With lan88xx based devices the lan78xx driver can get stuck in an
interrupt loop while bringing the device up, flooding the kernel log
with messages like the following:
lan78xx 2-3:1.0 enp1s0u3: kevent 4 may have been dropped
Removing interrupt support from the lan88xx PHY driver forces the
driver to use polling instead, which avoids the problem.
The issue has been observed with Raspberry Pi devices at least since
4.14 (see [1], bug report for their downstream kernel), as well as
with Nvidia devices [2] in 2020, where disabling interrupts was the
vendor-suggested workaround (together with the claim that phylib
changes in 4.9 made the interrupt handling in lan78xx incompatible).
Iperf reports well over 900Mbits/sec per direction with client in
--dualtest mode, so there does not seem to be a significant impact on
throughput (lan88xx device connected via switch to the peer).
Commit c7e73b5051d6 ("ARM: imx: mach-imx6ul: remove 14x14 EVK specific
PHY fixup") removed a PHY fixup that setted the clock mode and the LED
mode.
Make the Ethernet interface work again by doing as advised in the
commit's log, set clock mode and the LED mode in the device tree.
Completion of the FFA_PARTITION_INFO_GET ABI transfers the ownership of
the caller’s Rx buffer from the producer(typically partition mnager) to
the consumer(this driver/OS). FFA_RX_RELEASE transfers the ownership
from the consumer back to the producer.
However, when we set the flag to just return the count of partitions
deployed in the system corresponding to the specified UUID while
invoking FFA_PARTITION_INFO_GET, the Rx buffer ownership shouldn't be
transferred to this driver. We must be able to skip transferring back
the ownership to the partition manager when we request just to get the
count of the partitions as the buffers are not acquired in this case.
Firmware may return FFA_RET_DENIED or other error for the ffa_rx_release()
in such cases.
Using device_find_child() to lookup the proper SCMI device to destroy
causes an unbalance in device refcount, since device_find_child() calls an
implicit get_device(): this, in turns, inhibits the call of the provided
release methods upon devices destruction.
As a consequence, one of the structures that is not freed properly upon
destruction is the internal struct device_private dev->p populated by the
drivers subsystem core.
KMemleak detects this situation since loading/unloding some SCMI driver
causes related devices to be created/destroyed without calling any
device_release method.
est_qlen_notify() deletes its class from its active list with
list_del() when qlen is 0, therefore, it is not idempotent and
not friendly to its callers, like fq_codel_dequeue().
Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers'
life. Also change other list_del()'s to list_del_init() just to be
extra safe.
Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Link: https://patch.msgid.link/20250403211033.166059-6-xiyou.wangcong@gmail.com Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
qfq_qlen_notify() always deletes its class from its active list
with list_del_init() _and_ calls qfq_deactivate_agg() when the whole list
becomes empty.
To make it idempotent, just skip everything when it is not in the active
list.
Also change other list_del()'s to list_del_init() just to be extra safe.
Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250403211033.166059-5-xiyou.wangcong@gmail.com Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
hfsc_qlen_notify() is not idempotent either and not friendly
to its callers, like fq_codel_dequeue(). Let's make it idempotent
to ease qdisc_tree_reduce_backlog() callers' life:
1. update_vf() decreases cl->cl_nactive, so we can check whether it is
non-zero before calling it.
2. eltree_remove() always removes RB node cl->el_node, but we can use
RB_EMPTY_NODE() + RB_CLEAR_NODE() to make it safe.
Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250403211033.166059-4-xiyou.wangcong@gmail.com Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drr_qlen_notify() always deletes the DRR class from its active list
with list_del(), therefore, it is not idempotent and not friendly
to its callers, like fq_codel_dequeue().
Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers'
life. Also change other list_del()'s to list_del_init() just to be
extra safe.
Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250403211033.166059-3-xiyou.wangcong@gmail.com Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
htb_qlen_notify() always deactivates the HTB class and in fact could
trigger a warning if it is already deactivated. Therefore, it is not
idempotent and not friendly to its callers, like fq_codel_dequeue().
Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers'
life.
Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250403211033.166059-2-xiyou.wangcong@gmail.com Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The i.MX7D only has one PCIe controller, so controller_id should always be
0. The previous code is incorrect although yielding the correct result.
Fix by removing "IMX7D" from the switch case branch.
Fixes: 2d8ed461dbc9 ("PCI: imx6: Add support for i.MX8MQ") Link: https://lore.kernel.org/r/20241126075702.4099164-5-hongxing.zhu@nxp.com Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com> Signed-off-by: Krzysztof Wilczyński <kwilczynski@kernel.org> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Reviewed-by: Frank Li <Frank.Li@nxp.com>
[Because this switch case does more than just controller_id
logic, move the "IMX7D" case label instead of removing it entirely.] Signed-off-by: Ryan Matthews <ryanmatthews@fastmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
'io_acct_set' is only used for raid0 and raid456, prepare to use it for
raid1 and raid10, so that io accounting from different levels can be
consistent.
By the way, follow up patches will also use this io clone mechanism to
make sure 'active_io' represents in flight io, not io that is dispatching,
so that mddev_suspend will wait for io to be done as designed.
In case the CMD_RTS got corrupted by interferences, the MSE102x
doesn't allow a retransmission of the command. Instead the Ethernet
frame must be shifted out of the SPI FIFO. Since the actual length is
unknown, assume the maximum possible value.
Fixes: 2f207cbf0dd4 ("net: vertexcom: Add MSE102x SPI support") Signed-off-by: Stefan Wahren <wahrenst@gmx.net> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://patch.msgid.link/20250430133043.7722-5-wahrenst@gmx.net Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Since there is no protection in the SPI protocol against electrical
interferences, the driver shouldn't blindly trust the length payload
of CMD_RTS. So introduce a bounds check for incoming frames.
Fixes: 2f207cbf0dd4 ("net: vertexcom: Add MSE102x SPI support") Signed-off-by: Stefan Wahren <wahrenst@gmx.net> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://patch.msgid.link/20250430133043.7722-4-wahrenst@gmx.net Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The MSE102x doesn't provide any SPI commands for interrupt handling.
So in case the interrupt fired before the driver requests the IRQ,
the interrupt will never fire again. In order to fix this always poll
for pending packets after opening the interface.
Fixes: 2f207cbf0dd4 ("net: vertexcom: Add MSE102x SPI support") Signed-off-by: Stefan Wahren <wahrenst@gmx.net> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://patch.msgid.link/20250430133043.7722-2-wahrenst@gmx.net Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently the ptp_clock_register() is called before relative
ptp resource ready. It may cause unexpected result when upper
layer called the ptp API during the timewindow. Fix it by
moving the ptp_clock_register() to the function end.
Fixes: 0bf5eb788512 ("net: hns3: add support for PTP") Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Jijie Shao <shaojijie@huawei.com> Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> Link: https://patch.msgid.link/20250430093052.2400464-5-shaojijie@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The size of the tm_qset file of debugfs is limited to 64 KB,
which is too small in the scenario with 1280 qsets.
The size needs to be expanded to 1 MB.
Fixes: 5e69ea7ee2a6 ("net: hns3: refactor the debugfs process") Signed-off-by: Hao Lan <lanhao@huawei.com> Signed-off-by: Peiyang Wang <wangpeiyang1@huawei.com> Signed-off-by: Jijie Shao <shaojijie@huawei.com> Link: https://patch.msgid.link/20250430093052.2400464-4-shaojijie@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When a VF is passthrough to a VM, and the VM is killed, the reported
interrupt may not been handled, it will remain, and won't be clear by
the nic engine even with a flr or tqp reset. When the VM restart, the
interrupt of the first vector may be dropped by the second enable_irq
in vfio, see the issue below:
https://gitlab.com/qemu-project/qemu/-/issues/2884#note_2423361621
We notice that the vfio has always behaved this way, and the interrupt
is a residue of the nic engine, so we fix the problem by moving the
vector enable process out of the enable_irq loop.
The VF driver missed to store the rx VLAN tag strip state when
user change the rx VLAN tag offload state. And it will default
to enable the rx vlan tag strip when re-init VF device after
reset. So if user disable rx VLAN tag offload, and trig reset,
then the HW will still strip the VLAN tag from packet nad fill
into RX BD, but the VF driver will ignore it for rx VLAN tag
offload disabled. It may cause the rx VLAN tag dropped.
Fixes: b2641e2ad456 ("net: hns3: Add support of hardware rx-vlan-offload to HNS3 VF driver") Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Jijie Shao <shaojijie@huawei.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250430093052.2400464-2-shaojijie@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Activate TX hang workaround also in
fec_enet_txq_submit_skb() when TSO is not enabled.
Errata: ERR007885
Symptoms: NETDEV WATCHDOG: eth0 (fec): transmit queue 0 timed out
commit 37d6017b84f7 ("net: fec: Workaround for imx6sx enet tx hang when enable three queues")
There is a TDAR race condition for mutliQ when the software sets TDAR
and the UDMA clears TDAR simultaneously or in a small window (2-4 cycles).
This will cause the udma_tx and udma_tx_arbiter state machines to hang.
So, the Workaround is checking TDAR status four time, if TDAR cleared by
hardware and then write TDAR, otherwise don't set TDAR.
Fixes: 53bb20d1faba ("net: fec: add variable reg_desc_active to speed things up") Signed-off-by: Mattias Barthel <mattias.barthel@atlascopco.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://patch.msgid.link/20250429090826.3101258-1-mattiasbarthel@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Always map the `skb` to the LS descriptor. Previously skb was
mapped to EXT descriptor when the number of fragments is zero with
GSO enabled. Mapping the skb to EXT descriptor prevents it from
being freed, leading to a memory leak
Fixes: 23f0703c125b ("lan743x: Add main source files for new lan743x driver") Signed-off-by: Thangaraj Samynathan <thangaraj.s@microchip.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Link: https://patch.msgid.link/20250429052527.10031-1-thangaraj.s@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch addresses a data corruption issue observed in nvme-tcp during
testing.
In an NVMe native multipath setup, when an I/O timeout occurs, all
inflight I/Os are canceled almost immediately after the kernel socket is
shut down. These canceled I/Os are reported as host path errors,
triggering a failover that succeeds on a different path.
However, at this point, the original I/O may still be outstanding in the
host's network transmission path (e.g., the NIC’s TX queue). From the
user-space app's perspective, the buffer associated with the I/O is
considered completed since they're acked on the different path and may
be reused for new I/O requests.
Because nvme-tcp enables zero-copy by default in the transmission path,
this can lead to corrupted data being sent to the original target,
ultimately causing data corruption.
We can reproduce this data corruption by injecting delay on one path and
triggering i/o timeout.
To prevent this issue, this change ensures that all inflight
transmissions are fully completed from host's perspective before
returning from queue stop. To handle concurrent I/O timeout from multiple
namespaces under the same controller, always wait in queue stop
regardless of queue's state.
This aligns with the behavior of queue stopping in other NVMe fabric
transports.
Fixes: 3f2304f8c6d6 ("nvme-tcp: add NVMe over TCP host driver") Signed-off-by: Michael Liang <mliang@purestorage.com> Reviewed-by: Mohamed Khalfella <mkhalfella@purestorage.com> Reviewed-by: Randy Jennings <randyj@purestorage.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
For version 1 register dump that includes the PCIe stats, the existing
code incorrectly assumes that all PCIe stats are 64-bit values. Fix it
by using an array containing the starting and ending index of the 32-bit
values. The loop in bnxt_get_regs() will use the array to do proper
endian swap for the 32-bit values.
Fixes: b5d600b027eb ("bnxt_en: Add support for 'ethtool -d'") Reviewed-by: Shruti Parab <shruti.parab@broadcom.com> Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com> Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
This happens when copying the coredump segment list in
bnxt_hwrm_dbg_dma_data() with the HWRM_DBG_COREDUMP_LIST FW command.
The info->dest_buf buffer is allocated based on the number of coredump
segments returned by the FW. The segment list is then DMA'ed by
the FW and the length of the DMA is returned by FW. The driver then
copies this DMA'ed segment list to info->dest_buf.
In some cases, this DMA length may exceed the info->dest_buf length
and cause the above BUG condition. Fix it by capping the copy
length to not exceed the length of info->dest_buf. The extra
DMA data contains no useful information.
This code path is shared for the HWRM_DBG_COREDUMP_LIST and the
HWRM_DBG_COREDUMP_RETRIEVE FW commands. The buffering is different
for these 2 FW commands. To simplify the logic, we need to move
the line to adjust the buffer length for HWRM_DBG_COREDUMP_RETRIEVE
up, so that the new check to cap the copy length will work for both
commands.
Fixes: c74751f4c392 ("bnxt_en: Return error if FW returns more data than dump length") Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com> Signed-off-by: Shruti Parab <shruti.parab@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When handling HWRM_DBG_COREDUMP_LIST FW command in
bnxt_hwrm_dbg_dma_data(), the allocated buffer info->dest_buf is
not freed in the error path. In the normal path, info->dest_buf
is assigned to coredump->data and it will eventually be freed after
the coredump is collected.
Free info->dest_buf immediately inside bnxt_hwrm_dbg_dma_data() in
the error path.
Fixes: c74751f4c392 ("bnxt_en: Return error if FW returns more data than dump length") Reported-by: Michael Chan <michael.chan@broadcom.com> Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com> Signed-off-by: Shruti Parab <shruti.parab@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Simplest setup to reproduce the issue: connect 2 ports of the
LS1028A-RDB together (eno0 with swp0) and run:
$ ip link set eno0 up && ip link set swp0 up
$ tc qdisc replace dev swp0 parent root handle 100 taprio num_tc 8 \
queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 map 0 1 2 3 4 5 6 7 \
base-time 0 sched-entry S 20 300000 sched-entry S 10 200000 \
sched-entry S 20 300000 sched-entry S 48 200000 \
sched-entry S 20 300000 sched-entry S 83 200000 \
sched-entry S 40 300000 sched-entry S 00 200000 flags 2
$ ptp4l -i eno0 -f /etc/linuxptp/configs/gPTP.cfg -m &
$ ptp4l -i swp0 -f /etc/linuxptp/configs/gPTP.cfg -m
One will observe that the PTP state machine on swp0 starts
synchronizing, then it attempts to do a clock step, and after that, it
never fails to recover from the condition below.
ptp4l[82.427]: selected best master clock 00049f.fffe.05f627
ptp4l[82.428]: port 1 (swp0): MASTER to UNCALIBRATED on RS_SLAVE
ptp4l[83.252]: port 1 (swp0): UNCALIBRATED to SLAVE on MASTER_CLOCK_SELECTED
ptp4l[83.886]: rms 4537731277 max 9075462553 freq -18518 +/- 11467 delay 818 +/- 0
ptp4l[84.170]: timed out while polling for tx timestamp
ptp4l[84.171]: increasing tx_timestamp_timeout or increasing kworker priority may correct this issue, but a driver bug likely causes it
ptp4l[84.172]: port 1 (swp0): send peer delay request failed
ptp4l[84.173]: port 1 (swp0): clearing fault immediately
ptp4l[84.269]: port 1 (swp0): SLAVE to LISTENING on INIT_COMPLETE
ptp4l[85.303]: timed out while polling for tx timestamp
ptp4l[84.171]: increasing tx_timestamp_timeout or increasing kworker priority may correct this issue, but a driver bug likely causes it
ptp4l[84.172]: port 1 (swp0): send peer delay request failed
ptp4l[84.173]: port 1 (swp0): clearing fault immediately
ptp4l[84.269]: port 1 (swp0): SLAVE to LISTENING on INIT_COMPLETE
ptp4l[85.303]: timed out while polling for tx timestamp
ptp4l[85.304]: increasing tx_timestamp_timeout or increasing kworker priority may correct this issue, but a driver bug likely causes it
ptp4l[85.305]: port 1 (swp0): send peer delay response failed
ptp4l[85.306]: port 1 (swp0): clearing fault immediately
ptp4l[86.304]: timed out while polling for tx timestamp
A hint is given by the non-zero statistics for dropped packets which
were expecting hardware TX timestamps:
We know that when PTP clock stepping takes place (from ocelot_ptp_settime64()
or from ocelot_ptp_adjtime()), vsc9959_tas_clock_adjust() is called.
Another interesting hint is that placing an early return in
vsc9959_tas_clock_adjust(), so as to neutralize this function, fixes the
issue and TX timestamps are no longer dropped.
The debugging function written by me and included below is intended to
read the GCL RAM, after the admin schedule became operational, through
the two status registers available for this purpose:
QSYS_GCL_STATUS_REG_1 and QSYS_GCL_STATUS_REG_2.
static void vsc9959_print_tas_gcl(struct ocelot *ocelot)
{
u32 val, list_length, interval, gate_state;
int i, err;
err = read_poll_timeout(ocelot_read, val,
!(val & QSYS_PARAM_STATUS_REG_8_CONFIG_PENDING),
10, 100000, false, ocelot, QSYS_PARAM_STATUS_REG_8);
if (err) {
dev_err(ocelot->dev,
"Failed to wait for TAS config pending bit to clear: %pe\n",
ERR_PTR(err));
return;
}
val = ocelot_read(ocelot, QSYS_PARAM_STATUS_REG_3);
list_length = QSYS_PARAM_STATUS_REG_3_LIST_LENGTH_X(val);
for (i = 0; i < list_length; i++) {
ocelot_rmw(ocelot,
QSYS_GCL_STATUS_REG_1_GCL_ENTRY_NUM(i),
QSYS_GCL_STATUS_REG_1_GCL_ENTRY_NUM_M,
QSYS_GCL_STATUS_REG_1);
interval = ocelot_read(ocelot, QSYS_GCL_STATUS_REG_2);
val = ocelot_read(ocelot, QSYS_GCL_STATUS_REG_1);
gate_state = QSYS_GCL_STATUS_REG_1_GATE_STATE_X(val);
dev_info(ocelot->dev, "GCL entry %d: states 0x%x interval %u\n",
i, gate_state, interval);
}
}
Calling it from two places: after the initial QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE
performed by vsc9959_qos_port_tas_set(), and after the one done by
vsc9959_tas_clock_adjust(), I notice the following difference.
From the tc-taprio process context, where the schedule was initially
configured, the GCL looks like this:
mscc_felix 0000:00:00.5: GCL length: 8
mscc_felix 0000:00:00.5: GCL entry 0: states 0x20 interval 300000
mscc_felix 0000:00:00.5: GCL entry 1: states 0x10 interval 200000
mscc_felix 0000:00:00.5: GCL entry 2: states 0x20 interval 300000
mscc_felix 0000:00:00.5: GCL entry 3: states 0x48 interval 200000
mscc_felix 0000:00:00.5: GCL entry 4: states 0x20 interval 300000
mscc_felix 0000:00:00.5: GCL entry 5: states 0x83 interval 200000
mscc_felix 0000:00:00.5: GCL entry 6: states 0x40 interval 300000
mscc_felix 0000:00:00.5: GCL entry 7: states 0x0 interval 200000
But from the ptp4l clock stepping process context, when the
vsc9959_tas_clock_adjust() hook is called, the GCL RAM of the
operational schedule now looks like this:
mscc_felix 0000:00:00.5: GCL length: 8
mscc_felix 0000:00:00.5: GCL entry 0: states 0x0 interval 0
mscc_felix 0000:00:00.5: GCL entry 1: states 0x0 interval 0
mscc_felix 0000:00:00.5: GCL entry 2: states 0x0 interval 0
mscc_felix 0000:00:00.5: GCL entry 3: states 0x0 interval 0
mscc_felix 0000:00:00.5: GCL entry 4: states 0x0 interval 0
mscc_felix 0000:00:00.5: GCL entry 5: states 0x0 interval 0
mscc_felix 0000:00:00.5: GCL entry 6: states 0x0 interval 0
mscc_felix 0000:00:00.5: GCL entry 7: states 0x0 interval 0
I do not have a formal explanation, just experimental conclusions.
It appears that after triggering QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE
for a port's TAS, the GCL entry RAM is updated anyway, despite what the
documentation claims: "Specify the time interval in
QSYS::GCL_CFG_REG_2.TIME_INTERVAL. This triggers the actual RAM
write with the gate state and the time interval for the entry number
specified". We don't touch that register (through vsc9959_tas_gcl_set())
from vsc9959_tas_clock_adjust(), yet the GCL RAM is updated anyway.
It seems to be updated with effectively stale memory, which in my
testing can hold a variety of things, including even pieces of the
previously applied schedule, for particular schedule lengths.
As such, in most circumstances it is very difficult to pinpoint this
issue, because the newly updated schedule would "behave strangely",
but ultimately might still pass traffic to some extent, due to some
gate entries still being present in the stale GCL entry RAM. It is easy
to miss.
With the particular schedule given at the beginning, the GCL RAM
"happens" to be reproducibly rewritten with all zeroes, and this is
consistent with what we see: when the time-aware shaper has gate entries
with all gates closed, traffic is dropped on TX, no wonder we can't
retrieve TX timestamps.
Rewriting the GCL entry RAM when reapplying the new base time fixes the
observed issue.
Fixes: 8670dc33f48b ("net: dsa: felix: update base time of time-aware shaper when adjusting PTP time") Reported-by: Richie Pearn <richard.pearn@nxp.com> Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20250426144859.3128352-2-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
As it's name suggests, parse_eeprom() parses EEPROM data.
This is done by reading data, 16 bits at a time as follows:
for (i = 0; i < 128; i++)
((__le16 *) sromdata)[i] = cpu_to_le16(read_eeprom(np, i));
sromdata is at the same memory location as psrom.
And the type of psrom is a pointer to struct t_SROM.
As can be seen in the loop above, data is stored in sromdata, and thus psrom,
as 16-bit little-endian values.
However, the integer fields of t_SROM are host byte order integers.
And in the case of led_mode this leads to a little endian value
being incorrectly treated as host byte order.
Looking at rio_set_led_mode, this does appear to be a bug as that code
masks led_mode with 0x1, 0x2 and 0x8. Logic that would be effected by a
reversed byte order.
This problem would only manifest on big endian hosts.
Found by inspection while investigating a sparse warning
regarding the crc field of t_SROM.
I believe that warning is a false positive. And although I plan
to send a follow-up to use little-endian types for other the integer
fields of PSROM_t I do not believe that will involve any bug fixes.
As mentioned in the commit baeb705fd6a7 ("ice: always check VF VSI
pointer values"), we need to perform a null pointer check on the return
value of ice_get_vf_vsi() before using it.
Fixes: 6ebbe97a4881 ("ice: Add a per-VF limit on number of FDIR filters") Signed-off-by: Xuanqiang Luo <luoxuanqiang@kylinos.cn> Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Link: https://patch.msgid.link/20250425222636.3188441-3-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
As described in Gerrard's report [1], there are use cases where a netem
child qdisc will make the parent qdisc's enqueue callback reentrant.
In the case of qfq, there won't be a UAF, but the code will add the same
classifier to the list twice, which will cause memory corruption.
This patch checks whether the class was already added to the agg->active
list (cl_is_active) before doing the addition to cater for the reentrant
case.
As described in Gerrard's report [1], there are use cases where a netem
child qdisc will make the parent qdisc's enqueue callback reentrant.
In the case of ets, there won't be a UAF, but the code will add the same
classifier to the list twice, which will cause memory corruption.
In addition to checking for qlen being zero, this patch checks whether
the class was already added to the active_list (cl_is_active) before
doing the addition to cater for the reentrant case.
As described in Gerrard's report [1], we have a UAF case when an hfsc class
has a netem child qdisc. The crux of the issue is that hfsc is assuming
that checking for cl->qdisc->q.qlen == 0 guarantees that it hasn't inserted
the class in the vttree or eltree (which is not true for the netem
duplicate case).
This patch checks the n_active class variable to make sure that the code
won't insert the class in the vttree or eltree twice, catering for the
reentrant case.
As described in Gerrard's report [1], there are use cases where a netem
child qdisc will make the parent qdisc's enqueue callback reentrant.
In the case of drr, there won't be a UAF, but the code will add the same
classifier to the list twice, which will cause memory corruption.
In addition to checking for qlen being zero, this patch checks whether the
class was already added to the active_list (cl_is_active) before adding
to the list to cover for the reentrant case.
In mtk_star_rx_poll function, on event processing completion, the
mtk_star_emac driver calls napi_complete_done but ignores its return
code and enable RX DMA interrupts inconditionally. This return code
gives the info if a device should avoid rearming its interrupts or not,
so fix this behaviour by taking it into account.
Use spin_lock_irqsave and spin_unlock_irqrestore instead of spin_lock
and spin_unlock in mtk_star_emac driver to avoid spinlock recursion
occurrence that can happen when enabling the DMA interrupts again in
rx/tx poll.
ip link add br0 type bridge vlan_filtering 1 # vlan_default_pvid 1 is implicit
ip link set swp0 master br0
bridge vlan add dev swp0 vid 1
should result in the dropping of untagged and 802.1p-tagged traffic, but
we see that it continues to be accepted. Whereas, had we deleted VID 1
instead, the aforementioned dropping would have worked
This is because the ANA_PORT_DROP_CFG update logic doesn't run, because
ocelot_vlan_add() only calls ocelot_port_set_pvid() if the new VLAN has
the BRIDGE_VLAN_INFO_PVID flag.
Similar to other drivers like mt7530_port_vlan_add() which handle this
case correctly, we need to test whether the VLAN we're changing used to
have the BRIDGE_VLAN_INFO_PVID flag, but lost it now. That amounts to a
PVID deletion and should be treated as such.
Regarding blame attribution: this never worked properly since the
introduction of bridge VLAN filtering in commit 7142529f1688 ("net:
mscc: ocelot: add VLAN filtering"). However, there was a significant
paradigm shift which aligned the ANA_PORT_DROP_CFG register with the
PVID concept rather than with the native VLAN concept, and that change
wasn't targeted for 'stable'. Realistically, that is as far as this fix
needs to be propagated to.
Fixes: be0576fed6d3 ("net: mscc: ocelot: move the logic to drop 802.1p traffic to the pvid deletion") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20250424223734.3096202-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
I was revisiting the topic of 802.1ad treatment in the Ocelot switch [0]
and realized that not only is its basic VLAN classification pipeline
improper for offloading vlan_protocol 802.1ad bridges, but also improper
for offloading regular 802.1Q bridges already.
Namely, 802.1ad-tagged traffic should be treated as VLAN-untagged by
bridged ports, but this switch treats it as if it was 802.1Q-tagged with
the same VID as in the 802.1ad header. This is markedly different to
what the Linux bridge expects; see the "other_tpid()" function in
tools/testing/selftests/net/forwarding/bridge_vlan_aware.sh.
An idea came to me that the VCAP IS1 TCAM is more powerful than I'm
giving it credit for, and that it actually overwrites the classified VID
before the VLAN Table lookup takes place. In other words, it can be
used even to save a packet from being dropped on ingress due to VLAN
membership.
Add a sophisticated TCAM rule hardcoded into the driver to force the
switch to behave like a Linux bridge with vlan_filtering 1 vlan_protocol
802.1Q.
Regarding the lifetime of the filter: eventually the bridge will
disappear, and vlan_filtering on the port will be restored to 0 for
standalone mode. Then the filter will be deleted.
Fixes: 7142529f1688 ("net: mscc: ocelot: add VLAN filtering") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: 5ec6d7d737a4 ("net: mscc: ocelot: delete PVID VLAN when readding it as non-PVID") Signed-off-by: Sasha Levin <sashal@kernel.org>
The cited commit assumes enabling roce always succeeds. But it is
not true. Add error handling for it.
Fixes: 80f09dfc237f ("net/mlx5: Eswitch, enable RoCE loopback traffic") Signed-off-by: Chris Mi <cmi@nvidia.com> Reviewed-by: Roi Dayan <roid@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Mark Bloch <mbloch@nvidia.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Link: https://patch.msgid.link/20250423083611.324567-6-mbloch@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Initialize the source MAC address when creating the default GID entry.
Since this entry is used only for loopback traffic, it only needs to
be a unicast address. A zeroed-out MAC address is sufficient for this
purpose.
Without this fix, random bits would be assigned as the source address.
If these bits formed a multicast address, the firmware would return an
error, preventing the user from switching to switchdev mode:
When a VNI is deleted from a VXLAN device in 'vnifilter' mode, the FDB
entry associated with the default remote (assuming one was configured)
is deleted without holding the hash lock. This is wrong and will result
in a warning [1] being generated by the lockdep annotation that was
added by commit ebe642067455 ("vxlan: Create wrappers for FDB lookup").
Reproducer:
# ip link add vx0 up type vxlan dstport 4789 external vnifilter local 192.0.2.1
# bridge vni add vni 10010 remote 198.51.100.1 dev vx0
# bridge vni del vni 10010 dev vx0
Fix by acquiring the hash lock before the deletion and releasing it
afterwards. Blame the original commit that introduced the issue rather
than the one that exposed it.
plfxlc_mac_release() asserts that mac->lock is held. This assertion is
incorrect, because even if it was possible, it would not be the valid
behaviour. The function is used when probe fails or after the device is
disconnected. In both cases mac->lock can not be held as the driver is
not working with the device at the moment. All functions that use mac->lock
unlock it just after it was held. There is also no need to hold mac->lock
for plfxlc_mac_release() itself, as mac data is not affected, except for
mac->flags, which is modified atomically.
Issue:
When multiple audio streams share a common BE DAI, the BE DAI
widget can be powered up before its hardware parameters are configured.
This incorrect sequence leads to intermittent pcm_write errors.
For example, the below Tegra use-case throws an error:
aplay(2 streams) -> AMX(mux) -> ADX(demux) -> arecord(2 streams),
here, 'AMX TX' and 'ADX RX' are common BE DAIs.
For above usecase when failure happens below sequence is observed:
aplay(1) FE open()
- BE DAI callbacks added to the list
- BE DAI state = SND_SOC_DPCM_STATE_OPEN
aplay(2) FE open()
- BE DAI callbacks are not added to the list as the state is
already SND_SOC_DPCM_STATE_OPEN during aplay(1) FE open().
aplay(2) FE hw_params()
- BE DAI hw_params() callback ignored
aplay(2) FE prepare()
- Widget is powered ON without BE DAI hw_params() call
aplay(1) FE hw_params()
- BE DAI hw_params() is now called
Fix:
Add BE DAIs in the list if its state is either SND_SOC_DPCM_STATE_OPEN
or SND_SOC_DPCM_STATE_HW_PARAMS as well.
It ensures the widget is powered ON after BE DAI hw_params() callback.
Fixes: 0c25db3f7621 ("ASoC: soc-pcm: Don't reconnect an already active BE") Signed-off-by: Sheetal <sheetal@nvidia.com> Link: https://patch.msgid.link/20250404105953.2784819-1-sheetal@nvidia.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Move the conditional loading of hardware DR6 with the guest's DR6 value
out of the core .vcpu_run() loop to fix a bug where KVM can load hardware
with a stale vcpu->arch.dr6.
When the guest accesses a DR and host userspace isn't debugging the guest,
KVM disables DR interception and loads the guest's values into hardware on
VM-Enter and saves them on VM-Exit. This allows the guest to access DRs
at will, e.g. so that a sequence of DR accesses to configure a breakpoint
only generates one VM-Exit.
For DR0-DR3, the logic/behavior is identical between VMX and SVM, and also
identical between KVM_DEBUGREG_BP_ENABLED (userspace debugging the guest)
and KVM_DEBUGREG_WONT_EXIT (guest using DRs), and so KVM handles loading
DR0-DR3 in common code, _outside_ of the core kvm_x86_ops.vcpu_run() loop.
But for DR6, the guest's value doesn't need to be loaded into hardware for
KVM_DEBUGREG_BP_ENABLED, and SVM provides a dedicated VMCB field whereas
VMX requires software to manually load the guest value, and so loading the
guest's value into DR6 is handled by {svm,vmx}_vcpu_run(), i.e. is done
_inside_ the core run loop.
Unfortunately, saving the guest values on VM-Exit is initiated by common
x86, again outside of the core run loop. If the guest modifies DR6 (in
hardware, when DR interception is disabled), and then the next VM-Exit is
a fastpath VM-Exit, KVM will reload hardware DR6 with vcpu->arch.dr6 and
clobber the guest's actual value.
The bug shows up primarily with nested VMX because KVM handles the VMX
preemption timer in the fastpath, and the window between hardware DR6
being modified (in guest context) and DR6 being read by guest software is
orders of magnitude larger in a nested setup. E.g. in non-nested, the
VMX preemption timer would need to fire precisely between #DB injection
and the #DB handler's read of DR6, whereas with a KVM-on-KVM setup, the
window where hardware DR6 is "dirty" extends all the way from L1 writing
DR6 to VMRESUME (in L1).
L1's view:
==========
<L1 disables DR interception>
CPU 0/KVM-7289 [023] d.... 2925.640961: kvm_entry: vcpu 0
A: L1 Writes DR6
CPU 0/KVM-7289 [023] d.... 2925.640963: <hack>: Set DRs, DR6 = 0xffff0ff1
Reported-by: John Stultz <jstultz@google.com> Closes: https://lkml.kernel.org/r/CANDhNCq5_F3HfFYABqFGCA1bPd_%2BxgNj-iDQhH4tDk%2Bwi8iZZg%40mail.gmail.com Fixes: 375e28ffc0cf ("KVM: X86: Set host DR6 only on VMX and for KVM_DEBUGREG_WONT_EXIT") Fixes: d67668e9dd76 ("KVM: x86, SVM: isolate vcpu->arch.dr6 from vmcb->save.dr6") Cc: stable@vger.kernel.org Cc: Jim Mattson <jmattson@google.com> Tested-by: John Stultz <jstultz@google.com> Link: https://lore.kernel.org/r/20250125011833.3644371-1-seanjc@google.com Signed-off-by: Sean Christopherson <seanjc@google.com>
[jth: Handled conflicts with kvm_x86_ops reshuffle] Signed-off-by: James Houghton <jthoughton@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xfs/205 produces the following failure when always_cow is enabled:
# --- a/tests/xfs/205.out 2024-02-28 16:20:24.437887970 -0800
# +++ b/tests/xfs/205.out.bad 2024-06-03 21:13:40.584000000 -0700
# @@ -1,4 +1,5 @@
# QA output created by 205
# *** one file
# + !!! disk full (expected)
# *** one file, a few bytes at a time
# *** done
This is the result of overly aggressive attempts to align cow fork
delalloc reservations to the CoW extent size hint. Looking at the trace
data, we're trying to append a single fsblock to the "fred" file.
Trying to create a speculative post-eof reservation fails because
there's not enough space.
We then set @prealloc_blocks to zero and try again, but the cowextsz
alignment code triggers, which expands our request for a 1-fsblock
reservation into a 39-block reservation. There's not enough space for
that, so the whole write fails with ENOSPC even though there's
sufficient space in the filesystem to allocate the single block that we
need to land the write.
There are two things wrong here -- first, we shouldn't be attempting
speculative preallocations beyond what was requested when we're low on
space. Second, if we've already computed a posteof preallocation, we
shouldn't bother trying to align that to the cowextsize hint.
Fix both of these problems by adding a flag that only enables the
expansion of the delalloc reservation to the cowextsize if we're doing a
non-extending write, and only if we're not doing an ENOSPC retry. This
requires us to move the ENOSPC retry logic to xfs_bmapi_reserve_delalloc.
I probably should have caught this six years ago when 6ca30729c206d was
being reviewed, but oh well. Update the comments to reflect what the
code does now.
Fixes: 6ca30729c206d ("xfs: bmap code cleanup") Signed-off-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For a very very long time, inode inactivation has set the inode size to
zero before unmapping the extents associated with the data fork.
Unfortunately, commit 3c6f46eacd876 changed the inode verifier to
prohibit zero-length symlinks and directories. If an inode happens to
get logged in this state and the system crashes before freeing the
inode, log recovery will also fail on the broken inode.
Therefore, allow zero-size symlinks and directories as long as the link
count is zero; nobody will be able to open these files by handle so
there isn't any risk of data exposure.
Fixes: 3c6f46eacd876 ("xfs: sanity check directory inode di_size") Signed-off-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xfs_can_free_eofblocks returns false for files that have persistent
preallocations unless the force flag is passed and there are delayed
blocks. This means it won't free delalloc reservations for files
with persistent preallocations unless the force flag is set, and it
will also free the persistent preallocations if the force flag is
set and the file happens to have delayed allocations.
Both of these are bad, so do away with the force flag and always free
only post-EOF delayed allocations for files with the XFS_DIFLAG_PREALLOC
or APPEND flags set.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A user with a completely full filesystem experienced an unexpected
shutdown when the filesystem tried to write the superblock during
runtime.
kernel shows the following dmesg:
When xfs_log_sb writes super block to disk, b_fdblocks is fetched from
m_fdblocks without any lock. As m_fdblocks can experience a positive ->
negative -> positive changing when the FS reaches fullness (see
xfs_mod_fdblocks). So there is a chance that sb_fdblocks is negative, and
because sb_fdblocks is type of unsigned long long, it reads super big.
And sb_fdblocks being bigger than sb_dblocks is a problem during log
recovery, xfs_validate_sb_write() complains.
Fix:
As sb_fdblocks will be re-calculated during mount when lazysbcount is
enabled, We just need to make xfs_validate_sb_write() happy -- make sure
sb_fdblocks is not nenative. This patch also takes care of other percpu
counters in xfs_log_sb.
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is a symbolic link with a 297-byte target stored in a disk block,
which is to say this is a symlink with a remote target. The forkoff is
0, which is to say that there's 512 - 176 == 336 bytes in the inode core
to store the data fork.
Eventually, testing of generic/388 failed with the same inode corruption
message during inode recovery. In writing a debugging patch to call
xfs_dinode_verify on dirty inode log items when we're committing
transactions, I observed that xfs/298 can reproduce the problem quite
quickly.
xfs/298 creates a symbolic link, adds some extended attributes, then
deletes them all. The test failure occurs when the final removexattr
also deletes the attr fork because that does not convert the remote
symlink back into a shortform symlink. That is how we trip this test.
The only reason why xfs/298 only triggers with the debug patch added is
that it deletes the symlink, so the final iflush shows the inode as
free.
I wrote a quick fstest to emulate the behavior of xfs/298, except that
it leaves the symlinks on the filesystem after inducing the "corrupt"
state. Kernels going back at least as far as 4.18 have written out
symlink inodes in this manner and prior to 1eb70f54c445f they did not
object to reading them back in.
Because we've been writing out inodes this way for quite some time, the
only way to fix this is to relax the check for symbolic links.
Directories don't have this problem because di_size is bumped to
blocksize during the sf->data conversion.
Fixes: 1eb70f54c445f ("xfs: validate inode fork size against fork format") Signed-off-by: "Darrick J. Wong" <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Current clone operation could be non-atomic if the destination of a file
is beyond EOF, user could get a file with corrupted (zeroed) data on
crash.
The problem is about preallocations. If you write some data into a file:
[A...B)
and XFS decides to preallocate some post-eof blocks, then it can create
a delayed allocation reservation:
[A.........D)
The writeback path tries to convert delayed extents to real ones by
allocating blocks. If there aren't enough contiguous free space, we can
end up with two extents, the first real and the second still delalloc:
[A....C)[C.D)
After that, both the in-memory and the on-disk file sizes are still B.
If we clone into the range [E...F) from another file:
[A....C)[C.D) [E...F)
then xfs_reflink_zero_posteof() calls iomap_zero_range() to zero out the
range [B, E) beyond EOF and flush it. Since [C, D) is still a delalloc
extent, its pagecache will be zeroed and both the in-memory and on-disk
size will be updated to D after flushing but before cloning. This is
wrong, because the user can see the size change and read the zeroes
while the clone operation is ongoing.
We need to keep the in-memory and on-disk size before the clone
operation starts, so instead of writing zeroes through the page cache
for delayed ranges beyond EOF, we convert these ranges to unwritten and
invalidate any cached data over that range beyond EOF.
Suggested-by: Dave Chinner <david@fromorbit.com> Signed-off-by: Zhang Yi <yi.zhang@huawei.com> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since xfs_bmapi_convert_delalloc() only attempts to allocate the entire
delalloc extent and require multiple invocations to allocate the target
offset. So xfs_convert_blocks() add a loop to do this job and we call it
in the write back path, but xfs_convert_blocks() isn't a common helper.
Let's do it in xfs_bmapi_convert_delalloc() and drop
xfs_convert_blocks(), preparing for the post EOF delalloc blocks
converting in the buffered write begin path.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 1aa91d9c9933 ("xfs: Add async buffered write support") replace
xfs_ilock(XFS_ILOCK_EXCL) with xfs_ilock_for_iomap() when locking the
writing inode, and a new variable lockmode is used to indicate the lock
mode. Although the lockmode should always be XFS_ILOCK_EXCL, it's still
better to use this variable instead of useing XFS_ILOCK_EXCL directly
when unlocking the inode.
Fixes: 1aa91d9c9933 ("xfs: Add async buffered write support") Signed-off-by: Zhang Yi <yi.zhang@huawei.com> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In my haste to fix what I thought was a performance problem in the attr
scrub code, I neglected to notice that the xfs_attr_get_ilocked also had
the effect of checking that attributes can actually be looked up through
the attr dabtree. Fix this.
Fixes: 44af6c7e59b12 ("xfs: don't load local xattr values during scrub") Signed-off-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Strengthen the xattri log item recovery code by checking that we
actually have the required name and newname buffers for whatever
operation we're replaying.
Signed-off-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The XFS_SB_FEAT_INCOMPAT_LOG_XATTRS feature bit protects a filesystem
from old kernels that do not know how to recover extended attribute log
intent items. Make this check mandatory instead of a debugging assert.
Fixes: fd920008784ea ("xfs: Set up infrastructure for log attribute replay") Signed-off-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Accessing if_bytes without the ilock is racy. Remove the initial
if_bytes == 0 check in xfs_reflink_end_cow_extent and let
ext_iext_lookup_extent fail for this case after we've taken the ilock.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xfs_bmap_add_extent_delay_real takes parts or all of a delalloc extent
and converts them to a real extent. It is written to deal with any
potential overlap of the to be converted range with the delalloc extent,
but it turns out that currently only converting the entire extents, or a
part starting at the beginning is actually exercised, as the only caller
always tries to convert the entire delalloc extent, and either succeeds
or at least progresses partially from the start.
If it only converts a tiny part of a delalloc extent, the indirect block
calculation for the new delalloc extent (da_new) might be equivalent to that
of the existing delalloc extent (da_old). If this extent conversion now
requires allocating an indirect block that gets accounted into da_new,
leading to the assert that da_new must be smaller or equal to da_new
unless we split the extent to trigger.
Except for the assert that case is actually handled by just trying to
allocate more space, as that already handled for the split case (which
currently can't be reached at all), so just reusing it should be fine.
Except that without dipping into the reserved block pool that would make
it a bit too easy to trigger a fs shutdown due to ENOSPC. So in addition
to adjusting the assert, also dip into the reserved block pool.
Note that I could only reproduce the assert with a change to only convert
the actually asked range instead of the full delalloc extent from
xfs_bmapi_write.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xfs_bmapi_write can return 0 without actually returning a mapping in
mval in two different cases:
1) when there is absolutely no space available to do an allocation
2) when converting delalloc space, and the allocation is so small
that it only covers parts of the delalloc extent before the
range requested by the caller
Callers at best can handle one of these cases, but in many cases can't
cope with either one. Switch xfs_bmapi_write to always return a
mapping or return an error code instead. For case 1) above ENOSPC is
the obvious choice which is very much what the callers expect anyway.
For case 2) there is no really good error code, so pick a funky one
from the SysV streams portfolio.
which uses reserved blocks to create file systems that are gravely
out of space and thus cause at least xfs_file_alloc_space to hang
and trigger the lack of ENOSPC handling in xfs_dquot_disk_alloc.
Note that this patch does not actually make any caller but
xfs_alloc_file_space deal intelligently with case 2) above.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reported-by: 刘通 <lyutoon@gmail.com> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Chandan Babu R <chandanbabu@kernel.org> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com> Acked-by: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
syzbot reported this bug:
==================================================================
BUG: KASAN: slab-out-of-bounds in trace_seq_to_buffer kernel/trace/trace.c:1830 [inline]
BUG: KASAN: slab-out-of-bounds in tracing_splice_read_pipe+0x6be/0xdd0 kernel/trace/trace.c:6822
Write of size 4507 at addr ffff888032b6b000 by task syz.2.320/7260
It has been reported that trace_seq_to_buffer() tries to copy more data
than PAGE_SIZE to buf. Therefore, to prevent this, we should use the
smaller of trace_seq_used(&iter->seq) and PAGE_SIZE as an argument.
Link: https://lore.kernel.org/20250422113026.13308-1-aha310510@gmail.com Reported-by: syzbot+c8cd2d2c412b868263fb@syzkaller.appspotmail.com Fixes: 3c56819b14b0 ("tracing: splice support for tracing_pipe") Suggested-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Jeongjun Park <aha310510@gmail.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 7491cdf46b5c ("cpufreq: Avoid using inconsistent policy->min and
policy->max") overlooked the fact that policy->min and policy->max were
accessed directly in cpufreq_frequency_table_target() and in the
functions called by it. Consequently, the changes made by that commit
led to problems with setting policy limits.
Address this by passing the target frequency limits to __resolve_freq()
and cpufreq_frequency_table_target() and propagating them to the
functions called by the latter.
Fixes: 7491cdf46b5c ("cpufreq: Avoid using inconsistent policy->min and policy->max") Cc: 5.16+ <stable@vger.kernel.org> # 5.16+ Closes: https://lore.kernel.org/linux-pm/aAplED3IA_J0eZN0@linaro.org/ Reported-by: Stephan Gerhold <stephan.gerhold@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Tested-by: Stephan Gerhold <stephan.gerhold@linaro.org> Reviewed-by: Lifeng Zheng <zhenglifeng1@huawei.com> Link: https://patch.msgid.link/5896780.DvuYhMxLoT@rjwysocki.net Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since cpufreq_driver_resolve_freq() can run in parallel with
cpufreq_set_policy() and there is no synchronization between them,
the former may access policy->min and policy->max while the latter
is updating them and it may see intermediate values of them due
to the way the update is carried out. Also the compiler is free
to apply any optimizations it wants both to the stores in
cpufreq_set_policy() and to the loads in cpufreq_driver_resolve_freq()
which may result in additional inconsistencies.
To address this, use WRITE_ONCE() when updating policy->min and
policy->max in cpufreq_set_policy() and use READ_ONCE() for reading
them in cpufreq_driver_resolve_freq(). Moreover, rearrange the update
in cpufreq_set_policy() to avoid storing intermediate values in
policy->min and policy->max with the help of the observation that
their new values are expected to be properly ordered upfront.
Also modify cpufreq_driver_resolve_freq() to take the possible reverse
ordering of policy->min and policy->max, which may happen depending on
the ordering of operations when this function and cpufreq_set_policy()
run concurrently, into account by always honoring the max when it
turns out to be less than the min (in case it comes from thermal
throttling or similar).
Fixes: 151717690694 ("cpufreq: Make policy min/max hard requirements") Cc: 5.16+ <stable@vger.kernel.org> # 5.16+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Christian Loehle <christian.loehle@arm.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Link: https://patch.msgid.link/5907080.DvuYhMxLoT@rjwysocki.net Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Setting sess->user = NULL was introduced to fix the dangling pointer
created by ksmbd_free_user. However, it is possible another thread could
be operating on the session and make use of sess->user after it has been
passed to ksmbd_free_user but before sess->user is set to NULL.
Cc: stable@vger.kernel.org Signed-off-by: Sean Heelan <seanheelan@gmail.com> Acked-by: Namjae Jeon <linkinjeon@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In certain situations, the sysfs for uncore may not be present when all
CPUs in a package are offlined and then brought back online after boot.
This issue can occur if there is an error in adding the sysfs entry due
to a memory allocation failure. Retrying to bring the CPUs online will
not resolve the issue, as the uncore_cpu_mask is already set for the
package before the failure condition occurs.
This issue does not occur if the failure happens during module
initialization, as the module will fail to load in the event of any
error.
To address this, ensure that the uncore_cpu_mask is not set until the
successful return of uncore_freq_add_entry().
Fixes: dbce412a7733 ("platform/x86/intel-uncore-freq: Split common and enumeration part") Signed-off-by: Shouye Liu <shouyeliu@tencent.com> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250417032321.75580-1-shouyeliu@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On the Lenovo ThinkPad X201, when Intel VT-d is enabled in the BIOS, the
kernel boots with errors related to DMAR, the graphical interface appeared
quite choppy, and the system resets erratically within a minute after it
booted:
DMAR: DRHD: handling fault status reg 3
DMAR: [DMA Write NO_PASID] Request device [00:02.0] fault addr 0xb97ff000
[fault reason 0x05] PTE Write access is not set
Upon comparing boot logs with VT-d on/off, I found that the Intel Calpella
quirk (`quirk_calpella_no_shadow_gtt()') correctly applied the igfx IOMMU
disable/quirk correctly:
pci 0000:00:00.0: DMAR: BIOS has allocated no shadow GTT; disabling IOMMU
for graphics
Whereas with VT-d on, it went into the "else" branch, which then
triggered the DMAR handling fault above:
... else if (!disable_igfx_iommu) {
/* we have to ensure the gfx device is idle before we flush */
pci_info(dev, "Disabling batched IOTLB flush on Ironlake\n");
iommu_set_dma_strict();
}
Now, this is not exactly scientific, but moving 0x0044 to quirk_iommu_igfx
seems to have fixed the aforementioned issue. Running a few `git blame'
runs on the function, I have found that the quirk was originally
introduced as a fix specific to ThinkPad X201:
commit 9eecabcb9a92 ("intel-iommu: Abort IOMMU setup for igfx if BIOS gave
no shadow GTT space")
Which was later revised twice to the "else" branch we saw above:
- 2011: commit 6fbcfb3e467a ("intel-iommu: Workaround IOTLB hang on
Ironlake GPU")
- 2024: commit ba00196ca41c ("iommu/vt-d: Decouple igfx_off from graphic
identity mapping")
I'm uncertain whether further testings on this particular laptops were
done in 2011 and (honestly I'm not sure) 2024, but I would be happy to do
some distro-specific testing if that's what would be required to verify
this patch.
P.S., I also see IDs 0x0040, 0x0062, and 0x006a listed under the same
`quirk_calpella_no_shadow_gtt()' quirk, but I'm not sure how similar these
chipsets are (if they share the same issue with VT-d or even, indeed, if
this issue is specific to a bug in the Lenovo BIOS). With regards to
0x0062, it seems to be a Centrino wireless card, but not a chipset?
I have also listed a couple (distro and kernel) bug reports below as
references (some of them are from 7-8 years ago!), as they seem to be
similar issue found on different Westmere/Ironlake, Haswell, and Broadwell
hardware setups.
There is a string parsing logic error which can lead to an overflow of hid
or uid buffers. Comparing ACPIID_LEN against a total string length doesn't
take into account the lengths of individual hid and uid buffers so the
check is insufficient in some cases. For example if the length of hid
string is 4 and the length of the uid string is 260, the length of str
will be equal to ACPIID_LEN + 1 but uid string will overflow uid buffer
which size is 256.
The same applies to the hid string with length 13 and uid string with
length 250.
Check the length of hid and uid strings separately to prevent
buffer overflow.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
realloc_argv() was only updating the array size if it was called with
old_argv already allocated. The first time it was called to create an
argv array, it would allocate the array but return the array size as
zero. dm_split_args() would think that it couldn't store any arguments
in the array and would call realloc_argv() again, causing it to
reallocate the initial slots (this time using GPF_KERNEL) and finally
return a size. Aside from being wasteful, this could cause deadlocks on
targets that need to process messages without starting new IO. Instead,
realloc_argv should always update the allocated array size on success.
Fixes: a0651926553c ("dm table: don't copy from a NULL pointer in realloc_argv()") Cc: stable@vger.kernel.org Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The function brcmf_usb_dl_writeimage() calls the function
brcmf_usb_dl_cmd() but dose not check its return value. The
'state.state' and the 'state.bytes' are uninitialized if the
function brcmf_usb_dl_cmd() fails. It is dangerous to use
uninitialized variables in the conditions.
Add error handling for brcmf_usb_dl_cmd() to jump to error
handling path if the brcmf_usb_dl_cmd() fails and the
'state.state' and the 'state.bytes' are uninitialized.
Improve the error message to report more detailed error
information.
Fixes: 71bb244ba2fd ("brcm80211: fmac: add USB support for bcm43235/6/8 chipsets") Cc: stable@vger.kernel.org # v3.4+ Signed-off-by: Wentao Liang <vulab@iscas.ac.cn> Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Link: https://patch.msgid.link/20250422042203.2259-1-vulab@iscas.ac.cn Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On Qualcomm chipsets not all GPIOs are wakeup capable. Those GPIOs do not
have a corresponding MPM pin and should not be handled inside the MPM
driver. The IRQ domain hierarchy is always applied, so it's required to
explicitly disconnect the hierarchy for those. The pinctrl-msm driver marks
these with GPIO_NO_WAKE_IRQ. qcom-pdc has a check for this, but
irq-qcom-mpm is currently missing the check. This is causing crashes when
setting up interrupts for non-wake GPIOs:
root@rb1:~# gpiomon -c gpiochip1 10
irq: IRQ159: trimming hierarchy from :soc@0:interrupt-controller@f200000-1
Unable to handle kernel paging request at virtual address ffff8000a1dc3820
Hardware name: Qualcomm Technologies, Inc. Robotics RB1 (DT)
pc : mpm_set_type+0x80/0xcc
lr : mpm_set_type+0x5c/0xcc
Call trace:
mpm_set_type+0x80/0xcc (P)
qcom_mpm_set_type+0x64/0x158
irq_chip_set_type_parent+0x20/0x38
msm_gpio_irq_set_type+0x50/0x530
__irq_set_trigger+0x60/0x184
__setup_irq+0x304/0x6bc
request_threaded_irq+0xc8/0x19c
edge_detector_setup+0x260/0x364
linereq_create+0x420/0x5a8
gpio_ioctl+0x2d4/0x6c0
Fix this by copying the check for GPIO_NO_WAKE_IRQ from qcom-pdc.c, so that
MPM is removed entirely from the hierarchy for non-wake GPIOs.
According to the XGMAC specification, enabling features such as Layer 3
and Layer 4 Packet Filtering, Split Header and Virtualized Network support
automatically selects the IPC Full Checksum Offload Engine on the receive
side.
When RX checksum offload is disabled, these dependent features must also
be disabled to prevent abnormal behavior caused by mismatched feature
dependencies.
Ensure that toggling RX checksum offload (disabling or enabling) properly
disables or enables all dependent features, maintaining consistent and
expected behavior in the network device.
Cc: stable@vger.kernel.org Fixes: 1a510ccf5869 ("amd-xgbe: Add support for VXLAN offload capabilities") Signed-off-by: Vishal Badole <Vishal.Badole@amd.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250424130248.428865-1-Vishal.Badole@amd.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When generating the MSR_IA32_PEBS_ENABLE value that will be loaded on
VM-Entry to a KVM guest, mask the value with the vCPU's desired PEBS_ENABLE
value. Consulting only the host kernel's host vs. guest masks results in
running the guest with PEBS enabled even when the guest doesn't want to use
PEBS. Because KVM uses perf events to proxy the guest virtual PMU, simply
looking at exclude_host can't differentiate between events created by host
userspace, and events created by KVM on behalf of the guest.
Running the guest with PEBS unexpectedly enabled typically manifests as
crashes due to a near-infinite stream of #PFs. E.g. if the guest hasn't
written MSR_IA32_DS_AREA, the CPU will hit page faults on address '0' when
trying to record PEBS events.
The issue is most easily reproduced by running `perf kvm top` from before
commit 7b100989b4f6 ("perf evlist: Remove __evlist__add_default") (after
which, `perf kvm top` effectively stopped using PEBS). The userspace side
of perf creates a guest-only PEBS event, which intel_guest_get_msrs()
misconstrues a guest-*owned* PEBS event.
Arguably, this is a userspace bug, as enabling PEBS on guest-only events
simply cannot work, and userspace can kill VMs in many other ways (there
is no danger to the host). However, even if this is considered to be bad
userspace behavior, there's zero downside to perf/KVM restricting PEBS to
guest-owned events.
Note, commit 854250329c02 ("KVM: x86/pmu: Disable guest PEBS temporarily
in two rare situations") fixed the case where host userspace is profiling
KVM *and* userspace, but missed the case where userspace is profiling only
KVM.
Camm noticed that on parisc a SIGFPE exception will crash an application with
a second SIGFPE in the signal handler. Dave analyzed it, and it happens
because glibc uses a double-word floating-point store to atomically update
function descriptors. As a result of lazy binding, we hit a floating-point
store in fpe_func almost immediately.
When the T bit is set, an assist exception trap occurs when when the
co-processor encounters *any* floating-point instruction except for a double
store of register %fr0. The latter cancels all pending traps. Let's fix this
by clearing the Trap (T) bit in the FP status register before returning to the
signal handler in userspace.
The issue can be reproduced with this test program:
Commit a5951389e58d ("arm64: errata: Add newer ARM cores to the
spectre_bhb_loop_affected() lists") added some additional CPUs to the
Spectre-BHB workaround, including some new arrays for designs that
require new 'k' values for the workaround to be effective.
Unfortunately, the new arrays omitted the sentinel entry and so
is_midr_in_range_list() will walk off the end when it doesn't find a
match. With UBSAN enabled, this leads to a crash during boot when
is_midr_in_range_list() is inlined (which was more common prior to c8c2647e69be ("arm64: Make _midr_in_range_list() an exported
function")):
Cc: Lee Jones <lee@kernel.org> Cc: James Morse <james.morse@arm.com> Cc: Doug Anderson <dianders@chromium.org> Cc: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com> Cc: <stable@vger.kernel.org> Reported-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Fixes: a5951389e58d ("arm64: errata: Add newer ARM cores to the spectre_bhb_loop_affected() lists") Signed-off-by: Will Deacon <will@kernel.org> Reviewed-by: Lee Jones <lee@kernel.org> Reviewed-by: Douglas Anderson <dianders@chromium.org> Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Link: https://lore.kernel.org/r/20250501104747.28431-1-will@kernel.org Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Deferred probe with pm_runtime_put() may delay clock disable, causing
incorrect clock usage count. Use pm_runtime_put_sync() to ensure the
clock is disabled immediately.
Fixes: 13d6eb20fc79 ("i2c: imx-lpi2c: add runtime pm support") Signed-off-by: Clark Wang <xiaoning.wang@nxp.com> Signed-off-by: Carlos Song <carlos.song@nxp.com> Cc: <stable@vger.kernel.org> # v4.16+ Link: https://lore.kernel.org/r/20250421062341.2471922-1-carlos.song@nxp.com Signed-off-by: Andi Shyti <andi.shyti@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Nouveau is mostly designed in a way that it's expected that fences only
ever get signaled through nouveau_fence_signal(). However, in at least
one other place, nouveau_fence_done(), can signal fences, too. If that
happens (race) a signaled fence remains in the pending list for a while,
until it gets removed by nouveau_fence_update().
Should nouveau_fence_context_kill() run in the meantime, this would be
a bug because the function would attempt to set an error code on an
already signaled fence.
Have nouveau_fence_context_kill() check for a fence being signaled.
Cc: stable@vger.kernel.org # v5.10+ Fixes: ea13e5abf807 ("drm/nouveau: signal pending fences when channel has been killed") Suggested-by: Christian König <christian.koenig@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://lore.kernel.org/r/20250415121900.55719-3-phasta@kernel.org Signed-off-by: Danilo Krummrich <dakr@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There seem to be multiple USB device IDs used for these;
the one I have reports as 0b0e:030c when powered on.
(When powered off, it reports as 0b0e:0311.)
This reverts commit 67d1a8956d2d62fe6b4c13ebabb57806098511d8. Since this
commit has been proven to be problematic for the setup of USB-tethered
ethernet connections and the related breakage is very noticeable for
users it should be reverted until a fixed version of the change can be
rolled out.
Huacai Chen [Sun, 4 May 2025 02:10:54 +0000 (10:10 +0800)]
LoongArch: Fix build error due to backport
In 6.1 there is no pmdp_get() definition, so use *pmd directly, in order
to avoid such build error due to a recently backport:
arch/loongarch/mm/hugetlbpage.c: In function 'huge_pte_offset':
arch/loongarch/mm/hugetlbpage.c:50:25: error: implicit declaration of function 'pmdp_get'; did you mean 'ptep_get'? [-Wimplicit-function-declaration]
50 | return pmd_none(pmdp_get(pmd)) ? NULL : (pte_t *) pmd;
| ^~~~~~~~
| ptep_get
With recent changes to add more display ports did not change the Stream
name in q6afe-dai. This results in below error
"ASoC: Failed to add route DISPLAY_PORT_RX -> Display Port Playback(*)"
and sound card fails to probe.
Fix this by adding correct stream name.
Fixes: 90848a2557fe ("ASoC: qcom: q6dsp: add support to more display ports") Reported-by: Amit Pundir <amit.pundir@linaro.org> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> Link: https://lore.kernel.org/r/20230705124850.40069-1-srinivas.kandagatla@linaro.org Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit c14f7ccc9f5d ("PCI: Assign PCI domain IDs by ida_alloc()")
introduced a use-after-free bug in the bus removal cleanup. The issue was
found with kfence:
[ 19.293351] BUG: KFENCE: use-after-free read in pci_bus_release_domain_nr+0x10/0x70
The stack trace is a bit misleading as dw_pcie_host_deinit() doesn't
directly call pci_bus_release_domain_nr(). The issue turns out to be in
pci_remove_root_bus() which first calls pci_remove_bus() which frees the
struct pci_bus when its struct device is released. Then
pci_bus_release_domain_nr() is called and accesses the freed struct
pci_bus. Reordering these fixes the issue.
To differentiate between long arrays and cpumasks, the __cpumask() field
was created. Part of the TRACE_EVENT() macros test if the type is signed
or not by using the is_signed_type() macro. The __cpumask() field used the
__dynamic_array() helper but because cpumask_t is a structure, it could
not be used in the is_signed_type() macro as that would fail to build, so
instead it passed in the pointer to cpumask_t.
When enable initcall_debug together with higher debug level below.
CONFIG_CONSOLE_LOGLEVEL_DEFAULT=9
CONFIG_CONSOLE_LOGLEVEL_QUIET=9
CONFIG_MESSAGE_LOGLEVEL_DEFAULT=7
The initialization of i.MX8MP PCIe PHY might be timeout failed randomly.
To fix this issue, adjust the sequence of the resets refer to the power
up sequence listed below.
i.MX8MP PCIe PHY power up sequence:
/---------------------------------------------
1.8v supply ---------/
/---------------------------------------------------
0.8v supply ---/
Logs:
imx6q-pcie 33800000.pcie: host bridge /soc@0/pcie@33800000 ranges:
imx6q-pcie 33800000.pcie: IO 0x001ff80000..0x001ff8ffff -> 0x0000000000
imx6q-pcie 33800000.pcie: MEM 0x0018000000..0x001fefffff -> 0x0018000000
probe of clk_imx8mp_audiomix.reset.0 returned 0 after 1052 usecs
probe of 30e20000.clock-controller returned 0 after 32971 usecs
phy phy-32f00000.pcie-phy.4: phy poweron failed --> -110
probe of 30e10000.dma-controller returned 0 after 10235 usecs
imx6q-pcie 33800000.pcie: waiting for PHY ready timeout!
dwhdmi-imx 32fd8000.hdmi: Detected HDMI TX controller v2.13a with HDCP (samsung_dw_hdmi_phy2)
imx6q-pcie 33800000.pcie: probe with driver imx6q-pcie failed with error -110
Fixes: dce9edff16ee ("phy: freescale: imx8m-pcie: Add i.MX8MP PCIe PHY support") Cc: stable@vger.kernel.org Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com> Signed-off-by: Frank Li <Frank.Li@nxp.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
v2 changes:
- Rebase to latest fixes branch of linux-phy git repo.
- Richard's environment have problem and can't sent out patch. So I help
post this fix patch.
Commit 62baf70c3274 caused the ANA log page to be re-read, even on
controllers that do not support ANA. While this should generally
harmless, some controllers hang on the unsupported log page and
never finish probing.
Fixes: 62baf70c3274 ("nvme: re-read ANA log page after ns scan completes") Signed-off-by: Hannes Reinecke <hare@kernel.org> Tested-by: Srikanth Aithal <sraithal@amd.com>
[hch: more detailed commit message] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xdp: Reset bpf_redirect_info before running a xdp's BPF prog.
Ricardo reported a KASAN discovered use after free in v6.6-stable.
The syzbot starts a BPF program via xdp_test_run_batch() which assigns
ri->tgt_value via dev_hash_map_redirect() and the return code isn't
XDP_REDIRECT it looks like nonsense. So the output in
bpf_warn_invalid_xdp_action() appears once.
Then the TUN driver runs another BPF program (on the same CPU) which
returns XDP_REDIRECT without setting ri->tgt_value first. It invokes
bpf_trace_printk() to print four characters and obtain the required
return value. This is enough to get xdp_do_redirect() invoked which
then accesses the pointer in tgt_value which might have been already
deallocated.
This problem does not affect upstream because since commit 401cb7dae8130 ("net: Reference bpf_redirect_info via task_struct on PREEMPT_RT.")
the per-CPU variable is referenced via task's task_struct and exists on
the stack during NAPI callback. Therefore it is cleared once before the
first invocation and remains valid within the RCU section of the NAPI
callback.
Instead of performing the huge backport of the commit (plus its fix ups)
here is an alternative version which only resets the variable in
question prior invoking the BPF program.
Commit c050f5e91b47 ("net: dsa: mv88e6xxx: Fill in STU support for all
supported chips") introduced STU methods, but did not add them to the
6320 family. Fix it.
Fixes: c050f5e91b47 ("net: dsa: mv88e6xxx: Fill in STU support for all supported chips") Signed-off-by: Marek Behún <kabel@kernel.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://patch.msgid.link/20250317173250.28780-6-kabel@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If we're redirecting the skb, and haven't called tcf_mirred_forward(),
yet, we need to tell the core to drop the skb by setting the retcode
to SHOT. If we have called tcf_mirred_forward(), however, the skb
is out of our hands and returning SHOT will lead to UaF.
Move the retval override to the error path which actually need it.
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Fixes: e5cf1baf92cb ("act_mirred: use TC_ACT_REINSERT when possible") Signed-off-by: Jakub Kicinski <kuba@kernel.org> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[Minor conflict resolved due to code context change.] Signed-off-by: Jianqi Ren <jianqi.ren.cn@windriver.com> Signed-off-by: He Zhe <zhe.he@windriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / thirdparty / kernel / stable.git / log
