Michael Liang ec00ea5645 nvme-tcp: fix premature queue removal and I/O failover ... [ Upstream commit 77e40bbce93059658aee02786a32c5c98a240a8a ] 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>		2025-05-09 09:44:01 +02:00
..
accel	accel/qaic: Fix integer overflow in qaic_validate_req()	2025-03-28 21:59:54 +01:00
accessibility
acpi	ACPI PPTT: Fix coding mistakes in a couple of sizeof() calls	2025-05-02 07:50:58 +02:00
amba
android
ata	ata: libata-scsi: Fix ata_msense_control_ata_feature()	2025-05-02 07:50:48 +02:00
atm
auxdisplay	auxdisplay: hd44780: Fix an API misuse in hd44780.c	2025-05-02 07:50:38 +02:00
base	driver core: fix potential NULL pointer dereference in dev_uevent()	2025-05-02 07:51:03 +02:00
bcma
block	loop: aio inherit the ioprio of original request	2025-05-02 07:51:01 +02:00
bluetooth	Bluetooth: btusb: avoid NULL pointer dereference in skb_dequeue()	2025-05-09 09:43:57 +02:00
bus	bus: mhi: host: Fix race between unprepare and queue_buf	2025-04-25 10:45:26 +02:00
cache
cdrom
cdx	cdx: Fix possible UAF error in driver_override_show()	2025-03-13 12:58:36 +01:00
char	char: misc: register chrdev region with all possible minors	2025-05-02 07:50:49 +02:00
clk	clk: check for disabled clock-provider in of_clk_get_hw_from_clkspec()	2025-05-02 07:50:53 +02:00
clocksource	clocksource/drivers/stm32-lptimer: Use wakeup capable instead of init wakeup	2025-04-25 10:45:25 +02:00
comedi	comedi: jr3_pci: Fix synchronous deletion of timer	2025-05-02 07:51:03 +02:00
connector
counter	counter: microchip-tcb-capture: Fix undefined counter channel state on probe	2025-04-07 10:06:37 +02:00
cpufreq	cpufreq: Fix setting policy limits when frequency tables are used	2025-05-09 09:43:53 +02:00
cpuidle	cpuidle: riscv-sbi: fix device node release in early exit of for_each_possible_cpu	2025-01-17 13:36:16 +01:00
crypto	crypto: ccp - Add support for PCI device 0x1134	2025-05-02 07:50:52 +02:00
cxl	cxl/core/regs.c: Skip Memory Space Enable check for RCD and RCH Ports	2025-05-02 07:50:47 +02:00
dax
dca
devfreq
dio
dma	dmaengine: dmatest: Fix dmatest waiting less when interrupted	2025-05-02 07:50:55 +02:00
dma-buf	udmabuf: fix a buf size overflow issue during udmabuf creation	2025-05-02 07:50:57 +02:00
edac	EDAC/altera: Set DDR and SDMMC interrupt mask before registration	2025-05-09 09:43:50 +02:00
eisa
extcon
firewire
firmware	efi/libstub: Bump up EFI_MMAP_NR_SLACK_SLOTS to 32	2025-04-25 10:45:55 +02:00
fpga
fsi
gnss
gpio	gpiolib: of: Move Atmel HSMCI quirk up out of the regulator comment	2025-05-02 07:50:59 +02:00
gpu	drm/i915/pxp: fix undefined reference to `intel_pxp_gsccs_is_ready_for_sessions'	2025-05-09 09:43:56 +02:00
greybus
hid	HID: pidff: Fix null pointer dereference in pidff_find_fields	2025-04-25 10:45:12 +02:00
hsi	HSI: ssi_protocol: Fix use after free vulnerability in ssi_protocol Driver Due to Race Condition	2025-04-25 10:45:38 +02:00
hte
hv	Drivers: hv: vmbus: Don't release fb_mmio resource in vmbus_free_mmio()	2025-03-22 12:50:38 -07:00
hwmon	hwmon: (nct6775-core) Fix out of bounds access for NCT679{8,9}	2025-04-10 14:37:38 +02:00
hwspinlock
hwtracing	coresight-etm4x: add isb() before reading the TRCSTATR	2025-04-10 14:37:32 +02:00
i2c	i2c: imx-lpi2c: Fix clock count when probe defers	2025-05-09 09:43:50 +02:00
i3c	i3c: Add NULL pointer check in i3c_master_queue_ibi()	2025-04-25 10:45:28 +02:00
idle	intel_idle: Handle older CPUs, which stop the TSC in deeper C states, correctly	2025-03-07 16:45:49 +01:00
iio	iio: adc: ad7768-1: Fix conversion result sign	2025-05-02 07:50:39 +02:00
infiniband	qibfs: fix _another_ leak	2025-05-02 07:50:56 +02:00
input	Input: i8042 - swap old quirk combination with new quirk for more devices	2025-03-22 12:50:46 -07:00
interconnect
iommu	iommu: Handle race with default domain setup	2025-05-09 09:43:56 +02:00
ipack
irqchip	irqchip/qcom-mpm: Prevent crash when trying to handle non-wake GPIOs	2025-05-09 09:43:51 +02:00
isdn
leds	leds: rgb: leds-qcom-lpg: Fix calculation of best period Hi-Res PWMs	2025-04-25 10:45:28 +02:00
macintosh
mailbox	mailbox: pcc: Always clear the platform ack interrupt first	2025-05-02 07:50:54 +02:00
mcb	mcb: fix a double free bug in chameleon_parse_gdd()	2025-05-02 07:50:47 +02:00
md	dm: always update the array size in realloc_argv on success	2025-05-09 09:43:52 +02:00
media	media: vimc: skip .s_stream() for stopped entities	2025-05-02 07:50:37 +02:00
memory	memory: omap-gpmc: drop no compatible check	2025-04-10 14:37:38 +02:00
memstick	memstick: rtsx_usb_ms: Fix slab-use-after-free in rtsx_usb_ms_drv_remove	2025-04-07 10:06:37 +02:00
message
mfd	mfd: ene-kb3930: Fix a potential NULL pointer dereference	2025-04-25 10:45:28 +02:00
misc	objtool, lkdtm: Obfuscate the do_nothing() pointer	2025-05-02 07:50:56 +02:00
mmc	mmc: renesas_sdhi: Fix error handling in renesas_sdhi_probe	2025-05-09 09:43:51 +02:00
most
mtd	mtd: rawnand: Add status chack in r852_ready()	2025-04-25 10:45:29 +02:00
mux
net	bnxt_en: Fix ethtool -d byte order for 32-bit values	2025-05-09 09:44:01 +02:00
nfc
ntb	ntb_hw_amd: Add NTB PCI ID for new gen CPU	2025-05-02 07:50:56 +02:00
nubus
nvdimm
nvme	nvme-tcp: fix premature queue removal and I/O failover	2025-05-09 09:44:01 +02:00
nvmem	nvmem: imx-ocotp-ele: fix MAC address byte order	2025-02-27 04:10:47 -08:00
of	of: resolver: Fix device node refcount leakage in of_resolve_phandles()	2025-05-02 07:50:41 +02:00
opp	OPP: OF: Fix an OF node leak in _opp_add_static_v2()	2025-02-08 09:51:55 +01:00
parisc
parport
pci	PCI: imx6: Skip controller_id generation logic for i.MX7D	2025-05-09 09:43:55 +02:00
pcmcia
peci
perf	perf: arm_pmu: Don't disable counter in armpmu_add()	2025-04-25 10:45:10 +02:00
phy	phy: freescale: imx8m-pcie: assert phy reset and perst in power off	2025-04-25 10:45:36 +02:00
pinctrl	pinctrl: renesas: rza2: Fix potential NULL pointer dereference	2025-05-02 07:50:52 +02:00
platform	platform/x86/intel-uncore-freq: Fix missing uncore sysfs during CPU hotplug	2025-05-09 09:43:53 +02:00
pmdomain	pmdomain: imx8mp-blk-ctrl: add missing loop break condition	2025-01-23 17:21:17 +01:00
pnp
power	power: supply: max77693: Fix wrong conversion of charge input threshold value	2025-04-10 14:37:31 +02:00
powercap	powercap: call put_device() on an error path in powercap_register_control_type()	2025-03-22 12:50:40 -07:00
pps	pps: Fix a use-after-free	2025-02-08 09:52:38 +01:00
ps3
ptp	ptp: ocp: fix start time alignment in ptp_ocp_signal_set	2025-04-25 10:45:44 +02:00
pwm	pwm: fsl-ftm: Handle clk_get_rate() returning 0	2025-04-25 10:45:21 +02:00
rapidio	rapidio: fix an API misues when rio_add_net() fails	2025-03-13 12:58:27 +01:00
ras
regulator	objtool, regulator: rk808: Remove potential undefined behavior in rk806_set_mode_dcdc()	2025-05-02 07:50:56 +02:00
remoteproc	remoteproc: qcom_q6v5_mss: Handle platforms with one power domain	2025-04-10 14:37:31 +02:00
reset	reset: starfive: jh71x0: Fix accessing the empty member on JH7110 SoC	2025-04-07 10:06:36 +02:00
rpmsg
rtc	rtc: pcf85063: do a SW reset if POR failed	2025-05-02 07:50:57 +02:00
s390	s390/tty: Fix a potential memory leak bug	2025-05-02 07:50:53 +02:00
sbus
scsi	scsi: pm80xx: Set phy_attached to zero when device is gone	2025-05-02 07:51:01 +02:00
sh
siox
slimbus	slimbus: messaging: Free transaction ID in delayed interrupt scenario	2025-03-13 12:58:37 +01:00
soc	soc: qcom: ice: introduce devm_of_qcom_ice_get	2025-05-02 07:50:37 +02:00
soundwire	soundwire: slave: fix an OF node reference leak in soundwire slave device	2025-04-10 14:37:32 +02:00
spi	spi: tegra114: Don't fail set_cs_timing when delays are zero	2025-05-09 09:43:51 +02:00
spmi
ssb
staging	staging: rtl8723bs: select CONFIG_CRYPTO_LIB_AES	2025-04-10 14:37:34 +02:00
target	scsi: target: spc: Fix RSOC parameter data header size	2025-04-25 10:45:14 +02:00
tc
tee	tee: optee: Fix supplicant wait loop	2025-02-27 04:10:51 -08:00
thermal	thermal/drivers/rockchip: Add missing rk3328 mapping entry	2025-04-25 10:45:32 +02:00
thunderbolt	thunderbolt: Scan retimers after device router has been enumerated	2025-05-02 07:50:55 +02:00
tty	serial: sifive: lock port in startup()/shutdown() callbacks	2025-05-02 07:50:49 +02:00
ufs	scsi: ufs: exynos: Ensure pre_link() executes before exynos_ufs_phy_init()	2025-05-02 07:51:01 +02:00
uio
usb	usb: host: xhci-plat: mvebu: use ->quirks instead of ->init_quirk() func	2025-05-02 07:50:55 +02:00
vdpa	vdpa/mlx5: Fix oversized null mkey longer than 32bit	2025-04-25 10:45:27 +02:00
vfio	Revert "vfio/platform: check the bounds of read/write syscalls"	2025-02-21 13:57:27 +01:00
vhost	vhost-scsi: Fix handling of multiple calls to vhost_scsi_set_endpoint	2025-04-10 14:37:32 +02:00
video	backlight: led_bl: Hold led_access lock when calling led_sysfs_disable()	2025-04-25 10:45:30 +02:00
virt	drivers: virt: acrn: hsm: Use kzalloc to avoid info leak in pmcmd_ioctl	2025-03-13 12:58:37 +01:00
virtio
vlynq
w1
watchdog	watchdog: rti_wdt: Fix an OF node leak in rti_wdt_probe()	2025-02-08 09:52:25 +01:00
xen	xen: Change xen-acpi-processor dom0 dependency	2025-05-02 07:50:58 +02:00
zorro
Kconfig
Makefile