| Age | Commit message (Collapse) | Author |
|---|
|
Signed-off-by: Hector Martin <marcan@marcan.st>
Reviewed-by: Neal Gompa <neal@gompa.dev>
Reviewed-by: Sven Peter <sven@svenpeter.dev>
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
Change the definition of the inline functions nvmet_cc_en(),
nvmet_cc_css(), nvmet_cc_mps(), nvmet_cc_ams(), nvmet_cc_shn(),
nvmet_cc_iosqes(), and nvmet_cc_iocqes() to use the enum difinitions in
include/linux/nvme.h instead of hardcoded values.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
Reorganized the enum used to define the fields of the contrller
configuration (CC) register in include/linux/nvme.h to:
1) Group together all the values defined for each field.
2) Add the missing field masks definitions.
3) Add comments to describe the enum and each field.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
nvme_validate_passthru_nsid() logs an err message whose format string is
split over 2 lines. There is a missing space between the two pieces,
resulting in log lines like "... does not match nsid (1)of namespace".
Add the missing space between ")" and "of". Also combine the format
string pieces onto a single line to make the err message easier to grep.
Fixes: e7d4b5493a2d ("nvme: factor out a nvme_validate_passthru_nsid helper")
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
nvme_tcp_init_connection() attempts to receive an ICResp PDU but only
checks that the return value from recvmsg() is non-negative. If the
sender closes the TCP connection or sends fewer than 128 bytes, this
check will pass even though the full PDU wasn't received.
Ensure the full ICResp PDU is received by checking that recvmsg()
returns the expected 128 bytes.
Additionally set the MSG_WAITALL flag for recvmsg(), as a sender could
split the ICResp over multiple TCP frames. Without MSG_WAITALL,
recvmsg() could return prematurely with only part of the PDU.
Fixes: 3f2304f8c6d6 ("nvme-tcp: add NVMe over TCP host driver")
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
When compiling with W=1, a warning result for the function
nvme_tcp_set_queue_io_cpu():
host/tcp.c:1578: warning: Function parameter or struct member 'queue'
not described in 'nvme_tcp_set_queue_io_cpu'
host/tcp.c:1578: warning: expecting prototype for Track the number of
queues assigned to each cpu using a global per(). Prototype was for
nvme_tcp_set_queue_io_cpu() instead
Avoid this warning by using the regular comment format for the function
nvme_tcp_set_queue_io_cpu() instead of the kdoc comment format.
Fixes: 32193789878c ("nvme-tcp: Fix I/O queue cpu spreading for multiple controllers")
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
The delayed work item function nvmet_pci_epf_poll_sqs_work() polls all
submission queues and keeps running in a loop as long as commands are
being submitted by the host. Depending on the preemption configuration
of the kernel, under heavy command workload, this function can thus run
for more than RCU_CPU_STALL_TIMEOUT seconds, leading to a RCU stall:
rcu: INFO: rcu_sched self-detected stall on CPU
rcu: 5-....: (20998 ticks this GP) idle=4244/1/0x4000000000000000 softirq=301/301 fqs=5132
rcu: (t=21000 jiffies g=-443 q=12 ncpus=8)
CPU: 5 UID: 0 PID: 82 Comm: kworker/5:1 Not tainted 6.14.0-rc2 #1
Hardware name: Radxa ROCK 5B (DT)
Workqueue: events nvmet_pci_epf_poll_sqs_work [nvmet_pci_epf]
pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : dw_edma_device_tx_status+0xb8/0x130
lr : dw_edma_device_tx_status+0x9c/0x130
sp : ffff800080b5bbb0
x29: ffff800080b5bbb0 x28: ffff0331c5c78400 x27: ffff0331c1cd1960
x26: ffff0331c0e39010 x25: ffff0331c20e4000 x24: ffff0331c20e4a90
x23: 0000000000000000 x22: 0000000000000001 x21: 00000000005aca33
x20: ffff800080b5bc30 x19: ffff0331c123e370 x18: 000000000ab29e62
x17: ffffb2a878c9c118 x16: ffff0335bde82040 x15: 0000000000000000
x14: 000000000000017b x13: 00000000ee601780 x12: 0000000000000018
x11: 0000000000000000 x10: 0000000000000001 x9 : 0000000000000040
x8 : 00000000ee601780 x7 : 0000000105c785c0 x6 : ffff0331c1027d80
x5 : 0000000001ee7ad6 x4 : ffff0335bdea16c0 x3 : ffff0331c123e438
x2 : 00000000005aca33 x1 : 0000000000000000 x0 : ffff0331c123e410
Call trace:
dw_edma_device_tx_status+0xb8/0x130 (P)
dma_sync_wait+0x60/0xbc
nvmet_pci_epf_dma_transfer+0x128/0x264 [nvmet_pci_epf]
nvmet_pci_epf_poll_sqs_work+0x2a0/0x2e0 [nvmet_pci_epf]
process_one_work+0x144/0x390
worker_thread+0x27c/0x458
kthread+0xe8/0x19c
ret_from_fork+0x10/0x20
The solution for this is simply to explicitly allow rescheduling using
cond_resched(). However, since doing so for every loop of
nvmet_pci_epf_poll_sqs_work() significantly degrades performance
(for 4K random reads using 4 I/O queues, the maximum IOPS goes down from
137 KIOPS to 110 KIOPS), call cond_resched() every second to avoid the
RCU stalls.
Fixes: 0faa0fe6f90e ("nvmet: New NVMe PCI endpoint function target driver")
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
The function nvmet_pci_epf_poll_cc_work() will do nothing if there are
no changes to the controller configuration (CC) register. However, even
for such case, this function still calls nvmet_update_cc() and uselessly
writes the CSTS register. Avoid this by simply rescheduling the poll_cc
work if the CC register has not changed.
Also reschedule the poll_cc work if the function
nvmet_pci_epf_enable_ctrl() fails to allow the host the chance to try
again enabling the controller.
While at it, since there is no point in trying to handle the CC register
as quickly as possible, change the poll_cc work scheduling interval to
10 ms (from 5ms), to avoid excessive read accesses to that register.
Fixes: 0faa0fe6f90e ("nvmet: New NVMe PCI endpoint function target driver")
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
The function nvmet_pci_epf_poll_cc_work() sets the NVME_CSTS_RDY bit of
the controller status register (CSTS) when nvmet_pci_epf_enable_ctrl()
returns success. However, since this function can be called several
times (e.g. if the host reboots), instead of setting the bit in
ctrl->csts, initialize this field to only have NVME_CSTS_RDY set.
Conversely, if nvmet_pci_epf_enable_ctrl() fails, make sure to clear all
bits from ctrl->csts.
To simplify nvmet_pci_epf_poll_cc_work(), initialize ctrl->csts to
NVME_CSTS_RDY directly inside nvmet_pci_epf_enable_ctrl() and clear this
field in that function as well in case of a failure. To be consistent,
move clearing the NVME_CSTS_RDY bit from ctrl->csts when the controller
is being disabled from nvmet_pci_epf_poll_cc_work() into
nvmet_pci_epf_disable_ctrl().
Fixes: 0faa0fe6f90e ("nvmet: New NVMe PCI endpoint function target driver")
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
The queue state checking in nvmet_rdma_recv_done is not in queue state
lock.Queue state can transfer to LIVE in cm establish handler between
state checking and state lock here, cause a silent drop of nvme connect
cmd.
Recheck queue state whether in LIVE state in state lock to prevent this
issue.
Signed-off-by: Ruozhu Li <david.li@jaguarmicro.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
The namespace percpu counter protects pending I/O, and we can
only safely diable the namespace once the counter drop to zero.
Otherwise we end up with a crash when running blktests/nvme/058
(eg for loop transport):
[ 2352.930426] [ T53909] Oops: general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] PREEMPT SMP KASAN PTI
[ 2352.930431] [ T53909] KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f]
[ 2352.930434] [ T53909] CPU: 3 UID: 0 PID: 53909 Comm: kworker/u16:5 Tainted: G W 6.13.0-rc6 #232
[ 2352.930438] [ T53909] Tainted: [W]=WARN
[ 2352.930440] [ T53909] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-3.fc41 04/01/2014
[ 2352.930443] [ T53909] Workqueue: nvmet-wq nvme_loop_execute_work [nvme_loop]
[ 2352.930449] [ T53909] RIP: 0010:blkcg_set_ioprio+0x44/0x180
as the queue is already torn down when calling submit_bio();
So we need to init the percpu counter in nvmet_ns_enable(), and
wait for it to drop to zero in nvmet_ns_disable() to avoid having
I/O pending after the namespace has been disabled.
Fixes: 74d16965d7ac ("nvmet-loop: avoid using mutex in IO hotpath")
Signed-off-by: Hannes Reinecke <hare@kernel.org>
Reviewed-by: Nilay Shroff <nilay@linux.ibm.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
Previously, the NVMe/TCP host driver did not handle the C2HTermReq PDU,
instead printing "unsupported pdu type (3)" when received. This patch adds
support for processing the C2HTermReq PDU, allowing the driver
to print the Fatal Error Status field.
Example of output:
nvme nvme4: Received C2HTermReq (FES = Invalid PDU Header Field)
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
In order for two Acer FA100 SSDs to work in one PC (in the case of
myself, a Lenovo Legion T5 28IMB05), and not show one drive and not
the other, and sometimes mix up what drive shows up (randomly), these
two lines of code need to be added, and then both of the SSDs will
show up and not conflict when booting off of one of them. If you boot
up your computer with both SSDs installed without this patch, you may
also randomly get into a kernel panic (if the initrd is not set up) or
stuck in the initrd "/init" process, it is set up, however, if you do
apply this patch, there should not be problems with booting or seeing
both contents of the drive. Tested with the btrfs filesystem with a
RAID configuration of having the root drive '/' combined to make two
256GB Acer FA100 SSDs become 512GB in total storage.
Kernel Logs with patch applied (`dmesg -t | grep -i nvm`):
```
...
nvme 0000:04:00.0: platform quirk: setting simple suspend
nvme nvme0: pci function 0000:04:00.0
nvme 0000:05:00.0: platform quirk: setting simple suspend
nvme nvme1: pci function 0000:05:00.0
nvme nvme1: missing or invalid SUBNQN field.
nvme nvme1: allocated 64 MiB host memory buffer.
nvme nvme0: missing or invalid SUBNQN field.
nvme nvme0: allocated 64 MiB host memory buffer.
nvme nvme1: 8/0/0 default/read/poll queues
nvme nvme1: Ignoring bogus Namespace Identifiers
nvme nvme0: 8/0/0 default/read/poll queues
nvme nvme0: Ignoring bogus Namespace Identifiers
nvme0n1: p1 p2
...
```
Kernel Logs with patch not applied (`dmesg -t | grep -i nvm`):
```
...
nvme 0000:04:00.0: platform quirk: setting simple suspend
nvme nvme0: pci function 0000:04:00.0
nvme 0000:05:00.0: platform quirk: setting simple suspend
nvme nvme1: pci function 0000:05:00.0
nvme nvme0: missing or invalid SUBNQN field.
nvme nvme1: missing or invalid SUBNQN field.
nvme nvme0: allocated 64 MiB host memory buffer.
nvme nvme1: allocated 64 MiB host memory buffer.
nvme nvme0: 8/0/0 default/read/poll queues
nvme nvme1: 8/0/0 default/read/poll queues
nvme nvme1: globally duplicate IDs for nsid 1
nvme nvme1: VID:DID 1dbe:5216 model:Acer SSD FA100 256GB firmware:1.Z.J.2X
nvme0n1: p1 p2
...
```
Signed-off-by: Christopher Lentocha <christopherericlentocha@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
|
|
When adding LED support for mv88q222x devices the PHY private data
structure was added to the mv88q211x code path, the data structure is
however only allocated during mv88q222x probe. This results in a nullptr
deference for mv88q2110 devices.
Unable to handle kernel NULL pointer dereference at virtual address 0000000000000001
Mem abort info:
ESR = 0x0000000096000004
EC = 0x25: DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
FSC = 0x04: level 0 translation fault
Data abort info:
ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
CM = 0, WnR = 0, TnD = 0, TagAccess = 0
GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[0000000000000001] user address but active_mm is swapper
Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP
CPU: 3 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.14.0-rc1-arm64-renesas-00342-ga3783dbf2574 #7
Hardware name: Renesas White Hawk Single board based on r8a779g2 (DT)
pstate: 20400005 (nzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : mv88q2xxx_config_init+0x28/0x84
lr : mv88q2110_config_init+0x98/0xb0
sp : ffff8000823eb9d0
x29: ffff8000823eb9d0 x28: ffff000440942000 x27: ffff80008144e400
x26: 0000000000001002 x25: 0000000000000000 x24: 0000000000000000
x23: 0000000000000009 x22: ffff8000810534f0 x21: ffff800081053550
x20: 0000000000000000 x19: ffff0004437d6800 x18: 0000000000000018
x17: 00000000000961c8 x16: ffff0006bef75ec0 x15: 0000000000000001
x14: 0000000000000001 x13: ffff000440218080 x12: 071c71c71c71c71c
x11: ffff000440218080 x10: 0000000000001420 x9 : ffff8000823eb770
x8 : ffff8000823eb650 x7 : ffff8000823eb750 x6 : ffff8000823eb710
x5 : 0000000000000000 x4 : 0000000000000800 x3 : 0000000000000001
x2 : 0000000000000000 x1 : 00000000ffffffff x0 : ffff0004437d6800
Call trace:
mv88q2xxx_config_init+0x28/0x84 (P)
mv88q2110_config_init+0x98/0xb0
phy_init_hw+0x64/0x9c
phy_attach_direct+0x118/0x320
phy_connect_direct+0x24/0x80
of_phy_connect+0x5c/0xa0
rtsn_open+0x5bc/0x78c
__dev_open+0xf8/0x1fc
__dev_change_flags+0x198/0x220
dev_change_flags+0x20/0x64
ip_auto_config+0x270/0xefc
do_one_initcall+0xe4/0x22c
kernel_init_freeable+0x2a8/0x308
kernel_init+0x20/0x130
ret_from_fork+0x10/0x20
Code: b907e404 f9432814 3100083f 540000e3 (39400680)
---[ end trace 0000000000000000 ]---
Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b
SMP: stopping secondary CPUs
Kernel Offset: disabled
CPU features: 0x000,00000070,00801250,8200700b
Memory Limit: none
---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b ]---
Fix this by using a generic probe function for both mv88q211x and
mv88q222x devices that allocates the PHY private data structure, while
only the mv88q222x probes for LED support.
Fixes: a3783dbf2574 ("net: phy: marvell-88q2xxx: Add support for PHY LEDs on 88q2xxx")
Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://patch.msgid.link/20250214174650.2056949-1-niklas.soderlund+renesas@ragnatech.se
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Add a lightweight tracepoint to monitor TCP congestion window
adjustments via tcp_cwnd_reduction(). This tracepoint enables tracking
of:
- TCP window size fluctuations
- Active socket behavior
- Congestion window reduction events
Meta has been using BPF programs to monitor this function for years.
Adding a proper tracepoint provides a stable API for all users who need
to monitor TCP congestion window behavior.
Use DECLARE_TRACE instead of TRACE_EVENT to avoid creating trace event
infrastructure and exporting to tracefs, keeping the implementation
minimal. (Thanks Steven Rostedt)
Given that this patch creates a rawtracepoint, you could hook into it
using regular tooling, like bpftrace, using regular rawtracepoint
infrastructure, such as:
rawtracepoint:tcp_cwnd_reduction_tp {
....
}
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250214-cwnd_tracepoint-v2-1-ef8d15162d95@debian.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Dimitri Fedrau says:
====================
net: phy: marvell-88q2xxx: cleanup
- align defines
- order includes alphabetically
- enable temperature sensor in mv88q2xxx_config_init
Signed-off-by: Dimitri Fedrau <dima.fedrau@gmail.com>
====================
Link: https://patch.msgid.link/20250214-marvell-88q2xxx-cleanup-v1-0-71d67c20f308@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Temperature sensor gets enabled for 88Q222X devices in
mv88q222x_config_init. Move enabling to mv88q2xxx_config_init because
all 88Q2XXX devices support the temperature sensor.
Signed-off-by: Dimitri Fedrau <dima.fedrau@gmail.com>
Tested-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Order includes alphabetically.
Signed-off-by: Dimitri Fedrau <dima.fedrau@gmail.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Align some defines.
Signed-off-by: Dimitri Fedrau <dima.fedrau@gmail.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
We fixed the UAF issue in USB MIDI code by canceling the pending work
at closing each MIDI output device in the commit below. However, this
assumed that it's the only one that is tied with the endpoint, and it
resulted in unexpected data truncations when multiple devices are
assigned to a single endpoint and opened simultaneously.
For addressing the unexpected MIDI message drops, simply replace
cancel_work_sync() with flush_work(). The drain callback should have
been already invoked before the close callback, hence the port->active
flag must be already cleared. So this just assures that the pending
work is finished before freeing the resources.
Fixes: 0125de38122f ("ALSA: usb-audio: Cancel pending work at closing a MIDI substream")
Reported-and-tested-by: John Keeping <jkeeping@inmusicbrands.com>
Closes: https://lore.kernel.org/20250217111647.3368132-1-jkeeping@inmusicbrands.com
Link: https://patch.msgid.link/20250218114024.23125-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
Petr Machata says:
====================
vxlan: Join / leave MC group when reconfigured
When a vxlan netdevice is brought up, if its default remote is a multicast
address, the device joins the indicated group.
Therefore when the multicast remote address changes, the device should
leave the current group and subscribe to the new one. Similarly when the
interface used for endpoint communication is changed in a situation when
multicast remote is configured. This is currently not done.
Both vxlan_igmp_join() and vxlan_igmp_leave() can however fail. So it is
possible that with such fix, the netdevice will end up in an inconsistent
situation where the old group is not joined anymore, but joining the
new group fails. Should we join the new group first, and leave the old one
second, we might end up in the opposite situation, where both groups are
joined. Undoing any of this during rollback is going to be similarly
problematic.
One solution would be to just forbid the change when the netdevice is up.
However in vnifilter mode, changing the group address is allowed, and these
problems are simply ignored (see vxlan_vni_update_group()):
# ip link add name br up type bridge vlan_filtering 1
# ip link add vx1 up master br type vxlan external vnifilter local 192.0.2.1 dev lo dstport 4789
# bridge vni add dev vx1 vni 200 group 224.0.0.1
# tcpdump -i lo &
# bridge vni add dev vx1 vni 200 group 224.0.0.2
18:55:46.523438 IP 0.0.0.0 > 224.0.0.22: igmp v3 report, 1 group record(s)
18:55:46.943447 IP 0.0.0.0 > 224.0.0.22: igmp v3 report, 1 group record(s)
# bridge vni
dev vni group/remote
vx1 200 224.0.0.2
Having two different modes of operation for conceptually the same interface
is silly, so in this patchset, just do what the vnifilter code does and
deal with the errors by crossing fingers real hard.
====================
Link: https://patch.msgid.link/cover.1739548836.git.petrm@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Changes to MC remote need to be reflected in actual group memberships.
Add a test to verify that it is the case.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Instead of inlining equivalents, use lib.sh-provided primitives.
Use defer to manage vx lifetime.
This will make it easier to extend the test in the next patch.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
This helper could be useful to more than just forwarding tests.
Move it upstairs and port over to log_test_skip().
Split the function into two parts: the bit that actually checks and
reports skip, which is in a new function check_command(). And a bit
that exits the test script if the check fails. This allows users
consistent checking behavior while giving an option to bail out from
a single test without bailing out of the whole script.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
When a vxlan netdevice is brought up, if its default remote is a multicast
address, the device joins the indicated group.
Therefore when the multicast remote address changes, the device should
leave the current group and subscribe to the new one. Similarly when the
interface used for endpoint communication is changed in a situation when
multicast remote is configured. This is currently not done.
Both vxlan_igmp_join() and vxlan_igmp_leave() can however fail. So it is
possible that with such fix, the netdevice will end up in an inconsistent
situation where the old group is not joined anymore, but joining the new
group fails. Should we join the new group first, and leave the old one
second, we might end up in the opposite situation, where both groups are
joined. Undoing any of this during rollback is going to be similarly
problematic.
One solution would be to just forbid the change when the netdevice is up.
However in vnifilter mode, changing the group address is allowed, and these
problems are simply ignored (see vxlan_vni_update_group()):
# ip link add name br up type bridge vlan_filtering 1
# ip link add vx1 up master br type vxlan external vnifilter local 192.0.2.1 dev lo dstport 4789
# bridge vni add dev vx1 vni 200 group 224.0.0.1
# tcpdump -i lo &
# bridge vni add dev vx1 vni 200 group 224.0.0.2
18:55:46.523438 IP 0.0.0.0 > 224.0.0.22: igmp v3 report, 1 group record(s)
18:55:46.943447 IP 0.0.0.0 > 224.0.0.22: igmp v3 report, 1 group record(s)
# bridge vni
dev vni group/remote
vx1 200 224.0.0.2
Having two different modes of operation for conceptually the same interface
is silly, so in this patch, just do what the vnifilter code does and deal
with the errors by crossing fingers real hard.
The vnifilter code leaves old before joining new, and in case of join /
leave failures does not roll back the configuration changes that have
already been applied, but bails out of joining if it could not leave. Do
the same here: leave before join, apply changes unconditionally and do not
attempt to join if we couldn't leave.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
vxlan_dev_configure() only has a single caller that passes false
for the changelink parameter. Drop the parameter and inline the
sole value.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
We noticed the kworker in page_pool_release_retry() was waken
up repeatedly and infinitely in production because of the
buggy driver causing the inflight less than 0 and warning
us in page_pool_inflight()[1].
Since the inflight value goes negative, it means we should
not expect the whole page_pool to get back to work normally.
This patch mitigates the adverse effect by not rescheduling
the kworker when detecting the inflight negative in
page_pool_release_retry().
[1]
[Mon Feb 10 20:36:11 2025] ------------[ cut here ]------------
[Mon Feb 10 20:36:11 2025] Negative(-51446) inflight packet-pages
...
[Mon Feb 10 20:36:11 2025] Call Trace:
[Mon Feb 10 20:36:11 2025] page_pool_release_retry+0x23/0x70
[Mon Feb 10 20:36:11 2025] process_one_work+0x1b1/0x370
[Mon Feb 10 20:36:11 2025] worker_thread+0x37/0x3a0
[Mon Feb 10 20:36:11 2025] kthread+0x11a/0x140
[Mon Feb 10 20:36:11 2025] ? process_one_work+0x370/0x370
[Mon Feb 10 20:36:11 2025] ? __kthread_cancel_work+0x40/0x40
[Mon Feb 10 20:36:11 2025] ret_from_fork+0x35/0x40
[Mon Feb 10 20:36:11 2025] ---[ end trace ebffe800f33e7e34 ]---
Note: before this patch, the above calltrace would flood the
dmesg due to repeated reschedule of release_dw kworker.
Signed-off-by: Jason Xing <kerneljasonxing@gmail.com>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Link: https://patch.msgid.link/20250214064250.85987-1-kerneljasonxing@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Michal Luczaj says:
====================
sockmap, vsock: For connectible sockets allow only connected
Series deals with one more case of vsock surprising BPF/sockmap by being
inconsistency about (having an) assigned transport.
KASAN: null-ptr-deref in range [0x0000000000000120-0x0000000000000127]
CPU: 7 UID: 0 PID: 56 Comm: kworker/7:0 Not tainted 6.14.0-rc1+
Workqueue: vsock-loopback vsock_loopback_work
RIP: 0010:vsock_read_skb+0x4b/0x90
Call Trace:
sk_psock_verdict_data_ready+0xa4/0x2e0
virtio_transport_recv_pkt+0x1ca8/0x2acc
vsock_loopback_work+0x27d/0x3f0
process_one_work+0x846/0x1420
worker_thread+0x5b3/0xf80
kthread+0x35a/0x700
ret_from_fork+0x2d/0x70
ret_from_fork_asm+0x1a/0x30
This bug, similarly to commit f6abafcd32f9 ("vsock/bpf: return early if
transport is not assigned"), could be fixed with a single NULL check. But
instead, let's explore another approach: take a hint from
vsock_bpf_update_proto() and teach sockmap to accept only vsocks that are
already connected (no risk of transport being dropped or reassigned). At
the same time straight reject the listeners (vsock listening sockets do not
carry any transport anyway). This way BPF does not have to worry about
vsk->transport becoming NULL.
Signed-off-by: Michal Luczaj <mhal@rbox.co>
====================
Link: https://patch.msgid.link/20250213-vsock-listen-sockmap-nullptr-v1-0-994b7cd2f16b@rbox.co
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Verify that for a connectible AF_VSOCK socket, merely having a transport
assigned is insufficient; socket must be connected for the sockmap to
accept.
This does not test datagram vsocks. Even though it hardly matters. VMCI is
the only transport that features VSOCK_TRANSPORT_F_DGRAM, but it has an
unimplemented vsock_transport::readskb() callback, making it unsupported by
BPF/sockmap.
Signed-off-by: Michal Luczaj <mhal@rbox.co>
Acked-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Commit 515745445e92 ("selftest/bpf: Add test for vsock removal from sockmap
on close()") added test that checked if proto::close() callback was invoked
on AF_VSOCK socket release. I.e. it verified that a close()d vsock does
indeed get removed from the sockmap.
It was done simply by creating a socket pair and attempting to replace a
close()d one with its peer. Since, due to a recent change, sockmap does not
allow updating index with a non-established connectible vsock, redo it with
a freshly established one.
Signed-off-by: Michal Luczaj <mhal@rbox.co>
Acked-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
In the spirit of commit 91751e248256 ("vsock: prevent null-ptr-deref in
vsock_*[has_data|has_space]"), armorize the "impossible" cases with a
warning.
Fixes: 634f1a7110b4 ("vsock: support sockmap")
Signed-off-by: Michal Luczaj <mhal@rbox.co>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
sockmap expects all vsocks to have a transport assigned, which is expressed
in vsock_proto::psock_update_sk_prot(). However, there is an edge case
where an unconnected (connectible) socket may lose its previously assigned
transport. This is handled with a NULL check in the vsock/BPF recv path.
Another design detail is that listening vsocks are not supposed to have any
transport assigned at all. Which implies they are not supported by the
sockmap. But this is complicated by the fact that a socket, before
switching to TCP_LISTEN, may have had some transport assigned during a
failed connect() attempt. Hence, we may end up with a listening vsock in a
sockmap, which blows up quickly:
KASAN: null-ptr-deref in range [0x0000000000000120-0x0000000000000127]
CPU: 7 UID: 0 PID: 56 Comm: kworker/7:0 Not tainted 6.14.0-rc1+
Workqueue: vsock-loopback vsock_loopback_work
RIP: 0010:vsock_read_skb+0x4b/0x90
Call Trace:
sk_psock_verdict_data_ready+0xa4/0x2e0
virtio_transport_recv_pkt+0x1ca8/0x2acc
vsock_loopback_work+0x27d/0x3f0
process_one_work+0x846/0x1420
worker_thread+0x5b3/0xf80
kthread+0x35a/0x700
ret_from_fork+0x2d/0x70
ret_from_fork_asm+0x1a/0x30
For connectible sockets, instead of relying solely on the state of
vsk->transport, tell sockmap to only allow those representing established
connections. This aligns with the behaviour for AF_INET and AF_UNIX.
Fixes: 634f1a7110b4 ("vsock: support sockmap")
Signed-off-by: Michal Luczaj <mhal@rbox.co>
Acked-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Suman Ghosh says:
====================
Add af_xdp support for cn10k
This patchset includes changes to support AF_XDP for cn10k chipsets. Both
non-zero copy and zero copy will be supported after these changes. Also,
the RSS will be reconfigured once a particular receive queue is
added/removed to/from AF_XDP support.
Patch #1: octeontx2-pf: use xdp_return_frame() to free xdp buffers
Patch #2: octeontx2-pf: Add AF_XDP non-zero copy support
Patch #3: octeontx2-pf: AF_XDP zero copy receive support
Patch #4: octeontx2-pf: Reconfigure RSS table after enabling AF_XDP
zerocopy on rx queue
Patch #5: octeontx2-pf: Prepare for AF_XDP transmit
Patch #6: octeontx2-pf: AF_XDP zero copy transmit support
====================
Link: https://patch.msgid.link/20250213053141.2833254-1-sumang@marvell.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
This patch implements below changes,
1. To avoid concurrency with normal traffic uses
XDP queues.
2. Since there are chances that XDP and AF_XDP can
fall under same queue uses separate flags to handle
dma buffers.
Signed-off-by: Hariprasad Kelam <hkelam@marvell.com>
Signed-off-by: Suman Ghosh <sumang@marvell.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Implement necessary APIs required for AF_XDP transmit.
Signed-off-by: Hariprasad Kelam <hkelam@marvell.com>
Signed-off-by: Suman Ghosh <sumang@marvell.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
RSS table needs to be reconfigured once a rx queue is enabled or
disabled for AF_XDP zerocopy support. After enabling UMEM on a rx queue,
that queue should not be part of RSS queue selection algorithm.
Similarly the queue should be considered again after UMEM is disabled.
Signed-off-by: Suman Ghosh <sumang@marvell.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
This patch adds support to AF_XDP zero copy for CN10K.
This patch specifically adds receive side support. In this approach once
a xdp program with zero copy support on a specific rx queue is enabled,
then that receive quse is disabled/detached from the existing kernel
queue and re-assigned to the umem memory.
Signed-off-by: Suman Ghosh <sumang@marvell.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Set xdp rx ring memory type as MEM_TYPE_PAGE_POOL for
af-xdp to work. This is needed since xdp_return_frame
internally will use page pools.
Fixes: 06059a1a9a4a ("octeontx2-pf: Add XDP support to netdev PF")
Signed-off-by: Suman Ghosh <sumang@marvell.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
xdp_return_frames() will help to free the xdp frames and their
associated pages back to page pool.
Signed-off-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Suman Ghosh <sumang@marvell.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
During the locking rework in GPIOLIB, we omitted one important use-case,
namely: setting and getting values for GPIO descriptor arrays with
array_info present.
This patch does two things: first it makes struct gpio_array store the
address of the underlying GPIO device and not chip. Next: it protects
the chip with SRCU from removal in gpiod_get_array_value_complex() and
gpiod_set_array_value_complex().
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20250215095655.23152-1-brgl@bgdev.pl
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
|
|
Restricted pointers ("%pK") are not meant to be used through printk().
It can unintentionally expose security sensitive, raw pointer values.
Use regular pointer formatting instead.
For more background, see:
https://lore.kernel.org/lkml/20250113171731-dc10e3c1-da64-4af0-b767-7c7070468023@linutronix.de/
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250217-restricted-pointers-uprobes-v1-1-e8cbe5bb22a7@linutronix.de
|
|
When a process reduces its number of threads or clears bits in its CPU
affinity mask, the mm_cid allocation should eventually converge towards
smaller values.
However, the change introduced by:
commit 7e019dcc470f ("sched: Improve cache locality of RSEQ concurrency
IDs for intermittent workloads")
adds a per-mm/CPU recent_cid which is never unset unless a thread
migrates.
This is a tradeoff between:
A) Preserving cache locality after a transition from many threads to few
threads, or after reducing the hamming weight of the allowed CPU mask.
B) Making the mm_cid upper bounds wrt nr threads and allowed CPU mask
easy to document and understand.
C) Allowing applications to eventually react to mm_cid compaction after
reduction of the nr threads or allowed CPU mask, making the tracking
of mm_cid compaction easier by shrinking it back towards 0 or not.
D) Making sure applications that periodically reduce and then increase
again the nr threads or allowed CPU mask still benefit from good
cache locality with mm_cid.
Introduce the following changes:
* After shrinking the number of threads or reducing the number of
allowed CPUs, reduce the value of max_nr_cid so expansion of CID
allocation will preserve cache locality if the number of threads or
allowed CPUs increase again.
* Only re-use a recent_cid if it is within the max_nr_cid upper bound,
else find the first available CID.
Fixes: 7e019dcc470f ("sched: Improve cache locality of RSEQ concurrency IDs for intermittent workloads")
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Gabriele Monaco <gmonaco@redhat.com>
Link: https://lkml.kernel.org/r/20250210153253.460471-2-gmonaco@redhat.com
|
|
ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse
Pull fuse fixes from Miklos Szeredi:
This contains a fix for fuse readahead.
* tag 'fuse-fixes-6.14-rc4' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
fuse: revert back to __readahead_folio() for readahead
Link: https://lore.kernel.org/r/CAJfpegv=+M4hy=hfBKEgBN8vfWULWT9ApbQzCnPopnMqyjpkzA@mail.gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
In case CONFIG_XARRAY_MULTI is not defined, xa_store_order can store a
multi-index entry but xas_for_each can't tell sbiling entry from valid
entry. So the check_pause failed when we store a multi-index entry and
wish xas_for_each can handle it normally. Avoid to store multi-index
entry when CONFIG_XARRAY_MULTI is disabled to fix the failure.
Link: https://lkml.kernel.org/r/20250213163659.414309-1-shikemeng@huaweicloud.com
Fixes: c9ba5249ef8b ("Xarray: move forward index correctly in xas_pause()")
Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Closes: https://lore.kernel.org/r/CAMuHMdU_bfadUO=0OZ=AoQ9EAmQPA4wsLCBqohXR+QCeCKRn4A@mail.gmail.com
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The following bug report was found when running a PREEMPT_RT debug kernel.
BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:48
in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 140605, name: kunit_try_catch
preempt_count: 1, expected: 0
Call trace:
rt_spin_lock+0x70/0x140
find_vmap_area+0x84/0x168
find_vm_area+0x1c/0x50
print_address_description.constprop.0+0x2a0/0x320
print_report+0x108/0x1f8
kasan_report+0x90/0xc8
Since commit e30a0361b851 ("kasan: make report_lock a raw spinlock"),
report_lock was changed to raw_spinlock_t to fix another similar
PREEMPT_RT problem. That alone isn't enough to cover other corner cases.
print_address_description() is always invoked under the report_lock. The
context under this lock is always atomic even on PREEMPT_RT.
find_vm_area() acquires vmap_node::busy.lock which is a spinlock_t,
becoming a sleeping lock on PREEMPT_RT and must not be acquired in atomic
context.
Don't invoke find_vm_area() on PREEMPT_RT and just print the address.
Non-PREEMPT_RT builds remain unchanged. Add a DEFINE_WAIT_OVERRIDE_MAP()
macro to tell lockdep that this lock nesting is allowed because the
PREEMPT_RT part (which is invalid) has been taken care of. This macro was
first introduced in commit 0cce06ba859a ("debugobjects,locking: Annotate
debug_object_fill_pool() wait type violation").
Link: https://lkml.kernel.org/r/20250217204402.60533-1-longman@redhat.com
Fixes: e30a0361b851 ("kasan: make report_lock a raw spinlock")
Signed-off-by: Waiman Long <longman@redhat.com>
Suggested-by: Andrey Konovalov <andreyknvl@gmail.com>
Reviewed-by: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Dmitriy Vyukov <dvyukov@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Mariano Pache <npache@redhat.com>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Vincenzo Frascino <vincenzo.frascino@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Updated .mailmap, but forgot these other places.
Link: https://lkml.kernel.org/r/20250212173523.3979840-1-ndesaulniers@google.com
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
When testing if we should try to compact memory or drop caches before we
run the THP or HugeTLB tests we use | as an or operator. This doesn't
work since run_vmtests.sh is written in shell where this is used to pipe
the output of the first argument into the second. Instead use the shell's
-o operator.
Link: https://lkml.kernel.org/r/20250212-kselftest-mm-no-hugepages-v1-1-44702f538522@kernel.org
Fixes: b433ffa8dbac ("selftests: mm: perform some system cleanup before using hugepages")
Signed-off-by: Mark Brown <broonie@kernel.org>
Reviewed-by: Nico Pache <npache@redhat.com>
Cc: Mariano Pache <npache@redhat.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
When using the HugeTLB kernel command-line to allocate 1G pages from a
specific node, such as:
default_hugepagesz=1G hugepages=1:1
If node 1 happens to not have enough memory for the requested number of 1G
pages, the allocation falls back to other nodes. A quick way to reproduce
this is by creating a KVM guest with a memory-less node and trying to
allocate 1 1G page from it. Instead of failing, the allocation will
fallback to other nodes.
This defeats the purpose of node specific allocation. Also, specific node
allocation for 2M pages don't have this behavior: the allocation will just
fail for the pages it can't satisfy.
This issue happens because HugeTLB calls memblock_alloc_try_nid_raw() for
1G boot-time allocation as this function falls back to other nodes if the
allocation can't be satisfied. Use memblock_alloc_exact_nid_raw()
instead, which ensures that the allocation will only be satisfied from the
specified node.
Link: https://lkml.kernel.org/r/20250211034856.629371-1-luizcap@redhat.com
Fixes: b5389086ad7b ("hugetlbfs: extend the definition of hugepages parameter to support node allocation")
Signed-off-by: Luiz Capitulino <luizcap@redhat.com>
Acked-by: Oscar Salvador <osalvador@suse.de>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: "Mike Rapoport (IBM)" <rppt@kernel.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Zhenguo Yao <yaozhenguo1@gmail.com>
Cc: Frank van der Linden <fvdl@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
A softlockup issue was found with stress test:
watchdog: BUG: soft lockup - CPU#27 stuck for 26s! [migration/27:181]
CPU: 27 UID: 0 PID: 181 Comm: migration/27 6.14.0-rc2-next-20250210 #1
Stopper: multi_cpu_stop <- stop_machine_from_inactive_cpu
RIP: 0010:stop_machine_yield+0x2/0x10
RSP: 0000:ff4a0dcecd19be48 EFLAGS: 00000246
RAX: ffffffff89c0108f RBX: ff4a0dcec03afe44 RCX: 0000000000000000
RDX: ff1cdaaf6eba5808 RSI: 0000000000000282 RDI: ff1cda80c1775a40
RBP: 0000000000000001 R08: 00000011620096c6 R09: 7fffffffffffffff
R10: 0000000000000001 R11: 0000000000000100 R12: ff1cda80c1775a40
R13: 0000000000000000 R14: 0000000000000001 R15: ff4a0dcec03afe20
FS: 0000000000000000(0000) GS:ff1cdaaf6eb80000(0000)
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 00000025e2c2a001 CR4: 0000000000773ef0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
PKRU: 55555554
Call Trace:
multi_cpu_stop+0x8f/0x100
cpu_stopper_thread+0x90/0x140
smpboot_thread_fn+0xad/0x150
kthread+0xc2/0x100
ret_from_fork+0x2d/0x50
The stress test involves CPU hotplug operations and memory control group
(memcg) operations. The scenario can be described as follows:
echo xx > memory.max cache_ap_online oom_reaper
(CPU23) (CPU50)
xx < usage stop_machine_from_inactive_cpu
for(;;) // all active cpus
trigger OOM queue_stop_cpus_work
// waiting oom_reaper
multi_cpu_stop(migration/xx)
// sync all active cpus ack
// waiting cpu23 ack
// CPU50 loops in multi_cpu_stop
waiting cpu50
Detailed explanation:
1. When the usage is larger than xx, an OOM may be triggered. If the
process does not handle with ths kill signal immediately, it will loop
in the memory_max_write.
2. When cache_ap_online is triggered, the multi_cpu_stop is queued to the
active cpus. Within the multi_cpu_stop function, it attempts to
synchronize the CPU states. However, the CPU23 didn't acknowledge
because it is stuck in a loop within the for(;;).
3. The oom_reaper process is blocked because CPU50 is in a loop, waiting
for CPU23 to acknowledge the synchronization request.
4. Finally, it formed cyclic dependency and lead to softlockup and dead
loop.
To fix this issue, add cond_resched() in the memory_max_write, so that it
will not block migration task.
Link: https://lkml.kernel.org/r/20250211081819.33307-1-chenridong@huaweicloud.com
Fixes: b6e6edcfa405 ("mm: memcontrol: reclaim and OOM kill when shrinking memory.max below usage")
Signed-off-by: Chen Ridong <chenridong@huawei.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Wang Weiyang <wangweiyang2@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Link: https://lkml.kernel.org/r/20250211212117.3195265-1-ndesaulniers@google.com
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
