ring (pg_vec) and both are swapped together.
If pg_vec is NULL, meaning no packet ring was allocated, then neither
was rx_owner_map. And the field may be old state from a tpacket_v3.
Fixes: 61fad6816fc1 ("net/packet: tpacket_rcv: avoid a producer race condition") Reported-by: Syzbot <syzbot+1ac0994a0a0c55151121@syzkaller.appspotmail.com> Signed-off-by: Willem de Bruijn <willemb@google.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20211215143937.106178-1-willemdebruijn.kernel@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Zero-initialize memory for new map's value in function nsim_bpf_map_alloc
since it may cause a potential kernel information leak issue, as follows:
1. nsim_bpf_map_alloc calls nsim_map_alloc_elem to allocate elements for
a new map.
2. nsim_map_alloc_elem uses kmalloc to allocate map's value, but doesn't
zero it.
3. A user application can use IOCTL BPF_MAP_LOOKUP_ELEM to get specific
element's information in the map.
4. The kernel function map_lookup_elem will call bpf_map_copy_value to get
the information allocated at step-2, then use copy_to_user to copy to the
user buffer.
This can only leak information for an array map.
Fixes: 395cacb5f1a0 ("netdevsim: bpf: support fake map offload") Suggested-by: Jakub Kicinski <kuba@kernel.org> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Haimin Zhang <tcs.kernel@gmail.com> Link: https://lore.kernel.org/r/20211215111530.72103-1-tcs.kernel@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The MDIO bus speed must be initialized before talking to the PHY the first
time in order to avoid talking to it using a speed that the PHY doesn't
support.
This fixes HW initialization error -17 (IXGBE_ERR_PHY_ADDR_INVALID) on
Denverton CPUs (a.k.a. the Atom C3000 family) on ports with a 10Gb network
plugged in. On those devices, HLREG0[MDCSPD] resets to 1, which combined
with the 10Gb network results in a 24MHz MDIO speed, which is apparently
too fast for the connected PHY. PHY register reads over MDIO bus return
garbage, leading to initialization failure.
Reproduced with Linux kernel 4.19 and 5.15-rc7. Can be reproduced using
the following setup:
* Use an Atom C3000 family system with at least one X552 LAN on the SoC
* Disable PXE or other BIOS network initialization if possible
(the interface must not be initialized before Linux boots)
* Connect a live 10Gb Ethernet cable to an X550 port
* Power cycle (not reset, doesn't always work) the system and boot Linux
* Observe: ixgbe interfaces w/ 10GbE cables plugged in fail with error -17
Fixes: e84db7272798 ("ixgbe: Introduce function to control MDIO speed") Signed-off-by: Cyril Novikov <cnovikov@lynx.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit a296d665eae1 ("ixgbe: Add ethtool support to enable 2.5 and 5.0
Gbps support") introduced suppression of the advertisement of NBASE-T
speeds by default, according to Todd Fujinaka to accommodate customers
with network switches which could not cope with advertised NBASE-T
speeds, as posted in the E1000-devel mailing list:
However, the suppression was not documented at all, nor was how to
enable NBASE-T support.
Properly document the NBASE-T suppression and how to enable NBASE-T
support.
Fixes: a296d665eae1 ("ixgbe: Add ethtool support to enable 2.5 and 5.0 Gbps support") Reported-by: Robert Schlabbach <robert_s@gmx.net> Signed-off-by: Robert Schlabbach <robert_s@gmx.net> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The LTR maximum value was incorrectly written using the scale from
the LTR minimum value. This would cause incorrect values to be sent,
in cases where the initial calculation lead to different min/max scales.
In `igbvf_probe`, if register_netdev() fails, the program will go to
label err_hw_init, and then to label err_ioremap. In free_netdev() which
is just below label err_ioremap, there is `list_for_each_entry_safe` and
`netif_napi_del` which aims to delete all entries in `dev->napi_list`.
The program has added an entry `adapter->rx_ring->napi` which is added by
`netif_napi_add` in igbvf_alloc_queues(). However, adapter->rx_ring has
been freed below label err_hw_init. So this a UAF.
In terms of how to patch the problem, we can refer to igbvf_remove() and
delete the entry before `adapter->rx_ring`.
Fixes: d4e0fe01a38a0 (igbvf: add new driver to support 82576 virtual functions) Reported-by: Zheyu Ma <zheyuma97@gmail.com> Signed-off-by: Letu Ren <fantasquex@gmail.com> Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Move checking condition of VF MAC filter before clearing
or adding MAC filter to VF to prevent potential blackout caused
by removal of necessary and working VF's MAC filter.
A new warning in clang points out two instances where boolean
expressions are being used with a bitwise OR instead of logical OR:
drivers/soc/tegra/fuse/speedo-tegra20.c:72:9: warning: use of bitwise '|' with boolean operands [-Wbitwise-instead-of-logical]
reg = tegra_fuse_read_spare(i) |
^~~~~~~~~~~~~~~~~~~~~~~~~~
||
drivers/soc/tegra/fuse/speedo-tegra20.c:72:9: note: cast one or both operands to int to silence this warning
drivers/soc/tegra/fuse/speedo-tegra20.c:87:9: warning: use of bitwise '|' with boolean operands [-Wbitwise-instead-of-logical]
reg = tegra_fuse_read_spare(i) |
^~~~~~~~~~~~~~~~~~~~~~~~~~
||
drivers/soc/tegra/fuse/speedo-tegra20.c:87:9: note: cast one or both operands to int to silence this warning
2 warnings generated.
The motivation for the warning is that logical operations short circuit
while bitwise operations do not.
In this instance, tegra_fuse_read_spare() is not semantically returning
a boolean, it is returning a bit value. Use u32 for its return type so
that it can be used with either bitwise or boolean operators without any
warnings.
The mptcp ULP extension relies on sk->sk_sock_kern being set correctly:
It prevents setsockopt(fd, IPPROTO_TCP, TCP_ULP, "mptcp", 6); from
working for plain tcp sockets (any userspace-exposed socket).
But in case of fallback, accept() can return a plain tcp sk.
In such case, sk is still tagged as 'kernel' and setsockopt will work.
This will crash the kernel, The subflow extension has a NULL ctx->conn
mptcp socket:
BUG: KASAN: null-ptr-deref in subflow_data_ready+0x181/0x2b0
Call Trace:
tcp_data_ready+0xf8/0x370
[..]
Fixes: cf7da0d66cc1 ("mptcp: Create SUBFLOW socket for incoming connections") Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
We need to return EOPNOTSUPP for the unsupported mpls action type when
setup the flow action.
In the original implement, we will return 0 for the unsupported mpls
action type, actually we do not setup it and the following actions
to the flow action entry.
Fixes: 9838b20a7fb2 ("net: sched: take rtnl lock in tc_setup_flow_action()") Signed-off-by: Baowen Zheng <baowen.zheng@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
We should be doing the HE capabilities lookup based on the full
interface type so if P2P doesn't have HE but client has it doesn't
get confused. Fix that.
When we call ieee80211_agg_start_txq(), that will in turn call
schedule_and_wake_txq(). Called from ieee80211_stop_tx_ba_cb()
this is done under sta->lock, which leads to certain circular
lock dependencies, as reported by Chris Murphy:
https://lore.kernel.org/r/CAJCQCtSXJ5qA4bqSPY=oLRMbv-irihVvP7A2uGutEbXQVkoNaw@mail.gmail.com
In general, ieee80211_agg_start_txq() is usually not called
with sta->lock held, only in this one place. But it's always
called with sta->ampdu_mlme.mtx held, and that's therefore
clearly sufficient.
Change ieee80211_stop_tx_ba_cb() to also call it without the
sta->lock held, by factoring it out of ieee80211_remove_tid_tx()
(which is only called in this one place).
This breaks the locking chain and makes it less likely that
we'll have similar locking chain problems in the future.
Fixes: ba8c3d6f16a1 ("mac80211: add an intermediate software queue implementation") Reported-by: Chris Murphy <lists@colorremedies.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211202152554.f519884c8784.I555fef8e67d93fff3d9a304886c4a9f8b322e591@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The recent GRE selftests defined NUM_NETIFS=10. If the users copy
forwarding.config.sample to forwarding.config directly, they will get
error "Command line is not complete" when run the GRE tests, because
create_netif_veth() failed with no interface name defined.
Fix it by extending the NETIFS with p9 and p10.
Fixes: 2800f2485417 ("selftests: forwarding: Test multipath hashing on inner IP pkts for GRE tunnel") Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
we can remove 'q->classes[i].alist' only if DRR class 'i' was part of the
active list. In the ETS scheduler DRR classes belong to that list only if
the queue length is greater than zero: we need to test for non-zero value
of 'q->classes[i].qdisc->q.qlen' before removing from the list, similarly
to what has been done elsewhere in the ETS code.
Fixes: de6d25924c2a ("net/sched: sch_ets: don't peek at classes beyond 'nbands'") Reported-by: Shuang Li <shuali@redhat.com> Signed-off-by: Davide Caratti <dcaratti@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
IPv6 allows binding a socket to a device then binding to an address
not on the device (__inet6_bind -> ipv6_chk_addr with strict flag
not set). Update the bind tests to reflect legacy behavior.
Fixes: 34d0302ab861 ("selftests: Add ipv6 address bind tests to fcnal-test") Reported-by: Li Zhijian <lizhijian@fujitsu.com> Signed-off-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit referenced below added negative socket bind tests for VRF. The
socket binds should fail since the address to bind to is in a VRF yet
the socket is not bound to the VRF or a device within it. Update the
expected return code to check for 1 (bind failure) so the test passes
when the bind fails as expected. Add a 'show_hint' comment to explain
why the bind is expected to fail.
Fixes: 75b2b2b3db4c ("selftests: Add ipv4 address bind tests to fcnal-test") Reported-by: Li Zhijian <lizhijian@fujitsu.com> Signed-off-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit referenced below added configuration in the default VRF that
duplicates a VRF to check MD5 passwords are properly used and fail
when expected. That config should not be added all the time as it
can cause tests to pass that should not (by matching on default VRF
setup when it should not). Move the duplicate setup to a function
that is only called for the MD5 tests and add a cleanup function
to remove it after the MD5 tests.
Fixes: 5cad8bce26e0 ("fcnal-test: Add TCP MD5 tests for VRF") Signed-off-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently, the hns3_remove function firstly uninstall client instance,
and then uninstall acceletion engine device. The netdevice is freed in
client instance uninstall process, but acceletion engine device uninstall
process still use it to trace runtime information. This causes a use after
free problem.
So fixes it by check the instance register state to avoid use after free.
Fixes: d8355240cf8f ("net: hns3: add trace event support for PF/VF mailbox") Signed-off-by: Jie Wang <wangjie125@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
KMSAN reported a kernel-infoleak [1], that can exploited
by unpriv users.
After analysis it turned out UDP was not initializing
r->idiag_expires. Other users of inet_sk_diag_fill()
might make the same mistake in the future, so fix this
in inet_sk_diag_fill().
arch_kexec_apply_relocations_add currently ignores all errors returned
by arch_kexec_do_relocs. This means that every unknown relocation is
silently skipped causing unpredictable behavior while the relocated code
runs. Fix this by checking for errors and fail kexec_file_load if an
unknown relocation type is encountered.
The problem was found after gcc changed its behavior and used
R_390_PLT32DBL relocations for brasl instruction and relied on ld to
resolve the relocations in the final link in case direct calls are
possible. As the purgatory code is only linked partially (option -r)
ld didn't resolve the relocations leaving them for arch_kexec_do_relocs.
But arch_kexec_do_relocs doesn't know how to handle R_390_PLT32DBL
relocations so they were silently skipped. This ultimately caused an
endless loop in the purgatory as the brasl instructions kept branching
to itself.
Fixes: 71406883fd35 ("s390/kexec_file: Add kexec_file_load system call") Reported-by: Tao Liu <ltao@redhat.com> Signed-off-by: Philipp Rudo <prudo@redhat.com> Link: https://lore.kernel.org/r/20211208130741.5821-3-prudo@redhat.com Signed-off-by: Heiko Carstens <hca@linux.ibm.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
ping6 is failing as it should.
COMMAND: ip netns exec ns-A /bin/ping6 -c1 -w1 fe80::7c4c:bcff:fe66:a63a%red
strace of ping6 shows it is failing with '1',
so change the expected rc from 2 to 1.
Fixes: c0644e71df33 ("selftests: Add ipv6 ping tests to fcnal-test") Reported-by: kernel test robot <lkp@intel.com> Suggested-by: David Ahern <dsahern@gmail.com> Signed-off-by: Jie2x Zhou <jie2x.zhou@intel.com> Link: https://lore.kernel.org/r/20211209020230.37270-1-jie2x.zhou@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The VMADDR_CID_ANY flag used by a socket means that the socket isn't bound
to any specific CID. For example, a host vsock server may want to be bound
with VMADDR_CID_ANY, so that a guest vsock client can connect to the host
server with CID=VMADDR_CID_HOST (i.e. 2), and meanwhile, a host vsock
client can connect to the same local server with CID=VMADDR_CID_LOCAL
(i.e. 1).
The current implementation sets the destination socket's svm_cid to a
fixed CID value after the first client's connection, which isn't an
expected operation. For example, if the guest client first connects to the
host server, the server's svm_cid gets set to VMADDR_CID_HOST, then other
host clients won't be able to connect to the server anymore.
Reproduce steps:
1. Run the host server:
socat VSOCK-LISTEN:1234,fork -
2. Run a guest client to connect to the host server:
socat - VSOCK-CONNECT:2:1234
3. Run a host client to connect to the host server:
socat - VSOCK-CONNECT:1:1234
Without this patch, step 3. above fails to connect, and socat complains
"socat[1720] E connect(5, AF=40 cid:1 port:1234, 16): Connection
reset by peer".
With this patch, the above works well.
Fixes: c0cfa2d8a788 ("vsock: add multi-transports support") Signed-off-by: Wei Wang <wei.w.wang@intel.com> Link: https://lore.kernel.org/r/20211126011823.1760-1-wei.w.wang@intel.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
At the moment, using the ARM32 multi_v7_defconfig always results in two
SoCs being exposed in sysfs. This is wrong, as far as I'm aware the
Qualcomm DragonBoard 410c does not actually make use of a i.MX SoC. :)
This happens because imx_soc_device_init() registers the soc device
unconditionally, even when running on devices that do not make use of i.MX.
Arnd already reported this more than a year ago and even suggested a fix
similar to this commit, but for some reason it was never submitted.
Fix it by checking if the "__mxc_cpu_type" variable was actually
initialized by earlier platform code. On devices without i.MX it will
simply stay 0.
Cc: Peng Fan <peng.fan@nxp.com> Fixes: d2199b34871b ("ARM: imx: use device_initcall for imx_soc_device_init") Reported-by: Arnd Bergmann <arnd@arndb.de> Link: https://lore.kernel.org/r/CAK8P3a0hxO1TmK6oOMQ70AHSWJnP_CAq57YMOutrxkSYNjFeuw@mail.gmail.com/ Signed-off-by: Stephan Gerhold <stephan@gerhold.net> Reviewed-by: Fabio Estevam <festevam@gmail.com> Reviewed-by: Peng Fan <peng.fan@nxp.com> Signed-off-by: Shawn Guo <shawnguo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Before commit fc0c209c147f ("clk: Allow parents to be specified without
string names") child clks couldn't find their parent until the parent
clk was added to a list in __clk_core_init(). After that commit, child
clks can reference their parent clks directly via a clk_hw pointer, or
they can lookup that clk_hw pointer via DT if the parent clk is
registered with an OF clk provider.
The common clk framework treats hw->core being non-NULL as "the clk is
registered" per the logic within clk_core_fill_parent_index():
parent = entry->hw->core;
/*
* We have a direct reference but it isn't registered yet?
* Orphan it and let clk_reparent() update the orphan status
* when the parent is registered.
*/
if (!parent)
Therefore we need to be extra careful to not set hw->core until the clk
is fully registered with the clk framework. Otherwise we can get into a
situation where a child finds a parent clk and we move the child clk off
the orphan list when the parent isn't actually registered, wrecking our
enable accounting and breaking critical clks.
At this point, 'parent' points to clkBad even though clkBad hasn't been
fully registered yet. Ouch! A similar problem can happen if a clk
controller registers orphan clks that are referenced in the DT node of
another clk controller.
Let's fix all this by only setting the hw->core pointer underneath the
clk prepare lock in __clk_core_init(). This way we know that
clk_core_fill_parent_index() can't see hw->core be non-NULL until the
clk is fully registered.
Fixes: fc0c209c147f ("clk: Allow parents to be specified without string names") Signed-off-by: Mike Tipton <quic_mdtipton@quicinc.com> Link: https://lore.kernel.org/r/20211109043438.4639-1-quic_mdtipton@quicinc.com
[sboyd@kernel.org: Reword commit text, update comment] Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The smatch static checker warned about an uninitialized symbol usage in
this function, in the case where ceph_mdsc_build_path returns an error.
It turns out that that case is harmless, but it just looks sketchy.
Initialize the variable at declaration time, and remove the unneeded
setting of it later.
Fixes: a33f6432b3a6 ("ceph: encode inodes' parent/d_name in cap reconnect message") Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Xiubo Li <xiubli@redhat.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
opened_inodes is incremented twice when the same inode is opened twice
with O_RDONLY and O_WRONLY respectively.
To reproduce, run this python script, then check the metrics:
import os
for _ in range(10000):
fd_r = os.open('a', os.O_RDONLY)
fd_w = os.open('a', os.O_WRONLY)
os.close(fd_r)
os.close(fd_w)
Fixes: 1dd8d4708136 ("ceph: metrics for opened files, pinned caps and opened inodes") Signed-off-by: Hu Weiwen <sehuww@mail.scut.edu.cn> Reviewed-by: Xiubo Li <xiubli@redhat.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
For admission control, obviously all of that only works for
QoS data frames, otherwise we cannot even access the QoS
field in the header.
Syzbot reported (see below) an uninitialized value here due
to a status of a non-QoS nullfunc packet, which isn't even
long enough to contain the QoS header.
Fix this to only do anything for QoS data packets.
As stated in the schematics [1] and [2] P5 the APIO5 domain is supplied
by RK808-D Buck4, which in our case vcc1v8_codec - i.e. a 1.8 V regulator.
Currently only white noise comes from the ES8316's output, which - for
whatever reason - came up only after the the correct switch from i2s0_8ch_bus
to i2s0_2ch_bus for i2s0's pinctrl was done.
Fix this by setting the correct regulator for audio-supply.
Correct a typo in the vin-supply property. The input supply is
always-on, so this mistake doesn't affect whether the supply is actually
enabled correctly.
Correct a typo in the vin-supply property. The input supply is
always-on, so this mistake doesn't affect whether the supply is actually
enabled correctly.
According to the datasheet RTL8211, it must be asserted low for at least
10ms and at least 72ms "for internal circuits settling time" before
accessing the PHY registers.
If the audit daemon were ever to get stuck in a stopped state the
kernel's kauditd_thread() could get blocked attempting to send audit
records to the userspace audit daemon. With the kernel thread
blocked it is possible that the audit queue could grow unbounded as
certain audit record generating events must be exempt from the queue
limits else the system enter a deadlock state.
This patch resolves this problem by lowering the kernel thread's
socket sending timeout from MAX_SCHEDULE_TIMEOUT to HZ/10 and tweaks
the kauditd_send_queue() function to better manage the various audit
queues when connection problems occur between the kernel and the
audit daemon. With this patch, the backlog may temporarily grow
beyond the defined limits when the audit daemon is stopped and the
system is under heavy audit pressure, but kauditd_thread() will
continue to make progress and drain the queues as it would for other
connection problems. For example, with the audit daemon put into a
stopped state and the system configured to audit every syscall it
was still possible to shutdown the system without a kernel panic,
deadlock, etc.; granted, the system was slow to shutdown but that is
to be expected given the extreme pressure of recording every syscall.
The timeout value of HZ/10 was chosen primarily through
experimentation and this developer's "gut feeling". There is likely
no one perfect value, but as this scenario is limited in scope (root
privileges would be needed to send SIGSTOP to the audit daemon), it
is likely not worth exposing this as a tunable at present. This can
always be done at a later date if it proves necessary.
Cc: stable@vger.kernel.org Fixes: 5b52330bbfe63 ("audit: fix auditd/kernel connection state tracking") Reported-by: Gaosheng Cui <cuigaosheng1@huawei.com> Tested-by: Gaosheng Cui <cuigaosheng1@huawei.com> Reviewed-by: Richard Guy Briggs <rgb@redhat.com> Signed-off-by: Paul Moore <paul@paul-moore.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On s390, recordmcount.pl is looking for "bcrl 0,<xxx>" instructions in
the objdump -d outpout. However since binutils 2.37, objdump -d
display "jgnop <xxx>" for the same instruction. Update the
mcount_regex so that it accepts both.
In this function "c->off" is a u32 and "size" is a long. On 64bit systems
if "c->off" is greater than "size" then "size - c->off" is a negative and
we always return -E2BIG. But on 32bit systems the subtraction is type
promoted to a high positive u32 value and basically any "c->len" is
accepted.
Fixes: 4c8cf31885f6 ("vhost: introduce vDPA-based backend") Reported-by: Xie Yongji <xieyongji@bytedance.com> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Link: https://lore.kernel.org/r/20211208103337.GA4047@kili Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
virtio_max_dma_size() returns the maximum DMA mapping size of the virtio
device by querying dma_max_mapping_size() for the device when the DMA
API is in use for the vring. Unfortunately, the device passed is
initialised by register_virtio_device() and does not inherit the DMA
configuration from its parent, resulting in SWIOTLB errors when bouncing
is enabled and the default 256K mapping limit (IO_TLB_SEGSIZE) is not
respected:
| virtio-pci 0000:00:01.0: swiotlb buffer is full (sz: 294912 bytes), total 1024 (slots), used 725 (slots)
Follow the pattern used elsewhere in the virtio_ring code when calling
into the DMA layer and pass the parent device to dma_max_mapping_size()
instead.
Cc: Marc Zyngier <maz@kernel.org> Cc: Quentin Perret <qperret@google.com> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Jason Wang <jasowang@redhat.com> Signed-off-by: Will Deacon <will@kernel.org> Link: https://lore.kernel.org/r/20211201112018.25276-1-will@kernel.org Acked-by: Jason Wang <jasowang@redhat.com> Tested-by: Suzuki K Poulose <suzuki.poulose@arm.com> Fixes: e6d6dd6c875e ("virtio: Introduce virtio_max_dma_size()") Cc: Joerg Roedel <jroedel@suse.de> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Steven Price <steven.price@arm.com> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com> Cc: stable@vger.kernel.org Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Make the bounds propagation in __reg_assign_32_into_64() slightly more
robust and readable by aligning it similarly as we did back in the
__reg_combine_64_into_32() counterpart. Meaning, only propagate or
pessimize them as a smin/smax pair.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Reviewed-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For the case where both s32_{min,max}_value bounds are positive, the
__reg_assign_32_into_64() directly propagates them to their 64 bit
counterparts, otherwise it pessimises them into [0,u32_max] universe and
tries to refine them later on by learning through the tnum as per comment
in mentioned function. However, that does not always happen, for example,
in mov32 operation we call zext_32_to_64(dst_reg) which invokes the
__reg_assign_32_into_64() as is without subsequent bounds update as
elsewhere thus no refinement based on tnum takes place.
Thus, not calling into the __update_reg_bounds() / __reg_deduce_bounds() /
__reg_bound_offset() triplet as we do, for example, in case of ALU ops via
adjust_scalar_min_max_vals(), will lead to more pessimistic bounds when
dumping the full register state:
Technically, the smin_value=0 and smax_value=4294967295 bounds are not
incorrect, but given the register is still a constant, they break assumptions
about const scalars that smin_value == smax_value and umin_value == umax_value.
Without the smin_value == smax_value and umin_value == umax_value invariant
being intact for const scalars, it is possible to leak out kernel pointers
from unprivileged user space if the latter is enabled. For example, when such
registers are involved in pointer arithmtics, then adjust_ptr_min_max_vals()
will taint the destination register into an unknown scalar, and the latter
can be exported and stored e.g. into a BPF map value.
Fixes: 3f50f132d840 ("bpf: Verifier, do explicit ALU32 bounds tracking") Reported-by: Kuee K1r0a <liulin063@gmail.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Reviewed-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Without the bound checks for scpi_pd->name, it could result in the buffer
overflow when copying the SCPI device name from the corresponding device
tree node as the name string is set at maximum size of 30.
Let us fix it by using devm_kasprintf so that the string buffer is
allocated dynamically.
Fixes: 8bec4337ad40 ("firmware: scpi: add device power domain support using genpd") Reported-by: Pedro Batista <pedbap.g@gmail.com> Signed-off-by: Sudeep Holla <sudeep.holla@arm.com> Cc: stable@vger.kernel.org Cc: Cristian Marussi <cristian.marussi@arm.com> Link: https://lore.kernel.org/r/20211209120456.696879-1-sudeep.holla@arm.com' Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Sending them out on a different queue can cause a race condition where a
number of packets in the queue may be discarded by the receiver, because
the ADDBA request is sent too early.
This affects any driver with software A-MPDU setup which does not allocate
packet seqno in hardware on tx, regardless of whether iTXQ is used or not.
The only driver I've seen that explicitly deals with this issue internally
is mwl8k.
Cc: stable@vger.kernel.org Signed-off-by: Felix Fietkau <nbd@nbd.name> Link: https://lore.kernel.org/r/20211202124533.80388-1-nbd@nbd.name Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mark TXQs as having seen transmit while they were stopped if
we bail out of drv_wake_tx_queue() due to reconfig, so that
the queue wake after this will make them catch up. This is
particularly necessary for when TXQs are used for management
packets since those TXQs won't see a lot of traffic that'd
make them catch up later.
Cc: stable@vger.kernel.org Fixes: 4856bfd23098 ("mac80211: do not call driver wake_tx_queue op during reconfig") Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211129152938.4573a221c0e1.I0d1d5daea3089be3fc0dccc92991b0f8c5677f0c@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some drivers that do their own sequence number allocation (e.g. ath9k) rely
on being able to modify params->ssn on starting tx ampdu sessions.
This was broken by a change that modified it to use sta->tid_seq[tid] instead.
Cc: stable@vger.kernel.org Fixes: 31d8bb4e07f8 ("mac80211: agg-tx: refactor sending addba") Reported-by: Eneas U de Queiroz <cotequeiroz@gmail.com> Signed-off-by: Felix Fietkau <nbd@nbd.name> Link: https://lore.kernel.org/r/20211124094024.43222-1-nbd@nbd.name Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is not an unrecoverable situation. Users of kvm_read_guest_offset_cached
and kvm_write_guest_offset_cached must expect the read/write to fail, and
therefore it is possible to just return early with an error value.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
With the elevated 'KVM_CAP_MAX_VCPUS' value kvm_create_max_vcpus test
may hit RLIMIT_NOFILE limits:
# ./kvm_create_max_vcpus
KVM_CAP_MAX_VCPU_ID: 4096
KVM_CAP_MAX_VCPUS: 1024
Testing creating 1024 vCPUs, with IDs 0...1023.
/dev/kvm not available (errno: 24), skipping test
Adjust RLIMIT_NOFILE limits to make sure KVM_CAP_MAX_VCPUS fds can be
opened. Note, raising hard limit ('rlim_max') requires CAP_SYS_RESOURCE
capability which is generally not needed to run kvm selftests (but without
raising the limit the test is doomed to fail anyway).
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20211123135953.667434-1-vkuznets@redhat.com>
[Skip the test if the hard limit can be raised. - Paolo] Reviewed-by: Sean Christopherson <seanjc@google.com> Tested-by: Sean Christopherson <seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>