do {
ret = read(fd, buf, BUF_SIZE);
printf("ret=%d\n",ret);
if (ret < 0) {
perror("write test.jar failed");
}
} while (ret > 0);
free(buf);
close(fd);
}
(3) Compile the source file:
$gcc direct_io_read_file.c -D_GNU_SOURCE
(4) Run the test program:
$./a.out
The result is as following:
ret=1024
ret=1024
ret=952
ret=-1
write test.jar failed: Invalid argument.
I have tested this program on XFS filesystem, XFS does not have
this problem, because XFS use iomap_dio_rw() to do direct I/O
read. And the comparing between read offset and file size is done
in iomap_dio_rw(), the code is as following:
...only when "pos < size", direct I/O can be done, or 0 will be return.
I have tested the fix patch on Ext4, it is up to the mustard of
EINVAL in man2(read) as following:
#include <unistd.h>
ssize_t read(int fd, void *buf, size_t count);
EINVAL
fd is attached to an object which is unsuitable for reading;
or the file was opened with the O_DIRECT flag, and either the
address specified in buf, the value specified in count, or the
current file offset is not suitably aligned.
So I think this patch can be applied to fix ext4 direct I/O error.
However Ext4 introduces direct I/O read using iomap infrastructure
on kernel 5.5, the patch is commit <b1b4705d54ab>
("ext4: introduce direct I/O read using iomap infrastructure"),
then Ext4 will be the same as XFS, they all use iomap_dio_rw() to do direct
I/O read. So this problem does not exist on kernel 5.5 for Ext4.
>From above description, we can see this problem exists on all the kernel
versions between kernel 3.14 and kernel 5.4. It will cause the Applications
to fail to read. For example, when the search service downloads a new full
index file, the search engine is loading the previous index file and is
processing the search request, it can not use buffer io that may squeeze
the previous index file in use from pagecache, so the serch service must
use direct I/O read.
Please apply this patch on these kernel versions, or please use the method
on kernel 5.5 to fix this problem.
Fixes: 9fe55eea7e4b ("Fix race when checking i_size on direct i/o read") Reviewed-by: Jan Kara <jack@suse.cz> Co-developed-by: Wang Long <wanglong19@meituan.com> Signed-off-by: Wang Long <wanglong19@meituan.com> Signed-off-by: Jiang Ying <jiangying8582@126.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The addition of percpu.h to the list of includes in random.h revealed
some circular dependencies on arm64 and possibly other platforms. This
include was added solely for the pseudo-random definitions, which have
nothing to do with the rest of the definitions in this file but are
still there for legacy reasons.
This patch moves the pseudo-random parts to linux/prandom.h and the
percpu.h include with it, which is now guarded by _LINUX_PRANDOM_H and
protected against recursive inclusion.
A further cleanup step would be to remove this from <linux/random.h>
entirely, and make people who use the prandom infrastructure include
just the new header file. That's a bit of a churn patch, but grepping
for "prandom_" and "next_pseudo_random32" "struct rnd_state" should
catch most users.
But it turns out that that nice cleanup step is fairly painful, because
a _lot_ of code currently seems to depend on the implicit include of
<linux/random.h>, which can currently come in a lot of ways, including
such fairly core headfers as <linux/net.h>.
So the "nice cleanup" part may or may never happen.
Fixes: 1c9df907da83 ("random: fix circular include dependency on arm64 after addition of percpu.h") Tested-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Willy Tarreau <w@1wt.eu> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It turns out that the plugin right now ends up being really unhappy
about the change from 'static' to 'extern' storage that happened in
commit f227e3ec3b5c ("random32: update the net random state on interrupt
and activity").
This is probably a trivial fix for the latent_entropy plugin, but for
now, just remove net_rand_state from the list of things the plugin
worries about.
Daniel Díaz and Kees Cook independently reported that commit f227e3ec3b5c ("random32: update the net random state on interrupt and
activity") broke arm64 due to a circular dependency on include files
since the addition of percpu.h in random.h.
The correct fix would definitely be to move all the prandom32 stuff out
of random.h but for backporting, a smaller solution is preferred.
This one replaces linux/percpu.h with asm/percpu.h, and this fixes the
problem on x86_64, arm64, arm, and mips. Note that moving percpu.h
around didn't change anything and that removing it entirely broke
differently. When backporting, such options might still be considered
if this patch fails to help.
[ It turns out that an alternate fix seems to be to just remove the
troublesome <asm/pointer_auth.h> remove from the arm64 <asm/smp.h>
that causes the circular dependency.
But we might as well do the whole belt-and-suspenders thing, and
minimize inclusion in <linux/random.h> too. Either will fix the
problem, and both are good changes. - Linus ]
Reported-by: Daniel Díaz <daniel.diaz@linaro.org> Reported-by: Kees Cook <keescook@chromium.org> Tested-by: Marc Zyngier <maz@kernel.org> Fixes: f227e3ec3b5c Cc: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Willy Tarreau <w@1wt.eu> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix build error for the case:
defined(CONFIG_SMP) && !defined(CONFIG_CPU_V6)
config: keystone_defconfig
CC arch/arm/kernel/signal.o
In file included from ../include/linux/random.h:14,
from ../arch/arm/kernel/signal.c:8:
../arch/arm/include/asm/percpu.h: In function ‘__my_cpu_offset’:
../arch/arm/include/asm/percpu.h:29:34: error: ‘current_stack_pointer’ undeclared (first use in this function); did you mean ‘user_stack_pointer’?
: "Q" (*(const unsigned long *)current_stack_pointer));
^~~~~~~~~~~~~~~~~~~~~
user_stack_pointer
Fixes: f227e3ec3b5c ("random32: update the net random state on interrupt and activity") Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This modifies the first 32 bits out of the 128 bits of a random CPU's
net_rand_state on interrupt or CPU activity to complicate remote
observations that could lead to guessing the network RNG's internal
state.
Note that depending on some network devices' interrupt rate moderation
or binding, this re-seeding might happen on every packet or even almost
never.
In addition, with NOHZ some CPUs might not even get timer interrupts,
leaving their local state rarely updated, while they are running
networked processes making use of the random state. For this reason, we
also perform this update in update_process_times() in order to at least
update the state when there is user or system activity, since it's the
only case we care about.
Reported-by: Amit Klein <aksecurity@gmail.com> Suggested-by: Linus Torvalds <torvalds@linux-foundation.org> Cc: Eric Dumazet <edumazet@google.com> Cc: "Jason A. Donenfeld" <Jason@zx2c4.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Kees Cook <keescook@chromium.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: <stable@vger.kernel.org> Signed-off-by: Willy Tarreau <w@1wt.eu> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The reason for this is a printk() in the i8259 interrupt chip driver
which is invoked with the irq descriptor lock held, which reverses the
lock operations vs. printk() from arbitrary contexts.
Switch the printk() to printk_deferred() to avoid that.
Prevent setting the tscdeadline timer if the lapic is hw disabled.
Fixes: bce87cce88 (KVM: x86: consolidate different ways to test for in-kernel LAPIC) Cc: <stable@vger.kernel.org> Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Message-Id: <1596165141-28874-1-git-send-email-wanpengli@tencent.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There's a potential race in xennet_remove(); this is what the driver is
doing upon unregistering a network device:
1. state = read bus state
2. if state is not "Closed":
3. request to set state to "Closing"
4. wait for state to be set to "Closing"
5. request to set state to "Closed"
6. wait for state to be set to "Closed"
If the state changes to "Closed" immediately after step 1 we are stuck
forever in step 4, because the state will never go back from "Closed" to
"Closing".
Make sure to check also for state == "Closed" in step 4 to prevent the
deadlock.
Also add a 5 sec timeout any time we wait for the bus state to change,
to avoid getting stuck forever in wait_event().
Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
In the implementation of uld_send(), the skb is consumed on all
execution paths except one. Release skb when returning NET_XMIT_DROP.
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The ORC unwinder fails to unwind newly forked tasks which haven't yet
run on the CPU. It correctly reads the 'ret_from_fork' instruction
pointer from the stack, but it incorrectly interprets that value as a
call stack address rather than a "signal" one, so the address gets
incorrectly decremented in the call to orc_find(), resulting in bad ORC
data.
Fix it by forcing 'ret_from_fork' frames to be signal frames.
There are two issues with "i2c: cadence: Fix the hold bit setting" commit.
1. In case of combined message request from user space, when the HOLD
bit is cleared in cdns_i2c_mrecv function, a STOP condition is sent
on the bus even before the last message is started. This is because when
the HOLD bit is cleared, the FIFOS are empty and there is no pending
transfer. The STOP condition should occur only after the last message
is completed.
2. The code added by the commit is redundant. Driver is handling the
setting/clearing of HOLD bit in right way before the commit.
The setting of HOLD bit based on 'bus_hold_flag' is taken care in
cdns_i2c_master_xfer function even before cdns_i2c_msend/cdns_i2c_recv
functions.
The clearing of HOLD bit is taken care at the end of cdns_i2c_msend and
cdns_i2c_recv functions based on bus_hold_flag and byte count.
Since clearing of HOLD bit is done after the slave address is written to
the register (writing to address register triggers the message transfer),
it is ensured that STOP condition occurs at the right time after
completion of the pending transfer (last message).
Signed-off-by: Raviteja Narayanam <raviteja.narayanam@xilinx.com> Acked-by: Michal Simek <michal.simek@xilinx.com> Signed-off-by: Wolfram Sang <wsa@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
According to the report of [1], this driver is possible to cause
the following error in ravb_tx_timeout_work().
ravb e6800000.ethernet ethernet: failed to switch device to config mode
This error means that the hardware could not change the state
from "Operation" to "Configuration" while some tx and/or rx queue
are operating. After that, ravb_config() in ravb_dmac_init() will fail,
and then any descriptors will be not allocaled anymore so that NULL
pointer dereference happens after that on ravb_start_xmit().
To fix the issue, the ravb_tx_timeout_work() should check
the return values of ravb_stop_dma() and ravb_dmac_init().
If ravb_stop_dma() fails, ravb_tx_timeout_work() re-enables TX and RX
and just exits. If ravb_dmac_init() fails, just exits.
The kernel test bot reported[1] that using set_mask_bits on a u8 causes
the following issue on parisc:
hppa-linux-ld: drivers/phy/ti/phy-tusb1210.o: in function `tusb1210_probe':
>> (.text+0x2f4): undefined reference to `__cmpxchg_called_with_bad_pointer'
>> hppa-linux-ld: (.text+0x324): undefined reference to `__cmpxchg_called_with_bad_pointer'
hppa-linux-ld: (.text+0x354): undefined reference to `__cmpxchg_called_with_bad_pointer'
The implementation of s3fwrn5_recv_frame() is supposed to consume skb on
all execution paths. Release skb before returning -ENODEV.
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
This is likely firmware causing this but its starting to annoy customers.
Change the message level to verbose to prevent the spam.
Note that this seems to only show up with ISCSI enabled on the HBA via the
qedi driver.
Signed-off-by: Laurence Oberman <loberman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
On sparc32, tcflag_t is "unsigned long", unlike on all other
architectures, where it is "unsigned int":
drivers/net/usb/hso.c: In function ‘hso_serial_set_termios’:
include/linux/kern_levels.h:5:18: warning: format ‘%d’ expects argument of type ‘unsigned int’, but argument 4 has type ‘tcflag_t {aka long unsigned int}’ [-Wformat=]
drivers/net/usb/hso.c:1393:3: note: in expansion of macro ‘hso_dbg’
hso_dbg(0x16, "Termios called with: cflags new[%d] - old[%d]\n",
^~~~~~~
include/linux/kern_levels.h:5:18: warning: format ‘%d’ expects argument of type ‘unsigned int’, but argument 5 has type ‘tcflag_t {aka long unsigned int}’ [-Wformat=]
drivers/net/usb/hso.c:1393:3: note: in expansion of macro ‘hso_dbg’
hso_dbg(0x16, "Termios called with: cflags new[%d] - old[%d]\n",
^~~~~~~
As "unsigned long" is 32-bit on sparc32, fix this by casting all tcflag_t
parameters to "unsigned int".
While at it, use "%u" to format unsigned numbers.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The function invokes bpf_prog_inc(), which increases the reference
count of a bpf_prog object "rq->xdp_prog" if the object isn't NULL.
The refcount leak issues take place in two error handling paths. When
either mlx5_wq_ll_create() or mlx5_wq_cyc_create() fails, the function
simply returns the error code and forgets to drop the reference count
increased earlier, causing a reference count leak of "rq->xdp_prog".
Fix this issue by jumping to the error handling path err_rq_wq_destroy
while either function fails.
Fixes: 422d4c401edd ("net/mlx5e: RX, Split WQ objects for different RQ types") Signed-off-by: Xin Xiong <xiongx18@fudan.edu.cn> Signed-off-by: Xiyu Yang <xiyuyang19@fudan.edu.cn> Signed-off-by: Xin Tan <tanxin.ctf@gmail.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix the missing clk_disable_unprepare() before return
from gemini_ethernet_port_probe() in the error handling case.
Fixes: 4d5ae32f5e1e ("net: ethernet: Add a driver for Gemini gigabit ethernet") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Wang Hai <wanghai38@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix kernel oops observed when an ext adv data is larger than 31 bytes.
This can be reproduced by setting up an advertiser with advertisement
larger than 31 bytes. The issue is not sensitive to the advertisement
content. In particular, this was reproduced with an advertisement of
229 bytes filled with 'A'. See stack trace below.
This is fixed by not catching ext_adv as legacy adv are only cached to
be able to concatenate a scanable adv with its scan response before
sending it up through mgmt.
Although iph is expected to point to at least 20 bytes of valid memory,
ihl may be bogus, for example on reception of a corrupt packet. If it
happens to be less than 5, we really don't want to run away and
dereference 16GB worth of memory until it wraps back to exactly zero...
Fixes: 0e455d8e80aa ("arm64: Implement optimised IP checksum helpers") Reported-by: guodeqing <geffrey.guo@huawei.com> Signed-off-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Will Deacon <will@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit f7b93d42945c ("arm64/alternatives: use subsections for replacement
sequences") breaks LLVM's integrated assembler, because due to its
one-pass design, it cannot compute instruction sequence lengths before the
layout for the subsection has been finalized. This change fixes the build
by moving the .org directives inside the subsection, so they are processed
after the subsection layout is known.
A mpath object can hold reference on a list of skb that are waiting for
mpath resolution to be sent. When destroying a mpath this skb list
should be cleaned up in order to not leak memory.
Fix HASH_OF_MAPS bug of not putting inner map pointer on bpf_map_elem_update()
operation. This is due to per-cpu extra_elems optimization, which bypassed
free_htab_elem() logic doing proper clean ups. Make sure that inner map is put
properly in optimized case as well.
Fixes: 8c290e60fa2a ("bpf: fix hashmap extra_elems logic") Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Song Liu <songliubraving@fb.com> Link: https://lore.kernel.org/bpf/20200729040913.2815687-1-andriin@fb.com Signed-off-by: Sasha Levin <sashal@kernel.org>
RX queue IRQ mappings are disposed in both the TX IRQ and RX IRQ
error paths. Fix this and dispose of TX IRQ mappings correctly in
case of an error.
Fixes: ea22d51a7831 ("ibmvnic: simplify and improve driver probe function") Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The lifetime of EMAD transactions (i.e., 'struct mlxsw_reg_trans') is
managed using RCU. They are freed using kfree_rcu() once the transaction
ends.
However, in case the transaction failed it is freed immediately after being
removed from the active transactions list. This is problematic because it is
still possible for a different CPU to dereference the transaction from an RCU
read-side critical section while traversing the active transaction list in
mlxsw_emad_rx_listener_func(). In which case, a use-after-free is triggered
[1].
Fix this by freeing the transaction after a grace period by calling
kfree_rcu().
[1]
BUG: KASAN: use-after-free in mlxsw_emad_rx_listener_func+0x969/0xac0 drivers/net/ethernet/mellanox/mlxsw/core.c:671
Read of size 8 at addr ffff88800b7964e8 by task syz-executor.2/2881
Memory state around the buggy address: ffff88800b796380: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff88800b796400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff88800b796480: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^ ffff88800b796500: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88800b796580: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
Fixes: caf7297e7ab5 ("mlxsw: core: Introduce support for asynchronous EMAD register access") Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The lifetime of the Rx listener item ('rxl_item') is managed using RCU,
but is dereferenced outside of RCU read-side critical section, which can
lead to a use-after-free.
Fix this by increasing the scope of the RCU read-side critical section.
Fixes: 93c1edb27f9e ("mlxsw: Introduce Mellanox switch driver core") Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
This is a very similar problem to what had been fixed in:
commit 0d98ba8d70b0 ("scsi: hpsa: disable device during shutdown")
to address https://bugzilla.kernel.org/show_bug.cgi?id=199779.
Fixes: 2ba5fbd62b25 ("net/mlx4_core: Handle AER flow properly") Reported-by: Jake Lawrence <lawja@fb.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Reviewed-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The interrupt URB transfer-buffer was never freed on disconnect or after
probe errors.
Fixes: 55d7de9de6c3 ("Microchip's LAN7800 family USB 2/3 to 10/100/1000 Ethernet device driver") Cc: Woojung.Huh@microchip.com <Woojung.Huh@microchip.com> Signed-off-by: Johan Hovold <johan@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Add the missing endpoint sanity check to prevent a NULL-pointer
dereference should a malicious device lack the expected endpoints.
Note that the driver has a broken endpoint-lookup helper,
lan78xx_get_endpoints(), which can end up accepting interfaces in an
altsetting without endpoints as long as *some* altsetting has a bulk-in
and a bulk-out endpoint.
Fixes: 55d7de9de6c3 ("Microchip's LAN7800 family USB 2/3 to 10/100/1000 Ethernet device driver") Cc: Woojung.Huh@microchip.com <Woojung.Huh@microchip.com> Signed-off-by: Johan Hovold <johan@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The slow path for traced system call entries accessed a wrong memory
location to get the number of the maximum allowed system call number.
Renumber the numbered "local" label for the correct location to avoid
collisions with actual local labels.
Signed-off-by: Michael Karcher <kernel@mkarcher.dialup.fu-berlin.de> Tested-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> Fixes: f3a8308864f920d2 ("sh: Add a few missing irqflags tracing markers.") Signed-off-by: Rich Felker <dalias@libc.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Clang 9 threw:
warning: format specifies type 'unsigned short' but the argument has \
type 'int' [-Wformat]
typeflags, PORT_BASE, PORT_BASE + port_off);
Tested: make -C tools/testing/selftests TARGETS="net" run_tests
Fixes: 77f65ebdca50 ("packet: packet fanout rollover during socket overload") Signed-off-by: Tanner Love <tannerlove@google.com> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The signedness of char is implementation-dependent. Some systems
(including PowerPC and ARM) use unsigned char. Clang 9 threw:
warning: result of comparison of constant -1 with expression of type \
'char' is always true [-Wtautological-constant-out-of-range-compare]
&arg_index)) != -1) {
Tested: make -C tools/testing/selftests TARGETS="net" run_tests
Fixes: 16e781224198 ("selftests/net: Add a test to validate behavior of rx timestamps") Signed-off-by: Tanner Love <tannerlove@google.com> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The commits "xfrm: Move dst->path into struct xfrm_dst"
and "net: Create and use new helper xfrm_dst_child()."
changed xfrm bundle handling under the assumption
that xdst->path and dst->child are not a NULL pointer
only if dst->xfrm is not a NULL pointer. That is true
with one exception. If the xfrm hold queue is used
to wait until a SA is installed by the key manager,
we create a dummy bundle without a valid dst->xfrm
pointer. The current xfrm bundle handling crashes
in that case. Fix this by extending the NULL check
of dst->xfrm with a test of the DST_XFRM_QUEUE flag.
Fixes: 0f6c480f23f4 ("xfrm: Move dst->path into struct xfrm_dst") Fixes: b92cf4aab8e6 ("net: Create and use new helper xfrm_dst_child().") Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
x25_connect() invokes x25_get_neigh(), which returns a reference of the
specified x25_neigh object to "x25->neighbour" with increased refcnt.
When x25 connect success and returns, the reference still be hold by
"x25->neighbour", so the refcount should be decreased in
x25_disconnect() to keep refcount balanced.
The reference counting issue happens in x25_disconnect(), which forgets
to decrease the refcnt increased by x25_get_neigh() in x25_connect(),
causing a refcnt leak.
Fix this issue by calling x25_neigh_put() before x25_disconnect()
returns.
Signed-off-by: Xiyu Yang <xiyuyang19@fudan.edu.cn> Signed-off-by: Xin Tan <tanxin.ctf@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The code in xlog_wait uses the spinlock to make adding the task to
the wait queue, and setting the task state to UNINTERRUPTIBLE atomic
with respect to the waker.
Doing the wakeup after releasing the spinlock opens up the following
race condition:
Task 1 task 2
add task to wait queue
wake up task
set task state to UNINTERRUPTIBLE
This issue was found through code inspection as a result of kworkers
being observed stuck in UNINTERRUPTIBLE state with an empty
wait queue. It is rare and largely unreproducable.
Simply moving the spin_unlock to after the wake_up_all results
in the waker not being able to see a task on the waitqueue before
it has set its state to UNINTERRUPTIBLE.
This bug dates back to the conversion of this code to generic
waitqueue infrastructure from a counting semaphore back in 2008
which didn't place the wakeups consistently w.r.t. to the relevant
spin locks.
[dchinner: Also fix a similar issue in the shutdown path on
xc_commit_wait. Update commit log with more details of the issue.]
Fixes: d748c62367eb ("[XFS] Convert l_flushsema to a sv_t") Reported-by: Chris Mason <clm@fb.com> Signed-off-by: Rik van Riel <riel@surriel.com> Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Cc: stable@vger.kernel.org # 4.9.x-4.19.x
[modified for contextual change near xlog_state_do_callback()] Signed-off-by: Samuel Mendoza-Jonas <samjonas@amazon.com> Reviewed-by: Frank van der Linden <fllinden@amazon.com> Reviewed-by: Suraj Jitindar Singh <surajjs@amazon.com> Reviewed-by: Benjamin Herrenschmidt <benh@amazon.com> Reviewed-by: Anchal Agarwal <anchalag@amazon.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
rds_notify_queue_get() is potentially copying uninitialized kernel stack
memory to userspace since the compiler may leave a 4-byte hole at the end
of `cmsg`.
In 2016 we tried to fix this issue by doing `= { 0 };` on `cmsg`, which
unfortunately does not always initialize that 4-byte hole. Fix it by using
memset() instead.
Cc: stable@vger.kernel.org Fixes: f037590fff30 ("rds: fix a leak of kernel memory") Fixes: bdbe6fbc6a2f ("RDS: recv.c") Suggested-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Peilin Ye <yepeilin.cs@gmail.com> Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A use-after-free in drm_gem_open_ioctl can happen if the
GEM object handle is closed between the idr lookup and
retrieving the size from said object since a local reference
is not being held at that point. Hold the local reference
while the object can still be accessed to fix this and
plug the potential security hole.
Compiler leaves a 4-byte hole near the end of `dev_info`, causing
amdgpu_info_ioctl() to copy uninitialized kernel stack memory to userspace
when `size` is greater than 356.
In 2015 we tried to fix this issue by doing `= {};` on `dev_info`, which
unfortunately does not initialize that 4-byte hole. Fix it by using
memset() instead.
Cc: stable@vger.kernel.org Fixes: c193fa91b918 ("drm/amdgpu: information leak in amdgpu_info_ioctl()") Fixes: d38ceaf99ed0 ("drm/amdgpu: add core driver (v4)") Suggested-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Peilin Ye <yepeilin.cs@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unprivileged memory accesses generated by the so-called "translated"
instructions (e.g. LDRT) in kernel mode can cause user watchpoints to fire
unexpectedly. In such cases, the hw_breakpoint logic will invoke the user
overflow handler which will typically raise a SIGTRAP back to the current
task. This is futile when returning back to the kernel because (a) the
signal won't have been delivered and (b) userspace can't handle the thing
anyway.
Avoid invoking the user overflow handler for watchpoints triggered by
kernel uaccess routines, and instead single-step over the faulting
instruction as we would if no overflow handler had been installed.
Cc: <stable@vger.kernel.org> Fixes: f81ef4a920c8 ("ARM: 6356/1: hw-breakpoint: add ARM backend for the hw-breakpoint framework") Reported-by: Luis Machado <luis.machado@linaro.org> Tested-by: Luis Machado <luis.machado@linaro.org> Signed-off-by: Will Deacon <will@kernel.org> Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use offsetof to calculate offset of a field to take advantage of
compiler built-in version when possible, and avoid UBSAN warning when
compiling with Clang:
==================================================================
UBSAN: Undefined behaviour in net/wireless/wext-core.c:525:14
member access within null pointer of type 'struct iw_point'
CPU: 3 PID: 165 Comm: kworker/u16:3 Tainted: G S W 4.19.23 #43
Workqueue: cfg80211 __cfg80211_scan_done [cfg80211]
Call trace:
dump_backtrace+0x0/0x194
show_stack+0x20/0x2c
__dump_stack+0x20/0x28
dump_stack+0x70/0x94
ubsan_epilogue+0x14/0x44
ubsan_type_mismatch_common+0xf4/0xfc
__ubsan_handle_type_mismatch_v1+0x34/0x54
wireless_send_event+0x3cc/0x470
___cfg80211_scan_done+0x13c/0x220 [cfg80211]
__cfg80211_scan_done+0x28/0x34 [cfg80211]
process_one_work+0x170/0x35c
worker_thread+0x254/0x380
kthread+0x13c/0x158
ret_from_fork+0x10/0x18
===================================================================
Signed-off-by: Pi-Hsun Shih <pihsun@chromium.org> Reviewed-by: Nick Desaulniers <ndesaulniers@google.com> Link: https://lore.kernel.org/r/20191204081307.138765-1-pihsun@chromium.org Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Recently ASPM handling was changed to allow ASPM on PCIe-to-PCI/PCI-X
bridges. Unfortunately the ASMedia ASM1083/1085 PCIe to PCI bridge device
doesn't seem to function properly with ASPM enabled. On an Asus PRIME
H270-PRO motherboard, it causes errors like these:
In addition to flooding the kernel log, this also causes the machine to
wake up immediately after suspend is initiated.
The device advertises ASPM L0s and L1 support in the Link Capabilities
register, but the ASMedia web page for ASM1083 [1] claims "No PCIe ASPM
support".
Windows 10 (build 2004) enables L0s, but it also logs correctable PCIe
errors.
Some of the self tests create a test inode, setup some extents and then do
calls to btrfs_get_extent() to test that the corresponding extent maps
exist and are correct. However btrfs_get_extent(), since the 5.2 merge
window, now errors out when it finds a regular or prealloc extent for an
inode that does not correspond to a regular file (its ->i_mode is not
S_IFREG). This causes the self tests to fail sometimes, specially when
KASAN, slub_debug and page poisoning are enabled:
$ modprobe btrfs
modprobe: ERROR: could not insert 'btrfs': Invalid argument
This happens because the test inodes are created without ever initializing
the i_mode field of the inode, and neither VFS's new_inode() nor the btrfs
callback btrfs_alloc_inode() initialize the i_mode. Initialization of the
i_mode is done through the various callbacks used by the VFS to create
new inodes (regular files, directories, symlinks, tmpfiles, etc), which
all call btrfs_new_inode() which in turn calls inode_init_owner(), which
sets the inode's i_mode. Since the tests only uses new_inode() to create
the test inodes, the i_mode was never initialized.
This always happens on a VM I used with kasan, slub_debug and many other
debug facilities enabled. It also happened to someone who reported this
on bugzilla (on a 5.3-rc).
Fix this by setting i_mode to S_IFREG at btrfs_new_test_inode().
Fixes: 6bf9e4bd6a2778 ("btrfs: inode: Verify inode mode to avoid NULL pointer dereference")
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=204397 Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Now when sending packets, sk_mem_charge() and sk_mem_uncharge() have been
used to set sk_forward_alloc. We just need to call sk_wmem_schedule() to
check if the allocated should be raised, and call sk_mem_reclaim() to
check if the allocated should be reduced when it's under memory pressure.
If sk_wmem_schedule() returns false, which means no memory is allowed to
allocate, it will block and wait for memory to become available.
Note different from tcp, sctp wait_for_buf happens before allocating any
skb, so memory accounting check is done with the whole msg_len before it
too.
Reported-by: Matteo Croce <mcroce@redhat.com> Tested-by: Matteo Croce <mcroce@redhat.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
For symlink, we don't set BTRFS_I(inode)->io_tree.ops and leave it
empty, as symlink is only designed to have inlined extent, all handled
by tree block read. Thus no need to trigger btrfs_submit_bio_hook() for
inline file extent.
However end_bio_extent_readpage() expects tree->ops populated, as it's
reading regular data extent. This causes NULL pointer dereference.
[FIX]
This patch fixes the problem in two ways:
- Verify inode mode against its dir item when looking up inode
So in btrfs_lookup_dentry() if we find inode mode mismatch with dir
item, we error out so that corrupted inode will not be accessed.
- Verify inode mode when getting extent mapping
Only regular file should have regular or preallocated extent.
If we found regular/preallocated file extent for symlink or
the rest, we error out before submitting the read bio.
With this fix that crafted image can be rejected gracefully:
BTRFS critical (device loop0): inode mode mismatch with dir: inode mode=0121644 btrfs type=7 dir type=1
In htc_config_pipe_credits, htc_setup_complete, and htc_connect_service
if time out happens, the allocated buffer needs to be released.
Otherwise there will be memory leak.
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
In acp_hw_init there are some allocations that needs to be released in
case of failure:
1- adev->acp.acp_genpd should be released if any allocation attemp for
adev->acp.acp_cell, adev->acp.acp_res or i2s_pdata fails.
2- all of those allocations should be released if
mfd_add_hotplug_devices or pm_genpd_add_device fail.
3- Release is needed in case of time out values expire.
Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When executing the following command, we met kernel dump.
dmesg -c > /dev/null; cd /sys;
for i in `ls /sys/kernel/debug/regmap/* -d`; do
echo "Checking regmap in $i";
cat $i/registers;
done && grep -ri "0x02d0" *;
It is because the count value is too big, and kmalloc fails. So add an
upper bound check to allow max size `PAGE_SIZE << (MAX_ORDER - 1)`.
When vlan_newlink call register_vlan_dev fails, it might return error
with dev->reg_state = NETREG_UNREGISTERED. The rtnl_newlink should
free the memory. But currently rtnl_newlink only free the memory which
state is NETREG_UNINITIALIZED.
Currently, SO_REUSEPORT does not work well if connected sockets are in a
UDP reuseport group.
Then reuseport_has_conns() returns true and the result of
reuseport_select_sock() is discarded. Also, unconnected sockets have the
same score, hence only does the first unconnected socket in udp_hslot
always receive all packets sent to unconnected sockets.
So, the result of reuseport_select_sock() should be used for load
balancing.
The noteworthy point is that the unconnected sockets placed after
connected sockets in sock_reuseport.socks will receive more packets than
others because of the algorithm in reuseport_select_sock().
index | connected | reciprocal_scale | result
---------------------------------------------
0 | no | 20% | 40%
1 | no | 20% | 20%
2 | yes | 20% | 0%
3 | no | 20% | 40%
4 | yes | 20% | 0%
If most of the sockets are connected, this can be a problem, but it still
works better than now.
Fixes: acdcecc61285 ("udp: correct reuseport selection with connected sockets") CC: Willem de Bruijn <willemb@google.com> Reviewed-by: Benjamin Herrenschmidt <benh@amazon.com> Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If an unconnected socket in a UDP reuseport group connect()s, has_conns is
set to 1. Then, when a packet is received, udp[46]_lib_lookup2() scans all
sockets in udp_hslot looking for the connected socket with the highest
score.
However, when the number of sockets bound to the port exceeds max_socks,
reuseport_grow() resets has_conns to 0. It can cause udp[46]_lib_lookup2()
to return without scanning all sockets, resulting in that packets sent to
connected sockets may be distributed to unconnected sockets.
Therefore, reuseport_grow() should copy has_conns.
Fixes: acdcecc61285 ("udp: correct reuseport selection with connected sockets") CC: Willem de Bruijn <willemb@google.com> Reviewed-by: Benjamin Herrenschmidt <benh@amazon.com> Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When adding a stream with stream reconf, the new stream firstly is in
CLOSED state but new out chunks can still be enqueued. Then once gets
the confirmation from the peer, the state will change to OPEN.
However, if the peer denies, it needs to roll back the stream. But when
doing that, it only sets the stream outcnt back, and the chunks already
in the new stream don't get purged. It caused these chunks can still be
dequeued in sctp_outq_dequeue_data().
As its stream is still in CLOSE, the chunk will be enqueued to the head
again by sctp_outq_head_data(). This chunk will never be sent out, and
the chunks after it can never be dequeued. The assoc will be 'hung' in
a dead loop of sending this chunk.
To fix it, this patch is to purge these chunks already in the new
stream by calling sctp_stream_shrink_out() when failing to do the
addstream reconf.
Fixes: 11ae76e67a17 ("sctp: implement receiver-side procedures for the Reconf Response Parameter") Reported-by: Ying Xu <yinxu@redhat.com> Signed-off-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It's not necessary to go list_for_each for outq->out_chunk_list
when new outcnt >= old outcnt, as no chunk with higher sid than
new (outcnt - 1) exists in the outqueue.
While at it, also move the list_for_each code in a new function
sctp_stream_shrink_out(), which will be used in the next patch.
Signed-off-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We recently added some bounds checking in ax25_connect() and
ax25_sendmsg() and we so we removed the AX25_MAX_DIGIS checks because
they were no longer required.
Unfortunately, I believe they are required to prevent integer overflows
so I have added them back.
Fixes: 8885bb0621f0 ("AX.25: Prevent out-of-bounds read in ax25_sendmsg()") Fixes: 2f2a7ffad5c6 ("AX.25: Fix out-of-bounds read in ax25_connect()") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Previously TLP may send multiple probes of new data in one
flight. This happens when the sender is cwnd limited. After the
initial TLP containing new data is sent, the sender receives another
ACK that acks partial inflight. It may re-arm another TLP timer
to send more, if no further ACK returns before the next TLP timeout
(PTO) expires. The sender may send in theory a large amount of TLP
until send queue is depleted. This only happens if the sender sees
such irregular uncommon ACK pattern. But it is generally undesirable
behavior during congestion especially.
The original TLP design restrict only one TLP probe per inflight as
published in "Reducing Web Latency: the Virtue of Gentle Aggression",
SIGCOMM 2013. This patch changes TLP to send at most one probe
per inflight.
Note that if the sender is app-limited, TLP retransmits old data
and did not have this issue.
Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
rxrpc_sendmsg() returns EPIPE if there's an outstanding error, such as if
rxrpc_recvmsg() indicating ENODATA if there's nothing for it to read.
Change rxrpc_recvmsg() to return EAGAIN instead if there's nothing to read
as this particular error doesn't get stored in ->sk_err by the networking
core.
Also change rxrpc_sendmsg() so that it doesn't fail with delayed receive
errors (there's no way for it to report which call, if any, the error was
caused by).
Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both") Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We have to detach sock from socket in qrtr_release(),
otherwise skb->sk may still reference to this socket
when the skb is released in tun->queue, particularly
sk->sk_wq still points to &sock->wq, which leads to
a UAF.
Reported-and-tested-by: syzbot+6720d64f31c081c2f708@syzkaller.appspotmail.com Fixes: 28fb4e59a47d ("net: qrtr: Expose tunneling endpoint to user space") Cc: Bjorn Andersson <bjorn.andersson@linaro.org> Cc: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We can't use IS_UDPLITE to replace udp_sk->pcflag when UDPLITE_RECV_CC is
checked.
Fixes: b2bf1e2659b1 ("[UDP]: Clean up for IS_UDPLITE macro") Signed-off-by: Miaohe Lin <linmiaohe@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ip6gre_tunnel_uninit() has set 'ign->fb_tunnel_dev' to NULL, later
access to ign->fb_tunnel_dev cause null-ptr-deref. Fix it by saving
'ign->fb_tunnel_dev' to local variable ndev.
Fixes: dafabb6590cb ("ip6_gre: fix use-after-free in ip6gre_tunnel_lookup()") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This driver is not working because of problems of its receiving code.
This patch fixes it to make it work.
When the driver receives an LAPB frame, it should first pass the frame
to the LAPB module to process. After processing, the LAPB module passes
the data (the packet) back to the driver, the driver should then add a
one-byte pseudo header and pass the data to upper layers.
The changes to the "x25_asy_bump" function and the
"x25_asy_data_indication" function are to correctly implement this
procedure.
Also, the "x25_asy_unesc" function ignores any frame that is shorter
than 3 bytes. However the shortest frames are 2-byte long. So we need
to change it to allow 2-byte frames to pass.
Cc: Eric Dumazet <edumazet@google.com> Cc: Martin Schiller <ms@dev.tdt.de> Signed-off-by: Xie He <xie.he.0141@gmail.com> Reviewed-by: Martin Schiller <ms@dev.tdt.de> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
IRQs are disabled when freeing skbs in input queue.
Use the IRQ safe variant to free skbs here.
Fixes: 145dd5f9c88f ("net: flush the softnet backlog in process context") Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Checks on `addr_len` and `usax->sax25_ndigis` are insufficient.
ax25_sendmsg() can go out of bounds when `usax->sax25_ndigis` equals to 7
or 8. Fix it.
It is safe to remove `usax->sax25_ndigis > AX25_MAX_DIGIS`, since
`addr_len` is guaranteed to be less than or equal to
`sizeof(struct full_sockaddr_ax25)`
Signed-off-by: Peilin Ye <yepeilin.cs@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Checks on `addr_len` and `fsa->fsa_ax25.sax25_ndigis` are insufficient.
ax25_connect() can go out of bounds when `fsa->fsa_ax25.sax25_ndigis`
equals to 7 or 8. Fix it.
This issue has been reported as a KMSAN uninit-value bug, because in such
a case, ax25_connect() reaches into the uninitialized portion of the
`struct sockaddr_storage` statically allocated in __sys_connect().
It is safe to remove `fsa->fsa_ax25.sax25_ndigis > AX25_MAX_DIGIS` because
`addr_len` is guaranteed to be less than or equal to
`sizeof(struct full_sockaddr_ax25)`.
Reported-by: syzbot+c82752228ed975b0a623@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?id=55ef9d629f3b3d7d70b69558015b63b48d01af66 Signed-off-by: Peilin Ye <yepeilin.cs@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fix allows ath9k_htc modules to connect to WLAN once again.
Fixes: 2bbcaaee1fcb ("ath9k: Fix general protection fault in ath9k_hif_usb_rx_cb") Link: https://bugzilla.kernel.org/show_bug.cgi?id=208251 Signed-off-by: Mark O'Donovan <shiftee@posteo.net> Reported-by: Roman Mamedov <rm@romanrm.net> Tested-by: Viktor Jägersküpper <viktor_jaegerskuepper@freenet.de> Signed-off-by: Kalle Valo <kvalo@codeaurora.org> Link: https://lore.kernel.org/r/20200711043324.8079-1-shiftee@posteo.net Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In ath9k_hif_usb_rx_cb interface number is assumed to be 0.
usb_ifnum_to_if(urb->dev, 0)
But it isn't always true.
The case reported by syzbot:
https://lore.kernel.org/linux-usb/000000000000666c9c05a1c05d12@google.com
usb 2-1: new high-speed USB device number 2 using dummy_hcd
usb 2-1: config 1 has an invalid interface number: 2 but max is 0
usb 2-1: config 1 has no interface number 0
usb 2-1: New USB device found, idVendor=0cf3, idProduct=9271, bcdDevice=
1.08
usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
general protection fault, probably for non-canonical address
0xdffffc0000000015: 0000 [#1] SMP KASAN
KASAN: null-ptr-deref in range [0x00000000000000a8-0x00000000000000af]
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.6.0-rc5-syzkaller #0
The problem is dm_suspended() returns true not only during suspend,
but also during resume. So this race condition could occur:
1. dm_integrity_resume calls queue_work(ic->recalc_wq, &ic->recalc_work)
2. integrity_recalc (&ic->recalc_work) preempts the current thread
3. integrity_recalc calls if (unlikely(dm_suspended(ic->ti))) goto unlock_ret;
4. integrity_recalc exits and no recalculating is done.
To fix this race condition, add a function dm_post_suspending that is
only true during the postsuspend phase and use it instead of
dm_suspended().
When building on allyesconfig kernel for a NO_DMA=y platform (e.g.
Sun-3), CONFIG_SND_SOC_QCOM_COMMON=y, but CONFIG_SND_SOC_QDSP6_AFE=n,
leading to a link failure:
sound/soc/qcom/common.o: In function `qcom_snd_parse_of':
common.c:(.text+0x2e2): undefined reference to `q6afe_is_rx_port'
While SND_SOC_QDSP6 depends on HAS_DMA, SND_SOC_MSM8996 and SND_SOC_SDM845
don't, so the following warning is seen:
Until recently, this warning was harmless (from a compile-testing
point-of-view), but the new user of q6afe_is_rx_port() turned this into
a hard failure.
As the QDSP6 driver itself builds fine if NO_DMA=y, and it depends on
QCOM_APR (which in turns depends on ARCH_QCOM || COMPILE_TEST), it is
safe to increase compile testing coverage. Hence fix the link failure
by dropping the HAS_DMA dependency of SND_SOC_QDSP6.
The Lenovo Miix 2 10 has a keyboard dock with extra speakers in the dock.
Rather then the ACL5672's GPIO1 pin being used as IRQ to the CPU, it is
actually used to enable the amplifier for these speakers
(the IRQ to the CPU comes directly from the jack-detect switch).
Add a quirk for having an ext speaker-amplifier enable pin on GPIO1
and replace the Lenovo Miix 2 10's dmi_system_id table entry's wrong
GPIO_DEV quirk (which needs to be renamed to GPIO1_IS_IRQ) with the
new RT5670_GPIO1_IS_EXT_SPK_EN quirk, so that we enable the external
speaker-amplifier as necessary.
Also update the ident field for the dmi_system_id table entry, the
Miix models are not Thinkpads.
On x86-32 the idt_table with 256 entries needs only 2048 bytes. It is
page-aligned, but the end of the .bss..page_aligned section is not
guaranteed to be page-aligned.
As a result, objects from other .bss sections may end up on the same 4k
page as the idt_table, and will accidentially get mapped read-only during
boot, causing unexpected page-faults when the kernel writes to them.
This could be worked around by making the objects in the page aligned
sections page sized, but that's wrong.
Explicit sections which store only page aligned objects have an implicit
guarantee that the object is alone in the page in which it is placed. That
works for all objects except the last one. That's inconsistent.
Enforcing page sized objects for these sections would wreckage memory
sanitizers, because the object becomes artificially larger than it should
be and out of bound access becomes legit.
Align the end of the .bss..page_aligned and .data..page_aligned section on
page-size so all objects places in these sections are guaranteed to have
their own page.
The code was stuck in this loop in d_alloc_parallel:
4037d414: 0e 00 10 dc ldd 0(r16),ret0 4037d418: c7 fc 5f ed bb,< ret0,1f,4037d414 <d_alloc_parallel+0x384> 4037d41c: 08 00 02 40 nop
This is the inner loop of bit_spin_lock which is called by hlist_bl_unlock in
d_alloc_parallel:
static inline void bit_spin_lock(int bitnum, unsigned long *addr)
{
/*
* Assuming the lock is uncontended, this never enters
* the body of the outer loop. If it is contended, then
* within the inner loop a non-atomic test is used to
* busywait with less bus contention for a good time to
* attempt to acquire the lock bit.
*/
preempt_disable();
#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
while (unlikely(test_and_set_bit_lock(bitnum, addr))) {
preempt_enable();
do {
cpu_relax();
} while (test_bit(bitnum, addr));
preempt_disable();
}
#endif
__acquire(bitlock);
}
After consideration, I realized that we must be losing bit unlocks.
Then, I noticed that we missed defining atomic64_set_release().
Adding this define fixes the stalls in bit operations.
Avoid kernel crash when vddci_control is SMU7_VOLTAGE_CONTROL_NONE and
vddci_voltage_table is empty. It has been tested on Intel Hades Canyon
(i7-8809G).
Bug: https://bugzilla.kernel.org/show_bug.cgi?id=208489 Fixes: ac7822b0026f ("drm/amd/powerplay: add smumgr support for VEGAM (v2)") Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Qiu Wenbo <qiuwenbo@phytium.com.cn> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
NULL dereference occurs when string that is not ended with space or
newline is written to some dpm sysfs interface (for example pp_dpm_sclk).
This happens because strsep replaces the tmp with NULL if the delimiter
is not present in string, which is then dereferenced by tmp[0].
Reproduction example:
sudo sh -c 'echo -n 1 > /sys/class/drm/card0/device/pp_dpm_sclk'
Signed-off-by: Paweł Gronowski <me@woland.xyz> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixes: cafaf14a5d8f ("io-mapping: Always create a struct to hold metadata about the io-mapping") Signed-off-by: Michael J. Ruhl <michael.j.ruhl@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Mike Rapoport <rppt@linux.ibm.com> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: <stable@vger.kernel.org> Link: http://lkml.kernel.org/r/20200721171936.81563-1-michael.j.ruhl@intel.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the kmem_cache refcount is greater than one, we should not mark the
root kmem_cache as dying. If we mark the root kmem_cache dying
incorrectly, the non-root kmem_cache can never be destroyed. It
resulted in memory leak when memcg was destroyed. We can use the
following steps to reproduce.
1) Use kmem_cache_create() to create a new kmem_cache named A.
2) Coincidentally, the kmem_cache A is an alias for kmem_cache B,
so the refcount of B is just increased.
3) Use kmem_cache_destroy() to destroy the kmem_cache A, just
decrease the B's refcount but mark the B as dying.
4) Create a new memory cgroup and alloc memory from the kmem_cache
B. It leads to create a non-root kmem_cache for allocating memory.
5) When destroy the memory cgroup created in the step 4), the
non-root kmem_cache can never be destroyed.
If we repeat steps 4) and 5), this will cause a lot of memory leak. So
only when refcount reach zero, we mark the root kmem_cache as dying.
Fixes: 92ee383f6daa ("mm: fix race between kmem_cache destroy, create and deactivate") Signed-off-by: Muchun Song <songmuchun@bytedance.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: Shakeel Butt <shakeelb@google.com> Acked-by: Roman Gushchin <guro@fb.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: <stable@vger.kernel.org> Link: http://lkml.kernel.org/r/20200716165103.83462-1-songmuchun@bytedance.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently the memcg_params.dying flag and the corresponding workqueue used
for the asynchronous deactivation of kmem_caches is synchronized using the
slab_mutex.
It makes impossible to check this flag from the irq context, which will be
required in order to implement asynchronous release of kmem_caches.
So let's switch over to the irq-save flavor of the spinlock-based
synchronization.
Link: http://lkml.kernel.org/r/20190611231813.3148843-8-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Vladimir Davydov <vdavydov.dev@gmail.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Cc: Christoph Lameter <cl@linux.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Waiman Long <longman@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Andrei Vagin <avagin@gmail.com> Cc: Qian Cai <cai@lca.pw> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
It was hard to keep a test running, moving tasks between memcgs with
move_charge_at_immigrate, while swapping: mem_cgroup_id_get_many()'s
refcount is discovered to be 0 (supposedly impossible), so it is then
forced to REFCOUNT_SATURATED, and after thousands of warnings in quick
succession, the test is at last put out of misery by being OOM killed.
This is because of the way moved_swap accounting was saved up until the
task move gets completed in __mem_cgroup_clear_mc(), deferred from when
mem_cgroup_move_swap_account() actually exchanged old and new ids.
Concurrent activity can free up swap quicker than the task is scanned,
bringing id refcount down 0 (which should only be possible when
offlining).
Just skip that optimization: do that part of the accounting immediately.
Fixes: 615d66c37c75 ("mm: memcontrol: fix memcg id ref counter on swap charge move") Signed-off-by: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: Alex Shi <alex.shi@linux.alibaba.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Alex Shi <alex.shi@linux.alibaba.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Michal Hocko <mhocko@suse.com> Cc: <stable@vger.kernel.org> Link: http://lkml.kernel.org/r/alpine.LSU.2.11.2007071431050.4726@eggly.anvils Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When CROSS_COMPILE is set (e.g. aarch64-linux-gnu-), if
$(CROSS_COMPILE)elfedit is found at /usr/bin/aarch64-linux-gnu-elfedit,
GCC_TOOLCHAIN_DIR will be set to /usr/bin/. --prefix= will be set to
/usr/bin/ and Clang as of 11 will search for both
$(prefix)aarch64-linux-gnu-$needle and $(prefix)$needle.
GCC searchs for $(prefix)aarch64-linux-gnu/$version/$needle,
$(prefix)aarch64-linux-gnu/$needle and $(prefix)$needle. In practice,
$(prefix)aarch64-linux-gnu/$needle rarely contains executables.
To better model how GCC's -B/--prefix takes in effect in practice, newer
Clang (since
https://github.com/llvm/llvm-project/commit/3452a0d8c17f7166f479706b293caf6ac76ffd90)
only searches for $(prefix)$needle. Currently it will find /usr/bin/as
instead of /usr/bin/aarch64-linux-gnu-as.
Set --prefix= to $(GCC_TOOLCHAIN_DIR)$(notdir $(CROSS_COMPILE))
(/usr/bin/aarch64-linux-gnu-) so that newer Clang can find the
appropriate cross compiling GNU as (when -no-integrated-as is in
effect).
syzbot is reporting general protection fault in do_con_write() [1] caused
by vc->vc_screenbuf == ZERO_SIZE_PTR caused by vc->vc_screenbuf_size == 0
caused by vc->vc_cols == vc->vc_rows == vc->vc_size_row == 0 caused by
fb_set_var() from ioctl(FBIOPUT_VSCREENINFO) on /dev/fb0 , for
gotoxy(vc, 0, 0) from reset_terminal() from vc_init() from vc_allocate()
from con_install() from tty_init_dev() from tty_open() on such console
causes vc->vc_pos == 0x10000000e due to
((unsigned long) ZERO_SIZE_PTR) + -1U * 0 + (-1U << 1).
I don't think that a console with 0 column or 0 row makes sense. And it
seems that vc_do_resize() does not intend to allow resizing a console to
0 column or 0 row due to
Theoretically, cols and rows can be any range as long as
0 < cols * rows * 2 <= KMALLOC_MAX_SIZE is satisfied (e.g.
cols == 1048576 && rows == 2 is possible) because of
in visual_init() and kzalloc(vc->vc_screenbuf_size) in vc_allocate().
Since we can detect cols == 0 or rows == 0 via screenbuf_size = 0 in
visual_init(), we can reject kzalloc(0). Then, vc_allocate() will return
an error, and con_write() will not be called on a console with 0 column
or 0 row.
We need to make sure that integer overflow in visual_init() won't happen.
Since vc_do_resize() restricts cols <= 32767 and rows <= 32767, applying
1 <= cols <= 32767 and 1 <= rows <= 32767 restrictions to vc_allocate()
will be practically fine.
This patch does not touch con_init(), for returning -EINVAL there
does not help when we are not returning -ENOMEM.
syzbot is reporting general protection fault in bitfill_aligned() [1]
caused by integer underflow in bit_clear_margins(). The cause of this
problem is when and how do_vc_resize() updates vc->vc_{cols,rows}.
If vc_do_resize() fails (e.g. kzalloc() fails) when var.xres or var.yres
is going to shrink, vc->vc_{cols,rows} will not be updated. This allows
bit_clear_margins() to see info->var.xres < (vc->vc_cols * cw) or
info->var.yres < (vc->vc_rows * ch). Unexpectedly large rw or bh will
try to overrun the __iomem region and causes general protection fault.
Also, vc_resize(vc, 0, 0) does not set vc->vc_{cols,rows} = 0 due to
Of course, callers of vc_resize() are not handling vc_do_resize() failure
is bad. But we can't avoid vc_resize(vc, 0, 0) which returns 0. Therefore,
as a band-aid workaround, this patch checks integer underflow in
"struct fbcon_ops"->clear_margins call, assuming that
vc->vc_cols * vc->vc_font.width and vc->vc_rows * vc->vc_font.heigh do not
cause integer overflow.
Commit 7b668c064ec3 ("serial: 8250: Fix max baud limit in generic 8250
port") fixed limits of a baud rate setting for a generic 8250 port.
In other words since that commit the baud rate has been permitted to be
within [uartclk / 16 / UART_DIV_MAX; uartclk / 16], which is absolutely
normal for a standard 8250 UART port. But there are custom 8250 ports,
which provide extended baud rate limits. In particular the Mediatek 8250
port can work with baud rates up to "uartclk" speed.
Normally that and any other peculiarity is supposed to be handled in a
custom set_termios() callback implemented in the vendor-specific
8250-port glue-driver. Currently that is how it's done for the most of
the vendor-specific 8250 ports, but for some reason for Mediatek a
solution has been spread out to both the glue-driver and to the generic
8250-port code. Due to that a bug has been introduced, which permitted the
extended baud rate limit for all even for standard 8250-ports. The bug
has been fixed by the commit 7b668c064ec3 ("serial: 8250: Fix max baud
limit in generic 8250 port") by narrowing the baud rates limit back down to
the normal bounds. Unfortunately by doing so we also broke the
Mediatek-specific extended bauds feature.
A fix of the problem described above is twofold. First since we can't get
back the extended baud rate limits feature to the generic set_termios()
function and that method supports only a standard baud rates range, the
requested baud rate must be locally stored before calling it and then
restored back to the new termios structure after the generic set_termios()
finished its magic business. By doing so we still use the
serial8250_do_set_termios() method to set the LCR/MCR/FCR/etc. registers,
while the extended baud rate setting procedure will be performed later in
the custom Mediatek-specific set_termios() callback. Second since a true
baud rate is now fully calculated in the custom set_termios() method we
need to locally update the port timeout by calling the
uart_update_timeout() function. After the fixes described above are
implemented in the 8250_mtk.c driver, the Mediatek 8250-port should
get back to normally working with extended baud rates.
SERIAL_PORT_DFNS is not defined on each arch, if it's not defined,
serial8250_set_defaults() won't be called in serial8250_isa_init_ports(),
so the p->serial_in pointer won't be initialized, and it leads a null-ptr-deref.
Fix this problem by calling serial8250_set_defaults() after init uart port.
The `INSN_CONFIG` comedi instruction with sub-instruction code
`INSN_CONFIG_DIGITAL_TRIG` includes a base channel in `data[3]`. This is
used as a right shift amount for other bitmask values without being
checked. Shift amounts greater than or equal to 32 will result in
undefined behavior. Add code to deal with this.
The `INSN_CONFIG` comedi instruction with sub-instruction code
`INSN_CONFIG_DIGITAL_TRIG` includes a base channel in `data[3]`. This is
used as a right shift amount for other bitmask values without being
checked. Shift amounts greater than or equal to 32 will result in
undefined behavior. Add code to deal with this, adjusting the checks
for invalid channels so that enabled channel bits that would have been
lost by shifting are also checked for validity. Only channels 0 to 15
are valid.
Fixes: a8c66b684efaf ("staging: comedi: addi_apci_1500: rewrite the subdevice support functions") Cc: <stable@vger.kernel.org> #4.0+: ef75e14a6c93: staging: comedi: verify array index is correct before using it Cc: <stable@vger.kernel.org> #4.0+ Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Link: https://lore.kernel.org/r/20200717145257.112660-5-abbotti@mev.co.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
`ni6527_intr_insn_config()` processes `INSN_CONFIG` comedi instructions
for the "interrupt" subdevice. When `data[0]` is
`INSN_CONFIG_DIGITAL_TRIG` it is configuring the digital trigger. When
`data[2]` is `COMEDI_DIGITAL_TRIG_ENABLE_EDGES` it is configuring rising
and falling edge detection for the digital trigger, using a base channel
number (or shift amount) in `data[3]`, a rising edge bitmask in
`data[4]` and falling edge bitmask in `data[5]`.
If the base channel number (shift amount) is greater than or equal to
the number of channels (24) of the digital input subdevice, there are no
changes to the rising and falling edges, so the mask of channels to be
changed can be set to 0, otherwise the mask of channels to be changed,
and the rising and falling edge bitmasks are shifted by the base channel
number before calling `ni6527_set_edge_detection()` to change the
appropriate registers. Unfortunately, the code is comparing the base
channel (shift amount) to the interrupt subdevice's number of channels
(1) instead of the digital input subdevice's number of channels (24).
Fix it by comparing to 32 because all shift amounts for an `unsigned
int` must be less than that and everything from bit 24 upwards is
ignored by `ni6527_set_edge_detection()` anyway.