The intel-pt-events.py script displays only the last of consecutive switch
statements but that may not be the last switch event for the CPU. Fix by
keeping a dictionary of last context switch keyed by CPU, and make it
possible to see all switch events by adding option --all-switch-events.
Fixes: a92bf335fd82eeee ("perf scripts python: intel-pt-events.py: Add branches to script") Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Riccardo Mancini <rickyman7@gmail.com> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20211215080636.149562-4-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
$ perf record -e intel_pt//u uname
Linux
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.026 MB perf.data ]
$ perf inject -i perf.data --vm-time-correlation="dry-run 123"
$ perf inject -i perf.data --vm-time-correlation="dry-run 123:456"
Failed to parse VM Time Correlation options
0x620 [0x98]: failed to process type: 70 [Invalid argument]
$
Fixes: e3ff42bdebcfeb5f ("perf intel-pt: Parse VM Time Correlation options and set up decoding") Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Riccardo Mancini <rickyman7@gmail.com> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20211215080636.149562-2-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Make sure that finish_mount_kattr() is called after mount_kattr was
succesfully built in both the success and failure case to prevent
leaking any references we took when we built it. We returned early if
path lookup failed thereby risking to leak an additional reference we
took when building mount_kattr when an idmapped mount was requested.
This issue was also reported since 2017 in the thread [1],
unfortunately, the issue was still can be reproduced after fixing
DCCP.
The ipv4_mib_exit_net is called before tcp_sk_exit_batch when a net
namespace is destroyed since tcp_sk_ops is registered befrore
ipv4_mib_ops, which means tcp_sk_ops is in the front of ipv4_mib_ops
in the list of pernet_list. There will be a use-after-free on
net->mib.net_statistics in tw_timer_handler after ipv4_mib_exit_net
if there are some inflight time-wait timers.
This bug is not introduced by commit f2bf415cfed7 ("mib: add net to
NET_ADD_STATS_BH") since the net_statistics is a global variable
instead of dynamic allocation and freeing. Actually, commit 61a7e26028b9 ("mib: put net statistics on struct net") introduces
the bug since it put net statistics on struct net and free it when
net namespace is destroyed.
Moving init_ipv4_mibs() to the front of tcp_init() to fix this bug
and replace pr_crit() with panic() since continuing is meaningless
when init_ipv4_mibs() fails.
DAMON debugfs interface increases the reference counts of 'struct pid's
for targets from the 'target_ids' file write callback
('dbgfs_target_ids_write()'), but decreases the counts only in DAMON
monitoring termination callback ('dbgfs_before_terminate()').
Therefore, when 'target_ids' file is repeatedly written without DAMON
monitoring start/termination, the reference count is not decreased and
therefore memory for the 'struct pid' cannot be freed. This commit
fixes this issue by decreasing the reference counts when 'target_ids' is
written.
Link: https://lkml.kernel.org/r/20211229124029.23348-1-sj@kernel.org Fixes: 4bc05954d007 ("mm/damon: implement a debugfs-based user space interface") Signed-off-by: SeongJae Park <sj@kernel.org> Cc: <stable@vger.kernel.org> [5.15+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The spaceball.c module was not properly parsing the movement reports
coming from the device. The code read axis data as signed 16-bit
little-endian values starting at offset 2.
In fact, axis data in Spaceball movement reports are signed 16-bit
big-endian values starting at offset 3. This was determined first by
visually inspecting the data packets, and later verified by consulting:
http://spacemice.org/pdf/SpaceBall_2003-3003_Protocol.pdf
If this ever worked properly, it was in the time before Git...
Signed-off-by: Leo L. Schwab <ewhac@ewhac.org> Link: https://lore.kernel.org/r/20211221101630.1146385-1-ewhac@ewhac.org Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Syzbot has reported warning in __flush_work(). This warning is caused by
work->func == NULL, which means missing work initialization.
This may happen, since input_dev->close() calls
cancel_work_sync(&dev->work), but dev->work initalization happens _after_
input_register_device() call.
So this patch moves dev->work initialization before registering input
device
The PVSCSI implementation in the VMware hypervisor under specific
configuration ("SCSI Bus Sharing" set to "Physical") returns zero dataLen
in the completion descriptor for READ CAPACITY(16). As a result, the kernel
can not detect proper disk geometry. This can be recognized by the kernel
message:
The PVSCSI implementation in QEMU does not set dataLen at all, keeping it
zeroed. This leads to a boot hang as was reported by Shmulik Ladkani.
It is likely that the controller returns the garbage at the end of the
buffer. Residual length should be set by the driver in that case. The SCSI
layer will erase corresponding data. See commit bdb2b8cab439 ("[SCSI] erase
invalid data returned by device") for details.
Commit e662502b3a78 ("scsi: vmw_pvscsi: Set correct residual data length")
introduced the issue by setting residual length unconditionally, causing
the SCSI layer to erase the useful payload beyond dataLen when this value
is returned as 0.
As a result, considering existing issues in implementations of PVSCSI
controllers, we do not want to call scsi_set_resid() when dataLen ==
0. Calling scsi_set_resid() has no effect if dataLen equals buffer length.
Link: https://lore.kernel.org/lkml/20210824120028.30d9c071@blondie/ Link: https://lore.kernel.org/r/20211220190514.55935-1-amakhalov@vmware.com Fixes: e662502b3a78 ("scsi: vmw_pvscsi: Set correct residual data length") Cc: Matt Wang <wwentao@vmware.com> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: Vishal Bhakta <vbhakta@vmware.com> Cc: VMware PV-Drivers <pv-drivers@vmware.com> Cc: James E.J. Bottomley <jejb@linux.ibm.com> Cc: linux-scsi@vger.kernel.org Cc: stable@vger.kernel.org Reported-and-suggested-by: Shmulik Ladkani <shmulik.ladkani@gmail.com> Signed-off-by: Alexey Makhalov <amakhalov@vmware.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In 4.13, commit 74310e06be4d ("android: binder: Move buffer out of area shared with user space")
fixed a kernel structure visibility issue. As part of that patch,
sizeof(void *) was used as the buffer size for 0-length data payloads so
the driver could detect abusive clients sending 0-length asynchronous
transactions to a server by enforcing limits on async_free_size.
Unfortunately, on the "free" side, the accounting of async_free_space
did not add the sizeof(void *) back. The result was that up to 8-bytes of
async_free_space were leaked on every async transaction of 8-bytes or
less. These small transactions are uncommon, so this accounting issue
has gone undetected for several years.
The fix is to use "buffer_size" (the allocated buffer size) instead of
"size" (the logical buffer size) when updating the async_free_space
during the free operation. These are the same except for this
corner case of asynchronous transactions with payloads < 8 bytes.
Fixes: 74310e06be4d ("android: binder: Move buffer out of area shared with user space") Signed-off-by: Todd Kjos <tkjos@google.com> Cc: stable@vger.kernel.org # 4.14+ Link: https://lore.kernel.org/r/20211220190150.2107077-1-tkjos@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After commit 5b78ed24e8ec ("mm/pagemap: add mmap_assert_locked()
annotations to find_vma*()"), the call to get_user_pages() will trigger
the mmap assert.
Use get_user_pages_unlocked() when setting the enclave memory regions.
That's a similar pattern as mmap_read_lock() used together with
get_user_pages().
Fixes: 5b78ed24e8ec ("mm/pagemap: add mmap_assert_locked() annotations to find_vma*()") Cc: stable@vger.kernel.org Signed-off-by: Andra Paraschiv <andraprs@amazon.com> Link: https://lore.kernel.org/r/20211220195856.6549-1-andraprs@amazon.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is a seldom issue that the controller access invalid address
and trigger devapc or emimpu violation. That is due to memory access
is out of order and cause gpd data is not correct.
Add mb() to prohibit compiler or cpu from reordering to make sure GPD
is fully written before setting its HWO.
ffs_data_clear is indirectly called from both ffs_fs_kill_sb and
ffs_ep0_release, so it ends up being called twice when userland closes ep0
and then unmounts f_fs.
If userland provided an eventfd along with function's USB descriptors, it
ends up calling eventfd_ctx_put as many times, causing a refcount
underflow.
NULL-ify ffs_eventfd to prevent these extraneous eventfd_ctx_put calls.
Also, set epfiles to NULL right after de-allocating it, for readability.
For completeness, ffs_data_clear actually ends up being called thrice, the
last call being before the whole ffs structure gets freed, so when this
specific sequence happens there is a second underflow happening (but not
being reported):
The Fresco Logic FL1100 controller needs the TRUST_TX_LENGTH quirk like
other Fresco controllers, but should not have the BROKEN_MSI quirks set.
BROKEN_MSI quirk causes issues in detecting usb drives connected to docks
with this FL1100 controller.
The BROKEN_MSI flag was apparently accidentally set together with the
TRUST_TX_LENGTH quirk
Original patch went to stable so this should go there as well.
Fixes: ea0f69d82119 ("xhci: Enable trust tx length quirk for Fresco FL11 USB controller") Cc: stable@vger.kernel.org
cc: Nikolay Martynov <mar.kolya@gmail.com> Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com> Link: https://lore.kernel.org/r/20211221112825.54690-2-mathias.nyman@linux.intel.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Play a video on the raven (or PCO, raven2) platform, and then do the S3
test. When resume, the following error will be reported:
amdgpu 0000:02:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring
vcn_dec test failed (-110)
[drm:amdgpu_device_ip_resume_phase2 [amdgpu]] *ERROR* resume of IP block
<vcn_v1_0> failed -110
amdgpu 0000:02:00.0: amdgpu: amdgpu_device_ip_resume failed (-110).
PM: dpm_run_callback(): pci_pm_resume+0x0/0x90 returns -110
[why]
When playing the video: The power state flag of the vcn block is set to
POWER_STATE_ON.
When doing suspend: There is no change to the power state flag of the
vcn block, it is still POWER_STATE_ON.
When doing resume: Need to open the power gate of the vcn block and set
the power state flag of the VCN block to POWER_STATE_ON.
But at this time, the power state flag of the vcn block is already
POWER_STATE_ON. The power status flag check in the "8f2cdef drm/amd/pm:
avoid duplicate powergate/ungate setting" patch will return the
amdgpu_dpm_set_powergating_by_smu function directly.
As a result, the gate of the power was not opened, causing the
subsequent ring test to fail.
[how]
In the suspend function of the vcn block, explicitly change the power
state flag of the vcn block to POWER_STATE_OFF.
BugLink: https://gitlab.freedesktop.org/drm/amd/-/issues/1828 Signed-off-by: chen gong <curry.gong@amd.com> Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Always waiting for the exclusive fence resulted on some performance
regressions. So try to wait for the shared fences first, then the
exclusive fence should always be signaled already.
v2: fix incorrectly placed "(", add some comment why we do this.
Signed-off-by: Christian König <christian.koenig@amd.com> Tested-by: Stefan Fritsch <sf@sfritsch.de> Tested-by: Dan Moulding <dmoulding@me.com> Acked-by: Ben Skeggs <bskeggs@redhat.com> Signed-off-by: Christian König <christian.koenig@amd.com> Cc: <stable@vger.kernel.org> Link: https://patchwork.freedesktop.org/patch/msgid/20211209102335.18321-1-christian.koenig@amd.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Replace sa_family_t with __kernel_sa_family_t to fix the following
linux/nfc.h userspace compilation errors:
/usr/include/linux/nfc.h:266:2: error: unknown type name 'sa_family_t'
sa_family_t sa_family;
/usr/include/linux/nfc.h:274:2: error: unknown type name 'sa_family_t'
sa_family_t sa_family;
Fixes: 23b7869c0fd0 ("NFC: add the NFC socket raw protocol") Fixes: d646960f7986 ("NFC: Initial LLCP support") Cc: <stable@vger.kernel.org> Signed-off-by: Dmitry V. Levin <ldv@altlinux.org> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Wrong user data may cause warning in i2c_transfer(), ex: zero msgs.
Userspace should not be able to trigger warnings, so this patch adds
validation checks for user data in compact ioctl to prevent reported
warnings
Reported-and-tested-by: syzbot+e417648b303855b91d8a@syzkaller.appspotmail.com Fixes: 7d5cb45655f2 ("i2c compat ioctls: move to ->compat_ioctl()") Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Signed-off-by: Wolfram Sang <wsa@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The reference taken by 'of_find_device_by_node()' must be released when
not needed anymore.
Add the corresponding 'put_device()' in the and error handling paths.
Fixes: 18a6c85fcc78 ("fsl/fman: Add FMan Port Support") Signed-off-by: Miaoqian Lin <linmq006@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
As we can see from the comment of the nla_put() that it could return
-EMSGSIZE if the tailroom of the skb is insufficient.
Therefore, it should be better to check the return value of the
nla_put_u32 and return the error code if error accurs.
Also, there are many other functions have the same problem, and if this
patch is correct, I will commit a new version to fix all.
We need to first check if the context is a vlan one, then we need to
check the global bridge multicast vlan snooping flag, and finally the
vlan's multicast flag, otherwise we will unnecessarily enable vlan mcast
processing (e.g. querier timers).
Fixes: 7b54aaaf53cb ("net: bridge: multicast: add vlan state initialization and control") Signed-off-by: Nikolay Aleksandrov <nikolay@nvidia.com> Link: https://lore.kernel.org/r/20211228153142.536969-1-nikolay@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
udpgso_bench_tx call setup_sockaddr() for dest address before
parsing all arguments, if we specify "-p ${dst_port}" after "-D ${dst_ip}",
then ${dst_port} will be ignored, and using default cfg_port 8000.
This will cause test case "multiple GRO socks" failed in udpgro.sh.
Setup sockaddr after parsing all arguments.
Fixes: 3a687bef148d ("selftests: udp gso benchmark") Signed-off-by: Jianguo Wu <wujianguo@chinatelecom.cn> Reviewed-by: Willem de Bruijn <willemb@google.com> Link: https://lore.kernel.org/r/ff620d9f-5b52-06ab-5286-44b945453002@163.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
As reported[1] if startup query interval is set too low in combination with
large number of startup queries and we have multiple bridges or even a
single bridge with multiple querier vlans configured we can crash the
machine. Add a 1 second minimum which must be enforced by overwriting the
value if set lower (i.e. without returning an error) to avoid breaking
user-space. If that happens a log message is emitted to let the admin know
that the startup interval has been set to the minimum. It doesn't make
sense to make the startup interval lower than the normal query interval
so use the same value of 1 second. The issue has been present since these
intervals could be user-controlled.
As reported[1] if query interval is set too low and we have multiple
bridges or even a single bridge with multiple querier vlans configured
we can crash the machine. Add a 1 second minimum which must be enforced
by overwriting the value if set lower (i.e. without returning an error) to
avoid breaking user-space. If that happens a log message is emitted to let
the administrator know that the interval has been set to the minimum.
The issue has been present since these intervals could be user-controlled.
In case of an error in mlx5e_set_features(), 'netdev->features' must be
updated with the correct state of the device to indicate which features
were updated successfully.
To do that we maintain a copy of 'netdev->features' and update it after
successful feature changes, so we can assign it to back to
'netdev->features' if needed.
However, since not all netdev features are handled by the driver (e.g.
GRO/TSO/etc), some features may not be updated correctly in case of an
error updating another feature.
For example, while requesting to disable TSO (feature which is not
handled by the driver) and enable HW-GRO, if an error occurs during
HW-GRO enable, 'oper_features' will be assigned with 'netdev->features'
and HW-GRO turned off. TSO will remain enabled in such case, which is a
bug.
To solve that, instead of using 'netdev->features' as the baseline of
'oper_features' and changing it on set feature success, use 'features'
instead and update it in case of errors.
Fixes: 75b81ce719b7 ("net/mlx5e: Don't override netdev features field unless in error flow") Signed-off-by: Gal Pressman <gal@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
It was reported that when PCIe PTM is enabled, some lockups could
be observed with some integrated i225-V models.
While the issue is investigated, we can disable crosstimestamp for
those models and see no loss of functionality, because those models
don't have any support for time synchronization.
Fixes: a90ec8483732 ("igc: Add support for PTP getcrosststamp()") Link: https://lore.kernel.org/all/924175a188159f4e03bd69908a91e606b574139b.camel@gmx.de/ Reported-by: Stefan Dietrich <roots@gmx.de> Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Nechama Kraus <nechamax.kraus@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Though smc_cdc_tx_handler() checked the existence of smc connection,
smc_release() may have already dismissed and released the smc socket
before smc_cdc_tx_handler() further visits it.
smc_cdc_tx_handler() |smc_release()
if (!conn) |
|
|smc_cdc_tx_dismiss_slots()
| smc_cdc_tx_dismisser()
|
|sock_put(&smc->sk) <- last sock_put,
| smc_sock freed
bh_lock_sock(&smc->sk) (panic) |
To make sure we won't receive any CDC messages after we free the
smc_sock, add a refcount on the smc_connection for inflight CDC
message(posted to the QP but haven't received related CQE), and
don't release the smc_connection until all the inflight CDC messages
haven been done, for both success or failed ones.
Using refcount on CDC messages brings another problem: when the link
is going to be destroyed, smcr_link_clear() will reset the QP, which
then remove all the pending CQEs related to the QP in the CQ. To make
sure all the CQEs will always come back so the refcount on the
smc_connection can always reach 0, smc_ib_modify_qp_reset() was replaced
by smc_ib_modify_qp_error().
And remove the timeout in smc_wr_tx_wait_no_pending_sends() since we
need to wait for all pending WQEs done, or we may encounter use-after-
free when handling CQEs.
For IB device removal routine, we need to wait for all the QPs on that
device been destroyed before we can destroy CQs on the device, or
the refcount on smc_connection won't reach 0 and smc_sock cannot be
released.
Fixes: 5f08318f617b ("smc: connection data control (CDC)") Reported-by: Wen Gu <guwen@linux.alibaba.com> Signed-off-by: Dust Li <dust.li@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
We found smc_llc_send_link_delete_all() sometimes wait
for 2s timeout when testing with RDMA link up/down.
It is possible when a smc_link is in ACTIVATING state,
the underlaying QP is still in RESET or RTR state, which
cannot send any messages out.
smc_llc_send_link_delete_all() use smc_link_usable() to
checks whether the link is usable, if the QP is still in
RESET or RTR state, but the smc_link is in ACTIVATING, this
LLC message will always fail without any CQE entering the
CQ, and we will always wait 2s before timeout.
Since we cannot send any messages through the QP before
the QP enter RTS. I add a wrapper smc_link_sendable()
which checks the state of QP along with the link state.
And replace smc_link_usable() with smc_link_sendable()
in all LLC & CDC message sending routine.
Fixes: 5f08318f617b ("smc: connection data control (CDC)") Signed-off-by: Dust Li <dust.li@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Received frames have FCS truncated. There is no need
to subtract FCS length from the statistics.
Fixes: fe1a56420cf2 ("net: lantiq: Add Lantiq / Intel VRX200 Ethernet driver") Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
'ndev' is a managed resource allocated with devm_alloc_etherdev(), so there
is no need to call free_netdev() explicitly or there will be a double
free().
Simplify all error handling paths accordingly.
Fixes: d51b6ce441d3 ("net: ethernet: add ag71xx driver") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The D-Link DSB-650TX (2001:4002) is unable to receive Ethernet frames
that are longer than 1518 octets, for example, Ethernet frames that
contain 802.1Q VLAN tags.
The frames are sent to the pegasus driver via USB but the driver
discards them because they have the Long_pkt field set to 1 in the
received status report. The function read_bulk_callback of the pegasus
driver treats such received "packets" (in the terminology of the
hardware) as errors but the field simply does just indicate that the
Ethernet frame (MAC destination to FCS) is longer than 1518 octets.
It seems that in the 1990s there was a distinction between
"giant" (> 1518) and "runt" (< 64) frames and the hardware includes
flags to indicate this distinction. It seems that the purpose of the
distinction "giant" frames was to not allow infinitely long frames due
to transmission errors and to allow hardware to have an upper limit of
the frame size. However, the hardware already has such limit with its
2048 octet receive buffer and, therefore, Long_pkt is merely a
convention and should not be treated as a receive error.
Actually, the hardware is even able to receive Ethernet frames with 2048
octets which exceeds the claimed limit frame size limit of the driver of
1536 octets (PEGASUS_MTU).
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Matthias-Christian Ott <ott@mirix.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
In smc_wr_tx_send_wait() the completion on index specified by
pend->idx is initialized and after smc_wr_tx_send() was called the wait
for completion starts. pend->idx is used to get the correct index for
the wait, but the pend structure could already be cleared in
smc_wr_tx_process_cqe().
Introduce pnd_idx to hold and use a local copy of the correct index.
Fixes: 09c61d24f96d ("net/smc: wait for departure of an IB message") Signed-off-by: Karsten Graul <kgraul@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
This issue occurs when asoc is peeled off and the old sk is freed after
getting it by asoc->base.sk and before calling lock_sock(sk).
To prevent the sk free, as a holder of the sk, ep should be alive when
calling lock_sock(). This patch uses call_rcu() and moves sock_put and
ep free into sctp_endpoint_destroy_rcu(), so that it's safe to try to
hold the ep under rcu_read_lock in sctp_transport_traverse_process().
If sctp_endpoint_hold() returns true, it means this ep is still alive
and we have held it and can continue to dump it; If it returns false,
it means this ep is dead and can be freed after rcu_read_unlock, and
we should skip it.
In sctp_sock_dump(), after locking the sk, if this ep is different from
tsp->asoc->ep, it means during this dumping, this asoc was peeled off
before calling lock_sock(), and the sk should be skipped; If this ep is
the same with tsp->asoc->ep, it means no peeloff happens on this asoc,
and due to lock_sock, no peeloff will happen either until release_sock.
Note that delaying endpoint free won't delay the port release, as the
port release happens in sctp_endpoint_destroy() before calling call_rcu().
Also, freeing endpoint by call_rcu() makes it safe to access the sk by
asoc->base.sk in sctp_assocs_seq_show() and sctp_rcv().
Thanks Jones to bring this issue up.
v1->v2:
- improve the changelog.
- add kfree(ep) into sctp_endpoint_destroy_rcu(), as Jakub noticed.
Reported-by: syzbot+9276d76e83e3bcde6c99@syzkaller.appspotmail.com Reported-by: Lee Jones <lee.jones@linaro.org> Fixes: d25adbeb0cdb ("sctp: fix an use-after-free issue in sctp_sock_dump") 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>
The below referenced commit correctly updated the computation of number
of segments (gso_size) by using only the gso payload size and
removing the header lengths.
With this change the regression test started failing. Update
the tests to match this new behavior.
Both IPv4 and IPv6 tests are updated, as a separate patch in this series
will update udp_v6_send_skb to match this change in udp_send_skb.
Fixes: 158390e45612 ("udp: using datalen to cap max gso segments") Signed-off-by: Coco Li <lixiaoyan@google.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Link: https://lore.kernel.org/r/20211223222441.2975883-2-lixiaoyan@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The max number of UDP gso segments is intended to cap to
UDP_MAX_SEGMENTS, this is checked in udp_send_skb().
skb->len contains network and transport header len here, we should use
only data len instead.
This is the ipv6 counterpart to the below referenced commit,
which missed the ipv6 change
Fixes: 158390e45612 ("udp: using datalen to cap max gso segments") Signed-off-by: Coco Li <lixiaoyan@google.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Link: https://lore.kernel.org/r/20211223222441.2975883-1-lixiaoyan@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When there is ct or sample action, the ct or sample rule will be deleted
and return. But if there is an extra mirror action, the forward rule can't
be deleted because of the return.
Fix it by removing the return.
Fixes: 69e2916ebce4 ("net/mlx5: CT: Add support for mirroring") Fixes: f94d6389f6a8 ("net/mlx5e: TC, Add support to offload sample action") Signed-off-by: Chris Mi <cmi@nvidia.com> Reviewed-by: Roi Dayan <roid@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
There are two ICOSQs per channel: one is needed for RX, and the other
for async operations (XSK TX, kTLS offload). Currently, the recovery
flow for both is the same, and async ICOSQ is mistakenly treated like
the regular ICOSQ.
This patch prevents running the regular ICOSQ recovery on async ICOSQ.
The purpose of async ICOSQ is to handle XSK wakeup requests and post
kTLS offload RX parameters, it has nothing to do with RQ and XSKRQ UMRs,
so the regular recovery sequence is not applicable here.
Fixes: be5323c8379f ("net/mlx5e: Report and recover from CQE error on ICOSQ") Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com> Reviewed-by: Aya Levin <ayal@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Both regular RQ and XSKRQ use the same ICOSQ for UMRs. When doing
recovery for the ICOSQ, don't forget to deactivate XSKRQ.
XSK can be opened and closed while channels are active, so a new mutex
prevents the ICOSQ recovery from running at the same time. The ICOSQ
recovery deactivates and reactivates XSKRQ, so any parallel change in
XSK state would break consistency. As the regular RQ is running, it's
not enough to just flush the recovery work, because it can be
rescheduled.
Fixes: be5323c8379f ("net/mlx5e: Report and recover from CQE error on ICOSQ") Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Function mlx5e_tx_reporter_dump_sq() casts its void * argument to struct
mlx5e_txqsq *, but in TX-timeout-recovery flow the argument is actually
of type struct mlx5e_tx_timeout_ctx *.
mlx5_core 0000:08:00.1 enp8s0f1: TX timeout detected
mlx5_core 0000:08:00.1 enp8s0f1: TX timeout on queue: 1, SQ: 0x11ec, CQ: 0x146d, SQ Cons: 0x0 SQ Prod: 0x1, usecs since last trans: 21565000
BUG: stack guard page was hit at 0000000093f1a2de (stack is 00000000b66ea0dc..000000004d932dae)
kernel stack overflow (page fault): 0000 [#1] SMP NOPTI
CPU: 5 PID: 95 Comm: kworker/u20:1 Tainted: G W OE 5.13.0_mlnx #1
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
Workqueue: mlx5e mlx5e_tx_timeout_work [mlx5_core]
RIP: 0010:mlx5e_tx_reporter_dump_sq+0xd3/0x180
[mlx5_core]
Call Trace:
mlx5e_tx_reporter_dump+0x43/0x1c0 [mlx5_core]
devlink_health_do_dump.part.91+0x71/0xd0
devlink_health_report+0x157/0x1b0
mlx5e_reporter_tx_timeout+0xb9/0xf0 [mlx5_core]
? mlx5e_tx_reporter_err_cqe_recover+0x1d0/0x1d0
[mlx5_core]
? mlx5e_health_queue_dump+0xd0/0xd0 [mlx5_core]
? update_load_avg+0x19b/0x550
? set_next_entity+0x72/0x80
? pick_next_task_fair+0x227/0x340
? finish_task_switch+0xa2/0x280
mlx5e_tx_timeout_work+0x83/0xb0 [mlx5_core]
process_one_work+0x1de/0x3a0
worker_thread+0x2d/0x3c0
? process_one_work+0x3a0/0x3a0
kthread+0x115/0x130
? kthread_park+0x90/0x90
ret_from_fork+0x1f/0x30
--[ end trace 51ccabea504edaff ]---
RIP: 0010:mlx5e_tx_reporter_dump_sq+0xd3/0x180
PKRU: 55555554
Kernel panic - not syncing: Fatal exception
Kernel Offset: disabled
end Kernel panic - not syncing: Fatal exception
To fix this bug add a wrapper for mlx5e_tx_reporter_dump_sq() which
extracts the sq from struct mlx5e_tx_timeout_ctx and set it as the
TX-timeout-recovery flow dump callback.
Fixes: 5f29458b77d5 ("net/mlx5e: Support dump callback in TX reporter") Signed-off-by: Aya Levin <ayal@nvidia.com> Signed-off-by: Amir Tzin <amirtz@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Only prio 1 is supported if firmware doesn't support ignore flow
level for nic mode. The offending commit removed the check wrongly.
Add it back.
Fixes: 9a99c8f1253a ("net/mlx5e: E-Switch, Offload all chain 0 priorities when modify header and forward action is not supported") Signed-off-by: Chris Mi <cmi@nvidia.com> Reviewed-by: Roi Dayan <roid@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In case IRQ layer failed to find or to request irq, the driver is
printing the first cpu of the provided affinity as part of the error
print. Empty affinity is a valid input for the IRQ layer, and it is
an error to call cpumask_first() on empty affinity.
Remove the first cpu print from the error message.
hooks.c:5765:6: warning: 4th function call argument is an uninitialized
value
if (selinux_xfrm_postroute_last(sksec->sid, skb, &ad, proto))
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
selinux_parse_skb() can return ok without setting proto. The later call
to selinux_xfrm_postroute_last() does an early check of proto and can
return ok if the garbage proto value matches. So initialize proto.
Cc: stable@vger.kernel.org Fixes: eef9b41622f2 ("selinux: cleanup selinux_xfrm_sock_rcv_skb() and selinux_xfrm_postroute_last()") Signed-off-by: Tom Rix <trix@redhat.com>
[PM: typo/spelling and checkpatch.pl description fixes] Signed-off-by: Paul Moore <paul@paul-moore.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 8633ef82f101 ("drivers/firmware: consolidate EFI framebuffer setup
for all arches") made the Generic System Framebuffers (sysfb) driver able
to be built on non-x86 architectures.
But it left the efifb_setup_from_dmi() function prototype declaration in
the architecture specific headers. This could lead to the following
compiler warning as reported by the kernel test robot:
drivers/firmware/efi/sysfb_efi.c:70:6: warning: no previous prototype for function 'efifb_setup_from_dmi' [-Wmissing-prototypes]
void efifb_setup_from_dmi(struct screen_info *si, const char *opt)
^
drivers/firmware/efi/sysfb_efi.c:70:1: note: declare 'static' if the function is not intended to be used outside of this translation unit
void efifb_setup_from_dmi(struct screen_info *si, const char *opt)
Fixes: 8633ef82f101 ("drivers/firmware: consolidate EFI framebuffer setup for all arches") Reported-by: kernel test robot <lkp@intel.com> Cc: <stable@vger.kernel.org> # 5.15.x Signed-off-by: Javier Martinez Canillas <javierm@redhat.com> Acked-by: Thomas Zimmermann <tzimmermann@suse.de> Link: https://lore.kernel.org/r/20211126001333.555514-1-javierm@redhat.com Signed-off-by: Ard Biesheuvel <ardb@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 85bf17b28f97 ("recordmcount.pl: look for jgnop instruction as well
as bcrl on s390") added a new alternative mnemonic for the existing brcl
instruction. This is required for the combination old gcc version (pre 9.0)
and binutils since version 2.37.
However at the same time this commit introduced a typo, replacing brcl with
bcrl. As a result no mcount locations are detected anymore with old gcc
versions (pre 9.0) and binutils before version 2.37.
Fix this by using the correct mnemonic again.
Reported-by: Miroslav Benes <mbenes@suse.cz> Cc: Jerome Marchand <jmarchan@redhat.com> Cc: <stable@vger.kernel.org> Fixes: 85bf17b28f97 ("recordmcount.pl: look for jgnop instruction as well as bcrl on s390") Link: https://lore.kernel.org/r/alpine.LSU.2.21.2112230949520.19849@pobox.suse.cz Signed-off-by: Heiko Carstens <hca@linux.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In the HDAS ACPI scope, the SoundWire may not be the direct child of HDAS.
It needs to go through the ACPI table at max depth of 2 to find the
SoundWire device from HDAS.
The existing code currently sets a pointer to an ACPI handle before
checking that it's actually a SoundWire controller. This can lead to
issues where the graph walk continues and eventually fails, but the
pointer was set already.
This patch changes the logic so that the information provided to
the caller is set when a controller is found.
Fix modpost Section mismatch error in memblock_phys_alloc()
[...]
WARNING: modpost: vmlinux.o(.text.unlikely+0x1dcc): Section mismatch in reference
from the function memblock_phys_alloc() to the function .init.text:memblock_phys_alloc_range()
The function memblock_phys_alloc() references
the function __init memblock_phys_alloc_range().
This is often because memblock_phys_alloc lacks a __init
annotation or the annotation of memblock_phys_alloc_range is wrong.
ERROR: modpost: Section mismatches detected.
Set CONFIG_SECTION_MISMATCH_WARN_ONLY=y to allow them.
[...]
memblock_phys_alloc() is a one-line wrapper, make it __always_inline to
avoid these section mismatches.
Reported-by: k2ci <kernel-bot@kylinos.cn> Suggested-by: Mike Rapoport <rppt@kernel.org> Signed-off-by: Jackie Liu <liuyun01@kylinos.cn>
[rppt: slightly massaged changelog ] Signed-off-by: Mike Rapoport <rppt@linux.ibm.com> Link: https://lore.kernel.org/r/20211217020754.2874872-1-liu.yun@linux.dev Signed-off-by: Sasha Levin <sashal@kernel.org>
The devm_ioremap() function returns NULL on error, it doesn't return
error pointers. Also according to doc of device_property_read_u64_array,
values in info array are properties of device or NULL.
Signed-off-by: Miaoqian Lin <linmq006@gmail.com> Link: https://lore.kernel.org/r/20211210070753.10761-1-linmq006@gmail.com Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When a trap 7 (Instruction access rights) occurs, this means the CPU
couldn't execute an instruction due to missing execute permissions on
the memory region. In this case it seems the CPU didn't even fetched
the instruction from memory and thus did not store it in the cr19 (IIR)
register before calling the trap handler. So, the trap handler will find
some random old stale value in cr19.
This patch simply overwrites the stale IIR value with a constant magic
"bad food" value (0xbaadf00d), in the hope people don't start to try to
understand the various random IIR values in trap 7 dumps.
Noticed-by: John David Anglin <dave.anglin@bell.net> Signed-off-by: Helge Deller <deller@gmx.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
BPF layer extends the qdisc control block via struct bpf_skb_data_end
and because of that there is no more room to add variables to the
qdisc layer control block without going over the skb->cb size.
Extend the qdisc control block with a tc control block,
and move all tc related variables to there as a pre-step for
extending the tc control block with additional members.
Signed-off-by: Paul Blakey <paulb@nvidia.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
hwight16() is much faster. While we are at it, no need to include
"perm =" part into data_race() macro, for perm is a local variable
that cannot be accessed by other threads.
If tomoyo is used in a testing/fuzzing environment in learning mode,
for lots of domains the quota will be exceeded and stay exceeded
for prolonged periods of time. In such cases it's pointless (and slow)
to walk the whole acl list again and again just to rediscover that
the quota is exceeded. We already have the TOMOYO_DIF_QUOTA_WARNED flag
that notes the overflow condition. Check it early to avoid the slowdown.
[penguin-kernel]
This patch causes a user visible change that the learning mode will not be
automatically resumed after the quota is increased. To resume the learning
mode, administrator will need to explicitly clear TOMOYO_DIF_QUOTA_WARNED
flag after increasing the quota. But I think that this change is generally
preferable, for administrator likely wants to optimize the acl list for
that domain before increasing the quota, or that domain likely hits the
quota again. Therefore, don't try to care to clear TOMOYO_DIF_QUOTA_WARNED
flag automatically when the quota for that domain changed.
The ASUS UM325UA suffers from the same issue as the ASUS UX425UA, which
is a very similar laptop. The i8042 device is not usable immediately
after boot and fails to initialize, requiring a deferred retry.
We've got a bug report about the non-working keyboard on ASUS ZenBook
UX425UA. It seems that the PS/2 device isn't ready immediately at
boot but takes some seconds to get ready. Until now, the only
workaround is to defer the probe, but it's available only when the
driver is a module. However, many distros, including openSUSE as in
the original report, build the PS/2 input drivers into kernel, hence
it won't work easily.
This patch adds the support for the deferred probe for i8042 stuff as
a workaround of the problem above. When the deferred probe mode is
enabled and the device couldn't be probed, it'll be repeated with the
standard deferred probe mechanism.
The deferred probe mode is enabled either via the new option
i8042.probe_defer or via the quirk table entry. As of this patch, the
quirk table contains only ASUS ZenBook UX425UA.
The deferred probe part is based on Fabio's initial work.
This ioctl() implicitly assumed that the socket was already bound to
a valid local socket name, i.e. Phonet object. If the socket was not
bound, two separate problems would occur:
1) We'd send an pipe enablement request with an invalid source object.
2) Later socket calls could BUG on the socket unexpectedly being
connected yet not bound to a valid object.
Reported-by: syzbot+2dc91e7fc3dea88b1e8a@syzkaller.appspotmail.com Signed-off-by: Rémi Denis-Courmont <remi@remlab.net> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Avoid double free in tun_free_netdev() by moving the
dev->tstats and tun->security allocs to a new ndo_init routine
(tun_net_init()) that will be called by register_netdevice().
ndo_init is paired with the desctructor (tun_free_netdev()),
so if there's an error in register_netdevice() the destructor
will handle the frees.
BUG: KASAN: double-free or invalid-free in selinux_tun_dev_free_security+0x1a/0x20 security/selinux/hooks.c:5605
The previous commit 3e0588c291d6 ("hamradio: defer ax25 kfree after
unregister_netdev") reorder the kfree operations and unregister_netdev
operation to prevent UAF.
This commit improves the previous one by also deferring the nullify of
the ax->tty pointer. Otherwise, a NULL pointer dereference bug occurs.
Partial of the stack trace is shown below.
By placing the nullify action after the unregister_netdev, the ax->tty
pointer won't be assigned as NULL net_device framework layer is well
synchronized.
Signed-off-by: Lin Ma <linma@zju.edu.cn> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Even though there are two synchronization primitives before the kfree:
1. wait_for_completion(&ax->dead). This can prevent the race with
routines from mkiss_ioctl. However, it cannot stop the routine coming
from upper layer, i.e., the ax25_sendmsg.
2. netif_stop_queue(ax->dev). It seems that this line of code aims to
halt the transmit queue but it fails to stop the routine that already
being xmit.
This patch reorder the kfree after the unregister_netdev to avoid the
possible UAF as the unregister_netdev() is well synchronized and won't
return if there is a running routine.
Signed-off-by: Lin Ma <linma@zju.edu.cn> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Other syscall functions like ax25_getsockopt, ax25_getname,
ax25_info_show also suffer from similar races. To fix them, this patch
introduce lock_sock() into ax25_kill_by_device in order to guarantee
that the nullify action in cleanup routine cannot proceed when another
socket request is pending.
Signed-off-by: Hanjie Wu <nagi@zju.edu.cn> Signed-off-by: Lin Ma <linma@zju.edu.cn> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There are some chances that the actual base of hardware is different
from the value recorded by driver, so we have to reset the variable
of ocp_base to sync it.
Set ocp_base to -1. Then, it would be updated and the new base would be
set to the hardware next time.
Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Bit 7 of the status register indicates that the chip is busy
doing a conversion. It does not indicate an alarm status.
Stop reporting it as alarm status bit.
Tests with a real chip and a closer look into the datasheet reveals
that the local and remote critical alarm status bits are swapped for
MAX6680/MAX6681.
Some powers were changed during the jack insert detection
and clk's enable/disable in CCF.
If in parallel, the influence has a chance to detect
the wrong jack type, so add a lock.
The eKTH3900/eKTH5312 series do not support the firmware update rules of
Remark ID. Exclude these two series from checking it when updating the
firmware in touch controllers.
Some automated builds report a stack frame size in excess of 2 kB for
iqs626_probe(); the culprit appears to be the call to iqs626_parse_prop().
To solve this problem, specify noinline_for_stack for all of the
iqs626_parse_*() helper functions which are called inside a for loop
within iqs626_parse_prop().
As a result, a build with '-Wframe-larger-than' as low as 512 is free of
any such warnings.
Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Jeff LaBundy <jeff@labundy.com> Link: https://lore.kernel.org/r/20211129004104.453930-1-jeff@labundy.com Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
And it should have been released in the following process:
do_syscall_64
syscall_exit_to_user_mode
exit_to_user_mode_prepare
task_work_run
____fput
__fput
full_proxy_release ---> free here
However, the release function corresponding to file_operations is not
implemented in kfence. As a result, a memory leak occurs. Therefore,
the solution to this problem is to implement the corresponding release
function.
Link: https://lkml.kernel.org/r/20211206133628.2822545-1-libaokun1@huawei.com Fixes: 0ce20dd84089 ("mm: add Kernel Electric-Fence infrastructure") Signed-off-by: Baokun Li <libaokun1@huawei.com> Reported-by: Hulk Robot <hulkci@huawei.com> Acked-by: Marco Elver <elver@google.com> Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com> Cc: Alexander Potapenko <glider@google.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Yu Kuai <yukuai3@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Update the documentation for kvm-intel's emulate_invalid_guest_state to
rectify the description of KVM's default behavior, and to document that
the behavior and thus parameter only applies to L1.
Fixes: a27685c33acc ("KVM: VMX: Emulate invalid guest state by default") Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20211207193006.120997-4-seanjc@google.com> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The order of these two parameters is just reversed. gcc didn't warn on
that, probably because 'void *' can be converted from or to other
pointer types without warning.
Cc: stable@vger.kernel.org Fixes: 3d3c95046742 ("netfs: Provide readahead and readpage netfs helpers") Fixes: e1b1240c1ff5 ("netfs: Add write_begin helper") Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com> Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: Jeff Layton <jlayton@redhat.com> Link: https://lore.kernel.org/r/20211207031449.100510-1-jefflexu@linux.alibaba.com/ Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
- Overview
page fault in f2fs_setxattr() when mount and operate on corrupted image
- Reproduce
tested on kernel 5.16-rc3, 5.15.X under root
1. unzip tmp7.zip
2. ./single.sh f2fs 7
Sometimes need to run the script several times
- Kernel dump
loop0: detected capacity change from 0 to 131072
F2FS-fs (loop0): Found nat_bits in checkpoint
F2FS-fs (loop0): Mounted with checkpoint version = 7548c2ee
BUG: unable to handle page fault for address: ffffe47bc7123f48
RIP: 0010:kfree+0x66/0x320
Call Trace:
__f2fs_setxattr+0x2aa/0xc00 [f2fs]
f2fs_setxattr+0xfa/0x480 [f2fs]
__f2fs_set_acl+0x19b/0x330 [f2fs]
__vfs_removexattr+0x52/0x70
__vfs_removexattr_locked+0xb1/0x140
vfs_removexattr+0x56/0x100
removexattr+0x57/0x80
path_removexattr+0xa3/0xc0
__x64_sys_removexattr+0x17/0x20
do_syscall_64+0x37/0xb0
entry_SYSCALL_64_after_hwframe+0x44/0xae
The root cause is in __f2fs_setxattr(), we missed to do sanity check on
last xattr entry, result in out-of-bound memory access during updating
inconsistent xattr data of target inode.
After the fix, it can detect such xattr inconsistency as below:
F2FS-fs (loop11): inode (7) has invalid last xattr entry, entry_size: 60676
F2FS-fs (loop11): inode (8) has corrupted xattr
F2FS-fs (loop11): inode (8) has corrupted xattr
F2FS-fs (loop11): inode (8) has invalid last xattr entry, entry_size: 47736
Cc: stable@vger.kernel.org Reported-by: Wenqing Liu <wenqingliu0120@gmail.com> Signed-off-by: Chao Yu <chao@kernel.org> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pointer to the allocated pages (struct page *page) has already
progressed towards the end of allocation. It is incorrect to perform
__free_pages(page, order) using this pointer as we would free any
arbitrary pages. Fix this by stop modifying the page pointer.
DAMON debugfs interface iterates current monitoring targets in
'dbgfs_target_ids_read()' while holding the corresponding
'kdamond_lock'. However, it also destructs the monitoring targets in
'dbgfs_before_terminate()' without holding the lock. This can result in
a use_after_free bug. This commit avoids the race by protecting the
destruction with the corresponding 'kdamond_lock'.
Link: https://lkml.kernel.org/r/20211221094447.2241-1-sj@kernel.org Reported-by: Sangwoo Bae <sangwoob@amazon.com> Fixes: 4bc05954d007 ("mm/damon: implement a debugfs-based user space interface") Signed-off-by: SeongJae Park <sj@kernel.org> Cc: <stable@vger.kernel.org> [5.15.x] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hulk Robot reported a panic in put_page_testzero() when testing
madvise() with MADV_SOFT_OFFLINE. The BUG() is triggered when retrying
get_any_page(). This is because we keep MF_COUNT_INCREASED flag in
second try but the refcnt is not increased.
page dumped because: VM_BUG_ON_PAGE(page_ref_count(page) == 0)
------------[ cut here ]------------
kernel BUG at include/linux/mm.h:737!
invalid opcode: 0000 [#1] PREEMPT SMP
CPU: 5 PID: 2135 Comm: sshd Tainted: G B 5.16.0-rc6-dirty #373
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
RIP: release_pages+0x53f/0x840
Call Trace:
free_pages_and_swap_cache+0x64/0x80
tlb_flush_mmu+0x6f/0x220
unmap_page_range+0xe6c/0x12c0
unmap_single_vma+0x90/0x170
unmap_vmas+0xc4/0x180
exit_mmap+0xde/0x3a0
mmput+0xa3/0x250
do_exit+0x564/0x1470
do_group_exit+0x3b/0x100
__do_sys_exit_group+0x13/0x20
__x64_sys_exit_group+0x16/0x20
do_syscall_64+0x34/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
Modules linked in:
---[ end trace e99579b570fe0649 ]---
RIP: 0010:release_pages+0x53f/0x840
Link: https://lkml.kernel.org/r/20211221074908.3910286-1-liushixin2@huawei.com Fixes: b94e02822deb ("mm,hwpoison: try to narrow window race for free pages") Signed-off-by: Liu Shixin <liushixin2@huawei.com> Reported-by: Hulk Robot <hulkci@huawei.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When a memory error hits a tail page of a free hugepage,
__page_handle_poison() is expected to be called to isolate the error in
4kB unit, but it's not called due to the outdated if-condition in
memory_failure_hugetlb(). This loses the chance to isolate the error in
the finer unit, so it's not optimal. Drop the condition.
This "(p != head && TestSetPageHWPoison(head)" condition is based on the
old semantics of PageHWPoison on hugepage (where PG_hwpoison flag was
set on the subpage), so it's not necessray any more. By getting to set
PG_hwpoison on head page for hugepages, concurrent error events on
different subpages in a single hugepage can be prevented by
TestSetPageHWPoison(head) at the beginning of memory_failure_hugetlb().
So dropping the condition should not reopen the race window originally
mentioned in commit b985194c8c0a ("hwpoison, hugetlb:
lock_page/unlock_page does not match for handling a free hugepage")
However, this retry allocation completely ignores memory policy nodemask
allowing allocation to escape restrictions.
The first appearance of this bug seems to be the commit ac5b2c18911f
("mm: thp: relax __GFP_THISNODE for MADV_HUGEPAGE mappings").
The bug disappeared later in the commit 89c83fb539f9 ("mm, thp:
consolidate THP gfp handling into alloc_hugepage_direct_gfpmask") and
reappeared again in slightly different form in the commit 76e654cc91bb
("mm, page_alloc: allow hugepage fallback to remote nodes when
madvised")
Fix this by passing correct nodemask to the __alloc_pages() call.