6.1.128-rc1 review

This is the start of the stable review cycle for the 6.1.128 release. There are 49 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Sat, 01 Feb 2025 14:01:16 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.128-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

-------------
Pseudo-Shortlog of commits:

Greg Kroah-Hartman <[email protected]>

    Linux 6.1.128-rc1

Marek Szyprowski <[email protected]>
ASoC: samsung: midas_wm1811: Fix 'Headphone Switch' control creation

Paulo Alcantara <[email protected]>
smb: client: fix NULL ptr deref in crypto_aead_setkey()

Jack Greiner <[email protected]>
Input: xpad - add support for wooting two he (arm)

Nilton Perim Neto <[email protected]>
Input: xpad - add unofficial Xbox 360 wireless receiver clone

Mark Pearson <[email protected]>
Input: atkbd - map F23 key to support default copilot shortcut

Lianqin Hu <[email protected]>
ALSA: usb-audio: Add delay quirk for USB Audio Device

Greg Kroah-Hartman <[email protected]>
Revert "usb: gadget: u_serial: Disable ep before setting port to null to fix the crash caused by port being null"

Qasim Ijaz <[email protected]>
USB: serial: quatech2: fix null-ptr-deref in qt2_process_read_urb()

Enzo Matsumiya <[email protected]>
smb: client: fix UAF in async decryption

Anjaneyulu <[email protected]>
wifi: iwlwifi: add a few rate index validity checks

Easwar Hariharan <[email protected]>
scsi: storvsc: Ratelimit warning logs to prevent VM denial of service

Ido Schimmel <[email protected]>
ipv4: ip_tunnel: Fix suspicious RCU usage warning in ip_tunnel_find()

Luis Henriques (SUSE) <[email protected]>
ext4: fix access to uninitialised lock in fc replay path

Alex Williamson <[email protected]>
vfio/platform: check the bounds of read/write syscalls

Jiri Kosina <[email protected]>
Revert "HID: multitouch: Add support for lenovo Y9000P Touchpad"

Alexey Dobriyan <[email protected]>
block: fix integer overflow in BLKSECDISCARD

Jamal Hadi Salim <[email protected]>
net: sched: fix ets qdisc OOB Indexing

Pavel Begunkov <[email protected]>
io_uring: fix waiters missing wake ups

Andreas Gruenbacher <[email protected]>
gfs2: Truncate address space when flipping GFS2_DIF_JDATA flag

Christoph Hellwig <[email protected]>
xfs: respect the stable writes flag on the RT device

Christoph Hellwig <[email protected]>
xfs: clean up FS_XFLAG_REALTIME handling in xfs_ioctl_setattr_xflags

Darrick J. Wong <[email protected]>
xfs: dquot recovery does not validate the recovered dquot

Darrick J. Wong <[email protected]>
xfs: clean up dqblk extraction

Dave Chinner <[email protected]>
xfs: inode recovery does not validate the recovered inode

Omar Sandoval <[email protected]>
xfs: fix internal error from AGFL exhaustion

Leah Rumancik <[email protected]>
xfs: up(ic_sema) if flushing data device fails

Christoph Hellwig <[email protected]>
xfs: only remap the written blocks in xfs_reflink_end_cow_extent

Long Li <[email protected]>
xfs: abort intent items when recovery intents fail

Long Li <[email protected]>
xfs: factor out xfs_defer_pending_abort

Catherine Hoang <[email protected]>
xfs: allow read IO and FICLONE to run concurrently

Christoph Hellwig <[email protected]>
xfs: handle nimaps=0 from xfs_bmapi_write in xfs_alloc_file_space

Cheng Lin <[email protected]>
xfs: introduce protection for drop nlink

Darrick J. Wong <[email protected]>
xfs: make sure maxlen is still congruent with prod when rounding down

Darrick J. Wong <[email protected]>
xfs: fix units conversion error in xfs_bmap_del_extent_delay

Darrick J. Wong <[email protected]>
xfs: rt stubs should return negative errnos when rt disabled

Darrick J. Wong <[email protected]>
xfs: prevent rt growfs when quota is enabled

Darrick J. Wong <[email protected]>
xfs: hoist freeing of rt data fork extent mappings

Darrick J. Wong <[email protected]>
xfs: bump max fsgeom struct version

K Prateek Nayak <[email protected]>
softirq: Allow raising SCHED_SOFTIRQ from SMP-call-function on RT kernel

Omid Ehtemam-Haghighi <[email protected]>
ipv6: Fix soft lockups in fib6_select_path under high next hop churn

Cosmin Tanislav <[email protected]>
regmap: detach regmap from dev on regmap_exit

Charles Keepax <[email protected]>
ASoC: samsung: Add missing depends on I2C

Alper Nebi Yasak <[email protected]>
ASoC: samsung: midas_wm1811: Map missing jack kcontrols

Philippe Simons <[email protected]>
irqchip/sunxi-nmi: Add missing SKIP_WAKE flag

Tom Chung <[email protected]>
drm/amd/display: Use HW lock mgr for PSR1

Xiang Zhang <[email protected]>
scsi: iscsi: Fix redundant response for ISCSI_UEVENT_GET_HOST_STATS request

Linus Walleij <[email protected]>
seccomp: Stub for !CONFIG_SECCOMP

Charles Keepax <[email protected]>
ASoC: samsung: Add missing selects for MFD_WM8994

Charles Keepax <[email protected]>
ASoC: wm8994: Add depends on MFD core

-------------

Diffstat:

 Makefile                                           |   4 +-
 block/ioctl.c                                      |   9 +-
 drivers/base/regmap/regmap.c                       |  12 +
 .../gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c  |   3 +-
 drivers/hid/hid-ids.h                              |   1 -
 drivers/hid/hid-multitouch.c                       |   8 +-
 drivers/input/joystick/xpad.c                      |   2 +
 drivers/input/keyboard/atkbd.c                     |   2 +-
 drivers/irqchip/irq-sunxi-nmi.c                    |   3 +-
 drivers/net/wireless/intel/iwlwifi/dvm/rs.c        |   9 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c        |   9 +-
 drivers/scsi/scsi_transport_iscsi.c                |   4 +-
 drivers/scsi/storvsc_drv.c                         |   8 +-
 drivers/usb/gadget/function/u_serial.c             |   8 +-
 drivers/usb/serial/quatech2.c                      |   2 +-
 drivers/vfio/platform/vfio_platform_common.c       |  10 +
 fs/ext4/super.c                                    |   3 +-
 fs/gfs2/file.c                                     |   1 +
 fs/smb/client/smb2ops.c                            |  47 ++--
 fs/smb/client/smb2pdu.c                            |  10 +-
 fs/xfs/libxfs/xfs_alloc.c                          |  27 ++-
 fs/xfs/libxfs/xfs_bmap.c                           |  21 +-
 fs/xfs/libxfs/xfs_defer.c                          |  38 +--
 fs/xfs/libxfs/xfs_defer.h                          |   2 +-
 fs/xfs/libxfs/xfs_inode_buf.c                      |   3 +
 fs/xfs/libxfs/xfs_rtbitmap.c                       |  33 +++
 fs/xfs/libxfs/xfs_sb.h                             |   2 +-
 fs/xfs/xfs_bmap_util.c                             |  24 +-
 fs/xfs/xfs_dquot.c                                 |   5 +-
 fs/xfs/xfs_dquot_item_recover.c                    |  21 +-
 fs/xfs/xfs_file.c                                  |  63 ++++-
 fs/xfs/xfs_inode.c                                 |  24 ++
 fs/xfs/xfs_inode.h                                 |  17 ++
 fs/xfs/xfs_inode_item_recover.c                    |  14 +-
 fs/xfs/xfs_ioctl.c                                 |  34 ++-
 fs/xfs/xfs_iops.c                                  |   7 +
 fs/xfs/xfs_log.c                                   |  23 +-
 fs/xfs/xfs_log_recover.c                           |   2 +-
 fs/xfs/xfs_reflink.c                               |   5 +
 fs/xfs/xfs_rtalloc.c                               |  33 ++-
 fs/xfs/xfs_rtalloc.h                               |  27 ++-
 include/linux/seccomp.h                            |   2 +-
 io_uring/io_uring.c                                |   4 +-
 kernel/softirq.c                                   |  15 +-
 net/ipv4/ip_tunnel.c                               |   2 +-
 net/ipv6/ip6_fib.c                                 |   8 +-
 net/ipv6/route.c                                   |  45 ++--
 net/sched/sch_ets.c                                |   2 +
 sound/soc/codecs/Kconfig                           |   1 +
 sound/soc/samsung/Kconfig                          |   6 +-
 sound/soc/samsung/midas_wm1811.c                   |  24 +-
 sound/usb/quirks.c                                 |   2 +
 tools/testing/selftests/net/Makefile               |   1 +
 .../selftests/net/ipv6_route_update_soft_lockup.sh | 262 +++++++++++++++++++++
 54 files changed, 763 insertions(+), 191 deletions(-)

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Charles Keepax <[email protected]>

[ Upstream commit 5ed01155cea69801f1f0c908954a56a5a3474bed ]

The ASoC driver should not be used without the MFD component. This was causing randconfig issues with regmap IRQ which is selected by the MFD part of the wm8994 driver.

---
 sound/soc/codecs/Kconfig | 1 +
 1 file changed, 1 insertion(+)

index 0904827e2f3db..627e9fc92f0b8 100644
@@ -1982,6 +1982,7 @@ config SND_SOC_WM8993
 
 config SND_SOC_WM8994
 	tristate
+	depends on MFD_WM8994
 
 config SND_SOC_WM8995
 	tristate

2.39.5

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Charles Keepax <[email protected]>

[ Upstream commit fd55c6065bec5268740e944a1800e6fad00974d9 ]

Anything selecting SND_SOC_WM8994 should also select MFD_WM8994, as SND_SOC_WM8994 does not automatically do so. Add the missing selects.

---
 sound/soc/samsung/Kconfig | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

index 2a61e620cd3b0..6bf9fd720d4bf 100644
@@ -220,8 +220,9 @@ config SND_SOC_SAMSUNG_TM2_WM5110
 
 config SND_SOC_SAMSUNG_ARIES_WM8994
 	tristate "SoC I2S Audio support for WM8994 on Aries"
-	depends on SND_SOC_SAMSUNG && MFD_WM8994 && IIO && EXTCON
+	depends on SND_SOC_SAMSUNG && IIO && EXTCON
 	select SND_SOC_BT_SCO
+	select MFD_WM8994
 	select SND_SOC_WM8994
 	select SND_SAMSUNG_I2S
 	help

@@ -235,6 +236,7 @@ config SND_SOC_SAMSUNG_MIDAS_WM1811
 	tristate "SoC I2S Audio support for Midas boards"
 	depends on SND_SOC_SAMSUNG
 	select SND_SAMSUNG_I2S
+	select MFD_WM8994
 	select SND_SOC_WM8994
 	help
 	  Say Y if you want to add support for SoC audio on the Midas boards.

2.39.5

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Linus Walleij <[email protected]>

[ Upstream commit f90877dd7fb5085dd9abd6399daf63dd2969fc90 ]

When using !CONFIG_SECCOMP with CONFIG_GENERIC_ENTRY, the randconfig bots found the following snag:

kernel/entry/common.c: In function 'syscall_trace_enter':

of function '__secure_computing' [-Wimplicit-function-declaration]

      52 |                 ret = __secure_computing(NULL);
         |                       ^~~~~~~~~~~~~~~~~~

Since generic entry calls __secure_computing() unconditionally, fix this by moving the stub out of the ifdef clause for CONFIG_HAVE_ARCH_SECCOMP_FILTER so it's always available.

---
 include/linux/seccomp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index d31d76be49825..91ff537c62462 100644
@@ -69,10 +69,10 @@ struct seccomp_data;
 
 #ifdef CONFIG_HAVE_ARCH_SECCOMP_FILTER
 static inline int secure_computing(void) { return 0; }
-static inline int __secure_computing(const struct seccomp_data *sd) { return 0; }
 #else
 static inline void secure_computing_strict(int this_syscall) { return; }
 #endif
+static inline int __secure_computing(const struct seccomp_data *sd) { return 0; }
 
 static inline long prctl_get_seccomp(void)
 {

2.39.5

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiang Zhang <[email protected]>

[ Upstream commit 63ca02221cc5aa0731fe2b0cc28158aaa4b84982 ]

The ISCSI_UEVENT_GET_HOST_STATS request is already handled in iscsi_get_host_stats(). This fix ensures that redundant responses are skipped in iscsi_if_rx().

- On success: send reply and stats from iscsi_get_host_stats() within if_recv_msg().

- On error: fall through.

---
 drivers/scsi/scsi_transport_iscsi.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

index 49dbcd67579aa..687487ea4fd3b 100644
@@ -4102,7 +4102,7 @@ iscsi_if_rx(struct sk_buff *skb)
 		}
 		do {
 			/*
-			 * special case for GET_STATS:
+			 * special case for GET_STATS, GET_CHAP and GET_HOST_STATS:
 			 * on success - sending reply and stats from
 			 * inside of if_recv_msg(),
 			 * on error - fall through.

@@ -4111,6 +4111,8 @@ iscsi_if_rx(struct sk_buff *skb)
 				break;
 			if (ev->type == ISCSI_UEVENT_GET_CHAP && !err)
 				break;
+			if (ev->type == ISCSI_UEVENT_GET_HOST_STATS && !err)
+				break;
 			err = iscsi_if_send_reply(portid, nlh->nlmsg_type,
 						  ev, sizeof(*ev));
 			if (err == -EAGAIN && --retries < 0) {

2.39.5

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tom Chung <[email protected]>

[ Upstream commit b5c764d6ed556c4e81fbe3fd976da77ec450c08e ]

[Why]
Without the dmub hw lock, it may cause the lock timeout issue while do modeset on PSR1 eDP panel.

[How]
Allow dmub hw lock for PSR1.

---
 drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index 3f32e9c3fbaf4..8d7b2eee8c7c3 100644
@@ -65,7 +65,8 @@ void dmub_hw_lock_mgr_inbox0_cmd(struct dc_dmub_srv *dmub_srv,
 
 bool should_use_dmub_lock(struct dc_link *link)
 {
-	if (link->psr_settings.psr_version == DC_PSR_VERSION_SU_1)
+	if (link->psr_settings.psr_version == DC_PSR_VERSION_SU_1 ||
+	    link->psr_settings.psr_version == DC_PSR_VERSION_1)
 		return true;
 	return false;
 }

2.39.5

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Philippe Simons <[email protected]>

[ Upstream commit 3a748d483d80f066ca4b26abe45cdc0c367d13e9 ]

Some boards with Allwinner SoCs connect the PMIC's IRQ pin to the SoC's NMI pin instead of a normal GPIO. Since the power key is connected to the PMIC, and people expect to wake up a suspended system via this key, the NMI IRQ controller must stay alive when the system goes into suspend.

Add the SKIP_WAKE flag to prevent the sunxi NMI controller from going to sleep, so that the power key can wake up those systems.

[ tglx: Fixed up coding style ]

---
 drivers/irqchip/irq-sunxi-nmi.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index 21d49791f8552..83c7417611fa5 100644
@@ -187,7 +187,8 @@ static int __init sunxi_sc_nmi_irq_init(struct device_node *node,
 	gc->chip_types[0].chip.irq_unmask	= irq_gc_mask_set_bit;
 	gc->chip_types[0].chip.irq_eoi		= irq_gc_ack_set_bit;
 	gc->chip_types[0].chip.irq_set_type	= sunxi_sc_nmi_set_type;
-	gc->chip_types[0].chip.flags		= IRQCHIP_EOI_THREADED | IRQCHIP_EOI_IF_HANDLED;
+	gc->chip_types[0].chip.flags		= IRQCHIP_EOI_THREADED | IRQCHIP_EOI_IF_HANDLED |
+						  IRQCHIP_SKIP_SET_WAKE;
 	gc->chip_types[0].regs.ack		= reg_offs->pend;
 	gc->chip_types[0].regs.mask		= reg_offs->enable;
 	gc->chip_types[0].regs.type		= reg_offs->ctrl;

2.39.5

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alper Nebi Yasak <[email protected]>

[ Upstream commit d27224a45e5457ad89195d92decdd57596253428 ]

This driver does not map jack pins to kcontrols that PulseAudio/PipeWire need to handle jack detection events. The WM1811 codec used here seems to support detecting Headphone and Headset Mic connections. Expose each to userspace as a kcontrol and add the necessary widgets.

Stable-dep-of: 704dbe97a681 ("ASoC: samsung: Add missing depends on I2C")

---
 sound/soc/samsung/midas_wm1811.c | 25 ++++++++++++++++++++-----
 1 file changed, 20 insertions(+), 5 deletions(-)

index 6931b9a45b3e5..44b32f5cddcac 100644
@@ -38,6 +38,17 @@ struct midas_priv {
 	struct snd_soc_jack headset_jack;
 };
 
+static struct snd_soc_jack_pin headset_jack_pins[] = {
+	{
+		.pin = "Headphone",
+		.mask = SND_JACK_HEADPHONE,
+	},
+	{
+		.pin = "Headset Mic",
+		.mask = SND_JACK_MICROPHONE,
+	},
+};
+
 static int midas_start_fll1(struct snd_soc_pcm_runtime *rtd, unsigned int rate)
 {
 	struct snd_soc_card *card = rtd->card;

@@ -246,6 +257,7 @@ static const struct snd_kcontrol_new midas_controls[] = {
 	SOC_DAPM_PIN_SWITCH("Main Mic"),
 	SOC_DAPM_PIN_SWITCH("Sub Mic"),
 	SOC_DAPM_PIN_SWITCH("Headset Mic"),
+	SOC_DAPM_PIN_SWITCH("Headphone"),
 
 	SOC_DAPM_PIN_SWITCH("FM In"),
 };

@@ -261,6 +273,7 @@ static const struct snd_soc_dapm_widget midas_dapm_widgets[] = {
 	SND_SOC_DAPM_LINE("HDMI", NULL),
 	SND_SOC_DAPM_LINE("FM In", midas_fm_set),
 
+	SND_SOC_DAPM_HP("Headphone", NULL),
 	SND_SOC_DAPM_MIC("Headset Mic", NULL),
 	SND_SOC_DAPM_MIC("Main Mic", midas_mic_bias),
 	SND_SOC_DAPM_MIC("Sub Mic", midas_submic_bias),

@@ -305,11 +318,13 @@ static int midas_late_probe(struct snd_soc_card *card)
 		return ret;
 	}
 
-	ret = snd_soc_card_jack_new(card, "Headset",
-			SND_JACK_HEADSET | SND_JACK_MECHANICAL |
-			SND_JACK_BTN_0 | SND_JACK_BTN_1 | SND_JACK_BTN_2 |
-			SND_JACK_BTN_3 | SND_JACK_BTN_4 | SND_JACK_BTN_5,
-			&priv->headset_jack);
+	ret = snd_soc_card_jack_new_pins(card, "Headset",
+					 SND_JACK_HEADSET | SND_JACK_MECHANICAL |
+					 SND_JACK_BTN_0 | SND_JACK_BTN_1 | SND_JACK_BTN_2 |
+					 SND_JACK_BTN_3 | SND_JACK_BTN_4 | SND_JACK_BTN_5,
+					 &priv->headset_jack,
+					 headset_jack_pins,
+					 ARRAY_SIZE(headset_jack_pins));
 	if (ret)
 		return ret;
 

2.39.5

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Charles Keepax <[email protected]>

[ Upstream commit 704dbe97a68153a84319ad63f526e12ba868b88e ]

When switching to selects for MFD_WM8994 a dependency should have also been added for I2C, as the dependency on MFD_WM8994 will not be considered by the select.

---
 sound/soc/samsung/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

index 6bf9fd720d4bf..a529852144f98 100644
@@ -220,7 +220,7 @@ config SND_SOC_SAMSUNG_TM2_WM5110
 
 config SND_SOC_SAMSUNG_ARIES_WM8994
 	tristate "SoC I2S Audio support for WM8994 on Aries"
-	depends on SND_SOC_SAMSUNG && IIO && EXTCON
+	depends on SND_SOC_SAMSUNG && I2C && IIO && EXTCON
 	select SND_SOC_BT_SCO
 	select MFD_WM8994
 	select SND_SOC_WM8994

@@ -234,7 +234,7 @@ config SND_SOC_SAMSUNG_ARIES_WM8994
 
 config SND_SOC_SAMSUNG_MIDAS_WM1811
 	tristate "SoC I2S Audio support for Midas boards"
-	depends on SND_SOC_SAMSUNG
+	depends on SND_SOC_SAMSUNG && I2C
 	select SND_SAMSUNG_I2S
 	select MFD_WM8994
 	select SND_SOC_WM8994

2.39.5

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cosmin Tanislav <[email protected]>

commit 3061e170381af96d1e66799d34264e6414d428a7 upstream.

At the end of __regmap_init(), if dev is not NULL, regmap_attach_dev() is called, which adds a devres reference to the regmap, to be able to retrieve a dev's regmap by name using dev_get_regmap().

When calling regmap_exit, the opposite does not happen, and the reference is kept until the dev is detached.

Add a regmap_detach_dev() function and call it in regmap_exit() to make sure that the devres reference is not kept.

Rule: add

---
 drivers/base/regmap/regmap.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)

@@ -652,6 +652,17 @@ int regmap_attach_dev(struct device *dev
 }
 EXPORT_SYMBOL_GPL(regmap_attach_dev);
 
+static int dev_get_regmap_match(struct device *dev, void *res, void *data);
+
+static int regmap_detach_dev(struct device *dev, struct regmap *map)
+{
+	if (!dev)
+		return 0;
+
+	return devres_release(dev, dev_get_regmap_release,
+			      dev_get_regmap_match, (void *)map->name);
+}
+
 static enum regmap_endian regmap_get_reg_endian(const struct regmap_bus *bus,
 					const struct regmap_config *config)
 {

@@ -1536,6 +1547,7 @@ void regmap_exit(struct regmap *map)
 {
 	struct regmap_async *async;
 
+	regmap_detach_dev(map->dev, map);
 	regcache_exit(map);
 	regmap_debugfs_exit(map);
 	regmap_range_exit(map);

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Omid Ehtemam-Haghighi <[email protected]>

commit d9ccb18f83ea2bb654289b6ecf014fd267cc988b upstream.

Soft lockups have been observed on a cluster of Linux-based edge routers located in a highly dynamic environment. Using the bird service, these routers continuously update BGP-advertised routes due to frequently changing nexthop destinations, while also managing significant IPv6 traffic. The lockups occur during the traversal of the multipath circular linked-list in the fib6_select_path function, particularly while iterating through the siblings in the list. The issue typically arises when the nodes of the linked list are unexpectedly deleted concurrently on a different core—indicated by their 'next' and 'previous' elements pointing back to the node itself and their reference count dropping to zero. This results in an infinite loop, leading to a soft lockup that triggers a system panic via the watchdog timer.

Apply RCU primitives in the problematic code sections to resolve the issue. Where necessary, update the references to fib6_siblings to annotate or use the RCU APIs.

Include a test script that reproduces the issue. The script periodically updates the routing table while generating a heavy load of outgoing IPv6 traffic through multiple iperf3 clients. It consistently induces infinite soft lockups within a couple of minutes.

Kernel log:

0 [ffffbd13003e8d30] machine_kexec at ffffffff8ceaf3eb
1 [ffffbd13003e8d90] __crash_kexec at ffffffff8d0120e3
2 [ffffbd13003e8e58] panic at ffffffff8cef65d4
3 [ffffbd13003e8ed8] watchdog_timer_fn at ffffffff8d05cb03
4 [ffffbd13003e8f08] __hrtimer_run_queues at ffffffff8cfec62f 5 [ffffbd13003e8f70] hrtimer_interrupt at ffffffff8cfed756
6 [ffffbd13003e8fd0] __sysvec_apic_timer_interrupt at ffffffff8cea01af 7 [ffffbd13003e8ff0] sysvec_apic_timer_interrupt at ffffffff8df1b83d
-- <IRQ stack> --
8 [ffffbd13003d3708] asm_sysvec_apic_timer_interrupt at ffffffff8e000ecb

    [exception RIP: fib6_select_path+299]
    RIP: ffffffff8ddafe7b  RSP: ffffbd13003d37b8  RFLAGS: 00000287
    RAX: ffff975850b43600  RBX: ffff975850b40200  RCX: 0000000000000000
    RDX: 000000003fffffff  RSI: 0000000051d383e4  RDI: ffff975850b43618
    RBP: ffffbd13003d3800   R8: 0000000000000000   R9: ffff975850b40200
    R10: 0000000000000000  R11: 0000000000000000  R12: ffffbd13003d3830
    R13: ffff975850b436a8  R14: ffff975850b43600  R15: 0000000000000007
    ORIG_RAX: ffffffffffffffff  CS: 0010  SS: 0018

9 [ffffbd13003d3808] ip6_pol_route at ffffffff8ddb030c
10 [ffffbd13003d3888] ip6_pol_route_input at ffffffff8ddb068c 11 [ffffbd13003d3898] fib6_rule_lookup at ffffffff8ddf02b5
12 [ffffbd13003d3928] ip6_route_input at ffffffff8ddb0f47
13 [ffffbd13003d3a18] ip6_rcv_finish_core.constprop.0 at ffffffff8dd950d0 14 [ffffbd13003d3a30] ip6_list_rcv_finish.constprop.0 at ffffffff8dd96274 15 [ffffbd13003d3a98] ip6_sublist_rcv at ffffffff8dd96474
16 [ffffbd13003d3af8] ipv6_list_rcv at ffffffff8dd96615
17 [ffffbd13003d3b60] __netif_receive_skb_list_core at ffffffff8dc16fec 18 [ffffbd13003d3be0] netif_receive_skb_list_internal at ffffffff8dc176b3 19 [ffffbd13003d3c50] napi_gro_receive at ffffffff8dc565b9
20 [ffffbd13003d3c80] ice_receive_skb at ffffffffc087e4f5 [ice] 21 [ffffbd13003d3c90] ice_clean_rx_irq at ffffffffc0881b80 [ice] 22 [ffffbd13003d3d20] ice_napi_poll at ffffffffc088232f [ice] 23 [ffffbd13003d3d80] __napi_poll at ffffffff8dc18000
24 [ffffbd13003d3db8] net_rx_action at ffffffff8dc18581
25 [ffffbd13003d3e40] __do_softirq at ffffffff8df352e9
26 [ffffbd13003d3eb0] run_ksoftirqd at ffffffff8ceffe47
27 [ffffbd13003d3ec0] smpboot_thread_fn at ffffffff8cf36a30 28 [ffffbd13003d3ee8] kthread at ffffffff8cf2b39f
29 [ffffbd13003d3f28] ret_from_fork at ffffffff8ce5fa64
30 [ffffbd13003d3f50] ret_from_fork_asm at ffffffff8ce03cbb

---
 net/ipv6/ip6_fib.c                                           |    8 
 net/ipv6/route.c                                             |   45 +
 tools/testing/selftests/net/Makefile                         |    1 
 tools/testing/selftests/net/ipv6_route_update_soft_lockup.sh |  262 +++++++++++
 4 files changed, 297 insertions(+), 19 deletions(-)
 create mode 100755 tools/testing/selftests/net/ipv6_route_update_soft_lockup.sh

@@ -1180,8 +1180,8 @@ next_iter:
 		while (sibling) {
 			if (sibling->fib6_metric == rt->fib6_metric &&
 			    rt6_qualify_for_ecmp(sibling)) {
-				list_add_tail(&rt->fib6_siblings,
-					      &sibling->fib6_siblings);
+				list_add_tail_rcu(&rt->fib6_siblings,
+						  &sibling->fib6_siblings);
 				break;
 			}
 			sibling = rcu_dereference_protected(sibling->fib6_next,

@@ -1242,7 +1242,7 @@ add:
 							 fib6_siblings)
 					sibling->fib6_nsiblings--;
 				rt->fib6_nsiblings = 0;
-				list_del_init(&rt->fib6_siblings);
+				list_del_rcu(&rt->fib6_siblings);
 				rt6_multipath_rebalance(next_sibling);
 				return err;
 			}

@@ -1955,7 +1955,7 @@ static void fib6_del_route(struct fib6_t
 					 &rt->fib6_siblings, fib6_siblings)
 			sibling->fib6_nsiblings--;
 		rt->fib6_nsiblings = 0;
-		list_del_init(&rt->fib6_siblings);
+		list_del_rcu(&rt->fib6_siblings);
 		rt6_multipath_rebalance(next_sibling);
 	}
 

@@ -72,6 +72,7 @@ TEST_GEN_PROGS += sk_bind_sendto_listen
 TEST_GEN_PROGS += sk_connect_zero_addr
 TEST_PROGS += test_ingress_egress_chaining.sh
 TEST_GEN_FILES += nat6to4.o
+TEST_PROGS += ipv6_route_update_soft_lockup.sh
 
 TEST_FILES := settings
 

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: K Prateek Nayak <[email protected]>

commit 6675ce20046d149e1e1ffe7e9577947dee17aad5 upstream.

do_softirq_post_smp_call_flush() on PREEMPT_RT kernels carries a WARN_ON_ONCE() for any SOFTIRQ being raised from an SMP-call-function. Since do_softirq_post_smp_call_flush() is called with preempt disabled, raising a SOFTIRQ during flush_smp_call_function_queue() can lead to longer preempt disabled sections.

Since commit b2a02fc43a1f ("smp: Optimize send_call_function_single_ipi()") IPIs to an idle CPU in TIF_POLLING_NRFLAG mode can be optimized out by instead setting TIF_NEED_RESCHED bit in idle task's thread_info and relying on the flush_smp_call_function_queue() in the idle-exit path to run the SMP-call-function.

To trigger an idle load balancing, the scheduler queues nohz_csd_function() responsible for triggering an idle load balancing on a target nohz idle CPU and sends an IPI. Only now, this IPI is optimized out and the SMP-call-function is executed from flush_smp_call_function_queue() in do_idle() which can raise a SCHED_SOFTIRQ to trigger the balancing.

So far, this went undetected since, the need_resched() check in nohz_csd_function() would make it bail out of idle load balancing early as the idle thread does not clear TIF_POLLING_NRFLAG before calling flush_smp_call_function_queue(). The need_resched() check was added with the intent to catch a new task wakeup, however, it has recently discovered to be unnecessary and will be removed in the subsequent commit after which nohz_csd_function() can raise a SCHED_SOFTIRQ from flush_smp_call_function_queue() to trigger an idle load balance on an idle target in TIF_POLLING_NRFLAG mode.

nohz_csd_function() bails out early if "idle_cpu()" check for the target CPU, and does not lock the target CPU's rq until the very end, once it has found tasks to run on the CPU and will not inhibit the wakeup of, or running of a newly woken up higher priority task. Account for this and prevent a WARN_ON_ONCE() when SCHED_SOFTIRQ is raised from flush_smp_call_function_queue().

---
 kernel/softirq.c |   15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

@@ -294,17 +294,24 @@ static inline void invoke_softirq(void)
 		wakeup_softirqd();
 }
 
+#define SCHED_SOFTIRQ_MASK	BIT(SCHED_SOFTIRQ)
+
 /*
  * flush_smp_call_function_queue() can raise a soft interrupt in a function
- * call. On RT kernels this is undesired and the only known functionality
- * in the block layer which does this is disabled on RT. If soft interrupts
- * get raised which haven't been raised before the flush, warn so it can be
+ * call. On RT kernels this is undesired and the only known functionalities
+ * are in the block layer which is disabled on RT, and in the scheduler for
+ * idle load balancing. If soft interrupts get raised which haven't been
+ * raised before the flush, warn if it is not a SCHED_SOFTIRQ so it can be
  * investigated.
  */
 void do_softirq_post_smp_call_flush(unsigned int was_pending)
 {
-	if (WARN_ON_ONCE(was_pending != local_softirq_pending()))
+	unsigned int is_pending = local_softirq_pending();
+
+	if (unlikely(was_pending != is_pending)) {
+		WARN_ON_ONCE(was_pending != (is_pending & ~SCHED_SOFTIRQ_MASK));
 		invoke_softirq();
+	}
 }
 
 #else /* CONFIG_PREEMPT_RT */

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit 9488062805943c2d63350d3ef9e4dc093799789a ]

The latest version of the fs geometry structure is v5. Bump this constant so that xfs_db and mkfs calls to libxfs_fs_geometry will fill out all the fields.

IOWs, this commit is a no-op for the kernel, but will be useful for userspace reporting in later changes.

---
 fs/xfs/libxfs/xfs_sb.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -25,7 +25,7 @@ extern uint64_t	xfs_sb_version_to_featur
 
 extern int	xfs_update_secondary_sbs(struct xfs_mount *mp);
 
-#define XFS_FS_GEOM_MAX_STRUCT_VER	(4)
+#define XFS_FS_GEOM_MAX_STRUCT_VER	(5)
 extern void	xfs_fs_geometry(struct xfs_mount *mp, struct xfs_fsop_geom *geo,
 				int struct_version);
 extern int	xfs_sb_read_secondary(struct xfs_mount *mp,

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit 6c664484337b37fa0cf6e958f4019623e30d40f7 ]

Currently, xfs_bmap_del_extent_real contains a bunch of code to convert the physical extent of a data fork mapping for a realtime file into rt extents and pass that to the rt extent freeing function. Since the details of this aren't needed when CONFIG_XFS_REALTIME=n, move it to xfs_rtbitmap.c to reduce code size when realtime isn't enabled.

This will (one day) enable realtime EFIs to reuse the same unit-converting call with less code duplication.

---
 fs/xfs/libxfs/xfs_bmap.c     |   19 +++----------------
 fs/xfs/libxfs/xfs_rtbitmap.c |   33 +++++++++++++++++++++++++++++++++
 fs/xfs/xfs_rtalloc.h         |    5 +++++
 3 files changed, 41 insertions(+), 16 deletions(-)

@@ -5037,33 +5037,20 @@ xfs_bmap_del_extent_real(
 
 	flags = XFS_ILOG_CORE;
 	if (whichfork == XFS_DATA_FORK && XFS_IS_REALTIME_INODE(ip)) {
-		xfs_filblks_t	len;
-		xfs_extlen_t	mod;
-
-		len = div_u64_rem(del->br_blockcount, mp->m_sb.sb_rextsize,
-				  &mod);
-		ASSERT(mod == 0);
-
 		if (!(bflags & XFS_BMAPI_REMAP)) {
-			xfs_fsblock_t	bno;
-
-			bno = div_u64_rem(del->br_startblock,
-					mp->m_sb.sb_rextsize, &mod);
-			ASSERT(mod == 0);
-
-			error = xfs_rtfree_extent(tp, bno, (xfs_extlen_t)len);
+			error = xfs_rtfree_blocks(tp, del->br_startblock,
+					del->br_blockcount);
 			if (error)
 				goto done;
 		}
 
 		do_fx = 0;
-		nblks = len * mp->m_sb.sb_rextsize;
 		qfield = XFS_TRANS_DQ_RTBCOUNT;
 	} else {
 		do_fx = 1;
-		nblks = del->br_blockcount;
 		qfield = XFS_TRANS_DQ_BCOUNT;
 	}
+	nblks = del->br_blockcount;
 
 	del_endblock = del->br_startblock + del->br_blockcount;
 	if (cur) {

@@ -1005,6 +1005,39 @@ xfs_rtfree_extent(
 	return 0;
 }
 
+/*
+ * Free some blocks in the realtime subvolume.  rtbno and rtlen are in units of
+ * rt blocks, not rt extents; must be aligned to the rt extent size; and rtlen
+ * cannot exceed XFS_MAX_BMBT_EXTLEN.
+ */
+int
+xfs_rtfree_blocks(
+	struct xfs_trans	*tp,
+	xfs_fsblock_t		rtbno,
+	xfs_filblks_t		rtlen)
+{
+	struct xfs_mount	*mp = tp->t_mountp;
+	xfs_rtblock_t		bno;
+	xfs_filblks_t		len;
+	xfs_extlen_t		mod;
+
+	ASSERT(rtlen <= XFS_MAX_BMBT_EXTLEN);
+
+	len = div_u64_rem(rtlen, mp->m_sb.sb_rextsize, &mod);
+	if (mod) {
+		ASSERT(mod == 0);
+		return -EIO;
+	}
+
+	bno = div_u64_rem(rtbno, mp->m_sb.sb_rextsize, &mod);
+	if (mod) {
+		ASSERT(mod == 0);
+		return -EIO;
+	}
+
+	return xfs_rtfree_extent(tp, bno, len);
+}
+
 /* Find all the free records within a given range. */
 int
 xfs_rtalloc_query_range(

@@ -58,6 +58,10 @@ xfs_rtfree_extent(
 	xfs_rtblock_t		bno,	/* starting block number to free */
 	xfs_extlen_t		len);	/* length of extent freed */
 
+/* Same as above, but in units of rt blocks. */
+int xfs_rtfree_blocks(struct xfs_trans *tp, xfs_fsblock_t rtbno,
+		xfs_filblks_t rtlen);
+
 /*
  * Initialize realtime fields in the mount structure.
  */

@@ -139,6 +143,7 @@ int xfs_rtalloc_reinit_frextents(struct
 #else
 # define xfs_rtallocate_extent(t,b,min,max,l,f,p,rb)    (ENOSYS)
 # define xfs_rtfree_extent(t,b,l)                       (ENOSYS)
+# define xfs_rtfree_blocks(t,rb,rl)			(ENOSYS)
 # define xfs_rtpick_extent(m,t,l,rb)                    (ENOSYS)
 # define xfs_growfs_rt(mp,in)                           (ENOSYS)
 # define xfs_rtalloc_query_range(t,l,h,f,p)             (ENOSYS)

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit b73494fa9a304ab95b59f07845e8d7d36e4d23e0 ]

Quotas aren't (yet) supported with realtime, so we shouldn't allow userspace to set up a realtime section when quotas are enabled, even if they attached one via mount options. IOWS, you shouldn't be able to do:

# mkfs.xfs -f /dev/sda
# mount /dev/sda /mnt -o rtdev=/dev/sdb,usrquota
# xfs_growfs -r /mnt

---
 fs/xfs/xfs_rtalloc.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -954,7 +954,7 @@ xfs_growfs_rt(
 		return -EINVAL;
 
 	/* Unsupported realtime features. */
-	if (xfs_has_rmapbt(mp) || xfs_has_reflink(mp))
+	if (xfs_has_rmapbt(mp) || xfs_has_reflink(mp) || xfs_has_quota(mp))
 		return -EOPNOTSUPP;
 
 	nrblocks = in->newblocks;

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit c2988eb5cff75c02bc57e02c323154aa08f55b78 ]

When realtime support is not compiled into the kernel, these functions should return negative errnos, not positive errnos. While we're at it, fix a broken macro declaration.

---
 fs/xfs/xfs_rtalloc.h |   24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

@@ -141,17 +141,17 @@ int xfs_rtalloc_extent_is_free(struct xf
 			       bool *is_free);
 int xfs_rtalloc_reinit_frextents(struct xfs_mount *mp);
 #else
-# define xfs_rtallocate_extent(t,b,min,max,l,f,p,rb)    (ENOSYS)
-# define xfs_rtfree_extent(t,b,l)                       (ENOSYS)
-# define xfs_rtfree_blocks(t,rb,rl)			(ENOSYS)
-# define xfs_rtpick_extent(m,t,l,rb)                    (ENOSYS)
-# define xfs_growfs_rt(mp,in)                           (ENOSYS)
-# define xfs_rtalloc_query_range(t,l,h,f,p)             (ENOSYS)
-# define xfs_rtalloc_query_all(m,t,f,p)                 (ENOSYS)
-# define xfs_rtbuf_get(m,t,b,i,p)                       (ENOSYS)
-# define xfs_verify_rtbno(m, r)			(false)
-# define xfs_rtalloc_extent_is_free(m,t,s,l,i)          (ENOSYS)
-# define xfs_rtalloc_reinit_frextents(m)                (0)
+# define xfs_rtallocate_extent(t,b,min,max,l,f,p,rb)	(-ENOSYS)
+# define xfs_rtfree_extent(t,b,l)			(-ENOSYS)
+# define xfs_rtfree_blocks(t,rb,rl)			(-ENOSYS)
+# define xfs_rtpick_extent(m,t,l,rb)			(-ENOSYS)
+# define xfs_growfs_rt(mp,in)				(-ENOSYS)
+# define xfs_rtalloc_query_range(m,t,l,h,f,p)		(-ENOSYS)
+# define xfs_rtalloc_query_all(m,t,f,p)			(-ENOSYS)
+# define xfs_rtbuf_get(m,t,b,i,p)			(-ENOSYS)
+# define xfs_verify_rtbno(m, r)				(false)
+# define xfs_rtalloc_extent_is_free(m,t,s,l,i)		(-ENOSYS)
+# define xfs_rtalloc_reinit_frextents(m)		(0)
 static inline int		/* error */
 xfs_rtmount_init(
 	xfs_mount_t	*mp)	/* file system mount structure */

@@ -162,7 +162,7 @@ xfs_rtmount_init(
 	xfs_warn(mp, "Not built with CONFIG_XFS_RT");
 	return -ENOSYS;
 }
-# define xfs_rtmount_inodes(m)  (((mp)->m_sb.sb_rblocks == 0)? 0 : (ENOSYS))
+# define xfs_rtmount_inodes(m)  (((mp)->m_sb.sb_rblocks == 0)? 0 : (-ENOSYS))
 # define xfs_rtunmount_inodes(m)
 #endif	/* CONFIG_XFS_RT */
 

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit ddd98076d5c075c8a6c49d9e6e8ee12844137f23 ]

The unit conversions in this function do not make sense. First we convert a block count to bytes, then divide that bytes value by rextsize, which is in blocks, to get an rt extent count. You can't divide bytes by blocks to get a (possibly multiblock) extent value.

Fortunately nobody uses delalloc on the rt volume so this hasn't mattered.

---
 fs/xfs/libxfs/xfs_bmap.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -4807,7 +4807,7 @@ xfs_bmap_del_extent_delay(
 	ASSERT(got_endoff >= del_endoff);
 
 	if (isrt) {
-		uint64_t rtexts = XFS_FSB_TO_B(mp, del->br_blockcount);
+		uint64_t	rtexts = del->br_blockcount;
 
 		do_div(rtexts, mp->m_sb.sb_rextsize);
 		xfs_mod_frextents(mp, rtexts);

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit f6a2dae2a1f52ea23f649c02615d073beba4cc35 ]

In commit 2a6ca4baed62, we tried to fix an overflow problem in the realtime allocator that was caused by an overly large maxlen value causing xfs_rtcheck_range to run off the end of the realtime bitmap. Unfortunately, there is a subtle bug here -- maxlen (and minlen) both have to be aligned with @prod, but @prod can be larger than 1 if the user has set an extent size hint on the file, and that extent size hint is larger than the realtime extent size.

If the rt free space extents are not aligned to this file's extszhint because other files without extent size hints allocated space (or the number of rt extents is similarly not aligned), then it's possible that maxlen after clamping to sb_rextents will no longer be aligned to prod. The allocation will succeed just fine, but we still trip the assertion.

Fix the problem by reducing maxlen by any misalignment with prod. While we're at it, split the assertions into two so that we can tell which value had the bad alignment.

---
 fs/xfs/xfs_rtalloc.c |   31 ++++++++++++++++++++++++++-----
 1 file changed, 26 insertions(+), 5 deletions(-)

@@ -212,6 +212,23 @@ xfs_rtallocate_range(
 }
 
 /*
+ * Make sure we don't run off the end of the rt volume.  Be careful that
+ * adjusting maxlen downwards doesn't cause us to fail the alignment checks.
+ */
+static inline xfs_extlen_t
+xfs_rtallocate_clamp_len(
+	struct xfs_mount	*mp,
+	xfs_rtblock_t		startrtx,
+	xfs_extlen_t		rtxlen,
+	xfs_extlen_t		prod)
+{
+	xfs_extlen_t		ret;
+
+	ret = min(mp->m_sb.sb_rextents, startrtx + rtxlen) - startrtx;
+	return rounddown(ret, prod);
+}
+
+/*
  * Attempt to allocate an extent minlen<=len<=maxlen starting from
  * bitmap block bbno.  If we don't get maxlen then use prod to trim
  * the length, if given.  Returns error; returns starting block in *rtblock.

@@ -248,7 +265,7 @@ xfs_rtallocate_extent_block(
 	     i <= end;
 	     i++) {
 		/* Make sure we don't scan off the end of the rt volume. */
-		maxlen = min(mp->m_sb.sb_rextents, i + maxlen) - i;
+		maxlen = xfs_rtallocate_clamp_len(mp, i, maxlen, prod);
 
 		/*
 		 * See if there's a free extent of maxlen starting at i.

@@ -355,7 +372,8 @@ xfs_rtallocate_extent_exact(
 	int		isfree;		/* extent is free */
 	xfs_rtblock_t	next;		/* next block to try (dummy) */
 
-	ASSERT(minlen % prod == 0 && maxlen % prod == 0);
+	ASSERT(minlen % prod == 0);
+	ASSERT(maxlen % prod == 0);
 	/*
 	 * Check if the range in question (for maxlen) is free.
 	 */

@@ -438,7 +456,9 @@ xfs_rtallocate_extent_near(
 	xfs_rtblock_t	n;		/* next block to try */
 	xfs_rtblock_t	r;		/* result block */
 
-	ASSERT(minlen % prod == 0 && maxlen % prod == 0);
+	ASSERT(minlen % prod == 0);
+	ASSERT(maxlen % prod == 0);
+
 	/*
 	 * If the block number given is off the end, silently set it to
 	 * the last block.

@@ -447,7 +467,7 @@ xfs_rtallocate_extent_near(
 		bno = mp->m_sb.sb_rextents - 1;
 
 	/* Make sure we don't run off the end of the rt volume. */
-	maxlen = min(mp->m_sb.sb_rextents, bno + maxlen) - bno;
+	maxlen = xfs_rtallocate_clamp_len(mp, bno, maxlen, prod);
 	if (maxlen < minlen) {
 		*rtblock = NULLRTBLOCK;
 		return 0;

@@ -638,7 +658,8 @@ xfs_rtallocate_extent_size(
 	xfs_rtblock_t	r;		/* result block number */
 	xfs_suminfo_t	sum;		/* summary information for extents */
 
-	ASSERT(minlen % prod == 0 && maxlen % prod == 0);
+	ASSERT(minlen % prod == 0);
+	ASSERT(maxlen % prod == 0);
 	ASSERT(maxlen != 0);
 
 	/*

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cheng Lin <[email protected]>

[ Upstream commit 2b99e410b28f5a75ae417e6389e767c7745d6fce ]

When abnormal drop_nlink are detected on the inode, return error, to avoid corruption propagation.

---
 fs/xfs/xfs_inode.c |    7 +++++++
 1 file changed, 7 insertions(+)

@@ -919,6 +919,13 @@ xfs_droplink(
 	xfs_trans_t *tp,
 	xfs_inode_t *ip)
 {
+	if (VFS_I(ip)->i_nlink == 0) {
+		xfs_alert(ip->i_mount,
+			  "%s: Attempt to drop inode (%llu) with nlink zero.",
+			  __func__, ip->i_ino);
+		return -EFSCORRUPTED;
+	}
+
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
 
 	drop_nlink(VFS_I(ip));

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christoph Hellwig <[email protected]>

[ Upstream commit 35dc55b9e80cb9ec4bcb969302000b002b2ed850 ]

If xfs_bmapi_write finds a delalloc extent at the requested range, it tries to convert the entire delalloc extent to a real allocation.

But if the allocator cannot find a single free extent large enough to cover the start block of the requested range, xfs_bmapi_write will return 0 but leave *nimaps set to 0.

In that case we simply need to keep looping with the same startoffset_fsb so that one of the following allocations will eventually reach the requested range.

Note that this could affect any caller of xfs_bmapi_write that covers an existing delayed allocation. As far as I can tell we do not have any other such caller, though - the regular writeback path uses xfs_bmapi_convert_delalloc to convert delayed allocations to real ones, and direct I/O invalidates the page cache first.

---
 fs/xfs/xfs_bmap_util.c |   24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

@@ -780,12 +780,10 @@ xfs_alloc_file_space(
 {
 	xfs_mount_t		*mp = ip->i_mount;
 	xfs_off_t		count;
-	xfs_filblks_t		allocated_fsb;
 	xfs_filblks_t		allocatesize_fsb;
 	xfs_extlen_t		extsz, temp;
 	xfs_fileoff_t		startoffset_fsb;
 	xfs_fileoff_t		endoffset_fsb;
-	int			nimaps;
 	int			rt;
 	xfs_trans_t		*tp;
 	xfs_bmbt_irec_t		imaps[1], *imapp;

@@ -808,7 +806,6 @@ xfs_alloc_file_space(
 
 	count = len;
 	imapp = &imaps[0];
-	nimaps = 1;
 	startoffset_fsb	= XFS_B_TO_FSBT(mp, offset);
 	endoffset_fsb = XFS_B_TO_FSB(mp, offset + count);
 	allocatesize_fsb = endoffset_fsb - startoffset_fsb;

@@ -819,6 +816,7 @@ xfs_alloc_file_space(
 	while (allocatesize_fsb && !error) {
 		xfs_fileoff_t	s, e;
 		unsigned int	dblocks, rblocks, resblks;
+		int		nimaps = 1;
 
 		/*
 		 * Determine space reservations for data/realtime.

@@ -884,15 +882,19 @@ xfs_alloc_file_space(
 		if (error)
 			break;
 
-		allocated_fsb = imapp->br_blockcount;
-
-		if (nimaps == 0) {
-			error = -ENOSPC;
-			break;
+		/*
+		 * If the allocator cannot find a single free extent large
+		 * enough to cover the start block of the requested range,
+		 * xfs_bmapi_write will return 0 but leave *nimaps set to 0.
+		 *
+		 * In that case we simply need to keep looping with the same
+		 * startoffset_fsb so that one of the following allocations
+		 * will eventually reach the requested range.
+		 */
+		if (nimaps) {
+			startoffset_fsb += imapp->br_blockcount;
+			allocatesize_fsb -= imapp->br_blockcount;
 		}
-
-		startoffset_fsb += allocated_fsb;
-		allocatesize_fsb -= allocated_fsb;
 	}
 
 	return error;

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Catherine Hoang <[email protected]>

[ Upstream commit 14a537983b228cb050ceca3a5b743d01315dc4aa ]

One of our VM cluster management products needs to snapshot KVM image files so that they can be restored in case of failure. Snapshotting is done by redirecting VM disk writes to a sidecar file and using reflink on the disk image, specifically the FICLONE ioctl as used by "cp --reflink". Reflink locks the source and destination files while it operates, which means that reads from the main vm disk image are blocked, causing the vm to stall. When an image file is heavily fragmented, the copy process could take several minutes. Some of the vm image files have 50-100 million extent records, and duplicating that much metadata locks the file for 30 minutes or more. Having activities suspended for such a long time in a cluster node could result in node eviction.

Clone operations and read IO do not change any data in the source file, so they should be able to run concurrently. Demote the exclusive locks taken by FICLONE to shared locks to allow reads while cloning. While a clone is in progress, writes will take the IOLOCK_EXCL, so they block until the clone completes.

---
 fs/xfs/xfs_file.c    |   63 ++++++++++++++++++++++++++++++++++++++++-----------
 fs/xfs/xfs_inode.c   |   17 +++++++++++++
 fs/xfs/xfs_inode.h   |    9 +++++++
 fs/xfs/xfs_reflink.c |    4 +++
 4 files changed, 80 insertions(+), 13 deletions(-)

@@ -214,6 +214,43 @@ xfs_ilock_iocb(
 	return 0;
 }
 
+static int
+xfs_ilock_iocb_for_write(
+	struct kiocb		*iocb,
+	unsigned int		*lock_mode)
+{
+	ssize_t			ret;
+	struct xfs_inode	*ip = XFS_I(file_inode(iocb->ki_filp));
+
+	ret = xfs_ilock_iocb(iocb, *lock_mode);
+	if (ret)
+		return ret;
+
+	if (*lock_mode == XFS_IOLOCK_EXCL)
+		return 0;
+	if (!xfs_iflags_test(ip, XFS_IREMAPPING))
+		return 0;
+
+	xfs_iunlock(ip, *lock_mode);
+	*lock_mode = XFS_IOLOCK_EXCL;
+	return xfs_ilock_iocb(iocb, *lock_mode);
+}
+
+static unsigned int
+xfs_ilock_for_write_fault(
+	struct xfs_inode	*ip)
+{
+	/* get a shared lock if no remapping in progress */
+	xfs_ilock(ip, XFS_MMAPLOCK_SHARED);
+	if (!xfs_iflags_test(ip, XFS_IREMAPPING))
+		return XFS_MMAPLOCK_SHARED;
+
+	/* wait for remapping to complete */
+	xfs_iunlock(ip, XFS_MMAPLOCK_SHARED);
+	xfs_ilock(ip, XFS_MMAPLOCK_EXCL);
+	return XFS_MMAPLOCK_EXCL;
+}
+
 STATIC ssize_t
 xfs_file_dio_read(
 	struct kiocb		*iocb,

@@ -523,7 +560,7 @@ xfs_file_dio_write_aligned(
 	unsigned int		iolock = XFS_IOLOCK_SHARED;
 	ssize_t			ret;
 
-	ret = xfs_ilock_iocb(iocb, iolock);
+	ret = xfs_ilock_iocb_for_write(iocb, &iolock);
 	if (ret)
 		return ret;
 	ret = xfs_file_write_checks(iocb, from, &iolock);

@@ -590,7 +627,7 @@ retry_exclusive:
 		flags = IOMAP_DIO_FORCE_WAIT;
 	}
 
-	ret = xfs_ilock_iocb(iocb, iolock);
+	ret = xfs_ilock_iocb_for_write(iocb, &iolock);
 	if (ret)
 		return ret;
 

@@ -1158,7 +1195,7 @@ xfs_file_remap_range(
 	if (xfs_file_sync_writes(file_in) || xfs_file_sync_writes(file_out))
 		xfs_log_force_inode(dest);
 out_unlock:
-	xfs_iunlock2_io_mmap(src, dest);
+	xfs_iunlock2_remapping(src, dest);
 	if (ret)
 		trace_xfs_reflink_remap_range_error(dest, ret, _RET_IP_);
 	/*

@@ -1313,6 +1350,7 @@ __xfs_filemap_fault(
 	struct inode		*inode = file_inode(vmf->vma->vm_file);
 	struct xfs_inode	*ip = XFS_I(inode);
 	vm_fault_t		ret;
+	unsigned int		lock_mode = 0;
 
 	trace_xfs_filemap_fault(ip, pe_size, write_fault);
 

@@ -1321,25 +1359,24 @@ __xfs_filemap_fault(
 		file_update_time(vmf->vma->vm_file);
 	}
 
+	if (IS_DAX(inode) || write_fault)
+		lock_mode = xfs_ilock_for_write_fault(XFS_I(inode));
+
 	if (IS_DAX(inode)) {
 		pfn_t pfn;
 
-		xfs_ilock(XFS_I(inode), XFS_MMAPLOCK_SHARED);
 		ret = xfs_dax_fault(vmf, pe_size, write_fault, &pfn);
 		if (ret & VM_FAULT_NEEDDSYNC)
 			ret = dax_finish_sync_fault(vmf, pe_size, pfn);
-		xfs_iunlock(XFS_I(inode), XFS_MMAPLOCK_SHARED);
+	} else if (write_fault) {
+		ret = iomap_page_mkwrite(vmf, &xfs_page_mkwrite_iomap_ops);
 	} else {
-		if (write_fault) {
-			xfs_ilock(XFS_I(inode), XFS_MMAPLOCK_SHARED);
-			ret = iomap_page_mkwrite(vmf,
-					&xfs_page_mkwrite_iomap_ops);
-			xfs_iunlock(XFS_I(inode), XFS_MMAPLOCK_SHARED);
-		} else {
-			ret = filemap_fault(vmf);
-		}
+		ret = filemap_fault(vmf);
 	}
 
+	if (lock_mode)
+		xfs_iunlock(XFS_I(inode), lock_mode);
+
 	if (write_fault)
 		sb_end_pagefault(inode->i_sb);
 	return ret;

@@ -3644,6 +3644,23 @@ xfs_iunlock2_io_mmap(
 		inode_unlock(VFS_I(ip1));
 }
 
+/* Drop the MMAPLOCK and the IOLOCK after a remap completes. */
+void
+xfs_iunlock2_remapping(
+	struct xfs_inode	*ip1,
+	struct xfs_inode	*ip2)
+{
+	xfs_iflags_clear(ip1, XFS_IREMAPPING);
+
+	if (ip1 != ip2)
+		xfs_iunlock(ip1, XFS_MMAPLOCK_SHARED);
+	xfs_iunlock(ip2, XFS_MMAPLOCK_EXCL);
+
+	if (ip1 != ip2)
+		inode_unlock_shared(VFS_I(ip1));
+	inode_unlock(VFS_I(ip2));
+}
+
 /*
  * Reload the incore inode list for this inode.  Caller should ensure that
  * the link count cannot change, either by taking ILOCK_SHARED or otherwise

@@ -347,6 +347,14 @@ static inline bool xfs_inode_has_large_e
 /* Quotacheck is running but inode has not been added to quota counts. */
 #define XFS_IQUOTAUNCHECKED	(1 << 14)
 
+/*
+ * Remap in progress. Callers that wish to update file data while
+ * holding a shared IOLOCK or MMAPLOCK must drop the lock and retake
+ * the lock in exclusive mode. Relocking the file will block until
+ * IREMAPPING is cleared.
+ */
+#define XFS_IREMAPPING		(1U << 15)
+
 /* All inode state flags related to inode reclaim. */
 #define XFS_ALL_IRECLAIM_FLAGS	(XFS_IRECLAIMABLE | \
 				 XFS_IRECLAIM | \

@@ -595,6 +603,7 @@ void xfs_end_io(struct work_struct *work
 
 int xfs_ilock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2);
 void xfs_iunlock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2);
+void xfs_iunlock2_remapping(struct xfs_inode *ip1, struct xfs_inode *ip2);
 
 static inline bool
 xfs_inode_unlinked_incomplete(

@@ -1539,6 +1539,10 @@ xfs_reflink_remap_prep(
 	if (ret)
 		goto out_unlock;
 
+	xfs_iflags_set(src, XFS_IREMAPPING);
+	if (inode_in != inode_out)
+		xfs_ilock_demote(src, XFS_IOLOCK_EXCL | XFS_MMAPLOCK_EXCL);
+
 	return 0;
 out_unlock:
 	xfs_iunlock2_io_mmap(src, dest);

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Long Li <[email protected]>

[ Upstream commit 2a5db859c6825b5d50377dda9c3cc729c20cad43 ]

Factor out xfs_defer_pending_abort() from xfs_defer_trans_abort(), which not use transaction parameter, so it can be used after the transaction life cycle.

---
 fs/xfs/libxfs/xfs_defer.c |   23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

@@ -245,21 +245,18 @@ xfs_defer_create_intents(
 	return ret;
 }
 
-/* Abort all the intents that were committed. */
 STATIC void
-xfs_defer_trans_abort(
-	struct xfs_trans		*tp,
-	struct list_head		*dop_pending)
+xfs_defer_pending_abort(
+	struct xfs_mount		*mp,
+	struct list_head		*dop_list)
 {
 	struct xfs_defer_pending	*dfp;
 	const struct xfs_defer_op_type	*ops;
 
-	trace_xfs_defer_trans_abort(tp, _RET_IP_);
-
 	/* Abort intent items that don't have a done item. */
-	list_for_each_entry(dfp, dop_pending, dfp_list) {
+	list_for_each_entry(dfp, dop_list, dfp_list) {
 		ops = defer_op_types[dfp->dfp_type];
-		trace_xfs_defer_pending_abort(tp->t_mountp, dfp);
+		trace_xfs_defer_pending_abort(mp, dfp);
 		if (dfp->dfp_intent && !dfp->dfp_done) {
 			ops->abort_intent(dfp->dfp_intent);
 			dfp->dfp_intent = NULL;

@@ -267,6 +264,16 @@ xfs_defer_trans_abort(
 	}
 }
 
+/* Abort all the intents that were committed. */
+STATIC void
+xfs_defer_trans_abort(
+	struct xfs_trans		*tp,
+	struct list_head		*dop_pending)
+{
+	trace_xfs_defer_trans_abort(tp, _RET_IP_);
+	xfs_defer_pending_abort(tp->t_mountp, dop_pending);
+}
+
 /*
  * Capture resources that the caller said not to release ("held") when the
  * transaction commits.  Caller is responsible for zero-initializing @dres.

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Long Li <[email protected]>

[ Upstream commit f8f9d952e42dd49ae534f61f2fa7ca0876cb9848 ]

When recovering intents, we capture newly created intent items as part of committing recovered intent items. If intent recovery fails at a later point, we forget to remove those newly created intent items from the AIL and hang:

    [root@localhost ~]# cat /proc/539/stack
    [<0>] xfs_ail_push_all_sync+0x174/0x230
    [<0>] xfs_unmount_flush_inodes+0x8d/0xd0
    [<0>] xfs_mountfs+0x15f7/0x1e70
    [<0>] xfs_fs_fill_super+0x10ec/0x1b20
    [<0>] get_tree_bdev+0x3c8/0x730
    [<0>] vfs_get_tree+0x89/0x2c0
    [<0>] path_mount+0xecf/0x1800
    [<0>] do_mount+0xf3/0x110
    [<0>] __x64_sys_mount+0x154/0x1f0
    [<0>] do_syscall_64+0x39/0x80
    [<0>] entry_SYSCALL_64_after_hwframe+0x63/0xcd

When newly created intent items fail to commit via transaction, intent recovery hasn't created done items for these newly created intent items, so the capture structure is the sole owner of the captured intent items. We must release them explicitly or else they leak:

unreferenced object 0xffff888016719108 (size 432):
comm "mount", pid 529, jiffies 4294706839 (age 144.463s) hex dump (first 32 bytes):

    08 91 71 16 80 88 ff ff 08 91 71 16 80 88 ff ff  ..q.......q.....
    18 91 71 16 80 88 ff ff 18 91 71 16 80 88 ff ff  ..q.......q.....
  backtrace:
    [<ffffffff8230c68f>] xfs_efi_init+0x18f/0x1d0
    [<ffffffff8230c720>] xfs_extent_free_create_intent+0x50/0x150
    [<ffffffff821b671a>] xfs_defer_create_intents+0x16a/0x340
    [<ffffffff821bac3e>] xfs_defer_ops_capture_and_commit+0x8e/0xad0
    [<ffffffff82322bb9>] xfs_cui_item_recover+0x819/0x980
    [<ffffffff823289b6>] xlog_recover_process_intents+0x246/0xb70
    [<ffffffff8233249a>] xlog_recover_finish+0x8a/0x9a0
    [<ffffffff822eeafb>] xfs_log_mount_finish+0x2bb/0x4a0
    [<ffffffff822c0f4f>] xfs_mountfs+0x14bf/0x1e70
    [<ffffffff822d1f80>] xfs_fs_fill_super+0x10d0/0x1b20
    [<ffffffff81a21fa2>] get_tree_bdev+0x3d2/0x6d0
    [<ffffffff81a1ee09>] vfs_get_tree+0x89/0x2c0
    [<ffffffff81a9f35f>] path_mount+0xecf/0x1800
    [<ffffffff81a9fd83>] do_mount+0xf3/0x110
    [<ffffffff81aa00e4>] __x64_sys_mount+0x154/0x1f0
    [<ffffffff83968739>] do_syscall_64+0x39/0x80

Fix the problem above by abort intent items that don't have a done item when recovery intents fail.

---
 fs/xfs/libxfs/xfs_defer.c |    5 +++--
 fs/xfs/libxfs/xfs_defer.h |    2 +-
 fs/xfs/xfs_log_recover.c  |    2 +-
 3 files changed, 5 insertions(+), 4 deletions(-)

@@ -761,12 +761,13 @@ xfs_defer_ops_capture(
 
 /* Release all resources that we used to capture deferred ops. */
 void
-xfs_defer_ops_capture_free(
+xfs_defer_ops_capture_abort(
 	struct xfs_mount		*mp,
 	struct xfs_defer_capture	*dfc)
 {
 	unsigned short			i;
 
+	xfs_defer_pending_abort(mp, &dfc->dfc_dfops);
 	xfs_defer_cancel_list(mp, &dfc->dfc_dfops);
 
 	for (i = 0; i < dfc->dfc_held.dr_bufs; i++)

@@ -807,7 +808,7 @@ xfs_defer_ops_capture_and_commit(
 	/* Commit the transaction and add the capture structure to the list. */
 	error = xfs_trans_commit(tp);
 	if (error) {
-		xfs_defer_ops_capture_free(mp, dfc);
+		xfs_defer_ops_capture_abort(mp, dfc);
 		return error;
 	}
 

@@ -121,7 +121,7 @@ int xfs_defer_ops_capture_and_commit(str
 		struct list_head *capture_list);
 void xfs_defer_ops_continue(struct xfs_defer_capture *d, struct xfs_trans *tp,
 		struct xfs_defer_resources *dres);
-void xfs_defer_ops_capture_free(struct xfs_mount *mp,
+void xfs_defer_ops_capture_abort(struct xfs_mount *mp,
 		struct xfs_defer_capture *d);
 void xfs_defer_resources_rele(struct xfs_defer_resources *dres);
 

@@ -2514,7 +2514,7 @@ xlog_abort_defer_ops(
 
 	list_for_each_entry_safe(dfc, next, capture_list, dfc_list) {
 		list_del_init(&dfc->dfc_list);
-		xfs_defer_ops_capture_free(mp, dfc);
+		xfs_defer_ops_capture_abort(mp, dfc);
 	}
 }
 

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christoph Hellwig <[email protected]>

[ Upstream commit 55f669f34184ecb25b8353f29c7f6f1ae5b313d1 ]

xfs_reflink_end_cow_extent looks up the COW extent and the data fork extent at offset_fsb, and then proceeds to remap the common subset between the two.

It does however not limit the remapped extent to the passed in [*offset_fsbm end_fsb] range and thus potentially remaps more blocks than the one handled by the current I/O completion. This means that with sufficiently large data and COW extents we could be remapping COW fork mappings that have not been written to, leading to a stale data exposure on a powerfail event.

We use to have a xfs_trim_range to make the remap fit the I/O completion range, but that got (apparently accidentally) removed in commit df2fd88f8ac7 ("xfs: rewrite xfs_reflink_end_cow to use intents").

Note that I've only found this by code inspection, and a test case would probably require very specific delay and error injection.

---
 fs/xfs/xfs_reflink.c |    1 +
 1 file changed, 1 insertion(+)

@@ -783,6 +783,7 @@ xfs_reflink_end_cow_extent(
 		}
 	}
 	del = got;
+	xfs_trim_extent(&del, *offset_fsb, end_fsb - *offset_fsb);
 
 	/* Grab the corresponding mapping in the data fork. */
 	nmaps = 1;

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Leah Rumancik <[email protected]>

[ Upstream commit 471de20303dda0b67981e06d59cc6c4a83fd2a3c ]

We flush the data device cache before we issue external log IO. If the flush fails, we shut down the log immediately and return. However, the iclog->ic_sema is left in a decremented state so let's add an up(). Prior to this patch, xfs/438 would fail consistently when running with an external log device:

sync
-> xfs_log_force
-> xlog_write_iclog

      -> down(&iclog->ic_sema)
      -> blkdev_issue_flush (fail causes us to intiate shutdown)
          -> xlog_force_shutdown
          -> return

unmount
-> xfs_log_umount

      -> xlog_wait_iclog_completion
          -> down(&iclog->ic_sema) --------> HANG

There is a second early return / shutdown. Make sure the up() happens for it as well. Also make sure we cleanup the iclog state, xlog_state_done_syncing, before dropping the iclog lock.

---
 fs/xfs/xfs_log.c |   23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

@@ -1891,9 +1891,7 @@ xlog_write_iclog(
 		 * the buffer manually, the code needs to be kept in sync
 		 * with the I/O completion path.
 		 */
-		xlog_state_done_syncing(iclog);
-		up(&iclog->ic_sema);
-		return;
+		goto sync;
 	}
 
 	/*

@@ -1923,20 +1921,17 @@ xlog_write_iclog(
 		 * avoid shutdown re-entering this path and erroring out again.
 		 */
 		if (log->l_targ != log->l_mp->m_ddev_targp &&
-		    blkdev_issue_flush(log->l_mp->m_ddev_targp->bt_bdev)) {
-			xlog_force_shutdown(log, SHUTDOWN_LOG_IO_ERROR);
-			return;
-		}
+		    blkdev_issue_flush(log->l_mp->m_ddev_targp->bt_bdev))
+			goto shutdown;
 	}
 	if (iclog->ic_flags & XLOG_ICL_NEED_FUA)
 		iclog->ic_bio.bi_opf |= REQ_FUA;
 
 	iclog->ic_flags &= ~(XLOG_ICL_NEED_FLUSH | XLOG_ICL_NEED_FUA);
 
-	if (xlog_map_iclog_data(&iclog->ic_bio, iclog->ic_data, count)) {
-		xlog_force_shutdown(log, SHUTDOWN_LOG_IO_ERROR);
-		return;
-	}
+	if (xlog_map_iclog_data(&iclog->ic_bio, iclog->ic_data, count))
+		goto shutdown;
+
 	if (is_vmalloc_addr(iclog->ic_data))
 		flush_kernel_vmap_range(iclog->ic_data, count);
 

@@ -1957,6 +1952,12 @@ xlog_write_iclog(
 	}
 
 	submit_bio(&iclog->ic_bio);
+	return;
+shutdown:
+	xlog_force_shutdown(log, SHUTDOWN_LOG_IO_ERROR);
+sync:
+	xlog_state_done_syncing(iclog);
+	up(&iclog->ic_sema);
 }
 
 /*

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Omar Sandoval <[email protected]>

[ Upstream commit f63a5b3769ad7659da4c0420751d78958ab97675 ]

We've been seeing XFS errors like the following:

XFS: Internal error i != 1 at line 3526 of file fs/xfs/libxfs/xfs_btree.c. Caller xfs_btree_insert+0x1ec/0x280
...

Call Trace:
 xfs_corruption_error+0x94/0xa0
 xfs_btree_insert+0x221/0x280
 xfs_alloc_fixup_trees+0x104/0x3e0
 xfs_alloc_ag_vextent_size+0x667/0x820
 xfs_alloc_fix_freelist+0x5d9/0x750
 xfs_free_extent_fix_freelist+0x65/0xa0
 __xfs_free_extent+0x57/0x180

...

This is the XFS_IS_CORRUPT() check in xfs_btree_insert() when xfs_btree_insrec() fails.

After converting this into a panic and dissecting the core dump, I found that xfs_btree_insrec() is failing because it's trying to split a leaf node in the cntbt when the AG free list is empty. In particular, it's failing to get a block from the AGFL _while trying to refill the AGFL_.

If a single operation splits every level of the bnobt and the cntbt (and the rmapbt if it is enabled) at once, the free list will be empty. Then, when the next operation tries to refill the free list, it allocates space. If the allocation does not use a full extent, it will need to insert records for the remaining space in the bnobt and cntbt. And if those new records go in full leaves, the leaves (and potentially more nodes up to the old root) need to be split.

Fix it by accounting for the additional splits that may be required to refill the free list in the calculation for the minimum free list size.

P.S. As far as I can tell, this bug has existed for a long time -- maybe back to xfs-history commit afdf80ae7405 ("Add XFS_AG_MAXLEVELS macros ...") in April 1994! It requires a very unlucky sequence of events, and in fact we didn't hit it until a particular sparse mmap workload updated from 5.12 to 5.19. But this bug existed in 5.12, so it must've been exposed by some other change in allocation or writeback patterns. It's also much less likely to be hit with the rmapbt enabled, since that increases the minimum free list size and is unlikely to split at the same time as the bnobt and cntbt.

---
 fs/xfs/libxfs/xfs_alloc.c |   27 ++++++++++++++++++++++++---
 1 file changed, 24 insertions(+), 3 deletions(-)

@@ -2273,16 +2273,37 @@ xfs_alloc_min_freelist(
 
 	ASSERT(mp->m_alloc_maxlevels > 0);
 
+	/*
+	 * For a btree shorter than the maximum height, the worst case is that
+	 * every level gets split and a new level is added, then while inserting
+	 * another entry to refill the AGFL, every level under the old root gets
+	 * split again. This is:
+	 *
+	 *   (full height split reservation) + (AGFL refill split height)
+	 * = (current height + 1) + (current height - 1)
+	 * = (new height) + (new height - 2)
+	 * = 2 * new height - 2
+	 *
+	 * For a btree of maximum height, the worst case is that every level
+	 * under the root gets split, then while inserting another entry to
+	 * refill the AGFL, every level under the root gets split again. This is
+	 * also:
+	 *
+	 *   2 * (current height - 1)
+	 * = 2 * (new height - 1)
+	 * = 2 * new height - 2
+	 */
+
 	/* space needed by-bno freespace btree */
 	min_free = min_t(unsigned int, levels[XFS_BTNUM_BNOi] + 1,
-				       mp->m_alloc_maxlevels);
+				       mp->m_alloc_maxlevels) * 2 - 2;
 	/* space needed by-size freespace btree */
 	min_free += min_t(unsigned int, levels[XFS_BTNUM_CNTi] + 1,
-				       mp->m_alloc_maxlevels);
+				       mp->m_alloc_maxlevels) * 2 - 2;
 	/* space needed reverse mapping used space btree */
 	if (xfs_has_rmapbt(mp))
 		min_free += min_t(unsigned int, levels[XFS_BTNUM_RMAPi] + 1,
-						mp->m_rmap_maxlevels);
+						mp->m_rmap_maxlevels) * 2 - 2;
 
 	return min_free;
 }

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dave Chinner <[email protected]>

[ Upstream commit 038ca189c0d2c1570b4d922f25b524007c85cf94 ]

Discovered when trying to track down a weird recovery corruption issue that wasn't detected at recovery time.

The specific corruption was a zero extent count field when big extent counts are in use, and it turns out the dinode verifier doesn't detect that specific corruption case, either. So fix it too.

---
 fs/xfs/libxfs/xfs_inode_buf.c   |    3 +++
 fs/xfs/xfs_inode_item_recover.c |   14 +++++++++++++-
 2 files changed, 16 insertions(+), 1 deletion(-)

@@ -507,6 +507,9 @@ xfs_dinode_verify(
 	if (mode && nextents + naextents > nblocks)
 		return __this_address;
 
+	if (nextents + naextents == 0 && nblocks != 0)
+		return __this_address;
+
 	if (S_ISDIR(mode) && nextents > mp->m_dir_geo->max_extents)
 		return __this_address;
 

@@ -286,6 +286,7 @@ xlog_recover_inode_commit_pass2(
 	struct xfs_log_dinode		*ldip;
 	uint				isize;
 	int				need_free = 0;
+	xfs_failaddr_t			fa;
 
 	if (item->ri_buf[0].i_len == sizeof(struct xfs_inode_log_format)) {
 		in_f = item->ri_buf[0].i_addr;

@@ -530,8 +531,19 @@ out_owner_change:
 	    (dip->di_mode != 0))
 		error = xfs_recover_inode_owner_change(mp, dip, in_f,
 						       buffer_list);
-	/* re-generate the checksum. */
+	/* re-generate the checksum and validate the recovered inode. */
 	xfs_dinode_calc_crc(log->l_mp, dip);
+	fa = xfs_dinode_verify(log->l_mp, in_f->ilf_ino, dip);
+	if (fa) {
+		XFS_CORRUPTION_ERROR(
+			"Bad dinode after recovery",
+				XFS_ERRLEVEL_LOW, mp, dip, sizeof(*dip));
+		xfs_alert(mp,
+			"Metadata corruption detected at %pS, inode 0x%llx",
+			fa, in_f->ilf_ino);
+		error = -EFSCORRUPTED;
+		goto out_release;
+	}
 
 	ASSERT(bp->b_mount == mp);
 	bp->b_flags |= _XBF_LOGRECOVERY;

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit ed17f7da5f0c8b65b7b5f7c98beb0aadbc0546ee ]

Since the introduction of xfs_dqblk in V5, xfs really ought to find the dqblk pointer from the dquot buffer, then compute the xfs_disk_dquot pointer from the dqblk pointer. Fix the open-coded xfs_buf_offset calls and do the type checking in the correct order.

Note that this has made no practical difference since the start of the xfs_disk_dquot is coincident with the start of the xfs_dqblk.

---
 fs/xfs/xfs_dquot.c              |    5 +++--
 fs/xfs/xfs_dquot_item_recover.c |    7 ++++---
 2 files changed, 7 insertions(+), 5 deletions(-)

@@ -562,7 +562,8 @@ xfs_dquot_from_disk(
 	struct xfs_dquot	*dqp,
 	struct xfs_buf		*bp)
 {
-	struct xfs_disk_dquot	*ddqp = bp->b_addr + dqp->q_bufoffset;
+	struct xfs_dqblk	*dqb = xfs_buf_offset(bp, dqp->q_bufoffset);
+	struct xfs_disk_dquot	*ddqp = &dqb->dd_diskdq;
 
 	/*
 	 * Ensure that we got the type and ID we were looking for.

@@ -1250,7 +1251,7 @@ xfs_qm_dqflush(
 	}
 
 	/* Flush the incore dquot to the ondisk buffer. */
-	dqblk = bp->b_addr + dqp->q_bufoffset;
+	dqblk = xfs_buf_offset(bp, dqp->q_bufoffset);
 	xfs_dquot_to_disk(&dqblk->dd_diskdq, dqp);
 
 	/*

@@ -65,6 +65,7 @@ xlog_recover_dquot_commit_pass2(
 {
 	struct xfs_mount		*mp = log->l_mp;
 	struct xfs_buf			*bp;
+	struct xfs_dqblk		*dqb;
 	struct xfs_disk_dquot		*ddq, *recddq;
 	struct xfs_dq_logformat		*dq_f;
 	xfs_failaddr_t			fa;

@@ -130,14 +131,14 @@ xlog_recover_dquot_commit_pass2(
 		return error;
 
 	ASSERT(bp);
-	ddq = xfs_buf_offset(bp, dq_f->qlf_boffset);
+	dqb = xfs_buf_offset(bp, dq_f->qlf_boffset);
+	ddq = &dqb->dd_diskdq;
 
 	/*
 	 * If the dquot has an LSN in it, recover the dquot only if it's less
 	 * than the lsn of the transaction we are replaying.
 	 */
 	if (xfs_has_crc(mp)) {
-		struct xfs_dqblk *dqb = (struct xfs_dqblk *)ddq;
 		xfs_lsn_t	lsn = be64_to_cpu(dqb->dd_lsn);
 
 		if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0) {

@@ -147,7 +148,7 @@ xlog_recover_dquot_commit_pass2(
 
 	memcpy(ddq, recddq, item->ri_buf[1].i_len);
 	if (xfs_has_crc(mp)) {
-		xfs_update_cksum((char *)ddq, sizeof(struct xfs_dqblk),
+		xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk),
 				 XFS_DQUOT_CRC_OFF);
 	}
 

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit 9c235dfc3d3f901fe22acb20f2ab37ff39f2ce02 ]

When we're recovering ondisk quota records from the log, we need to validate the recovered buffer contents before writing them to disk.

---
 fs/xfs/xfs_dquot_item_recover.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

@@ -19,6 +19,7 @@
 #include "xfs_log.h"
 #include "xfs_log_priv.h"
 #include "xfs_log_recover.h"
+#include "xfs_error.h"
 
 STATIC void
 xlog_recover_dquot_ra_pass2(

@@ -152,6 +153,19 @@ xlog_recover_dquot_commit_pass2(
 				 XFS_DQUOT_CRC_OFF);
 	}
 
+	/* Validate the recovered dquot. */
+	fa = xfs_dqblk_verify(log->l_mp, dqb, dq_f->qlf_id);
+	if (fa) {
+		XFS_CORRUPTION_ERROR("Bad dquot after recovery",
+				XFS_ERRLEVEL_LOW, mp, dqb,
+				sizeof(struct xfs_dqblk));
+		xfs_alert(mp,
+ "Metadata corruption detected at %pS, dquot 0x%x",
+				fa, dq_f->qlf_id);
+		error = -EFSCORRUPTED;
+		goto out_release;
+	}
+
 	ASSERT(dq_f->qlf_size == 2);
 	ASSERT(bp->b_mount == mp);
 	bp->b_flags |= _XBF_LOGRECOVERY;

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christoph Hellwig <[email protected]>

[ Upstream commit c421df0b19430417a04f68919fc3d1943d20ac04 ]

Introduce a local boolean variable if FS_XFLAG_REALTIME to make the checks for it more obvious, and de-densify a few of the conditionals using it to make them more readable while at it.

---
 fs/xfs/xfs_ioctl.c |   22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

@@ -1120,23 +1120,25 @@ xfs_ioctl_setattr_xflags(
 	struct fileattr		*fa)
 {
 	struct xfs_mount	*mp = ip->i_mount;
+	bool			rtflag = (fa->fsx_xflags & FS_XFLAG_REALTIME);
 	uint64_t		i_flags2;
 
-	/* Can't change realtime flag if any extents are allocated. */
-	if ((ip->i_df.if_nextents || ip->i_delayed_blks) &&
-	    XFS_IS_REALTIME_INODE(ip) != (fa->fsx_xflags & FS_XFLAG_REALTIME))
-		return -EINVAL;
+	if (rtflag != XFS_IS_REALTIME_INODE(ip)) {
+		/* Can't change realtime flag if any extents are allocated. */
+		if (ip->i_df.if_nextents || ip->i_delayed_blks)
+			return -EINVAL;
+	}
 
-	/* If realtime flag is set then must have realtime device */
-	if (fa->fsx_xflags & FS_XFLAG_REALTIME) {
+	if (rtflag) {
+		/* If realtime flag is set then must have realtime device */
 		if (mp->m_sb.sb_rblocks == 0 || mp->m_sb.sb_rextsize == 0 ||
 		    (ip->i_extsize % mp->m_sb.sb_rextsize))
 			return -EINVAL;
-	}
 
-	/* Clear reflink if we are actually able to set the rt flag. */
-	if ((fa->fsx_xflags & FS_XFLAG_REALTIME) && xfs_is_reflink_inode(ip))
-		ip->i_diflags2 &= ~XFS_DIFLAG2_REFLINK;
+		/* Clear reflink if we are actually able to set the rt flag. */
+		if (xfs_is_reflink_inode(ip))
+			ip->i_diflags2 &= ~XFS_DIFLAG2_REFLINK;
+	}
 
 	/* Don't allow us to set DAX mode for a reflinked file for now. */
 	if ((fa->fsx_xflags & FS_XFLAG_DAX) && xfs_is_reflink_inode(ip))

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christoph Hellwig <[email protected]>

[ Upstream commit 9c04138414c00ae61421f36ada002712c4bac94a ]

Update the per-folio stable writes flag dependening on which device an inode resides on.

---
 fs/xfs/xfs_inode.h |    8 ++++++++
 fs/xfs/xfs_ioctl.c |    8 ++++++++
 fs/xfs/xfs_iops.c  |    7 +++++++
 3 files changed, 23 insertions(+)

@@ -569,6 +569,14 @@ extern void xfs_setup_inode(struct xfs_i
 extern void xfs_setup_iops(struct xfs_inode *ip);
 extern void xfs_diflags_to_iflags(struct xfs_inode *ip, bool init);
 
+static inline void xfs_update_stable_writes(struct xfs_inode *ip)
+{
+	if (bdev_stable_writes(xfs_inode_buftarg(ip)->bt_bdev))
+		mapping_set_stable_writes(VFS_I(ip)->i_mapping);
+	else
+		mapping_clear_stable_writes(VFS_I(ip)->i_mapping);
+}
+
 /*
  * When setting up a newly allocated inode, we need to call
  * xfs_finish_inode_setup() once the inode is fully instantiated at

@@ -1153,6 +1153,14 @@ xfs_ioctl_setattr_xflags(
 	ip->i_diflags2 = i_flags2;
 
 	xfs_diflags_to_iflags(ip, false);
+
+	/*
+	 * Make the stable writes flag match that of the device the inode
+	 * resides on when flipping the RT flag.
+	 */
+	if (rtflag != XFS_IS_REALTIME_INODE(ip) && S_ISREG(VFS_I(ip)->i_mode))
+		xfs_update_stable_writes(ip);
+
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 	XFS_STATS_INC(mp, xs_ig_attrchg);

@@ -1292,6 +1292,13 @@ xfs_setup_inode(
 	mapping_set_gfp_mask(inode->i_mapping, (gfp_mask & ~(__GFP_FS)));
 
 	/*
+	 * For real-time inodes update the stable write flags to that of the RT
+	 * device instead of the data device.
+	 */
+	if (S_ISREG(inode->i_mode) && XFS_IS_REALTIME_INODE(ip))
+		xfs_update_stable_writes(ip);
+
+	/*
 	 * If there is no attribute fork no ACL can exist on this inode,
 	 * and it can't have any file capabilities attached to it either.
 	 */

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andreas Gruenbacher <[email protected]>

commit 7c9d9223802fbed4dee1ae301661bf346964c9d2 upstream.

Truncate an inode's address space when flipping the GFS2_DIF_JDATA flag:
depending on that flag, the pages in the address space will either use buffer heads or iomap_folio_state structs, and we cannot mix the two.

---
 fs/gfs2/file.c |    1 +
 1 file changed, 1 insertion(+)

@@ -250,6 +250,7 @@ static int do_gfs2_set_flags(struct inod
 		error = filemap_fdatawait(inode->i_mapping);
 		if (error)
 			goto out;
+		truncate_inode_pages(inode->i_mapping, 0);
 		if (new_flags & GFS2_DIF_JDATA)
 			gfs2_ordered_del_inode(ip);
 	}

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pavel Begunkov <[email protected]>

There are reports of mariadb hangs, which is caused by a missing barrier in the waking code resulting in waiters losing events.

The problem was introduced in a backport
3ab9326f93ec4 ("io_uring: wake up optimisations"),
and the change restores the barrier present in the original commit 3ab9326f93ec4 ("io_uring: wake up optimisations")

Reported by: Xan Charbonnet <[email protected]>

---
 io_uring/io_uring.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -597,8 +597,10 @@ static inline void __io_cq_unlock_post_f
 	io_commit_cqring(ctx);
 	spin_unlock(&ctx->completion_lock);
 	io_commit_cqring_flush(ctx);
-	if (!(ctx->flags & IORING_SETUP_DEFER_TASKRUN))
+	if (!(ctx->flags & IORING_SETUP_DEFER_TASKRUN)) {
+		smp_mb();
 		__io_cqring_wake(ctx);
+	}
 }
 
 void io_cq_unlock_post(struct io_ring_ctx *ctx)

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jamal Hadi Salim <[email protected]>

commit d62b04fca4340a0d468d7853bd66e511935a18cb upstream.

Haowei Yan <[email protected]> found that ets_class_from_arg() can index an Out-Of-Bound class in ets_class_from_arg() when passed clid of
0. The overflow may cause local privilege escalation.

 [   18.852298] ------------[ cut here ]------------
 [   18.853271] UBSAN: array-index-out-of-bounds in net/sched/sch_ets.c:93:20
 [   18.853743] index 18446744073709551615 is out of range for type 'ets_class [16]'
 [   18.854254] CPU: 0 UID: 0 PID: 1275 Comm: poc Not tainted 6.12.6-dirty #17
 [   18.854821] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
 [   18.856532] Call Trace:
 [   18.857441]  <TASK>
 [   18.858227]  dump_stack_lvl+0xc2/0xf0
 [   18.859607]  dump_stack+0x10/0x20
 [   18.860908]  __ubsan_handle_out_of_bounds+0xa7/0xf0
 [   18.864022]  ets_class_change+0x3d6/0x3f0
 [   18.864322]  tc_ctl_tclass+0x251/0x910
 [   18.864587]  ? lock_acquire+0x5e/0x140
 [   18.865113]  ? __mutex_lock+0x9c/0xe70
 [   18.866009]  ? __mutex_lock+0xa34/0xe70
 [   18.866401]  rtnetlink_rcv_msg+0x170/0x6f0
 [   18.866806]  ? __lock_acquire+0x578/0xc10
 [   18.867184]  ? __pfx_rtnetlink_rcv_msg+0x10/0x10
 [   18.867503]  netlink_rcv_skb+0x59/0x110
 [   18.867776]  rtnetlink_rcv+0x15/0x30
 [   18.868159]  netlink_unicast+0x1c3/0x2b0
 [   18.868440]  netlink_sendmsg+0x239/0x4b0
 [   18.868721]  ____sys_sendmsg+0x3e2/0x410
 [   18.869012]  ___sys_sendmsg+0x88/0xe0
 [   18.869276]  ? rseq_ip_fixup+0x198/0x260
 [   18.869563]  ? rseq_update_cpu_node_id+0x10a/0x190
 [   18.869900]  ? trace_hardirqs_off+0x5a/0xd0
 [   18.870196]  ? syscall_exit_to_user_mode+0xcc/0x220
 [   18.870547]  ? do_syscall_64+0x93/0x150
 [   18.870821]  ? __memcg_slab_free_hook+0x69/0x290
 [   18.871157]  __sys_sendmsg+0x69/0xd0
 [   18.871416]  __x64_sys_sendmsg+0x1d/0x30
 [   18.871699]  x64_sys_call+0x9e2/0x2670
 [   18.871979]  do_syscall_64+0x87/0x150
 [   18.873280]  ? do_syscall_64+0x93/0x150
 [   18.874742]  ? lock_release+0x7b/0x160
 [   18.876157]  ? do_user_addr_fault+0x5ce/0x8f0
 [   18.877833]  ? irqentry_exit_to_user_mode+0xc2/0x210
 [   18.879608]  ? irqentry_exit+0x77/0xb0
 [   18.879808]  ? clear_bhb_loop+0x15/0x70
 [   18.880023]  ? clear_bhb_loop+0x15/0x70
 [   18.880223]  ? clear_bhb_loop+0x15/0x70
 [   18.880426]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
 [   18.880683] RIP: 0033:0x44a957
 [   18.880851] Code: ff ff e8 fc 00 00 00 66 2e 0f 1f 84 00 00 00 00 00 66 90 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 8974 24 10
 [   18.881766] RSP: 002b:00007ffcdd00fad8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
 [   18.882149] RAX: ffffffffffffffda RBX: 00007ffcdd010db8 RCX: 000000000044a957
 [   18.882507] RDX: 0000000000000000 RSI: 00007ffcdd00fb70 RDI: 0000000000000003
 [   18.885037] RBP: 00007ffcdd010bc0 R08: 000000000703c770 R09: 000000000703c7c0
 [   18.887203] R10: 0000000000000080 R11: 0000000000000246 R12: 0000000000000001
 [   18.888026] R13: 00007ffcdd010da8 R14: 00000000004ca7d0 R15: 0000000000000001
 [   18.888395]  </TASK>
 [   18.888610] ---[ end trace ]---

---
 net/sched/sch_ets.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -91,6 +91,8 @@ ets_class_from_arg(struct Qdisc *sch, un
 {
 	struct ets_sched *q = qdisc_priv(sch);
 
+	if (arg == 0 || arg > q->nbands)
+		return NULL;
 	return &q->classes[arg - 1];
 }
 

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexey Dobriyan <[email protected]>

commit 697ba0b6ec4ae04afb67d3911799b5e2043b4455 upstream.

I independently rediscovered

	commit 22d24a544b0d49bbcbd61c8c0eaf77d3c9297155
	block: fix overflow in blk_ioctl_discard()

but for secure erase.

Same problem:

	uint64_t r[2] = {512, 18446744073709551104ULL};
	ioctl(fd, BLKSECDISCARD, r);

will enter near infinite loop inside blkdev_issue_secure_erase():

a.out: attempt to access beyond end of device
loop0: rw=5, sector=3399043073, nr_sectors = 1024 limit=2048
bio_check_eod: 3286214 callbacks suppressed

---
 block/ioctl.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

@@ -115,7 +115,7 @@ static int blk_ioctl_discard(struct bloc
 		return -EINVAL;
 
 	filemap_invalidate_lock(inode->i_mapping);
-	err = truncate_bdev_range(bdev, mode, start, start + len - 1);
+	err = truncate_bdev_range(bdev, mode, start, end - 1);
 	if (err)
 		goto fail;
 	err = blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_KERNEL);

@@ -127,7 +127,7 @@ fail:
 static int blk_ioctl_secure_erase(struct block_device *bdev, fmode_t mode,
 		void __user *argp)
 {
-	uint64_t start, len;
+	uint64_t start, len, end;
 	uint64_t range[2];
 	int err;
 

@@ -142,11 +142,12 @@ static int blk_ioctl_secure_erase(struct
 	len = range[1];
 	if ((start & 511) || (len & 511))
 		return -EINVAL;
-	if (start + len > bdev_nr_bytes(bdev))
+	if (check_add_overflow(start, len, &end) ||
+	    end > bdev_nr_bytes(bdev))
 		return -EINVAL;
 
 	filemap_invalidate_lock(bdev->bd_inode->i_mapping);
-	err = truncate_bdev_range(bdev, mode, start, start + len - 1);
+	err = truncate_bdev_range(bdev, mode, start, end - 1);
 	if (!err)
 		err = blkdev_issue_secure_erase(bdev, start >> 9, len >> 9,
 						GFP_KERNEL);

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiri Kosina <[email protected]>

commit 3d88ba86ba6f35a0467f25a88c38aa5639190d04 upstream.

This reverts commit 251efae73bd46b097deec4f9986d926813aed744.

Quoting Wang Yuli:

"The 27C6:01E0 touchpad doesn't require the workaround and applying it would actually break functionality.

The initial report came from a BBS forum, but we suspect the information provided by the forum user may be incorrect which could happen sometimes. [1]

Further investigation showed that the Lenovo Y9000P 2024 doesn't even use a Goodix touchpad. [2]

For the broader issue of 27c6:01e0 being unusable on some devices, it just need to address it with a libinput quirk.

In conclusion, we should revert this commit, which is the best solution."

---
 drivers/hid/hid-ids.h        |    1 -
 drivers/hid/hid-multitouch.c |    8 ++------
 2 files changed, 2 insertions(+), 7 deletions(-)

@@ -507,7 +507,6 @@
 #define USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_E100 0xe100
 
 #define I2C_VENDOR_ID_GOODIX		0x27c6
-#define I2C_DEVICE_ID_GOODIX_01E0	0x01e0
 #define I2C_DEVICE_ID_GOODIX_01E8	0x01e8
 #define I2C_DEVICE_ID_GOODIX_01E9	0x01e9
 #define I2C_DEVICE_ID_GOODIX_01F0	0x01f0

@@ -1447,8 +1447,7 @@ static __u8 *mt_report_fixup(struct hid_
 {
 	if (hdev->vendor == I2C_VENDOR_ID_GOODIX &&
 	    (hdev->product == I2C_DEVICE_ID_GOODIX_01E8 ||
-	     hdev->product == I2C_DEVICE_ID_GOODIX_01E9 ||
-		 hdev->product == I2C_DEVICE_ID_GOODIX_01E0)) {
+	     hdev->product == I2C_DEVICE_ID_GOODIX_01E9)) {
 		if (rdesc[607] == 0x15) {
 			rdesc[607] = 0x25;
 			dev_info(

@@ -2073,10 +2072,7 @@ static const struct hid_device_id mt_dev
 		     I2C_DEVICE_ID_GOODIX_01E8) },
 	{ .driver_data = MT_CLS_WIN_8_FORCE_MULTI_INPUT_NSMU,
 	  HID_DEVICE(BUS_I2C, HID_GROUP_ANY, I2C_VENDOR_ID_GOODIX,
-		     I2C_DEVICE_ID_GOODIX_01E9) },
-	{ .driver_data = MT_CLS_WIN_8_FORCE_MULTI_INPUT_NSMU,
-	  HID_DEVICE(BUS_I2C, HID_GROUP_ANY, I2C_VENDOR_ID_GOODIX,
-		     I2C_DEVICE_ID_GOODIX_01E0) },
+		     I2C_DEVICE_ID_GOODIX_01E8) },
 
 	/* GoodTouch panels */
 	{ .driver_data = MT_CLS_NSMU,

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Williamson <[email protected]>

commit ce9ff21ea89d191e477a02ad7eabf4f996b80a69 upstream.

count and offset are passed from user space and not checked, only offset is capped to 40 bits, which can be used to read/write out of bounds of the device.

---
 drivers/vfio/platform/vfio_platform_common.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

@@ -391,6 +391,11 @@ static ssize_t vfio_platform_read_mmio(s
 {
 	unsigned int done = 0;
 
+	if (off >= reg->size)
+		return -EINVAL;
+
+	count = min_t(size_t, count, reg->size - off);
+
 	if (!reg->ioaddr) {
 		reg->ioaddr =
 			ioremap(reg->addr, reg->size);

@@ -470,6 +475,11 @@ static ssize_t vfio_platform_write_mmio(
 {
 	unsigned int done = 0;
 
+	if (off >= reg->size)
+		return -EINVAL;
+
+	count = min_t(size_t, count, reg->size - off);
+
 	if (!reg->ioaddr) {
 		reg->ioaddr =
 			ioremap(reg->addr, reg->size);

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luis Henriques (SUSE) <[email protected]>

commit 23dfdb56581ad92a9967bcd720c8c23356af74c1 upstream.

The following kernel trace can be triggered with fstest generic/629 when executed against a filesystem with fast-commit feature enabled:

INFO: trying to register non-static key.
The code is fine but needs lockdep annotation, or maybe you didn't initialize this object before use? turning off the locking correctness validator.

CPU: 0 PID: 866 Comm: mount Not tainted 6.10.0+ #11
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.2-3-gd478f380-prebuilt.qemu.org 04/01/2014
Call Trace:
 <TASK>
 dump_stack_lvl+0x66/0x90
 register_lock_class+0x759/0x7d0
 __lock_acquire+0x85/0x2630
 ? __find_get_block+0xb4/0x380
 lock_acquire+0xd1/0x2d0
 ? __ext4_journal_get_write_access+0xd5/0x160
 _raw_spin_lock+0x33/0x40
 ? __ext4_journal_get_write_access+0xd5/0x160
 __ext4_journal_get_write_access+0xd5/0x160
 ext4_reserve_inode_write+0x61/0xb0
 __ext4_mark_inode_dirty+0x79/0x270
 ? ext4_ext_replay_set_iblocks+0x2f8/0x450
 ext4_ext_replay_set_iblocks+0x330/0x450
 ext4_fc_replay+0x14c8/0x1540
 ? jread+0x88/0x2e0
 ? rcu_is_watching+0x11/0x40
 do_one_pass+0x447/0xd00
 jbd2_journal_recover+0x139/0x1b0
 jbd2_journal_load+0x96/0x390
 ext4_load_and_init_journal+0x253/0xd40
 ext4_fill_super+0x2cc6/0x3180

...

In the replay path there's an attempt to lock sbi->s_bdev_wb_lock in function ext4_check_bdev_write_error(). Unfortunately, at this point this spinlock has not been initialized yet. Moving it's initialization to an earlier point in __ext4_fill_super() fixes this splat.

---
 fs/ext4/super.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -5264,6 +5264,8 @@ static int __ext4_fill_super(struct fs_c
 	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
 	mutex_init(&sbi->s_orphan_lock);
 
+	spin_lock_init(&sbi->s_bdev_wb_lock);
+
 	ext4_fast_commit_init(sb);
 
 	sb->s_root = NULL;

@@ -5514,7 +5516,6 @@ static int __ext4_fill_super(struct fs_c
 	 * Save the original bdev mapping's wb_err value which could be
 	 * used to detect the metadata async write error.
 	 */
-	spin_lock_init(&sbi->s_bdev_wb_lock);
 	errseq_check_and_advance(&sb->s_bdev->bd_inode->i_mapping->wb_err,
 				 &sbi->s_bdev_wb_err);
 	sb->s_bdev->bd_super = sb;

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

commit 90e0569dd3d32f4f4d2ca691d3fa5a8a14a13c12 upstream.

The per-netns IP tunnel hash table is protected by the RTNL mutex and ip_tunnel_find() is only called from the control path where the mutex is taken.

Add a lockdep expression to hlist_for_each_entry_rcu() in ip_tunnel_find() in order to validate that the mutex is held and to silence the suspicious RCU usage warning [1].

[1]
WARNING: suspicious RCU usage 6.12.0-rc3-custom-gd95d9a31aceb #139 Not tainted
-----------------------------
net/ipv4/ip_tunnel.c:221 RCU-list traversed in non-reader section!!

other info that might help us debug this:

rcu_scheduler_active = 2, debug_locks = 1
1 lock held by ip/362:
#0: ffffffff86fc7cb0 (rtnl_mutex){+.+.}-{3:3}, at: rtnetlink_rcv_msg+0x377/0xf60

stack backtrace:
CPU: 12 UID: 0 PID: 362 Comm: ip Not tainted 6.12.0-rc3-custom-gd95d9a31aceb #139
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
Call Trace:
 <TASK>
 dump_stack_lvl+0xba/0x110
 lockdep_rcu_suspicious.cold+0x4f/0xd6
 ip_tunnel_find+0x435/0x4d0
 ip_tunnel_newlink+0x517/0x7a0
 ipgre_newlink+0x14c/0x170
 __rtnl_newlink+0x1173/0x19c0
 rtnl_newlink+0x6c/0xa0
 rtnetlink_rcv_msg+0x3cc/0xf60
 netlink_rcv_skb+0x171/0x450
 netlink_unicast+0x539/0x7f0
 netlink_sendmsg+0x8c1/0xd80
 ____sys_sendmsg+0x8f9/0xc20
 ___sys_sendmsg+0x197/0x1e0
 __sys_sendmsg+0x122/0x1f0
 do_syscall_64+0xbb/0x1d0
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

---
 net/ipv4/ip_tunnel.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -218,7 +218,7 @@ static struct ip_tunnel *ip_tunnel_find(
 	struct ip_tunnel *t = NULL;
 	struct hlist_head *head = ip_bucket(itn, parms);
 
-	hlist_for_each_entry_rcu(t, head, hash_node) {
+	hlist_for_each_entry_rcu(t, head, hash_node, lockdep_rtnl_is_held()) {
 		if (local == t->parms.iph.saddr &&
 		    remote == t->parms.iph.daddr &&
 		    link == READ_ONCE(t->parms.link) &&

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Easwar Hariharan <[email protected]>

commit d2138eab8cde61e0e6f62d0713e45202e8457d6d upstream.

If there's a persistent error in the hypervisor, the SCSI warning for failed I/O can flood the kernel log and max out CPU utilization, preventing troubleshooting from the VM side. Ratelimit the warning so it doesn't DoS the VM.

---
 drivers/scsi/storvsc_drv.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

@@ -171,6 +171,12 @@ do {								\
 		dev_warn(&(dev)->device, fmt, ##__VA_ARGS__);	\
 } while (0)
 
+#define storvsc_log_ratelimited(dev, level, fmt, ...)				\
+do {										\
+	if (do_logging(level))							\
+		dev_warn_ratelimited(&(dev)->device, fmt, ##__VA_ARGS__);	\
+} while (0)
+
 struct vmscsi_request {
 	u16 length;
 	u8 srb_status;

@@ -1168,7 +1174,7 @@ static void storvsc_on_io_completion(str
 		int loglevel = (stor_pkt->vm_srb.cdb[0] == TEST_UNIT_READY) ?
 			STORVSC_LOGGING_WARN : STORVSC_LOGGING_ERROR;
 
-		storvsc_log(device, loglevel,
+		storvsc_log_ratelimited(device, loglevel,
 			"tag#%d cmd 0x%x status: scsi 0x%x srb 0x%x hv 0x%x\n",
 			scsi_cmd_to_rq(request->cmd)->tag,
 			stor_pkt->vm_srb.cdb[0],

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Anjaneyulu <[email protected]>

commit efbe8f81952fe469d38655744627d860879dcde8 upstream.

Validate index before access iwl_rate_mcs to keep rate->index inside the valid boundaries. Use MCS_0_INDEX if index is less than MCS_0_INDEX and MCS_9_INDEX if index is greater then MCS_9_INDEX.

---
 drivers/net/wireless/intel/iwlwifi/dvm/rs.c |    9 ++++++---
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c |    9 ++++++---
 2 files changed, 12 insertions(+), 6 deletions(-)

@@ -2,7 +2,7 @@
 /******************************************************************************
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
- * Copyright (C) 2019 - 2020, 2022 Intel Corporation
+ * Copyright (C) 2019 - 2020, 2022 - 2023 Intel Corporation
  *****************************************************************************/
 #include <linux/kernel.h>
 #include <linux/skbuff.h>

@@ -125,7 +125,7 @@ static int iwl_hwrate_to_plcp_idx(u32 ra
 				return idx;
 	}
 
-	return -1;
+	return IWL_RATE_INVALID;
 }
 
 static void rs_rate_scale_perform(struct iwl_priv *priv,

@@ -3146,7 +3146,10 @@ static ssize_t rs_sta_dbgfs_scale_table_
 	for (i = 0; i < LINK_QUAL_MAX_RETRY_NUM; i++) {
 		index = iwl_hwrate_to_plcp_idx(
 			le32_to_cpu(lq_sta->lq.rs_table[i].rate_n_flags));
-		if (is_legacy(tbl->lq_type)) {
+		if (index == IWL_RATE_INVALID) {
+			desc += sprintf(buff + desc, " rate[%d] 0x%X invalid rate\n",
+				i, le32_to_cpu(lq_sta->lq.rs_table[i].rate_n_flags));
+		} else if (is_legacy(tbl->lq_type)) {
 			desc += sprintf(buff+desc, " rate[%d] 0x%X %smbps\n",
 				i, le32_to_cpu(lq_sta->lq.rs_table[i].rate_n_flags),
 				iwl_rate_mcs[index].mbps);

@@ -1072,10 +1072,13 @@ static void rs_get_lower_rate_down_colum
 
 		rate->bw = RATE_MCS_CHAN_WIDTH_20;
 
-		WARN_ON_ONCE(rate->index < IWL_RATE_MCS_0_INDEX ||
-			     rate->index > IWL_RATE_MCS_9_INDEX);
+		if (WARN_ON_ONCE(rate->index < IWL_RATE_MCS_0_INDEX))
+			rate->index = rs_ht_to_legacy[IWL_RATE_MCS_0_INDEX];
+		else if (WARN_ON_ONCE(rate->index > IWL_RATE_MCS_9_INDEX))
+			rate->index = rs_ht_to_legacy[IWL_RATE_MCS_9_INDEX];
+		else
+			rate->index = rs_ht_to_legacy[rate->index];
 
-		rate->index = rs_ht_to_legacy[rate->index];
 		rate->ldpc = false;
 	} else {
 		/* Downgrade to SISO with same MCS if in MIMO  */

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Enzo Matsumiya <[email protected]>

commit b0abcd65ec545701b8793e12bc27dc98042b151a upstream.

Doing an async decryption (large read) crashes with a slab-use-after-free way down in the crypto API.

Reproducer:

    # mount.cifs -o ...,seal,esize=1 //srv/share /mnt
    # dd if=/mnt/largefile of=/dev/null
    ...
    [  194.196391] ==================================================================
    [  194.196844] BUG: KASAN: slab-use-after-free in gf128mul_4k_lle+0xc1/0x110
    [  194.197269] Read of size 8 at addr ffff888112bd0448 by task kworker/u77:2/899
    [  194.197707]
    [  194.197818] CPU: 12 UID: 0 PID: 899 Comm: kworker/u77:2 Not tainted 6.11.0-lku-00028-gfca3ca14a17a-dirty #43
    [  194.198400] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.2-3-gd478f380-prebuilt.qemu.org 04/01/2014
    [  194.199046] Workqueue: smb3decryptd smb2_decrypt_offload [cifs]
    [  194.200032] Call Trace:
    [  194.200191]  <TASK>
    [  194.200327]  dump_stack_lvl+0x4e/0x70
    [  194.200558]  ? gf128mul_4k_lle+0xc1/0x110
    [  194.200809]  print_report+0x174/0x505
    [  194.201040]  ? __pfx__raw_spin_lock_irqsave+0x10/0x10
    [  194.201352]  ? srso_return_thunk+0x5/0x5f
    [  194.201604]  ? __virt_addr_valid+0xdf/0x1c0
    [  194.201868]  ? gf128mul_4k_lle+0xc1/0x110
    [  194.202128]  kasan_report+0xc8/0x150
    [  194.202361]  ? gf128mul_4k_lle+0xc1/0x110
    [  194.202616]  gf128mul_4k_lle+0xc1/0x110
    [  194.202863]  ghash_update+0x184/0x210
    [  194.203103]  shash_ahash_update+0x184/0x2a0
    [  194.203377]  ? __pfx_shash_ahash_update+0x10/0x10
    [  194.203651]  ? srso_return_thunk+0x5/0x5f
    [  194.203877]  ? crypto_gcm_init_common+0x1ba/0x340
    [  194.204142]  gcm_hash_assoc_remain_continue+0x10a/0x140
    [  194.204434]  crypt_message+0xec1/0x10a0 [cifs]
    [  194.206489]  ? __pfx_crypt_message+0x10/0x10 [cifs]
    [  194.208507]  ? srso_return_thunk+0x5/0x5f
    [  194.209205]  ? srso_return_thunk+0x5/0x5f
    [  194.209925]  ? srso_return_thunk+0x5/0x5f
    [  194.210443]  ? srso_return_thunk+0x5/0x5f
    [  194.211037]  decrypt_raw_data+0x15f/0x250 [cifs]
    [  194.212906]  ? __pfx_decrypt_raw_data+0x10/0x10 [cifs]
    [  194.214670]  ? srso_return_thunk+0x5/0x5f
    [  194.215193]  smb2_decrypt_offload+0x12a/0x6c0 [cifs]

This is because TFM is being used in parallel.

Fix this by allocating a new AEAD TFM for async decryption, but keep the existing one for synchronous READ cases (similar to what is done in smb3_calc_signature()).

Also remove the calls to aead_request_set_callback() and crypto_wait_req() since it's always going to be a synchronous operation.

---
 fs/smb/client/smb2ops.c |   47 ++++++++++++++++++++++++++++-------------------
 fs/smb/client/smb2pdu.c |    6 ++++++
 2 files changed, 34 insertions(+), 19 deletions(-)

@@ -4488,7 +4488,7 @@ smb2_get_enc_key(struct TCP_Server_Info
  */
 static int
 crypt_message(struct TCP_Server_Info *server, int num_rqst,
-	      struct smb_rqst *rqst, int enc)
+	      struct smb_rqst *rqst, int enc, struct crypto_aead *tfm)
 {
 	struct smb2_transform_hdr *tr_hdr =
 		(struct smb2_transform_hdr *)rqst[0].rq_iov[0].iov_base;

@@ -4499,8 +4499,6 @@ crypt_message(struct TCP_Server_Info *se
 	u8 key[SMB3_ENC_DEC_KEY_SIZE];
 	struct aead_request *req;
 	u8 *iv;
-	DECLARE_CRYPTO_WAIT(wait);
-	struct crypto_aead *tfm;
 	unsigned int crypt_len = le32_to_cpu(tr_hdr->OriginalMessageSize);
 	void *creq;
 

@@ -4511,14 +4509,6 @@ crypt_message(struct TCP_Server_Info *se
 		return rc;
 	}
 
-	rc = smb3_crypto_aead_allocate(server);
-	if (rc) {
-		cifs_server_dbg(VFS, "%s: crypto alloc failed\n", __func__);
-		return rc;
-	}
-
-	tfm = enc ? server->secmech.enc : server->secmech.dec;
-
 	if ((server->cipher_type == SMB2_ENCRYPTION_AES256_CCM) ||
 		(server->cipher_type == SMB2_ENCRYPTION_AES256_GCM))
 		rc = crypto_aead_setkey(tfm, key, SMB3_GCM256_CRYPTKEY_SIZE);

@@ -4557,11 +4547,7 @@ crypt_message(struct TCP_Server_Info *se
 	aead_request_set_crypt(req, sg, sg, crypt_len, iv);
 	aead_request_set_ad(req, assoc_data_len);
 
-	aead_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG,
-				  crypto_req_done, &wait);
-
-	rc = crypto_wait_req(enc ? crypto_aead_encrypt(req)
-				: crypto_aead_decrypt(req), &wait);
+	rc = enc ? crypto_aead_encrypt(req) : crypto_aead_decrypt(req);
 
 	if (!rc && enc)
 		memcpy(&tr_hdr->Signature, sign, SMB2_SIGNATURE_SIZE);

@@ -4650,7 +4636,7 @@ smb3_init_transform_rq(struct TCP_Server
 	/* fill the 1st iov with a transform header */
 	fill_transform_hdr(tr_hdr, orig_len, old_rq, server->cipher_type);
 
-	rc = crypt_message(server, num_rqst, new_rq, 1);
+	rc = crypt_message(server, num_rqst, new_rq, 1, server->secmech.enc);
 	cifs_dbg(FYI, "Encrypt message returned %d\n", rc);
 	if (rc)
 		goto err_free;

@@ -4676,8 +4662,9 @@ decrypt_raw_data(struct TCP_Server_Info
 		 unsigned int npages, unsigned int page_data_size,
 		 bool is_offloaded)
 {
-	struct kvec iov[2];
+	struct crypto_aead *tfm;
 	struct smb_rqst rqst = {NULL};
+	struct kvec iov[2];
 	int rc;
 
 	iov[0].iov_base = buf;

@@ -4692,9 +4679,31 @@ decrypt_raw_data(struct TCP_Server_Info
 	rqst.rq_pagesz = PAGE_SIZE;
 	rqst.rq_tailsz = (page_data_size % PAGE_SIZE) ? : PAGE_SIZE;
 
-	rc = crypt_message(server, 1, &rqst, 0);
+	if (is_offloaded) {
+		if ((server->cipher_type == SMB2_ENCRYPTION_AES128_GCM) ||
+		    (server->cipher_type == SMB2_ENCRYPTION_AES256_GCM))
+			tfm = crypto_alloc_aead("gcm(aes)", 0, 0);
+		else
+			tfm = crypto_alloc_aead("ccm(aes)", 0, 0);
+		if (IS_ERR(tfm)) {
+			rc = PTR_ERR(tfm);
+			cifs_server_dbg(VFS, "%s: Failed alloc decrypt TFM, rc=%d\n", __func__, rc);
+
+			return rc;
+		}
+	} else {
+		if (unlikely(!server->secmech.dec))
+			return -EIO;
+
+		tfm = server->secmech.dec;
+	}
+
+	rc = crypt_message(server, 1, &rqst, 0, tfm);
 	cifs_dbg(FYI, "Decrypt message returned %d\n", rc);
 
+	if (is_offloaded)
+		crypto_free_aead(tfm);
+
 	if (rc)
 		return rc;
 

@@ -1105,6 +1105,12 @@ SMB2_negotiate(const unsigned int xid,
 		else
 			cifs_server_dbg(VFS, "Missing expected negotiate contexts\n");
 	}
+
+	if (server->cipher_type && !rc) {
+		rc = smb3_crypto_aead_allocate(server);
+		if (rc)
+			cifs_server_dbg(VFS, "%s: crypto alloc failed, rc=%d\n", __func__, rc);
+	}
 neg_exit:
 	free_rsp_buf(resp_buftype, rsp);
 	return rc;

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Qasim Ijaz <[email protected]>

commit 575a5adf48b06a2980c9eeffedf699ed5534fade upstream.

This patch addresses a null-ptr-deref in qt2_process_read_urb() due to an incorrect bounds check in the following:

       if (newport > serial->num_ports) {
               dev_err(&port->dev,
                       "%s - port change to invalid port: %i\n",
                       __func__, newport);
               break;
       }

The condition doesn't account for the valid range of the serial->port buffer, which is from 0 to serial->num_ports - 1. When newport is equal to serial->num_ports, the assignment of "port" in the following code is out-of-bounds and NULL:

       serial_priv->current_port = newport;
       port = serial->port[serial_priv->current_port];

The fix checks if newport is greater than or equal to serial->num_ports indicating it is out-of-bounds.

---
 drivers/usb/serial/quatech2.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -511,7 +511,7 @@ static void qt2_process_read_urb(struct
 
 				newport = *(ch + 3);
 
-				if (newport > serial->num_ports) {
+				if (newport >= serial->num_ports) {
 					dev_err(&port->dev,
 						"%s - port change to invalid port: %i\n",
 						__func__, newport);

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Greg Kroah-Hartman <[email protected]>

commit 086fd062bc3883ae1ce4166cff5355db315ad879 upstream.

This reverts commit 13014969cbf07f18d62ceea40bd8ca8ec9d36cec.

It is reported to cause crashes on Tegra systems, so revert it for now.

---
 drivers/usb/gadget/function/u_serial.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

@@ -1393,10 +1393,6 @@ void gserial_disconnect(struct gserial *
 	/* REVISIT as above: how best to track this? */
 	port->port_line_coding = gser->port_line_coding;
 
-	/* disable endpoints, aborting down any active I/O */
-	usb_ep_disable(gser->out);
-	usb_ep_disable(gser->in);
-
 	port->port_usb = NULL;
 	gser->ioport = NULL;
 	if (port->port.count > 0) {

@@ -1408,6 +1404,10 @@ void gserial_disconnect(struct gserial *
 	spin_unlock(&port->port_lock);
 	spin_unlock_irqrestore(&serial_port_lock, flags);
 
+	/* disable endpoints, aborting down any active I/O */
+	usb_ep_disable(gser->out);
+	usb_ep_disable(gser->in);
+
 	/* finally, free any unused/unusable I/O buffers */
 	spin_lock_irqsave(&port->port_lock, flags);
 	if (port->port.count == 0)

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lianqin Hu <[email protected]>

commit ad5b205f9e022b407d91f952faddd05718be2866 upstream.

Audio control requests that sets sampling frequency sometimes fail on this card. Adding delay between control messages eliminates that problem.

usb 1-1: New USB device found, idVendor=0d8c, idProduct=0014 usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0 usb 1-1: Product: USB Audio Device
usb 1-1: Manufacturer: C-Media Electronics Inc.

---
 sound/usb/quirks.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -2137,6 +2137,8 @@ static const struct usb_audio_quirk_flag
 		   QUIRK_FLAG_CTL_MSG_DELAY_1M),
 	DEVICE_FLG(0x0c45, 0x6340, /* Sonix HD USB Camera */
 		   QUIRK_FLAG_GET_SAMPLE_RATE),
+	DEVICE_FLG(0x0d8c, 0x0014, /* USB Audio Device */
+		   QUIRK_FLAG_CTL_MSG_DELAY_1M),
 	DEVICE_FLG(0x0ecb, 0x205c, /* JBL Quantum610 Wireless */
 		   QUIRK_FLAG_FIXED_RATE),
 	DEVICE_FLG(0x0ecb, 0x2069, /* JBL Quantum810 Wireless */

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Pearson <[email protected]>

commit 907bc9268a5a9f823ffa751957a5c1dd59f83f42 upstream.

Microsoft defined Meta+Shift+F23 as the Copilot shortcut instead of a dedicated keycode, and multiple vendors have their keyboards emit this sequence in response to users pressing a dedicated "Copilot" key. Unfortunately the default keymap table in atkbd does not map scancode 0x6e (F23) and so the key combination does not work even if userspace is ready to handle it.

Because this behavior is common between multiple vendors and the scancode is currently unused map 0x6e to keycode 193 (KEY_F23) so that key sequence is generated properly.

MS documentation for the scan code:
https://learn.microsoft.com/en-us/windows/win32/inputdev/about-keyboard-input#scan-codes
Confirmed on Lenovo, HP and Dell machines by Canonical. Tested on Lenovo T14s G6 AMD.

---
 drivers/input/keyboard/atkbd.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -89,7 +89,7 @@ static const unsigned short atkbd_set2_k
 	  0, 46, 45, 32, 18,  5,  4, 95,  0, 57, 47, 33, 20, 19,  6,183,
 	  0, 49, 48, 35, 34, 21,  7,184,  0,  0, 50, 36, 22,  8,  9,185,
 	  0, 51, 37, 23, 24, 11, 10,  0,  0, 52, 53, 38, 39, 25, 12,  0,
-	  0, 89, 40,  0, 26, 13,  0,  0, 58, 54, 28, 27,  0, 43,  0, 85,
+	  0, 89, 40,  0, 26, 13,  0,193, 58, 54, 28, 27,  0, 43,  0, 85,
 	  0, 86, 91, 90, 92,  0, 14, 94,  0, 79,124, 75, 71,121,  0,  0,
 	 82, 83, 80, 76, 77, 72,  1, 69, 87, 78, 81, 74, 55, 73, 70, 99,
 

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nilton Perim Neto <[email protected]>

commit e4940fe6322c851659c17852b671c6e7b1aa9f56 upstream.

Although it mimics the Microsoft's VendorID, it is in fact a clone. Taking into account that the original Microsoft Receiver is not being manufactured anymore, this drive can solve dpad issues encontered by those who still use the original 360 Wireless controller but are using a receiver clone.

---
 drivers/input/joystick/xpad.c |    1 +
 1 file changed, 1 insertion(+)

@@ -145,6 +145,7 @@ static const struct xpad_device {
 	{ 0x045e, 0x028e, "Microsoft X-Box 360 pad", 0, XTYPE_XBOX360 },
 	{ 0x045e, 0x028f, "Microsoft X-Box 360 pad v2", 0, XTYPE_XBOX360 },
 	{ 0x045e, 0x0291, "Xbox 360 Wireless Receiver (XBOX)", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360W },
+	{ 0x045e, 0x02a9, "Xbox 360 Wireless Receiver (Unofficial)", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360W },
 	{ 0x045e, 0x02d1, "Microsoft X-Box One pad", 0, XTYPE_XBOXONE },
 	{ 0x045e, 0x02dd, "Microsoft X-Box One pad (Firmware 2015)", 0, XTYPE_XBOXONE },
 	{ 0x045e, 0x02e3, "Microsoft X-Box One Elite pad", MAP_PADDLES, XTYPE_XBOXONE },

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jack Greiner <[email protected]>

commit 222f3390c15c4452a9f7e26f5b7d9138e75d00d5 upstream.

Add Wooting Two HE (ARM) to the list of supported devices.

---
 drivers/input/joystick/xpad.c |    1 +
 1 file changed, 1 insertion(+)

@@ -367,6 +367,7 @@ static const struct xpad_device {
 	{ 0x31e3, 0x1200, "Wooting Two", 0, XTYPE_XBOX360 },
 	{ 0x31e3, 0x1210, "Wooting Lekker", 0, XTYPE_XBOX360 },
 	{ 0x31e3, 0x1220, "Wooting Two HE", 0, XTYPE_XBOX360 },
+	{ 0x31e3, 0x1230, "Wooting Two HE (ARM)", 0, XTYPE_XBOX360 },
 	{ 0x31e3, 0x1300, "Wooting 60HE (AVR)", 0, XTYPE_XBOX360 },
 	{ 0x31e3, 0x1310, "Wooting 60HE (ARM)", 0, XTYPE_XBOX360 },
 	{ 0x3285, 0x0607, "Nacon GC-100", 0, XTYPE_XBOX360 },

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paulo Alcantara <[email protected]>

commit 4bdec0d1f658f7c98749bd2c5a486e6cfa8565d2 upstream.

Neither SMB3.0 or SMB3.02 supports encryption negotiate context, so when SMB2_GLOBAL_CAP_ENCRYPTION flag is set in the negotiate response, the client uses AES-128-CCM as the default cipher. See MS-SMB2 3.3.5.4.

Commit b0abcd65ec54 ("smb: client: fix UAF in async decryption") added a @server->cipher_type check to conditionally call smb3_crypto_aead_allocate(), but that check would always be false as @server->cipher_type is unset for SMB3.02.

Fix the following KASAN splat by setting @server->cipher_type for SMB3.02 as well.

mount.cifs //srv/share /mnt -o vers=3.02,seal,...

BUG: KASAN: null-ptr-deref in crypto_aead_setkey+0x2c/0x130
Read of size 8 at addr 0000000000000020 by task mount.cifs/1095
CPU: 1 UID: 0 PID: 1095 Comm: mount.cifs Not tainted 6.12.0 #1
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-3.fc41
04/01/2014
Call Trace:
 <TASK>
 dump_stack_lvl+0x5d/0x80
 ? crypto_aead_setkey+0x2c/0x130
 kasan_report+0xda/0x110
 ? crypto_aead_setkey+0x2c/0x130
 crypto_aead_setkey+0x2c/0x130
 crypt_message+0x258/0xec0 [cifs]
 ? __asan_memset+0x23/0x50
 ? __pfx_crypt_message+0x10/0x10 [cifs]
 ? mark_lock+0xb0/0x6a0
 ? hlock_class+0x32/0xb0
 ? mark_lock+0xb0/0x6a0
 smb3_init_transform_rq+0x352/0x3f0 [cifs]
 ? lock_acquire.part.0+0xf4/0x2a0
 smb_send_rqst+0x144/0x230 [cifs]
 ? __pfx_smb_send_rqst+0x10/0x10 [cifs]
 ? hlock_class+0x32/0xb0
 ? smb2_setup_request+0x225/0x3a0 [cifs]
 ? __pfx_cifs_compound_last_callback+0x10/0x10 [cifs]
 compound_send_recv+0x59b/0x1140 [cifs]
 ? __pfx_compound_send_recv+0x10/0x10 [cifs]
 ? __create_object+0x5e/0x90
 ? hlock_class+0x32/0xb0
 ? do_raw_spin_unlock+0x9a/0xf0
 cifs_send_recv+0x23/0x30 [cifs]
 SMB2_tcon+0x3ec/0xb30 [cifs]
 ? __pfx_SMB2_tcon+0x10/0x10 [cifs]
 ? lock_acquire.part.0+0xf4/0x2a0
 ? __pfx_lock_release+0x10/0x10
 ? do_raw_spin_trylock+0xc6/0x120
 ? lock_acquire+0x3f/0x90
 ? _get_xid+0x16/0xd0 [cifs]
 ? __pfx_SMB2_tcon+0x10/0x10 [cifs]
 ? cifs_get_smb_ses+0xcdd/0x10a0 [cifs]
 cifs_get_smb_ses+0xcdd/0x10a0 [cifs]
 ? __pfx_cifs_get_smb_ses+0x10/0x10 [cifs]
 ? cifs_get_tcp_session+0xaa0/0xca0 [cifs]
 cifs_mount_get_session+0x8a/0x210 [cifs]
 dfs_mount_share+0x1b0/0x11d0 [cifs]
 ? __pfx___lock_acquire+0x10/0x10
 ? __pfx_dfs_mount_share+0x10/0x10 [cifs]
 ? lock_acquire.part.0+0xf4/0x2a0
 ? find_held_lock+0x8a/0xa0
 ? hlock_class+0x32/0xb0
 ? lock_release+0x203/0x5d0
 cifs_mount+0xb3/0x3d0 [cifs]
 ? do_raw_spin_trylock+0xc6/0x120
 ? __pfx_cifs_mount+0x10/0x10 [cifs]
 ? lock_acquire+0x3f/0x90
 ? find_nls+0x16/0xa0
 ? smb3_update_mnt_flags+0x372/0x3b0 [cifs]
 cifs_smb3_do_mount+0x1e2/0xc80 [cifs]
 ? __pfx_vfs_parse_fs_string+0x10/0x10
 ? __pfx_cifs_smb3_do_mount+0x10/0x10 [cifs]
 smb3_get_tree+0x1bf/0x330 [cifs]
 vfs_get_tree+0x4a/0x160
 path_mount+0x3c1/0xfb0
 ? kasan_quarantine_put+0xc7/0x1d0
 ? __pfx_path_mount+0x10/0x10
 ? kmem_cache_free+0x118/0x3e0
 ? user_path_at+0x74/0xa0
 __x64_sys_mount+0x1a6/0x1e0
 ? __pfx___x64_sys_mount+0x10/0x10
 ? mark_held_locks+0x1a/0x90
 do_syscall_64+0xbb/0x1d0
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

---
 fs/smb/client/smb2pdu.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -1070,7 +1070,9 @@ SMB2_negotiate(const unsigned int xid,
 	 * SMB3.0 supports only 1 cipher and doesn't have a encryption neg context
 	 * Set the cipher type manually.
 	 */
-	if (server->dialect == SMB30_PROT_ID && (server->capabilities & SMB2_GLOBAL_CAP_ENCRYPTION))
+	if ((server->dialect == SMB30_PROT_ID ||
+	     server->dialect == SMB302_PROT_ID) &&
+	    (server->capabilities & SMB2_GLOBAL_CAP_ENCRYPTION))
 		server->cipher_type = SMB2_ENCRYPTION_AES128_CCM;
 
 	security_blob = smb2_get_data_area_len(&blob_offset, &blob_length,

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Szyprowski <[email protected]>

commit 48c6253fefa38556e0c5c2942edd9181529407e4 upstream.

'Headphone Switch' control is already registered from sound/soc/codecs/wm_hubs.c:479, so duplicating it in midas_wm1811 causes following probe failure:

midas-audio sound: control 2:0:0:Headphone Switch:0 is already present midas-audio sound: ASoC: Failed to add Headphone Switch: -16 midas-audio sound: Failed to register card: -16
midas-audio: probe of sound failed with error -16

Fix this by dropping duplicated control.

---
 sound/soc/samsung/midas_wm1811.c |    1 -
 1 file changed, 1 deletion(-)

@@ -257,7 +257,6 @@ static const struct snd_kcontrol_new mid
 	SOC_DAPM_PIN_SWITCH("Main Mic"),
 	SOC_DAPM_PIN_SWITCH("Sub Mic"),
 	SOC_DAPM_PIN_SWITCH("Headset Mic"),
-	SOC_DAPM_PIN_SWITCH("Headphone"),
 
 	SOC_DAPM_PIN_SWITCH("FM In"),
 };

On ARCH_BRCMSTB using 32-bit and 64-bit ARM kernels, build tested on BMIPS_GENERIC:

Florian

All tests passing for Tegra ...

Test results for stable-v6.1:

    10 builds:	10 pass, 0 fail
    26 boots:	26 pass, 0 fail
    115 tests:	115 pass, 0 fail

Linux version: 6.1.128-rc1-gda19df6ebb6c
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra194-p2972-0000, tegra194-p3509-0000+p3668-0000,
                tegra20-ventana, tegra210-p2371-2180,
                tegra210-p3450-0000, tegra30-cardhu-a04

Jon

Hi!

CIP testing did not find any problems here:

https://gitlab.com/cip-project/cip-testing/linux-stable-rc-ci/-/tree/linux-6.1.y

Best regards,
Pavel

DENX Software Engineering GmbH,        Managing Director: Erika Unter
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

Built and booted successfully on RISC-V RV64 (HiFive Unmatched).

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

## Build
* kernel: 6.1.128-rc1
* git: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
* git commit: da19df6ebb6c09ded78f67e201f202979c1a5727
* git describe: v6.1.127-50-gda19df6ebb6c
* test details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-6.1.y/build/v6.1.127-50-gda19df6ebb6c

## Test Regressions (compared to v6.1.126-65-gad6747190c53)

## Metric Regressions (compared to v6.1.126-65-gad6747190c53)

## Test Fixes (compared to v6.1.126-65-gad6747190c53)

## Metric Fixes (compared to v6.1.126-65-gad6747190c53)

## Test result summary
total: 108475, pass: 71884, fail: 19274, skip: 16867, xfail: 450

## Build Summary
* arc: 6 total, 5 passed, 1 failed
* arm: 139 total, 139 passed, 0 failed
* arm64: 46 total, 44 passed, 2 failed
* i386: 31 total, 27 passed, 4 failed
* mips: 30 total, 25 passed, 5 failed
* parisc: 5 total, 5 passed, 0 failed
* powerpc: 36 total, 33 passed, 3 failed
* riscv: 14 total, 13 passed, 1 failed
* s390: 18 total, 17 passed, 1 failed
* sh: 12 total, 10 passed, 2 failed
* sparc: 9 total, 8 passed, 1 failed
* x86_64: 38 total, 38 passed, 0 failed

## Test suites summary
* boot
* commands
* kselftest-arm64
* kselftest-breakpoints
* kselftest-capabilities
* kselftest-cgroup
* kselftest-clone3
* kselftest-core
* kselftest-cpu-hotplug
* kselftest-cpufreq
* kselftest-efivarfs
* kselftest-exec
* kselftest-filesystems
* kselftest-filesystems-binderfs
* kselftest-filesystems-epoll
* kselftest-firmware
* kselftest-fpu
* kselftest-ftrace
* kselftest-futex
* kselftest-gpio
* kselftest-intel_pstate
* kselftest-ipc
* kselftest-kcmp
* kselftest-kvm
* kselftest-livepatch
* kselftest-membarrier
* kselftest-memfd
* kselftest-mincore
* kselftest-mqueue
* kselftest-net
* kselftest-net-mptcp
* kselftest-openat2
* kselftest-ptrace
* kselftest-rseq
* kselftest-rtc
* kselftest-seccomp
* kselftest-sigaltstack
* kselftest-size
* kselftest-tc-testing
* kselftest-timers
* kselftest-tmpfs
* kselftest-tpm2
* kselftest-user_events
* kselftest-vDSO
* kselftest-x86
* kunit
* kvm-unit-tests
* libgpiod
* libhugetlbfs
* log-parser-boot
* log-parser-build-clang
* log-parser-build-gcc
* log-parser-test
* ltp-capability
* ltp-commands
* ltp-containers
* ltp-controllers
* ltp-cpuhotplug
* ltp-crypto
* ltp-cve
* ltp-dio
* ltp-fcntl-locktests
* ltp-filecaps
* ltp-fs
* ltp-fs_bind
* ltp-fs_perms_simple
* ltp-hugetlb
* ltp-ipc
* ltp-math
* ltp-mm
* ltp-nptl
* ltp-pty
* ltp-sched
* ltp-smoke
* ltp-syscalls
* ltp-tracing
* perf
* rcutorture

--
Linaro LKFT
https://lkft.linaro.org

OVERVIEW

Builds: 40 passed, 0 failed

Boot tests: 593 passed, 0 failed

CI systems: broonie, maestro

REVISION

    Commit
        name: v6.1.127-50-gda19df6ebb6c
        hash: da19df6ebb6c09ded78f67e201f202979c1a5727
    Checked out from
        https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y

BUILDS

No build failures found

BOOT TESTS

No boot failures found

See complete and up-to-date report at:

    https://kcidb.kernelci.org/d/revision/revision?orgId=1&var-git_commit_hash=da19df6ebb6c09ded78f67e201f202979c1a5727&var-patchset_hash=

Thanks,
KernelCI team

The kernel, bpf tool, perf tool, and kselftest builds fine for v6.1.128-rc1 on x86 and arm64 Azure VM.

Thanks,
Hardik

Am 30.01.2025 um 15:01 schrieb Greg Kroah-Hartman:

Builds, boots and works on my 2-socket Ivy Bridge Xeon E5-2697 v2 server. No dmesg oddities or regressions found.

Beste Grüße,
Peter Schneider

Climb the mountain not to plant your flag, but to embrace the challenge,
enjoy the air and behold the view. Climb it so you can see the world,
not so the world can see you.                    -- David McCullough Jr.

OpenPGP:  0xA3828BD796CCE11A8CADE8866E3A92C92C3FF244
Download: https://www.peters-netzplatz.de/download/pschneider1968_pub.asc
https://keys.mailvelope.com/pks/lookup?op=get&[email protected]
https://keys.mailvelope.com/pks/lookup?op=get&[email protected]