5.4.168-rc1 review

This is the start of the stable review cycle for the 5.4.168 release. There are 71 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 Wed, 22 Dec 2021 14:30:09 +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/v5.x/stable-review/patch-5.4.168-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-5.4.y
and the diffstat can be found below.

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

    Linux 5.4.168-rc1

Juergen Gross <[email protected]>
xen/netback: don't queue unlimited number of packages

Juergen Gross <[email protected]>
xen/netback: fix rx queue stall detection

Juergen Gross <[email protected]>
xen/console: harden hvc_xen against event channel storms

Juergen Gross <[email protected]>
xen/netfront: harden netfront against event channel storms

Juergen Gross <[email protected]>
xen/blkfront: harden blkfront against event channel storms

Magnus Karlsson <[email protected]>
Revert "xsk: Do not sleep in poll() when need_wakeup set"

Leon Romanovsky <[email protected]>
net: sched: Fix suspicious RCU usage while accessing tcf_tunnel_info

Felix Fietkau <[email protected]>
mac80211: fix regression in SSN handling of addba tx

Paul E. McKenney <[email protected]>
rcu: Mark accesses to rcu_state.n_force_qs

George Kennedy <[email protected]>
scsi: scsi_debug: Sanity check block descriptor length in resp_mode_select()

Miklos Szeredi <[email protected]>
ovl: fix warning in ovl_create_real()

Miklos Szeredi <[email protected]>
fuse: annotate lock in fuse_reverse_inval_entry()

Pavel Skripkin <[email protected]>
media: mxl111sf: change mutex_init() location

Magnus Karlsson <[email protected]>
xsk: Do not sleep in poll() when need_wakeup set

Fabio Estevam <[email protected]>
ARM: dts: imx6ull-pinfunc: Fix CSI_DATA07__ESAI_TX0 pad name

Nathan Chancellor <[email protected]>
Input: touchscreen - avoid bitwise vs logical OR warning

Nathan Chancellor <[email protected]>
mwifiex: Remove unnecessary braces from HostCmd_SET_SEQ_NO_BSS_INFO

Johannes Berg <[email protected]>
mac80211: validate extended element ID is present

Le Ma <[email protected]>
drm/amdgpu: correct register access for RLC_JUMP_TABLE_RESTORE

George Kennedy <[email protected]>
libata: if T_LENGTH is zero, dma direction should be DMA_NONE

Yu Liao <[email protected]>
timekeeping: Really make sure wall_to_monotonic isn't positive

Daniele Palmas <[email protected]>
USB: serial: option: add Telit FN990 compositions

Johan Hovold <[email protected]>

    USB: serial: cp210x: fix CP2105 GPIO registration

Nehal Bakulchandra Shah <[email protected]>
usb: xhci: Extend support for runtime power management for AMD's Yellow carp.

Stefan Roese <[email protected]>
PCI/MSI: Mask MSI-X vectors only on success

Thomas Gleixner <[email protected]>
PCI/MSI: Clear PCI_MSIX_FLAGS_MASKALL on error

Jimmy Wang <[email protected]>
USB: NO_LPM quirk Lenovo USB-C to Ethernet Adapher(RTL8153-04)

Greg Kroah-Hartman <[email protected]>
USB: gadget: bRequestType is a bitfield, not a enum

Eric Dumazet <[email protected]>
sit: do not call ipip6_dev_free() from sit_init_net()

Florian Fainelli <[email protected]>
net: systemport: Add global locking for descriptor lifecycle

D. Wythe <[email protected]>
net/smc: Prevent smc_release() from long blocking

Gal Pressman <[email protected]>

    net: Fix double 0x prefix print in SKB dump

Willem de Bruijn <[email protected]>
net/packet: rx_owner_map depends on pg_vec

Haimin Zhang <[email protected]>
netdevsim: Zero-initialize memory for new map's value in function nsim_bpf_map_alloc

Cyril Novikov <[email protected]>
ixgbe: set X550 MDIO speed before talking to PHY

Letu Ren <[email protected]>
igbvf: fix double free in igbvf_probe

Karen Sornek <[email protected]>
igb: Fix removal of unicast MAC filters of VFs

Nathan Chancellor <[email protected]>
soc/tegra: fuse: Fix bitwise vs. logical OR warning

Hangyu Hua <[email protected]>
rds: memory leak in __rds_conn_create()

Baowen Zheng <[email protected]>
flow_offload: return EOPNOTSUPP for the unsupported mpls action type

Vlad Buslov <[email protected]>
net: sched: lock action when translating it to flow_action infra

Johannes Berg <[email protected]>
mac80211: fix lookup when adding AddBA extension element

Johannes Berg <[email protected]>
mac80211: accept aggregation sessions on 6 GHz

Johannes Berg <[email protected]>

    mac80211: agg-tx: don't schedule_and_wake_txq() under sta->lock

Mordechay Goodstein <[email protected]>
mac80211: agg-tx: refactor sending addba

Hangbin Liu <[email protected]>
selftest/net/forwarding: declare NETIFS p9 p10

Alyssa Ross <[email protected]>
dmaengine: st_fdma: fix MODULE_ALIAS

David Ahern <[email protected]>
selftests: Fix IPv6 address bind tests

David Ahern <[email protected]>
selftests: Fix raw socket bind tests with VRF

Eric Dumazet <[email protected]>
inet_diag: fix kernel-infoleak for UDP sockets

Eric Dumazet <[email protected]>
inet_diag: use jiffies_delta_to_msecs()

Eric Dumazet <[email protected]>
sch_cake: do not call cake_destroy() from cake_init()

Philipp Rudo <[email protected]>
s390/kexec_file: fix error handling when applying relocations

Jie2x Zhou <[email protected]>
selftests: net: Correct ping6 expected rc from 2 to 1

Mike Tipton <[email protected]>
clk: Don't parent clks until the parent is fully registered

Dinh Nguyen <[email protected]>
ARM: socfpga: dts: fix qspi node compatible

Randy Dunlap <[email protected]>
hv: utils: add PTP_1588_CLOCK to Kconfig to fix build

Johannes Berg <[email protected]>
mac80211: track only QoS data frames for admission control

Alex Bee <[email protected]>
arm64: dts: rockchip: fix audio-supply for Rock Pi 4

John Keeping <[email protected]>

    arm64: dts: rockchip: fix rk3399-leez-p710 vcc3v3-lan supply

Artem Lapkin <[email protected]>
arm64: dts: rockchip: remove mmc-hs400-enhanced-strobe from rk3399-khadas-edge

J. Bruce Fields <[email protected]>
nfsd: fix use-after-free due to delegation race

Fabrice Gasnier <[email protected]>
iio: adc: stm32: fix a current leak by resetting pcsel before disabling vdda

Paul Moore <[email protected]>
audit: improve robustness of the audit queue handling

Joe Thornber <[email protected]>
dm btree remove: fix use after free in rebalance_children()

Jerome Marchand <[email protected]>
recordmcount.pl: look for jgnop instruction as well as bcrl on s390

Will Deacon <[email protected]>
virtio_ring: Fix querying of maximum DMA mapping size for virtio device

Sudeep Holla <[email protected]>
firmware: arm_scpi: Fix string overflow in SCPI genpd driver

Felix Fietkau <[email protected]>
mac80211: send ADDBA requests using the tid/queue of the aggregation session

Johannes Berg <[email protected]>
mac80211: mark TX-during-stop for TX in in_reconfig

Vitaly Kuznetsov <[email protected]>
KVM: selftests: Make sure kvm_create_max_vcpus test won't hit RLIMIT_NOFILE

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

Diffstat:

 Makefile                                           |   4 +-
 arch/arm/boot/dts/imx6ull-pinfunc.h                |   2 +-
 arch/arm/boot/dts/socfpga_arria10_socdk_qspi.dts   |   2 +-
 arch/arm/boot/dts/socfpga_arria5_socdk.dts         |   2 +-
 arch/arm/boot/dts/socfpga_cyclone5_socdk.dts       |   2 +-
 arch/arm/boot/dts/socfpga_cyclone5_sockit.dts      |   2 +-
 arch/arm/boot/dts/socfpga_cyclone5_socrates.dts    |   2 +-
 arch/arm/boot/dts/socfpga_cyclone5_sodia.dts       |   2 +-
 arch/arm/boot/dts/socfpga_cyclone5_vining_fpga.dts |   4 +-
 .../boot/dts/rockchip/rk3399-khadas-edge.dtsi      |   1 -
 arch/arm64/boot/dts/rockchip/rk3399-leez-p710.dts  |   2 +-
 arch/arm64/boot/dts/rockchip/rk3399-rock-pi-4.dts  |   2 +-
 arch/s390/kernel/machine_kexec_file.c              |   7 +-
 drivers/ata/libata-scsi.c                          |  15 ++-
 drivers/block/xen-blkfront.c                       |  15 ++-
 drivers/clk/clk.c                                  |  15 ++-
 drivers/dma/st_fdma.c                              |   2 +-
 drivers/firmware/scpi_pm_domain.c                  |  10 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c              |   4 +-
 drivers/hv/Kconfig                                 |   1 +
 drivers/iio/adc/stm32-adc.c                        |   1 +
 drivers/input/touchscreen/of_touchscreen.c         |  18 +--
 drivers/md/persistent-data/dm-btree-remove.c       |   2 +-
 drivers/media/usb/dvb-usb-v2/mxl111sf.c            |  16 ++-
 drivers/net/ethernet/broadcom/bcmsysport.c         |   5 +-
 drivers/net/ethernet/broadcom/bcmsysport.h         |   1 +
 drivers/net/ethernet/intel/igb/igb_main.c          |  28 ++---
 drivers/net/ethernet/intel/igbvf/netdev.c          |   1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c      |   3 +
 drivers/net/netdevsim/bpf.c                        |   1 +
 drivers/net/wireless/marvell/mwifiex/cmdevt.c      |   4 +-
 drivers/net/wireless/marvell/mwifiex/fw.h          |   8 +-
 drivers/net/xen-netback/common.h                   |   1 +
 drivers/net/xen-netback/rx.c                       |  77 ++++++++-----
 drivers/net/xen-netfront.c                         | 125 ++++++++++++++++-----
 drivers/pci/msi.c                                  |  15 ++-
 drivers/scsi/scsi_debug.c                          |   4 +-
 drivers/soc/tegra/fuse/fuse-tegra.c                |   2 +-
 drivers/soc/tegra/fuse/fuse.h                      |   2 +-
 drivers/tty/hvc/hvc_xen.c                          |  30 ++++-
 drivers/usb/core/quirks.c                          |   3 +
 drivers/usb/gadget/composite.c                     |   6 +-
 drivers/usb/gadget/legacy/dbgp.c                   |   6 +-
 drivers/usb/gadget/legacy/inode.c                  |   6 +-
 drivers/usb/host/xhci-pci.c                        |   6 +-
 drivers/usb/serial/cp210x.c                        |   6 +-
 drivers/usb/serial/option.c                        |   8 ++
 drivers/virtio/virtio_ring.c                       |   2 +-
 fs/fuse/dir.c                                      |   2 +-
 fs/nfsd/nfs4state.c                                |   9 +-
 fs/overlayfs/dir.c                                 |   3 +-
 fs/overlayfs/overlayfs.h                           |   1 +
 fs/overlayfs/super.c                               |  12 +-
 include/net/tc_act/tc_tunnel_key.h                 |   7 +-
 kernel/audit.c                                     |  21 ++--
 kernel/rcu/tree.c                                  |  10 +-
 kernel/time/timekeeping.c                          |   3 +-
 net/core/skbuff.c                                  |   2 +-
 net/ipv4/inet_diag.c                               |  19 ++--
 net/ipv6/sit.c                                     |   1 -
 net/mac80211/agg-rx.c                              |   8 +-
 net/mac80211/agg-tx.c                              |  80 ++++++++-----
 net/mac80211/driver-ops.h                          |   5 +-
 net/mac80211/mlme.c                                |  13 ++-
 net/mac80211/sta_info.h                            |   1 +
 net/mac80211/util.c                                |   2 +
 net/packet/af_packet.c                             |   5 +-
 net/rds/connection.c                               |   1 +
 net/sched/act_sample.c                             |   2 -
 net/sched/cls_api.c                                |  18 ++-
 net/sched/sch_cake.c                               |   6 +-
 net/smc/af_smc.c                                   |   4 +-
 scripts/recordmcount.pl                            |   2 +-
 tools/testing/selftests/kvm/kvm_create_max_vcpus.c |  30 +++++
 tools/testing/selftests/net/fcnal-test.sh          |  23 ++--
 .../net/forwarding/forwarding.config.sample        |   2 +
 76 files changed, 525 insertions(+), 250 deletions(-)

From: Vitaly Kuznetsov <[email protected]>

[ Upstream commit 908fa88e420f30dde6d80f092795a18ec72ca6d3 ]

With the elevated 'KVM_CAP_MAX_VCPUS' value kvm_create_max_vcpus test may hit RLIMIT_NOFILE limits:

 # ./kvm_create_max_vcpus
 KVM_CAP_MAX_VCPU_ID: 4096
 KVM_CAP_MAX_VCPUS: 1024
 Testing creating 1024 vCPUs, with IDs 0...1023.
 /dev/kvm not available (errno: 24), skipping test

Adjust RLIMIT_NOFILE limits to make sure KVM_CAP_MAX_VCPUS fds can be opened. Note, raising hard limit ('rlim_max') requires CAP_SYS_RESOURCE capability which is generally not needed to run kvm selftests (but without raising the limit the test is doomed to fail anyway).

Message-Id: <[email protected]>

---
 tools/testing/selftests/kvm/kvm_create_max_vcpus.c |   30 +++++++++++++++++++++
 1 file changed, 30 insertions(+)

@@ -12,6 +12,7 @@
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
+#include <sys/resource.h>
 
 #include "test_util.h"
 

@@ -43,11 +44,40 @@ int main(int argc, char *argv[])
 {
 	int kvm_max_vcpu_id = kvm_check_cap(KVM_CAP_MAX_VCPU_ID);
 	int kvm_max_vcpus = kvm_check_cap(KVM_CAP_MAX_VCPUS);
+	/*
+	 * Number of file descriptors reqired, KVM_CAP_MAX_VCPUS for vCPU fds +
+	 * an arbitrary number for everything else.
+	 */
+	int nr_fds_wanted = kvm_max_vcpus + 100;
+	struct rlimit rl;
 
 	printf("KVM_CAP_MAX_VCPU_ID: %d\n", kvm_max_vcpu_id);
 	printf("KVM_CAP_MAX_VCPUS: %d\n", kvm_max_vcpus);
 
 	/*
+	 * Check that we're allowed to open nr_fds_wanted file descriptors and
+	 * try raising the limits if needed.
+	 */
+	TEST_ASSERT(!getrlimit(RLIMIT_NOFILE, &rl), "getrlimit() failed!");
+
+	if (rl.rlim_cur < nr_fds_wanted) {
+		rl.rlim_cur = nr_fds_wanted;
+		if (rl.rlim_max < nr_fds_wanted) {
+			int old_rlim_max = rl.rlim_max;
+			rl.rlim_max = nr_fds_wanted;
+
+			int r = setrlimit(RLIMIT_NOFILE, &rl);
+			if (r < 0) {
+				printf("RLIMIT_NOFILE hard limit is too low (%d, wanted %d)\n",
+				       old_rlim_max, nr_fds_wanted);
+				exit(KSFT_SKIP);
+			}
+		} else {
+			TEST_ASSERT(!setrlimit(RLIMIT_NOFILE, &rl), "setrlimit() failed!");
+		}
+	}
+
+	/*
 	 * Upstream KVM prior to 4.8 does not support KVM_CAP_MAX_VCPU_ID.
 	 * Userspace is supposed to use KVM_CAP_MAX_VCPUS as the maximum ID
 	 * in this case.

From: Johannes Berg <[email protected]>

commit db7205af049d230e7e0abf61c1e74c1aab40f390 upstream.

Mark TXQs as having seen transmit while they were stopped if we bail out of drv_wake_tx_queue() due to reconfig, so that the queue wake after this will make them catch up. This is particularly necessary for when TXQs are used for management packets since those TXQs won't see a lot of traffic that'd make them catch up later.

---
 net/mac80211/driver-ops.h |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

@@ -1202,8 +1202,11 @@ static inline void drv_wake_tx_queue(str
 {
 	struct ieee80211_sub_if_data *sdata = vif_to_sdata(txq->txq.vif);
 
-	if (local->in_reconfig)
+	/* In reconfig don't transmit now, but mark for waking later */
+	if (local->in_reconfig) {
+		set_bit(IEEE80211_TXQ_STOP_NETIF_TX, &txq->flags);
 		return;
+	}
 
 	if (!check_sdata_in_driver(sdata))
 		return;

From: Felix Fietkau <[email protected]>

commit 1fe98f5690c4219d419ea9cc190f94b3401cf324 upstream.

Sending them out on a different queue can cause a race condition where a number of packets in the queue may be discarded by the receiver, because the ADDBA request is sent too early.
This affects any driver with software A-MPDU setup which does not allocate packet seqno in hardware on tx, regardless of whether iTXQ is used or not. The only driver I've seen that explicitly deals with this issue internally is mwl8k.

---
 net/mac80211/agg-tx.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -106,7 +106,7 @@ static void ieee80211_send_addba_request
 	mgmt->u.action.u.addba_req.start_seq_num =
 					cpu_to_le16(start_seq_num << 4);
 
-	ieee80211_tx_skb(sdata, skb);
+	ieee80211_tx_skb_tid(sdata, skb, tid);
 }
 
 void ieee80211_send_bar(struct ieee80211_vif *vif, u8 *ra, u16 tid, u16 ssn)

From: Sudeep Holla <[email protected]>

commit 865ed67ab955428b9aa771d8b4f1e4fb7fd08945 upstream.

Without the bound checks for scpi_pd->name, it could result in the buffer overflow when copying the SCPI device name from the corresponding device tree node as the name string is set at maximum size of 30.

Let us fix it by using devm_kasprintf so that the string buffer is allocated dynamically.

---
 drivers/firmware/scpi_pm_domain.c |   10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

@@ -16,7 +16,6 @@ struct scpi_pm_domain {
 	struct generic_pm_domain genpd;
 	struct scpi_ops *ops;
 	u32 domain;
-	char name[30];
 };
 
 /*

@@ -110,8 +109,13 @@ static int scpi_pm_domain_probe(struct p
 
 		scpi_pd->domain = i;
 		scpi_pd->ops = scpi_ops;
-		sprintf(scpi_pd->name, "%pOFn.%d", np, i);
-		scpi_pd->genpd.name = scpi_pd->name;
+		scpi_pd->genpd.name = devm_kasprintf(dev, GFP_KERNEL,
+						     "%pOFn.%d", np, i);
+		if (!scpi_pd->genpd.name) {
+			dev_err(dev, "Failed to allocate genpd name:%pOFn.%d\n",
+				np, i);
+			continue;
+		}
 		scpi_pd->genpd.power_off = scpi_pd_power_off;
 		scpi_pd->genpd.power_on = scpi_pd_power_on;
 

From: Will Deacon <[email protected]>

commit 817fc978b5a29b039db0418a91072b31c9aab152 upstream.

virtio_max_dma_size() returns the maximum DMA mapping size of the virtio device by querying dma_max_mapping_size() for the device when the DMA API is in use for the vring. Unfortunately, the device passed is initialised by register_virtio_device() and does not inherit the DMA configuration from its parent, resulting in SWIOTLB errors when bouncing is enabled and the default 256K mapping limit (IO_TLB_SEGSIZE) is not respected:

| virtio-pci 0000:00:01.0: swiotlb buffer is full (sz: 294912 bytes), total 1024 (slots), used 725 (slots)

Follow the pattern used elsewhere in the virtio_ring code when calling into the DMA layer and pass the parent device to dma_max_mapping_size() instead.

---
 drivers/virtio/virtio_ring.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -263,7 +263,7 @@ size_t virtio_max_dma_size(struct virtio
 	size_t max_segment_size = SIZE_MAX;
 
 	if (vring_use_dma_api(vdev))
-		max_segment_size = dma_max_mapping_size(&vdev->dev);
+		max_segment_size = dma_max_mapping_size(vdev->dev.parent);
 
 	return max_segment_size;
 }

From: Jerome Marchand <[email protected]>

commit 85bf17b28f97ca2749968d8786dc423db320d9c2 upstream.

On s390, recordmcount.pl is looking for "bcrl 0,<xxx>" instructions in the objdump -d outpout. However since binutils 2.37, objdump -d display "jgnop <xxx>" for the same instruction. Update the mcount_regex so that it accepts both.

---
 scripts/recordmcount.pl |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -252,7 +252,7 @@ if ($arch eq "x86_64") {
 
 } elsif ($arch eq "s390" && $bits == 64) {
     if ($cc =~ /-DCC_USING_HOTPATCH/) {
-	$mcount_regex = "^\\s*([0-9a-fA-F]+):\\s*c0 04 00 00 00 00\\s*brcl\\s*0,[0-9a-f]+ <([^\+]*)>\$";
+	$mcount_regex = "^\\s*([0-9a-fA-F]+):\\s*c0 04 00 00 00 00\\s*(bcrl\\s*0,|jgnop\\s*)[0-9a-f]+ <([^\+]*)>\$";
 	$mcount_adjust = 0;
     } else {
 	$mcount_regex = "^\\s*([0-9a-fA-F]+):\\s*R_390_(PC|PLT)32DBL\\s+_mcount\\+0x2\$";

From: Joe Thornber <[email protected]>

commit 1b8d2789dad0005fd5e7d35dab26a8e1203fb6da upstream.

Move dm_tm_unlock() after dm_tm_dec().

---
 drivers/md/persistent-data/dm-btree-remove.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -423,9 +423,9 @@ static int rebalance_children(struct sha
 
 		memcpy(n, dm_block_data(child),
 		       dm_bm_block_size(dm_tm_get_bm(info->tm)));
-		dm_tm_unlock(info->tm, child);
 
 		dm_tm_dec(info->tm, dm_block_location(child));
+		dm_tm_unlock(info->tm, child);
 		return 0;
 	}
 

From: Paul Moore <[email protected]>

commit f4b3ee3c85551d2d343a3ba159304066523f730f upstream.

If the audit daemon were ever to get stuck in a stopped state the kernel's kauditd_thread() could get blocked attempting to send audit records to the userspace audit daemon. With the kernel thread blocked it is possible that the audit queue could grow unbounded as certain audit record generating events must be exempt from the queue limits else the system enter a deadlock state.

This patch resolves this problem by lowering the kernel thread's socket sending timeout from MAX_SCHEDULE_TIMEOUT to HZ/10 and tweaks the kauditd_send_queue() function to better manage the various audit queues when connection problems occur between the kernel and the audit daemon. With this patch, the backlog may temporarily grow beyond the defined limits when the audit daemon is stopped and the system is under heavy audit pressure, but kauditd_thread() will continue to make progress and drain the queues as it would for other connection problems. For example, with the audit daemon put into a stopped state and the system configured to audit every syscall it was still possible to shutdown the system without a kernel panic, deadlock, etc.; granted, the system was slow to shutdown but that is to be expected given the extreme pressure of recording every syscall.

The timeout value of HZ/10 was chosen primarily through experimentation and this developer's "gut feeling". There is likely no one perfect value, but as this scenario is limited in scope (root privileges would be needed to send SIGSTOP to the audit daemon), it is likely not worth exposing this as a tunable at present. This can always be done at a later date if it proves necessary.

---
 kernel/audit.c |   21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

@@ -712,7 +712,7 @@ static int kauditd_send_queue(struct soc
 {
 	int rc = 0;
 	struct sk_buff *skb;
-	static unsigned int failed = 0;
+	unsigned int failed = 0;
 
 	/* NOTE: kauditd_thread takes care of all our locking, we just use
 	 *       the netlink info passed to us (e.g. sk and portid) */

@@ -729,32 +729,30 @@ static int kauditd_send_queue(struct soc
 			continue;
 		}
 
+retry:
 		/* grab an extra skb reference in case of error */
 		skb_get(skb);
 		rc = netlink_unicast(sk, skb, portid, 0);
 		if (rc < 0) {
-			/* fatal failure for our queue flush attempt? */
+			/* send failed - try a few times unless fatal error */
 			if (++failed >= retry_limit ||
 			    rc == -ECONNREFUSED || rc == -EPERM) {
-				/* yes - error processing for the queue */
 				sk = NULL;
 				if (err_hook)
 					(*err_hook)(skb);
-				if (!skb_hook)
-					goto out;
-				/* keep processing with the skb_hook */
+				if (rc == -EAGAIN)
+					rc = 0;
+				/* continue to drain the queue */
 				continue;
 			} else
-				/* no - requeue to preserve ordering */
-				skb_queue_head(queue, skb);
+				goto retry;
 		} else {
-			/* it worked - drop the extra reference and continue */
+			/* skb sent - drop the extra reference and continue */
 			consume_skb(skb);
 			failed = 0;
 		}
 	}
 
-out:
 	return (rc >= 0 ? 0 : rc);
 }
 

@@ -1557,7 +1555,8 @@ static int __net_init audit_net_init(str
 		audit_panic("cannot initialize netlink socket in namespace");
 		return -ENOMEM;
 	}
-	aunet->sk->sk_sndtimeo = MAX_SCHEDULE_TIMEOUT;
+	/* limit the timeout in case auditd is blocked/stopped */
+	aunet->sk->sk_sndtimeo = HZ / 10;
 
 	return 0;
 }

From: Fabrice Gasnier <[email protected]>

commit f711f28e71e965c0d1141c830fa7131b41abbe75 upstream.

Some I/Os are connected to ADC input channels, when the corresponding bit in PCSEL register are set on STM32H7 and STM32MP15. This is done in the prepare routine of stm32-adc driver.
There are constraints here, as PCSEL shouldn't be set when VDDA supply is disabled. Enabling/disabling of VDDA supply in done via stm32-adc-core runtime PM routines (before/after ADC is enabled/disabled).

Currently, PCSEL remains set when disabling ADC. Later on, PM runtime can disable the VDDA supply. This creates some conditions on I/Os that can start to leak current.
So PCSEL needs to be cleared when disabling the ADC.

---
 drivers/iio/adc/stm32-adc.c |    1 +
 1 file changed, 1 insertion(+)

@@ -933,6 +933,7 @@ pwr_dwn:
 
 static void stm32h7_adc_unprepare(struct stm32_adc *adc)
 {
+	stm32_adc_writel(adc, STM32H7_ADC_PCSEL, 0);
 	stm32h7_adc_disable(adc);
 	stm32h7_adc_enter_pwr_down(adc);
 }

From: J. Bruce Fields <[email protected]>

commit 548ec0805c399c65ed66c6641be467f717833ab5 upstream.

A delegation break could arrive as soon as we've called vfs_setlease. A delegation break runs a callback which immediately (in nfsd4_cb_recall_prepare) adds the delegation to del_recall_lru. If we then exit nfs4_set_delegation without hashing the delegation, it will be freed as soon as the callback is done with it, without ever being removed from del_recall_lru.

Symptoms show up later as use-after-free or list corruption warnings, usually in the laundromat thread.

I suspect aba2072f4523 "nfsd: grant read delegations to clients holding writes" made this bug easier to hit, but I looked as far back as v3.0 and it looks to me it already had the same problem. So I'm not sure where the bug was introduced; it may have been there from the beginning.

[Salvatore Bonaccorso: Backport for context changes to versions which do not have 20b7d86f29d3 ("nfsd: use boottime for lease expiry calculation")]

---
 fs/nfsd/nfs4state.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

@@ -1041,6 +1041,11 @@ hash_delegation_locked(struct nfs4_deleg
 	return 0;
 }
 
+static bool delegation_hashed(struct nfs4_delegation *dp)
+{
+	return !(list_empty(&dp->dl_perfile));
+}
+
 static bool
 unhash_delegation_locked(struct nfs4_delegation *dp)
 {

@@ -1048,7 +1053,7 @@ unhash_delegation_locked(struct nfs4_del
 
 	lockdep_assert_held(&state_lock);
 
-	if (list_empty(&dp->dl_perfile))
+	if (!delegation_hashed(dp))
 		return false;
 
 	dp->dl_stid.sc_type = NFS4_CLOSED_DELEG_STID;

@@ -4406,7 +4411,7 @@ static void nfsd4_cb_recall_prepare(stru
 	 * queued for a lease break. Don't queue it again.
 	 */
 	spin_lock(&state_lock);
-	if (dp->dl_time == 0) {
+	if (delegation_hashed(dp) && dp->dl_time == 0) {
 		dp->dl_time = get_seconds();
 		list_add_tail(&dp->dl_recall_lru, &nn->del_recall_lru);
 	}

From: Artem Lapkin <[email protected]>

[ Upstream commit 6dd0053683804427529ef3523f7872f473440a19 ]

Remove mmc-hs400-enhanced-strobe from the rk3399-khadas-edge dts to improve compatibility with a wider range of eMMC chips.

Before (BJTD4R 29.1 GiB):

[    7.001493] mmc2: CQHCI version 5.10
[    7.027971] mmc2: SDHCI controller on fe330000.mmc [fe330000.mmc] using ADMA
.......
[    7.207086] mmc2: mmc_select_hs400es failed, error -110
[    7.207129] mmc2: error -110 whilst initialising MMC card
[    7.308893] mmc2: mmc_select_hs400es failed, error -110
[    7.308921] mmc2: error -110 whilst initialising MMC card
[    7.427524] mmc2: mmc_select_hs400es failed, error -110
[    7.427546] mmc2: error -110 whilst initialising MMC card
[    7.590993] mmc2: mmc_select_hs400es failed, error -110
[    7.591012] mmc2: error -110 whilst initialising MMC card

After:

[    6.960785] mmc2: CQHCI version 5.10
[    6.984672] mmc2: SDHCI controller on fe330000.mmc [fe330000.mmc] using ADMA
[    7.175021] mmc2: Command Queue Engine enabled
[    7.175053] mmc2: new HS400 MMC card at address 0001
[    7.175808] mmcblk2: mmc2:0001 BJTD4R 29.1 GiB
[    7.176033] mmcblk2boot0: mmc2:0001 BJTD4R 4.00 MiB
[    7.176245] mmcblk2boot1: mmc2:0001 BJTD4R 4.00 MiB
[    7.176495] mmcblk2rpmb: mmc2:0001 BJTD4R 4.00 MiB, chardev (242:0)

---
 arch/arm64/boot/dts/rockchip/rk3399-khadas-edge.dtsi | 1 -
 1 file changed, 1 deletion(-)

index e87a04477440e..292ca70c512b5 100644
@@ -685,7 +685,6 @@ &sdmmc {
 &sdhci {
 	bus-width = <8>;
 	mmc-hs400-1_8v;
-	mmc-hs400-enhanced-strobe;
 	non-removable;
 	status = "okay";
 };

2.33.0

From: John Keeping <[email protected]>

[ Upstream commit 2b454a90e2ccdd6e03f88f930036da4df577be76 ]

Correct a typo in the vin-supply property. The input supply is always-on, so this mistake doesn't affect whether the supply is actually enabled correctly.

---
 arch/arm64/boot/dts/rockchip/rk3399-leez-p710.dts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 73be38a537960..a72e77c261ef3 100644
@@ -49,7 +49,7 @@ vcc3v3_lan: vcc3v3-lan {
 		regulator-boot-on;
 		regulator-min-microvolt = <3300000>;
 		regulator-max-microvolt = <3300000>;
-		vim-supply = <&vcc3v3_sys>;
+		vin-supply = <&vcc3v3_sys>;
 	};
 
 	vcc3v3_sys: vcc3v3-sys {

2.33.0

From: Alex Bee <[email protected]>

[ Upstream commit 8240e87f16d17a9592c9d67857a3dcdbcb98f10d ]

As stated in the schematics [1] and [2] P5 the APIO5 domain is supplied by RK808-D Buck4, which in our case vcc1v8_codec - i.e. a 1.8 V regulator.

Currently only white noise comes from the ES8316's output, which - for whatever reason - came up only after the the correct switch from i2s0_8ch_bus to i2s0_2ch_bus for i2s0's pinctrl was done.

Fix this by setting the correct regulator for audio-supply.

[1] https://dl.radxa.com/rockpi4/docs/hw/rockpi4/rockpi4_v13_sch_20181112.pdf
[2] https://dl.radxa.com/rockpi4/docs/hw/rockpi4/rockpi_4c_v12_sch_20200620.pdf

---
 arch/arm64/boot/dts/rockchip/rk3399-rock-pi-4.dts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 1ae1ebd4efdd0..da3b031d4befa 100644
@@ -452,7 +452,7 @@ &io_domains {
 	status = "okay";
 
 	bt656-supply = <&vcc_3v0>;
-	audio-supply = <&vcc_3v0>;
+	audio-supply = <&vcc1v8_codec>;
 	sdmmc-supply = <&vcc_sdio>;
 	gpio1830-supply = <&vcc_3v0>;
 };

2.33.0

From: Johannes Berg <[email protected]>

[ Upstream commit d5e568c3a4ec2ddd23e7dc5ad5b0c64e4f22981a ]

For admission control, obviously all of that only works for QoS data frames, otherwise we cannot even access the QoS field in the header.

Syzbot reported (see below) an uninitialized value here due to a status of a non-QoS nullfunc packet, which isn't even long enough to contain the QoS header.

Fix this to only do anything for QoS data packets.

---
 net/mac80211/mlme.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

index ccaf2389ccc1d..5c727af01143f 100644
@@ -2418,11 +2418,18 @@ static void ieee80211_sta_tx_wmm_ac_notify(struct ieee80211_sub_if_data *sdata,
 					   u16 tx_time)
 {
 	struct ieee80211_if_managed *ifmgd = &sdata->u.mgd;
-	u16 tid = ieee80211_get_tid(hdr);
-	int ac = ieee80211_ac_from_tid(tid);
-	struct ieee80211_sta_tx_tspec *tx_tspec = &ifmgd->tx_tspec[ac];
+	u16 tid;
+	int ac;
+	struct ieee80211_sta_tx_tspec *tx_tspec;
 	unsigned long now = jiffies;
 
+	if (!ieee80211_is_data_qos(hdr->frame_control))
+		return;
+
+	tid = ieee80211_get_tid(hdr);
+	ac = ieee80211_ac_from_tid(tid);
+	tx_tspec = &ifmgd->tx_tspec[ac];
+
 	if (likely(!tx_tspec->admitted_time))
 		return;
 

2.33.0

From: Randy Dunlap <[email protected]>

[ Upstream commit 1dc2f2b81a6a9895da59f3915760f6c0c3074492 ]

The hyperv utilities use PTP clock interfaces and should depend a a kconfig symbol such that they will be built as a loadable module or builtin so that linker errors do not happen.

Prevents these build errors:

ld: drivers/hv/hv_util.o: in function `hv_timesync_deinit':
hv_util.c:(.text+0x37d): undefined reference to `ptp_clock_unregister'
ld: drivers/hv/hv_util.o: in function `hv_timesync_init':
hv_util.c:(.text+0x738): undefined reference to `ptp_clock_register'

---
 drivers/hv/Kconfig | 1 +
 1 file changed, 1 insertion(+)

index 79e5356a737a2..210e532ac277f 100644
@@ -17,6 +17,7 @@ config HYPERV_TIMER
 config HYPERV_UTILS
 	tristate "Microsoft Hyper-V Utilities driver"
 	depends on HYPERV && CONNECTOR && NLS
+	depends on PTP_1588_CLOCK_OPTIONAL
 	help
 	  Select this option to enable the Hyper-V Utilities.
 

2.33.0

From: Dinh Nguyen <[email protected]>

[ Upstream commit cb25b11943cbcc5a34531129952870420f8be858 ]

The QSPI flash node needs to have the required "jedec,spi-nor" in the compatible string.

---
 arch/arm/boot/dts/socfpga_arria10_socdk_qspi.dts   | 2 +-
 arch/arm/boot/dts/socfpga_arria5_socdk.dts         | 2 +-
 arch/arm/boot/dts/socfpga_cyclone5_socdk.dts       | 2 +-
 arch/arm/boot/dts/socfpga_cyclone5_sockit.dts      | 2 +-
 arch/arm/boot/dts/socfpga_cyclone5_socrates.dts    | 2 +-
 arch/arm/boot/dts/socfpga_cyclone5_sodia.dts       | 2 +-
 arch/arm/boot/dts/socfpga_cyclone5_vining_fpga.dts | 4 ++--
 7 files changed, 8 insertions(+), 8 deletions(-)

index b4c0a76a4d1af..4c2fcfcc7baed 100644
@@ -12,7 +12,7 @@ &qspi {
 	flash0: n25q00@0 {
 		#address-cells = <1>;
 		#size-cells = <1>;
-		compatible = "n25q00aa";
+		compatible = "micron,mt25qu02g", "jedec,spi-nor";
 		reg = <0>;
 		spi-max-frequency = <100000000>;
 

index 90e676e7019f2..1b02d46496a85 100644
@@ -119,7 +119,7 @@ &qspi {
 	flash: flash@0 {
 		#address-cells = <1>;
 		#size-cells = <1>;
-		compatible = "n25q256a";
+		compatible = "micron,n25q256a", "jedec,spi-nor";
 		reg = <0>;
 		spi-max-frequency = <100000000>;
 

index 6f138b2b26163..51bb436784e24 100644
@@ -124,7 +124,7 @@ &qspi {
 	flash0: n25q00@0 {
 		#address-cells = <1>;
 		#size-cells = <1>;
-		compatible = "n25q00";
+		compatible = "micron,mt25qu02g", "jedec,spi-nor";
 		reg = <0>;	/* chip select */
 		spi-max-frequency = <100000000>;
 

index c155ff02eb6e0..cae9ddd5ed38b 100644
@@ -169,7 +169,7 @@ &qspi {
 	flash: flash@0 {
 		#address-cells = <1>;
 		#size-cells = <1>;
-		compatible = "n25q00";
+		compatible = "micron,mt25qu02g", "jedec,spi-nor";
 		reg = <0>;
 		spi-max-frequency = <100000000>;
 

index 8d5d3996f6f27..ca18b959e6559 100644
@@ -80,7 +80,7 @@ &qspi {
 	flash: flash@0 {
 		#address-cells = <1>;
 		#size-cells = <1>;
-		compatible = "n25q256a";
+		compatible = "micron,n25q256a", "jedec,spi-nor";
 		reg = <0>;
 		spi-max-frequency = <100000000>;
 		m25p,fast-read;

index 99a71757cdf46..3f7aa7bf0863a 100644
@@ -116,7 +116,7 @@ &qspi {
 	flash0: n25q512a@0 {
 		#address-cells = <1>;
 		#size-cells = <1>;
-		compatible = "n25q512a";
+		compatible = "micron,n25q512a", "jedec,spi-nor";
 		reg = <0>;
 		spi-max-frequency = <100000000>;
 

index a060718758b67..25874e1b9c829 100644
@@ -224,7 +224,7 @@ &qspi {
 	n25q128@0 {
 		#address-cells = <1>;
 		#size-cells = <1>;
-		compatible = "n25q128";
+		compatible = "micron,n25q128", "jedec,spi-nor";
 		reg = <0>;		/* chip select */
 		spi-max-frequency = <100000000>;
 		m25p,fast-read;

@@ -241,7 +241,7 @@ n25q128@0 {
 	n25q00@1 {
 		#address-cells = <1>;
 		#size-cells = <1>;
-		compatible = "n25q00";
+		compatible = "micron,mt25qu02g", "jedec,spi-nor";
 		reg = <1>;		/* chip select */
 		spi-max-frequency = <100000000>;
 		m25p,fast-read;

2.33.0

From: Mike Tipton <[email protected]>

[ Upstream commit 54baf56eaa40aa5cdcd02b3c20d593e4e1211220 ]

Before commit fc0c209c147f ("clk: Allow parents to be specified without string names") child clks couldn't find their parent until the parent clk was added to a list in __clk_core_init(). After that commit, child clks can reference their parent clks directly via a clk_hw pointer, or they can lookup that clk_hw pointer via DT if the parent clk is registered with an OF clk provider.

The common clk framework treats hw->core being non-NULL as "the clk is registered" per the logic within clk_core_fill_parent_index():

	parent = entry->hw->core;
	/*
	 * We have a direct reference but it isn't registered yet?
	 * Orphan it and let clk_reparent() update the orphan status
	 * when the parent is registered.
	 */
	if (!parent)

Therefore we need to be extra careful to not set hw->core until the clk is fully registered with the clk framework. Otherwise we can get into a situation where a child finds a parent clk and we move the child clk off the orphan list when the parent isn't actually registered, wrecking our enable accounting and breaking critical clks.

Consider the following scenario:

CPU0 CPU1
---- ----
struct clk_hw clkBad;
struct clk_hw clkA;

  clkA.init.parent_hws = { &clkBad };

clk_hw_register(&clkA) clk_hw_register(&clkBad)
... __clk_register()

					     hw->core = core
					     ...
   __clk_register()
    __clk_core_init()
     clk_prepare_lock()
     __clk_init_parent()
      clk_core_get_parent_by_index()
       clk_core_fill_parent_index()
        if (entry->hw) {
	 parent = entry->hw->core;

At this point, 'parent' points to clkBad even though clkBad hasn't been fully registered yet. Ouch! A similar problem can happen if a clk controller registers orphan clks that are referenced in the DT node of another clk controller.

Let's fix all this by only setting the hw->core pointer underneath the clk prepare lock in __clk_core_init(). This way we know that clk_core_fill_parent_index() can't see hw->core be non-NULL until the clk is fully registered.

---
 drivers/clk/clk.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

index 6ff87cd867121..e4e1b4e94a67b 100644
@@ -3299,6 +3299,14 @@ static int __clk_core_init(struct clk_core *core)
 
 	clk_prepare_lock();
 
+	/*
+	 * Set hw->core after grabbing the prepare_lock to synchronize with
+	 * callers of clk_core_fill_parent_index() where we treat hw->core
+	 * being NULL as the clk not being registered yet. This is crucial so
+	 * that clks aren't parented until their parent is fully registered.
+	 */
+	core->hw->core = core;
+
 	ret = clk_pm_runtime_get(core);
 	if (ret)
 		goto unlock;

@@ -3452,8 +3460,10 @@ static int __clk_core_init(struct clk_core *core)
 out:
 	clk_pm_runtime_put(core);
 unlock:
-	if (ret)
+	if (ret) {
 		hlist_del_init(&core->child_node);
+		core->hw->core = NULL;
+	}
 
 	clk_prepare_unlock();
 

@@ -3699,7 +3709,6 @@ __clk_register(struct device *dev, struct device_node *np, struct clk_hw *hw)
 	core->num_parents = init->num_parents;
 	core->min_rate = 0;
 	core->max_rate = ULONG_MAX;
-	hw->core = core;
 
 	ret = clk_core_populate_parent_map(core, init);
 	if (ret)

@@ -3717,7 +3726,7 @@ __clk_register(struct device *dev, struct device_node *np, struct clk_hw *hw)
 		goto fail_create_clk;
 	}
 
-	clk_core_link_consumer(hw->core, hw->clk);
+	clk_core_link_consumer(core, hw->clk);
 
 	ret = __clk_core_init(core);
 	if (!ret)

2.33.0

From: Jie2x Zhou <[email protected]>

[ Upstream commit 92816e2629808726af015c7f5b14adc8e4f8b147 ]

./fcnal-test.sh -v -t ipv6_ping
TEST: ping out, VRF bind - ns-B IPv6 LLA [FAIL]
TEST: ping out, VRF bind - multicast IP [FAIL]

ping6 is failing as it should.
COMMAND: ip netns exec ns-A /bin/ping6 -c1 -w1 fe80::7c4c:bcff:fe66:a63a%red strace of ping6 shows it is failing with '1',
so change the expected rc from 2 to 1.

---
 tools/testing/selftests/net/fcnal-test.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 782a8da5d9500..475ac62373e92 100755
@@ -1884,7 +1884,7 @@ ipv6_ping_vrf()
 		log_start
 		show_hint "Fails since VRF device does not support linklocal or multicast"
 		run_cmd ${ping6} -c1 -w1 ${a}
-		log_test_addr ${a} $? 2 "ping out, VRF bind"
+		log_test_addr ${a} $? 1 "ping out, VRF bind"
 	done
 
 	for a in ${NSB_IP6} ${NSB_LO_IP6} ${NSB_LINKIP6}%${NSA_DEV} ${MCAST}%${NSA_DEV}

2.33.0

From: Philipp Rudo <[email protected]>

[ Upstream commit 41967a37b8eedfee15b81406a9f3015be90d3980 ]

arch_kexec_apply_relocations_add currently ignores all errors returned by arch_kexec_do_relocs. This means that every unknown relocation is silently skipped causing unpredictable behavior while the relocated code runs. Fix this by checking for errors and fail kexec_file_load if an unknown relocation type is encountered.

The problem was found after gcc changed its behavior and used R_390_PLT32DBL relocations for brasl instruction and relied on ld to resolve the relocations in the final link in case direct calls are possible. As the purgatory code is only linked partially (option -r) ld didn't resolve the relocations leaving them for arch_kexec_do_relocs. But arch_kexec_do_relocs doesn't know how to handle R_390_PLT32DBL relocations so they were silently skipped. This ultimately caused an endless loop in the purgatory as the brasl instructions kept branching to itself.

---
 arch/s390/kernel/machine_kexec_file.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

index e7435f3a3d2d2..76cd09879eaf4 100644
@@ -277,6 +277,7 @@ int arch_kexec_apply_relocations_add(struct purgatory_info *pi,
 {
 	Elf_Rela *relas;
 	int i, r_type;
+	int ret;
 
 	relas = (void *)pi->ehdr + relsec->sh_offset;
 

@@ -311,7 +312,11 @@ int arch_kexec_apply_relocations_add(struct purgatory_info *pi,
 		addr = section->sh_addr + relas[i].r_offset;
 
 		r_type = ELF64_R_TYPE(relas[i].r_info);
-		arch_kexec_do_relocs(r_type, loc, val, addr);
+		ret = arch_kexec_do_relocs(r_type, loc, val, addr);
+		if (ret) {
+			pr_err("Unknown rela relocation: %d\n", r_type);
+			return -ENOEXEC;
+		}
 	}
 	return 0;
 }

2.33.0

From: Eric Dumazet <[email protected]>

[ Upstream commit ab443c53916730862cec202078d36fd4008bea79 ]

qdiscs are not supposed to call their own destroy() method from init(), because core stack already does that.

syzbot was able to trigger use after free:

DEBUG_LOCKS_WARN_ON(lock->magic != lock)

WARNING: CPU: 0 PID: 21902 at kernel/locking/mutex.c:586 __mutex_lock_common kernel/locking/mutex.c:586 [inline]
WARNING: CPU: 0 PID: 21902 at kernel/locking/mutex.c:586 __mutex_lock+0x9ec/0x12f0 kernel/locking/mutex.c:740
Modules linked in:
CPU: 0 PID: 21902 Comm: syz-executor189 Not tainted 5.16.0-rc4-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:__mutex_lock_common kernel/locking/mutex.c:586 [inline]
RIP: 0010:__mutex_lock+0x9ec/0x12f0 kernel/locking/mutex.c:740
Code: 08 84 d2 0f 85 19 08 00 00 8b 05 97 38 4b 04 85 c0 0f 85 27 f7 ff ff 48 c7 c6 20 00 ac 89 48 c7 c7 a0 fe ab 89 e8 bf 76 ba ff <0f> 0b e9 0d f7 ff ff 48 8b 44 24 40 48 8d b8 c8 08 00 00 48 89 f8
RSP: 0018:ffffc9000627f290 EFLAGS: 00010282
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: ffff88802315d700 RSI: ffffffff815f1db8 RDI: fffff52000c4fe44
RBP: ffff88818f28e000 R08: 0000000000000000 R09: 0000000000000000
R10: ffffffff815ebb5e R11: 0000000000000000 R12: 0000000000000000
R13: dffffc0000000000 R14: ffffc9000627f458 R15: 0000000093c30000
FS:  0000555556abc400(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fda689c3303 CR3: 000000001cfbb000 CR4: 0000000000350ef0
Call Trace:
 <TASK>
 tcf_chain0_head_change_cb_del+0x2e/0x3d0 net/sched/cls_api.c:810
 tcf_block_put_ext net/sched/cls_api.c:1381 [inline]
 tcf_block_put_ext net/sched/cls_api.c:1376 [inline]
 tcf_block_put+0xbc/0x130 net/sched/cls_api.c:1394
 cake_destroy+0x3f/0x80 net/sched/sch_cake.c:2695
 qdisc_create.constprop.0+0x9da/0x10f0 net/sched/sch_api.c:1293
 tc_modify_qdisc+0x4c5/0x1980 net/sched/sch_api.c:1660
 rtnetlink_rcv_msg+0x413/0xb80 net/core/rtnetlink.c:5571
 netlink_rcv_skb+0x153/0x420 net/netlink/af_netlink.c:2496
 netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline]
 netlink_unicast+0x533/0x7d0 net/netlink/af_netlink.c:1345
 netlink_sendmsg+0x904/0xdf0 net/netlink/af_netlink.c:1921
 sock_sendmsg_nosec net/socket.c:704 [inline]
 sock_sendmsg+0xcf/0x120 net/socket.c:724
 ____sys_sendmsg+0x6e8/0x810 net/socket.c:2409
 ___sys_sendmsg+0xf3/0x170 net/socket.c:2463
 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2492
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f1bb06badb9
Code: Unable to access opcode bytes at RIP 0x7f1bb06bad8f.
RSP: 002b:00007fff3012a658 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f1bb06badb9
RDX: 0000000000000000 RSI: 00000000200007c0 RDI: 0000000000000003
RBP: 0000000000000000 R08: 0000000000000003 R09: 0000000000000003
R10: 0000000000000003 R11: 0000000000000246 R12: 00007fff3012a688
R13: 00007fff3012a6a0 R14: 00007fff3012a6e0 R15: 00000000000013c2
 </TASK>

---
 net/sched/sch_cake.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

index e8eebe40e0ae9..0eb4d4a568f77 100644
@@ -2724,7 +2724,7 @@ static int cake_init(struct Qdisc *sch, struct nlattr *opt,
 	q->tins = kvcalloc(CAKE_MAX_TINS, sizeof(struct cake_tin_data),
 			   GFP_KERNEL);
 	if (!q->tins)
-		goto nomem;
+		return -ENOMEM;
 
 	for (i = 0; i < CAKE_MAX_TINS; i++) {
 		struct cake_tin_data *b = q->tins + i;

@@ -2754,10 +2754,6 @@ static int cake_init(struct Qdisc *sch, struct nlattr *opt,
 	q->min_netlen = ~0;
 	q->min_adjlen = ~0;
 	return 0;
-
-nomem:
-	cake_destroy(sch);
-	return -ENOMEM;
 }
 
 static int cake_dump(struct Qdisc *sch, struct sk_buff *skb)

2.33.0

From: Eric Dumazet <[email protected]>

[ Upstream commit 3828a93f5cfdf5d8a4ff9dead741e9a2871ff57b ]

Use jiffies_delta_to_msecs() to avoid reporting 'infinite' timeouts and to cleanup code.

---
 net/ipv4/inet_diag.c | 15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)

index 4f71aca156662..6f8118b29ba51 100644
@@ -240,17 +240,17 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 		r->idiag_timer = 1;
 		r->idiag_retrans = icsk->icsk_retransmits;
 		r->idiag_expires =
-			jiffies_to_msecs(icsk->icsk_timeout - jiffies);
+			jiffies_delta_to_msecs(icsk->icsk_timeout - jiffies);
 	} else if (icsk->icsk_pending == ICSK_TIME_PROBE0) {
 		r->idiag_timer = 4;
 		r->idiag_retrans = icsk->icsk_probes_out;
 		r->idiag_expires =
-			jiffies_to_msecs(icsk->icsk_timeout - jiffies);
+			jiffies_delta_to_msecs(icsk->icsk_timeout - jiffies);
 	} else if (timer_pending(&sk->sk_timer)) {
 		r->idiag_timer = 2;
 		r->idiag_retrans = icsk->icsk_probes_out;
 		r->idiag_expires =
-			jiffies_to_msecs(sk->sk_timer.expires - jiffies);
+			jiffies_delta_to_msecs(sk->sk_timer.expires - jiffies);
 	} else {
 		r->idiag_timer = 0;
 		r->idiag_expires = 0;

@@ -338,16 +338,13 @@ static int inet_twsk_diag_fill(struct sock *sk,
 	r = nlmsg_data(nlh);
 	BUG_ON(tw->tw_state != TCP_TIME_WAIT);
 
-	tmo = tw->tw_timer.expires - jiffies;
-	if (tmo < 0)
-		tmo = 0;
-
 	inet_diag_msg_common_fill(r, sk);
 	r->idiag_retrans      = 0;
 
 	r->idiag_state	      = tw->tw_substate;
 	r->idiag_timer	      = 3;
-	r->idiag_expires      = jiffies_to_msecs(tmo);
+	tmo = tw->tw_timer.expires - jiffies;
+	r->idiag_expires      = jiffies_delta_to_msecs(tmo);
 	r->idiag_rqueue	      = 0;
 	r->idiag_wqueue	      = 0;
 	r->idiag_uid	      = 0;

@@ -381,7 +378,7 @@ static int inet_req_diag_fill(struct sock *sk, struct sk_buff *skb,
 		     offsetof(struct sock, sk_cookie));
 
 	tmo = inet_reqsk(sk)->rsk_timer.expires - jiffies;
-	r->idiag_expires = (tmo >= 0) ? jiffies_to_msecs(tmo) : 0;
+	r->idiag_expires = jiffies_delta_to_msecs(tmo);
 	r->idiag_rqueue	= 0;
 	r->idiag_wqueue	= 0;
 	r->idiag_uid	= 0;

2.33.0

From: Eric Dumazet <[email protected]>

[ Upstream commit 71ddeac8cd1d217744a0e060ff520e147c9328d1 ]

KMSAN reported a kernel-infoleak [1], that can exploited by unpriv users.

After analysis it turned out UDP was not initializing r->idiag_expires. Other users of inet_sk_diag_fill() might make the same mistake in the future, so fix this in inet_sk_diag_fill().

[1]

BUG: KMSAN: kernel-infoleak in instrument_copy_to_user include/linux/instrumented.h:121 [inline]
BUG: KMSAN: kernel-infoleak in copyout lib/iov_iter.c:156 [inline]
BUG: KMSAN: kernel-infoleak in _copy_to_iter+0x69d/0x25c0 lib/iov_iter.c:670
 instrument_copy_to_user include/linux/instrumented.h:121 [inline]
 copyout lib/iov_iter.c:156 [inline]
 _copy_to_iter+0x69d/0x25c0 lib/iov_iter.c:670
 copy_to_iter include/linux/uio.h:155 [inline]
 simple_copy_to_iter+0xf3/0x140 net/core/datagram.c:519
 __skb_datagram_iter+0x2cb/0x1280 net/core/datagram.c:425
 skb_copy_datagram_iter+0xdc/0x270 net/core/datagram.c:533
 skb_copy_datagram_msg include/linux/skbuff.h:3657 [inline]
 netlink_recvmsg+0x660/0x1c60 net/netlink/af_netlink.c:1974
 sock_recvmsg_nosec net/socket.c:944 [inline]
 sock_recvmsg net/socket.c:962 [inline]
 sock_read_iter+0x5a9/0x630 net/socket.c:1035
 call_read_iter include/linux/fs.h:2156 [inline]
 new_sync_read fs/read_write.c:400 [inline]
 vfs_read+0x1631/0x1980 fs/read_write.c:481
 ksys_read+0x28c/0x520 fs/read_write.c:619
 __do_sys_read fs/read_write.c:629 [inline]
 __se_sys_read fs/read_write.c:627 [inline]
 __x64_sys_read+0xdb/0x120 fs/read_write.c:627
 do_syscall_x64 arch/x86/entry/common.c:51 [inline]
 do_syscall_64+0x54/0xd0 arch/x86/entry/common.c:82
 entry_SYSCALL_64_after_hwframe+0x44/0xae

Uninit was created at:

 slab_post_alloc_hook mm/slab.h:524 [inline]
 slab_alloc_node mm/slub.c:3251 [inline]
 __kmalloc_node_track_caller+0xe0c/0x1510 mm/slub.c:4974
 kmalloc_reserve net/core/skbuff.c:354 [inline]
 __alloc_skb+0x545/0xf90 net/core/skbuff.c:426
 alloc_skb include/linux/skbuff.h:1126 [inline]
 netlink_dump+0x3d5/0x16a0 net/netlink/af_netlink.c:2245
 __netlink_dump_start+0xd1c/0xee0 net/netlink/af_netlink.c:2370
 netlink_dump_start include/linux/netlink.h:254 [inline]
 inet_diag_handler_cmd+0x2e7/0x400 net/ipv4/inet_diag.c:1343
 sock_diag_rcv_msg+0x24a/0x620
 netlink_rcv_skb+0x447/0x800 net/netlink/af_netlink.c:2491
 sock_diag_rcv+0x63/0x80 net/core/sock_diag.c:276
 netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline]
 netlink_unicast+0x1095/0x1360 net/netlink/af_netlink.c:1345
 netlink_sendmsg+0x16f3/0x1870 net/netlink/af_netlink.c:1916
 sock_sendmsg_nosec net/socket.c:704 [inline]
 sock_sendmsg net/socket.c:724 [inline]
 sock_write_iter+0x594/0x690 net/socket.c:1057
 do_iter_readv_writev+0xa7f/0xc70
 do_iter_write+0x52c/0x1500 fs/read_write.c:851
 vfs_writev fs/read_write.c:924 [inline]
 do_writev+0x63f/0xe30 fs/read_write.c:967
 __do_sys_writev fs/read_write.c:1040 [inline]
 __se_sys_writev fs/read_write.c:1037 [inline]
 __x64_sys_writev+0xe5/0x120 fs/read_write.c:1037
 do_syscall_x64 arch/x86/entry/common.c:51 [inline]
 do_syscall_64+0x54/0xd0 arch/x86/entry/common.c:82
 entry_SYSCALL_64_after_hwframe+0x44/0xae

Bytes 68-71 of 312 are uninitialized
Memory access of size 312 starts at ffff88812ab54000 Data copied to user address 0000000020001440

CPU: 1 PID: 6365 Comm: syz-executor801 Not tainted 5.16.0-rc3-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011

---
 net/ipv4/inet_diag.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

index 6f8118b29ba51..f8f79672cc5f3 100644
@@ -200,6 +200,7 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 	r->idiag_state = sk->sk_state;
 	r->idiag_timer = 0;
 	r->idiag_retrans = 0;
+	r->idiag_expires = 0;
 
 	if (inet_diag_msg_attrs_fill(sk, skb, r, ext, user_ns, net_admin))
 		goto errout;

@@ -251,9 +252,6 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 		r->idiag_retrans = icsk->icsk_probes_out;
 		r->idiag_expires =
 			jiffies_delta_to_msecs(sk->sk_timer.expires - jiffies);
-	} else {
-		r->idiag_timer = 0;
-		r->idiag_expires = 0;
 	}
 
 	if ((ext & (1 << (INET_DIAG_INFO - 1))) && handler->idiag_info_size) {

2.33.0

From: David Ahern <[email protected]>

[ Upstream commit 0f108ae4452025fef529671998f6c7f1c4526790 ]

Commit referenced below added negative socket bind tests for VRF. The socket binds should fail since the address to bind to is in a VRF yet the socket is not bound to the VRF or a device within it. Update the expected return code to check for 1 (bind failure) so the test passes when the bind fails as expected. Add a 'show_hint' comment to explain why the bind is expected to fail.

---
 tools/testing/selftests/net/fcnal-test.sh | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index 475ac62373e92..c8f28e847ee9e 100755
@@ -1491,8 +1491,9 @@ ipv4_addr_bind_vrf()
 	for a in ${NSA_IP} ${VRF_IP}
 	do
 		log_start
+		show_hint "Socket not bound to VRF, but address is in VRF"
 		run_cmd nettest -s -R -P icmp -l ${a} -b
-		log_test_addr ${a} $? 0 "Raw socket bind to local address"
+		log_test_addr ${a} $? 1 "Raw socket bind to local address"
 
 		log_start
 		run_cmd nettest -s -R -P icmp -l ${a} -d ${NSA_DEV} -b

2.33.0

From: David Ahern <[email protected]>

[ Upstream commit 28a2686c185e84b6aa6a4d9c9a972360eb7ca266 ]

IPv6 allows binding a socket to a device then binding to an address not on the device (__inet6_bind -> ipv6_chk_addr with strict flag not set). Update the bind tests to reflect legacy behavior.

---
 tools/testing/selftests/net/fcnal-test.sh | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

index c8f28e847ee9e..157822331954d 100755
@@ -2891,11 +2891,14 @@ ipv6_addr_bind_novrf()
 	run_cmd nettest -6 -s -l ${a} -d ${NSA_DEV} -t1 -b
 	log_test_addr ${a} $? 0 "TCP socket bind to local address after device bind"
 
+	# Sadly, the kernel allows binding a socket to a device and then
+	# binding to an address not on the device. So this test passes
+	# when it really should not
 	a=${NSA_LO_IP6}
 	log_start
-	show_hint "Should fail with 'Cannot assign requested address'"
-	run_cmd nettest -6 -s -l ${a} -d ${NSA_DEV} -t1 -b
-	log_test_addr ${a} $? 1 "TCP socket bind to out of scope local address"
+	show_hint "Tecnically should fail since address is not on device but kernel allows"
+	run_cmd nettest -6 -s -l ${a} -I ${NSA_DEV} -t1 -b
+	log_test_addr ${a} $? 0 "TCP socket bind to out of scope local address"
 }
 
 ipv6_addr_bind_vrf()

@@ -2936,10 +2939,15 @@ ipv6_addr_bind_vrf()
 	run_cmd nettest -6 -s -l ${a} -d ${NSA_DEV} -t1 -b
 	log_test_addr ${a} $? 0 "TCP socket bind to local address with device bind"
 
+	# Sadly, the kernel allows binding a socket to a device and then
+	# binding to an address not on the device. The only restriction
+	# is that the address is valid in the L3 domain. So this test
+	# passes when it really should not
 	a=${VRF_IP6}
 	log_start
-	run_cmd nettest -6 -s -l ${a} -d ${NSA_DEV} -t1 -b
-	log_test_addr ${a} $? 1 "TCP socket bind to VRF address with device bind"
+	show_hint "Tecnically should fail since address is not on device but kernel allows"
+	run_cmd nettest -6 -s -l ${a} -I ${NSA_DEV} -t1 -b
+	log_test_addr ${a} $? 0 "TCP socket bind to VRF address with device bind"
 
 	a=${NSA_LO_IP6}
 	log_start

2.33.0

From: Alyssa Ross <[email protected]>

[ Upstream commit 822c9f2b833c53fc67e8adf6f63ecc3ea24d502c ]

modprobe can't handle spaces in aliases.

---
 drivers/dma/st_fdma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 67087dbe2f9fa..f7393c19a1ba3 100644
@@ -873,4 +873,4 @@ MODULE_LICENSE("GPL v2");
 MODULE_DESCRIPTION("STMicroelectronics FDMA engine driver");
 MODULE_AUTHOR("Ludovic.barre <[email protected]>");
 MODULE_AUTHOR("Peter Griffin <[email protected]>");
-MODULE_ALIAS("platform: " DRIVER_NAME);
+MODULE_ALIAS("platform:" DRIVER_NAME);

2.33.0

From: Hangbin Liu <[email protected]>

[ Upstream commit 71da1aec215290e249d09c44c768df859f3a3bba ]

The recent GRE selftests defined NUM_NETIFS=10. If the users copy forwarding.config.sample to forwarding.config directly, they will get error "Command line is not complete" when run the GRE tests, because create_netif_veth() failed with no interface name defined.

Fix it by extending the NETIFS with p9 and p10.

---
 tools/testing/selftests/net/forwarding/forwarding.config.sample | 2 ++
 1 file changed, 2 insertions(+)

index e2adb533c8fcb..e71c61ee4cc67 100644
@@ -13,6 +13,8 @@ NETIFS[p5]=veth4
 NETIFS[p6]=veth5
 NETIFS[p7]=veth6
 NETIFS[p8]=veth7
+NETIFS[p9]=veth8
+NETIFS[p10]=veth9
 
 ##############################################################################
 # Defines

2.33.0

From: Mordechay Goodstein <[email protected]>

[ Upstream commit 31d8bb4e07f80935ee9bf599a9d99de7ca90fc5a ]

We move the actual arming the timer and sending ADDBA to a function for the use in different places calling the same logic.

---
 net/mac80211/agg-tx.c | 67 +++++++++++++++++++++++++------------------
 1 file changed, 39 insertions(+), 28 deletions(-)

index 3d2af1851bdf9..1a5768ae5f515 100644
@@ -9,7 +9,7 @@
  * Copyright 2007, Michael Wu <[email protected]>
  * Copyright 2007-2010, Intel Corporation
  * Copyright(c) 2015-2017 Intel Deutschland GmbH
- * Copyright (C) 2018 - 2019 Intel Corporation
+ * Copyright (C) 2018 - 2020 Intel Corporation
  */
 
 #include <linux/ieee80211.h>

@@ -448,6 +448,43 @@ static void sta_addba_resp_timer_expired(struct timer_list *t)
 	ieee80211_stop_tx_ba_session(&sta->sta, tid);
 }
 
+static void ieee80211_send_addba_with_timeout(struct sta_info *sta,
+					      struct tid_ampdu_tx *tid_tx)
+{
+	struct ieee80211_sub_if_data *sdata = sta->sdata;
+	struct ieee80211_local *local = sta->local;
+	u8 tid = tid_tx->tid;
+	u16 buf_size;
+
+	/* activate the timer for the recipient's addBA response */
+	mod_timer(&tid_tx->addba_resp_timer, jiffies + ADDBA_RESP_INTERVAL);
+	ht_dbg(sdata, "activated addBA response timer on %pM tid %d\n",
+	       sta->sta.addr, tid);
+
+	spin_lock_bh(&sta->lock);
+	sta->ampdu_mlme.last_addba_req_time[tid] = jiffies;
+	sta->ampdu_mlme.addba_req_num[tid]++;
+	spin_unlock_bh(&sta->lock);
+
+	if (sta->sta.he_cap.has_he) {
+		buf_size = local->hw.max_tx_aggregation_subframes;
+	} else {
+		/*
+		 * We really should use what the driver told us it will
+		 * transmit as the maximum, but certain APs (e.g. the
+		 * LinkSys WRT120N with FW v1.0.07 build 002 Jun 18 2012)
+		 * will crash when we use a lower number.
+		 */
+		buf_size = IEEE80211_MAX_AMPDU_BUF_HT;
+	}
+
+	/* send AddBA request */
+	ieee80211_send_addba_request(sdata, sta->sta.addr, tid,
+				     tid_tx->dialog_token,
+				     sta->tid_seq[tid] >> 4,
+				     buf_size, tid_tx->timeout);
+}
+
 void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid)
 {
 	struct tid_ampdu_tx *tid_tx;

@@ -462,7 +499,6 @@ void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid)
 		.timeout = 0,
 	};
 	int ret;
-	u16 buf_size;
 
 	tid_tx = rcu_dereference_protected_tid_tx(sta, tid);
 

@@ -501,32 +537,7 @@ void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid)
 		return;
 	}
 
-	/* activate the timer for the recipient's addBA response */
-	mod_timer(&tid_tx->addba_resp_timer, jiffies + ADDBA_RESP_INTERVAL);
-	ht_dbg(sdata, "activated addBA response timer on %pM tid %d\n",
-	       sta->sta.addr, tid);
-
-	spin_lock_bh(&sta->lock);
-	sta->ampdu_mlme.last_addba_req_time[tid] = jiffies;
-	sta->ampdu_mlme.addba_req_num[tid]++;
-	spin_unlock_bh(&sta->lock);
-
-	if (sta->sta.he_cap.has_he) {
-		buf_size = local->hw.max_tx_aggregation_subframes;
-	} else {
-		/*
-		 * We really should use what the driver told us it will
-		 * transmit as the maximum, but certain APs (e.g. the
-		 * LinkSys WRT120N with FW v1.0.07 build 002 Jun 18 2012)
-		 * will crash when we use a lower number.
-		 */
-		buf_size = IEEE80211_MAX_AMPDU_BUF_HT;
-	}
-
-	/* send AddBA request */
-	ieee80211_send_addba_request(sdata, sta->sta.addr, tid,
-				     tid_tx->dialog_token, params.ssn,
-				     buf_size, tid_tx->timeout);
+	ieee80211_send_addba_with_timeout(sta, tid_tx);
 }
 
 /*

2.33.0

From: Johannes Berg <[email protected]>

[ Upstream commit 06c41bda0ea14aa7fba932a9613c4ee239682cf0 ]

When we call ieee80211_agg_start_txq(), that will in turn call schedule_and_wake_txq(). Called from ieee80211_stop_tx_ba_cb() this is done under sta->lock, which leads to certain circular lock dependencies, as reported by Chris Murphy:
https://lore.kernel.org/r/CAJCQCtSXJ5qA4bqSPY=oLRMbv-irihVvP7A2uGutEbXQVkoNaw@mail.gmail.com

In general, ieee80211_agg_start_txq() is usually not called with sta->lock held, only in this one place. But it's always called with sta->ampdu_mlme.mtx held, and that's therefore clearly sufficient.

Change ieee80211_stop_tx_ba_cb() to also call it without the sta->lock held, by factoring it out of ieee80211_remove_tid_tx() (which is only called in this one place).

This breaks the locking chain and makes it less likely that we'll have similar locking chain problems in the future.

---
 net/mac80211/agg-tx.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

index 1a5768ae5f515..d801ceb2ed7fa 100644
@@ -9,7 +9,7 @@
  * Copyright 2007, Michael Wu <[email protected]>
  * Copyright 2007-2010, Intel Corporation
  * Copyright(c) 2015-2017 Intel Deutschland GmbH
- * Copyright (C) 2018 - 2020 Intel Corporation
+ * Copyright (C) 2018 - 2021 Intel Corporation
  */
 
 #include <linux/ieee80211.h>

@@ -213,6 +213,8 @@ ieee80211_agg_start_txq(struct sta_info *sta, int tid, bool enable)
 	struct ieee80211_txq *txq = sta->sta.txq[tid];
 	struct txq_info *txqi;
 
+	lockdep_assert_held(&sta->ampdu_mlme.mtx);
+
 	if (!txq)
 		return;
 

@@ -290,7 +292,6 @@ static void ieee80211_remove_tid_tx(struct sta_info *sta, int tid)
 	ieee80211_assign_tid_tx(sta, tid, NULL);
 
 	ieee80211_agg_splice_finish(sta->sdata, tid);
-	ieee80211_agg_start_txq(sta, tid, false);
 
 	kfree_rcu(tid_tx, rcu_head);
 }

@@ -871,6 +872,7 @@ void ieee80211_stop_tx_ba_cb(struct sta_info *sta, int tid,
 {
 	struct ieee80211_sub_if_data *sdata = sta->sdata;
 	bool send_delba = false;
+	bool start_txq = false;
 
 	ht_dbg(sdata, "Stopping Tx BA session for %pM tid %d\n",
 	       sta->sta.addr, tid);

@@ -888,10 +890,14 @@ void ieee80211_stop_tx_ba_cb(struct sta_info *sta, int tid,
 		send_delba = true;
 
 	ieee80211_remove_tid_tx(sta, tid);
+	start_txq = true;
 
  unlock_sta:
 	spin_unlock_bh(&sta->lock);
 
+	if (start_txq)
+		ieee80211_agg_start_txq(sta, tid, false);
+
 	if (send_delba)
 		ieee80211_send_delba(sdata, sta->sta.addr, tid,
 			WLAN_BACK_INITIATOR, WLAN_REASON_QSTA_NOT_USE);

2.33.0

From: Johannes Berg <[email protected]>

[ Upstream commit 93382a0d119b3ab95e3ebca51ea15aa87187b493 ]

On 6 GHz, stations don't have ht_supported set, but they can still do aggregation since they must have HE, allow that.

---
 net/mac80211/agg-rx.c | 5 +++--
 net/mac80211/agg-tx.c | 3 ++-
 2 files changed, 5 insertions(+), 3 deletions(-)

index 4d1c335e06e57..7f245e9f114c2 100644
@@ -9,7 +9,7 @@
  * Copyright 2007, Michael Wu <[email protected]>
  * Copyright 2007-2010, Intel Corporation
  * Copyright(c) 2015-2017 Intel Deutschland GmbH
- * Copyright (C) 2018        Intel Corporation
+ * Copyright (C) 2018-2020 Intel Corporation
  */
 
 /**

@@ -292,7 +292,8 @@ void ___ieee80211_start_rx_ba_session(struct sta_info *sta,
 		goto end;
 	}
 
-	if (!sta->sta.ht_cap.ht_supported) {
+	if (!sta->sta.ht_cap.ht_supported &&
+	    sta->sdata->vif.bss_conf.chandef.chan->band != NL80211_BAND_6GHZ) {
 		ht_dbg(sta->sdata,
 		       "STA %pM erroneously requests BA session on tid %d w/o QoS\n",
 		       sta->sta.addr, tid);

index d801ceb2ed7fa..8d3b905e551a3 100644
@@ -583,7 +583,8 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *pubsta, u16 tid,
 		 "Requested to start BA session on reserved tid=%d", tid))
 		return -EINVAL;
 
-	if (!pubsta->ht_cap.ht_supported)
+	if (!pubsta->ht_cap.ht_supported &&
+	    sta->sdata->vif.bss_conf.chandef.chan->band != NL80211_BAND_6GHZ)
 		return -EINVAL;
 
 	if (WARN_ON_ONCE(!local->ops->ampdu_action))

2.33.0

From: Johannes Berg <[email protected]>

[ Upstream commit 511ab0c1dfb260a6b17b8771109e8d63474473a7 ]

We should be doing the HE capabilities lookup based on the full interface type so if P2P doesn't have HE but client has it doesn't get confused. Fix that.

---
 net/mac80211/agg-rx.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

index 7f245e9f114c2..49ec9bfb6c8e6 100644
@@ -9,7 +9,7 @@
  * Copyright 2007, Michael Wu <[email protected]>
  * Copyright 2007-2010, Intel Corporation
  * Copyright(c) 2015-2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2020 Intel Corporation
+ * Copyright (C) 2018-2021 Intel Corporation
  */
 
 /**

@@ -191,7 +191,8 @@ static void ieee80211_add_addbaext(struct ieee80211_sub_if_data *sdata,
 	sband = ieee80211_get_sband(sdata);
 	if (!sband)
 		return;
-	he_cap = ieee80211_get_he_iftype_cap(sband, sdata->vif.type);
+	he_cap = ieee80211_get_he_iftype_cap(sband,
+					     ieee80211_vif_type_p2p(&sdata->vif));
 	if (!he_cap)
 		return;
 

2.33.0

From: Vlad Buslov <[email protected]>

[ Upstream commit 7a47281439ba00b11fc098f36695522184ce5a82 ]

In order to remove dependency on rtnl lock, take action's tcfa_lock when constructing its representation as flow_action_entry structure.

Refactor tcf_sample_get_group() to assume that caller holds tcf_lock and don't take it manually. This callback is only called from flow_action infra representation translator which now calls it with tcf_lock held, so this refactoring is necessary to prevent deadlock.

Allocate memory with GFP_ATOMIC flag for ip_tunnel_info copy because tcf_tunnel_info_copy() is only called from flow_action representation infra code with tcf_lock spinlock taken.

---
 include/net/tc_act/tc_tunnel_key.h |  2 +-
 net/sched/act_sample.c             |  2 --
 net/sched/cls_api.c                | 17 +++++++++++------
 3 files changed, 12 insertions(+), 9 deletions(-)

index 0689d9bcdf841..2b3df076e5b62 100644
@@ -69,7 +69,7 @@ tcf_tunnel_info_copy(const struct tc_action *a)
 	if (tun) {
 		size_t tun_size = sizeof(*tun) + tun->options_len;
 		struct ip_tunnel_info *tun_copy = kmemdup(tun, tun_size,
-							  GFP_KERNEL);
+							  GFP_ATOMIC);
 
 		return tun_copy;
 	}

index 74450b0f69fc5..214f4efdd9920 100644
@@ -265,14 +265,12 @@ tcf_sample_get_group(const struct tc_action *a,
 	struct tcf_sample *s = to_sample(a);
 	struct psample_group *group;
 
-	spin_lock_bh(&s->tcf_lock);
 	group = rcu_dereference_protected(s->psample_group,
 					  lockdep_is_held(&s->tcf_lock));
 	if (group) {
 		psample_group_take(group);
 		*destructor = tcf_psample_group_put;
 	}
-	spin_unlock_bh(&s->tcf_lock);
 
 	return group;
 }

index 7f20fd37e01e0..61aa63cc170b4 100644
@@ -3436,7 +3436,7 @@ static void tcf_sample_get_group(struct flow_action_entry *entry,
 int tc_setup_flow_action(struct flow_action *flow_action,
 			 const struct tcf_exts *exts, bool rtnl_held)
 {
-	const struct tc_action *act;
+	struct tc_action *act;
 	int i, j, k, err = 0;
 
 	if (!exts)

@@ -3450,6 +3450,7 @@ int tc_setup_flow_action(struct flow_action *flow_action,
 		struct flow_action_entry *entry;
 
 		entry = &flow_action->entries[j];
+		spin_lock_bh(&act->tcfa_lock);
 		if (is_tcf_gact_ok(act)) {
 			entry->id = FLOW_ACTION_ACCEPT;
 		} else if (is_tcf_gact_shot(act)) {

@@ -3490,13 +3491,13 @@ int tc_setup_flow_action(struct flow_action *flow_action,
 				break;
 			default:
 				err = -EOPNOTSUPP;
-				goto err_out;
+				goto err_out_locked;
 			}
 		} else if (is_tcf_tunnel_set(act)) {
 			entry->id = FLOW_ACTION_TUNNEL_ENCAP;
 			err = tcf_tunnel_encap_get_tunnel(entry, act);
 			if (err)
-				goto err_out;
+				goto err_out_locked;
 		} else if (is_tcf_tunnel_release(act)) {
 			entry->id = FLOW_ACTION_TUNNEL_DECAP;
 		} else if (is_tcf_pedit(act)) {

@@ -3510,7 +3511,7 @@ int tc_setup_flow_action(struct flow_action *flow_action,
 					break;
 				default:
 					err = -EOPNOTSUPP;
-					goto err_out;
+					goto err_out_locked;
 				}
 				entry->mangle.htype = tcf_pedit_htype(act, k);
 				entry->mangle.mask = tcf_pedit_mask(act, k);

@@ -3561,15 +3562,16 @@ int tc_setup_flow_action(struct flow_action *flow_action,
 				entry->mpls_mangle.ttl = tcf_mpls_ttl(act);
 				break;
 			default:
-				goto err_out;
+				goto err_out_locked;
 			}
 		} else if (is_tcf_skbedit_ptype(act)) {
 			entry->id = FLOW_ACTION_PTYPE;
 			entry->ptype = tcf_skbedit_ptype(act);
 		} else {
 			err = -EOPNOTSUPP;
-			goto err_out;
+			goto err_out_locked;
 		}
+		spin_unlock_bh(&act->tcfa_lock);
 
 		if (!is_tcf_pedit(act))
 			j++;

@@ -3583,6 +3585,9 @@ int tc_setup_flow_action(struct flow_action *flow_action,
 		tc_cleanup_flow_action(flow_action);
 
 	return err;
+err_out_locked:
+	spin_unlock_bh(&act->tcfa_lock);
+	goto err_out;
 }
 EXPORT_SYMBOL(tc_setup_flow_action);
 

2.33.0

From: Baowen Zheng <[email protected]>

[ Upstream commit 166b6a46b78bf8b9559a6620c3032f9fe492e082 ]

We need to return EOPNOTSUPP for the unsupported mpls action type when setup the flow action.

In the original implement, we will return 0 for the unsupported mpls action type, actually we do not setup it and the following actions to the flow action entry.

---
 net/sched/cls_api.c | 1 +
 1 file changed, 1 insertion(+)

index 61aa63cc170b4..a4c61205462ac 100644
@@ -3562,6 +3562,7 @@ int tc_setup_flow_action(struct flow_action *flow_action,
 				entry->mpls_mangle.ttl = tcf_mpls_ttl(act);
 				break;
 			default:
+				err = -EOPNOTSUPP;
 				goto err_out_locked;
 			}
 		} else if (is_tcf_skbedit_ptype(act)) {

2.33.0

From: Hangyu Hua <[email protected]>

[ Upstream commit 5f9562ebe710c307adc5f666bf1a2162ee7977c0 ]

__rds_conn_create() did not release conn->c_path when loop_trans != 0 and trans->t_prefer_loopback != 0 and is_outgoing == 0.

---
 net/rds/connection.c | 1 +
 1 file changed, 1 insertion(+)

index c85bd6340eaa7..92ff40e7a66cf 100644
@@ -253,6 +253,7 @@ static struct rds_connection *__rds_conn_create(struct net *net,
 				 * should end up here, but if it
 				 * does, reset/destroy the connection.
 				 */
+				kfree(conn->c_path);
 				kmem_cache_free(rds_conn_slab, conn);
 				conn = ERR_PTR(-EOPNOTSUPP);
 				goto out;

2.33.0

From: Nathan Chancellor <[email protected]>

[ Upstream commit a7083763619f7485ccdade160deb81737cf2732f ]

A new warning in clang points out two instances where boolean expressions are being used with a bitwise OR instead of logical OR:

drivers/soc/tegra/fuse/speedo-tegra20.c:72:9: warning: use of bitwise '|' with boolean operands [-Wbitwise-instead-of-logical]

                reg = tegra_fuse_read_spare(i) |
                      ^~~~~~~~~~~~~~~~~~~~~~~~~~
                                               ||
drivers/soc/tegra/fuse/speedo-tegra20.c:72:9: note: cast one or both operands to int to silence this warning
drivers/soc/tegra/fuse/speedo-tegra20.c:87:9: warning: use of bitwise '|' with boolean operands [-Wbitwise-instead-of-logical]
                reg = tegra_fuse_read_spare(i) |
                      ^~~~~~~~~~~~~~~~~~~~~~~~~~
                                               ||

drivers/soc/tegra/fuse/speedo-tegra20.c:87:9: note: cast one or both operands to int to silence this warning 2 warnings generated.

The motivation for the warning is that logical operations short circuit while bitwise operations do not.

In this instance, tegra_fuse_read_spare() is not semantically returning a boolean, it is returning a bit value. Use u32 for its return type so that it can be used with either bitwise or boolean operators without any warnings.

---
 drivers/soc/tegra/fuse/fuse-tegra.c | 2 +-
 drivers/soc/tegra/fuse/fuse.h       | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

index 3eb44e65b3261..1a54bac512b69 100644
@@ -172,7 +172,7 @@ static struct platform_driver tegra_fuse_driver = {
 };
 builtin_platform_driver(tegra_fuse_driver);
 
-bool __init tegra_fuse_read_spare(unsigned int spare)
+u32 __init tegra_fuse_read_spare(unsigned int spare)
 {
 	unsigned int offset = fuse->soc->info->spare + spare * 4;
 

index 7230cb3305033..6996cfc7cbca3 100644
@@ -53,7 +53,7 @@ struct tegra_fuse {
 void tegra_init_revision(void);
 void tegra_init_apbmisc(void);
 
-bool __init tegra_fuse_read_spare(unsigned int spare);
+u32 __init tegra_fuse_read_spare(unsigned int spare);
 u32 __init tegra_fuse_read_early(unsigned int offset);
 
 #ifdef CONFIG_ARCH_TEGRA_2x_SOC

2.33.0

From: Karen Sornek <[email protected]>

[ Upstream commit 584af82154f56e6b2740160fcc84a2966d969e15 ]

Move checking condition of VF MAC filter before clearing or adding MAC filter to VF to prevent potential blackout caused by removal of necessary and working VF's MAC filter.

---
 drivers/net/ethernet/intel/igb/igb_main.c | 28 +++++++++++------------
 1 file changed, 14 insertions(+), 14 deletions(-)

index c11244a9b7e69..3df25b231ab5c 100644
@@ -7374,6 +7374,20 @@ static int igb_set_vf_mac_filter(struct igb_adapter *adapter, const int vf,
 	struct vf_mac_filter *entry = NULL;
 	int ret = 0;
 
+	if ((vf_data->flags & IGB_VF_FLAG_PF_SET_MAC) &&
+	    !vf_data->trusted) {
+		dev_warn(&pdev->dev,
+			 "VF %d requested MAC filter but is administratively denied\n",
+			  vf);
+		return -EINVAL;
+	}
+	if (!is_valid_ether_addr(addr)) {
+		dev_warn(&pdev->dev,
+			 "VF %d attempted to set invalid MAC filter\n",
+			  vf);
+		return -EINVAL;
+	}
+
 	switch (info) {
 	case E1000_VF_MAC_FILTER_CLR:
 		/* remove all unicast MAC filters related to the current VF */

@@ -7387,20 +7401,6 @@ static int igb_set_vf_mac_filter(struct igb_adapter *adapter, const int vf,
 		}
 		break;
 	case E1000_VF_MAC_FILTER_ADD:
-		if ((vf_data->flags & IGB_VF_FLAG_PF_SET_MAC) &&
-		    !vf_data->trusted) {
-			dev_warn(&pdev->dev,
-				 "VF %d requested MAC filter but is administratively denied\n",
-				 vf);
-			return -EINVAL;
-		}
-		if (!is_valid_ether_addr(addr)) {
-			dev_warn(&pdev->dev,
-				 "VF %d attempted to set invalid MAC filter\n",
-				 vf);
-			return -EINVAL;
-		}
-
 		/* try to find empty slot in the list */
 		list_for_each(pos, &adapter->vf_macs.l) {
 			entry = list_entry(pos, struct vf_mac_filter, l);

2.33.0

From: Letu Ren <[email protected]>

[ Upstream commit b6d335a60dc624c0d279333b22c737faa765b028 ]

In igbvf_probe, if register_netdev() fails, the program will go to label err_hw_init, and then to label err_ioremap. In free_netdev() which is just below label err_ioremap, there is list_for_each_entry_safe and netif_napi_del which aims to delete all entries in dev->napi_list. The program has added an entry adapter->rx_ring->napi which is added by netif_napi_add in igbvf_alloc_queues(). However, adapter->rx_ring has been freed below label err_hw_init. So this a UAF.

In terms of how to patch the problem, we can refer to igbvf_remove() and delete the entry before adapter->rx_ring.

The KASAN logs are as follows:

[   35.126075] BUG: KASAN: use-after-free in free_netdev+0x1fd/0x450
[   35.127170] Read of size 8 at addr ffff88810126d990 by task modprobe/366
[   35.128360]
[   35.128643] CPU: 1 PID: 366 Comm: modprobe Not tainted 5.15.0-rc2+ #14
[   35.129789] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014
[   35.131749] Call Trace:
[   35.132199]  dump_stack_lvl+0x59/0x7b
[   35.132865]  print_address_description+0x7c/0x3b0
[   35.133707]  ? free_netdev+0x1fd/0x450
[   35.134378]  __kasan_report+0x160/0x1c0
[   35.135063]  ? free_netdev+0x1fd/0x450
[   35.135738]  kasan_report+0x4b/0x70
[   35.136367]  free_netdev+0x1fd/0x450
[   35.137006]  igbvf_probe+0x121d/0x1a10 [igbvf]
[   35.137808]  ? igbvf_vlan_rx_add_vid+0x100/0x100 [igbvf]
[   35.138751]  local_pci_probe+0x13c/0x1f0
[   35.139461]  pci_device_probe+0x37e/0x6c0
[   35.165526]
[   35.165806] Allocated by task 366:
[   35.166414]  ____kasan_kmalloc+0xc4/0xf0
[   35.167117]  foo_kmem_cache_alloc_trace+0x3c/0x50 [igbvf]
[   35.168078]  igbvf_probe+0x9c5/0x1a10 [igbvf]
[   35.168866]  local_pci_probe+0x13c/0x1f0
[   35.169565]  pci_device_probe+0x37e/0x6c0
[   35.179713]
[   35.179993] Freed by task 366:
[   35.180539]  kasan_set_track+0x4c/0x80
[   35.181211]  kasan_set_free_info+0x1f/0x40
[   35.181942]  ____kasan_slab_free+0x103/0x140
[   35.182703]  kfree+0xe3/0x250
[   35.183239]  igbvf_probe+0x1173/0x1a10 [igbvf]
[   35.184040]  local_pci_probe+0x13c/0x1f0

---
 drivers/net/ethernet/intel/igbvf/netdev.c | 1 +
 1 file changed, 1 insertion(+)

index 77cb2ab7dab40..1082e49ea0560 100644
@@ -2887,6 +2887,7 @@ static int igbvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	return 0;
 
 err_hw_init:
+	netif_napi_del(&adapter->rx_ring->napi);
 	kfree(adapter->tx_ring);
 	kfree(adapter->rx_ring);
 err_sw_init:

2.33.0

From: Cyril Novikov <[email protected]>

[ Upstream commit bf0a375055bd1afbbf02a0ef45f7655da7b71317 ]

The MDIO bus speed must be initialized before talking to the PHY the first time in order to avoid talking to it using a speed that the PHY doesn't support.

This fixes HW initialization error -17 (IXGBE_ERR_PHY_ADDR_INVALID) on Denverton CPUs (a.k.a. the Atom C3000 family) on ports with a 10Gb network plugged in. On those devices, HLREG0[MDCSPD] resets to 1, which combined with the 10Gb network results in a 24MHz MDIO speed, which is apparently too fast for the connected PHY. PHY register reads over MDIO bus return garbage, leading to initialization failure.

Reproduced with Linux kernel 4.19 and 5.15-rc7. Can be reproduced using the following setup:

* Use an Atom C3000 family system with at least one X552 LAN on the SoC
* Disable PXE or other BIOS network initialization if possible (the interface must not be initialized before Linux boots)
* Connect a live 10Gb Ethernet cable to an X550 port
* Power cycle (not reset, doesn't always work) the system and boot Linux
* Observe: ixgbe interfaces w/ 10GbE cables plugged in fail with error -17

---
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 3 +++
 1 file changed, 3 insertions(+)

index 9c42f741ed5ef..74728c0a44a81 100644
@@ -3405,6 +3405,9 @@ static s32 ixgbe_reset_hw_X550em(struct ixgbe_hw *hw)
 	/* flush pending Tx transactions */
 	ixgbe_clear_tx_pending(hw);
 
+	/* set MDIO speed before talking to the PHY in case it's the 1st time */
+	ixgbe_set_mdio_speed(hw);
+
 	/* PHY ops must be identified and initialized prior to reset */
 	status = hw->phy.ops.init(hw);
 	if (status == IXGBE_ERR_SFP_NOT_SUPPORTED ||

2.33.0

From: Haimin Zhang <[email protected]>

[ Upstream commit 481221775d53d6215a6e5e9ce1cce6d2b4ab9a46 ]

Zero-initialize memory for new map's value in function nsim_bpf_map_alloc since it may cause a potential kernel information leak issue, as follows:
1. nsim_bpf_map_alloc calls nsim_map_alloc_elem to allocate elements for a new map.
2. nsim_map_alloc_elem uses kmalloc to allocate map's value, but doesn't zero it.
3. A user application can use IOCTL BPF_MAP_LOOKUP_ELEM to get specific element's information in the map.
4. The kernel function map_lookup_elem will call bpf_map_copy_value to get the information allocated at step-2, then use copy_to_user to copy to the user buffer.
This can only leak information for an array map.

---
 drivers/net/netdevsim/bpf.c | 1 +
 1 file changed, 1 insertion(+)

index 2b74425822ab1..e0a4acc6144bf 100644
@@ -510,6 +510,7 @@ nsim_bpf_map_alloc(struct netdevsim *ns, struct bpf_offloaded_map *offmap)
 				goto err_free;
 			key = nmap->entry[i].key;
 			*key = i;
+			memset(nmap->entry[i].value, 0, offmap->map.value_size);
 		}
 	}
 

2.33.0

From: Willem de Bruijn <[email protected]>

[ Upstream commit ec6af094ea28f0f2dda1a6a33b14cd57e36a9755 ]

Packet sockets may switch ring versions. Avoid misinterpreting state between versions, whose fields share a union. rx_owner_map is only allocated with a packet ring (pg_vec) and both are swapped together. If pg_vec is NULL, meaning no packet ring was allocated, then neither was rx_owner_map. And the field may be old state from a tpacket_v3.

---
 net/packet/af_packet.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

index 0ffbf3d17911a..6062bd5bf132b 100644
@@ -4453,9 +4453,10 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u,
 	}
 
 out_free_pg_vec:
-	bitmap_free(rx_owner_map);
-	if (pg_vec)
+	if (pg_vec) {
+		bitmap_free(rx_owner_map);
 		free_pg_vec(pg_vec, order, req->tp_block_nr);
+	}
 out:
 	return err;
 }

2.33.0

From: Gal Pressman <[email protected]>

[ Upstream commit 8a03ef676ade55182f9b05115763aeda6dc08159 ]

When printing netdev features %pNF already takes care of the 0x prefix, remove the explicit one.

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

index 7dba091bc8617..ac083685214e0 100644
@@ -768,7 +768,7 @@ void skb_dump(const char *level, const struct sk_buff *skb, bool full_pkt)
 	       ntohs(skb->protocol), skb->pkt_type, skb->skb_iif);
 
 	if (dev)
-		printk("%sdev name=%s feat=0x%pNF\n",
+		printk("%sdev name=%s feat=%pNF\n",
 		       level, dev->name, &dev->features);
 	if (sk)
 		printk("%ssk family=%hu type=%u proto=%u\n",

2.33.0

From: D. Wythe <[email protected]>

[ Upstream commit 5c15b3123f65f8fbb1b445d9a7e8812e0e435df2 ]

In nginx/wrk benchmark, there's a hung problem with high probability on case likes that: (client will last several minutes to exit)

server: smc_run nginx

client: smc_run wrk -c 10000 -t 1 http://server

Client hangs with the following backtrace:

0 [ffffa7ce8Of3bbf8] __schedule at ffffffff9f9eOd5f
1 [ffffa7ce8Of3bc88] schedule at ffffffff9f9eløe6
2 [ffffa7ce8Of3bcaO] schedule_timeout at ffffffff9f9e3f3c
3 [ffffa7ce8Of3bd2O] wait_for_common at ffffffff9f9el9de
4 [ffffa7ce8Of3bd8O] __flush_work at ffffffff9fOfeOl3
5 [ffffa7ce8øf3bdfO] smc_release at ffffffffcO697d24 [smc]
6 [ffffa7ce8Of3be2O] __sock_release at ffffffff9f8O2e2d
7 [ffffa7ce8Of3be4ø] sock_close at ffffffff9f8ø2ebl
8 [ffffa7ce8øf3be48] __fput at ffffffff9f334f93
9 [ffffa7ce8Of3be78] task_work_run at ffffffff9flOlff5
10 [ffffa7ce8Of3beaO] do_exit at ffffffff9fOe5Ol2
11 [ffffa7ce8Of3bflO] do_group_exit at ffffffff9fOe592a
12 [ffffa7ce8Of3bf38] __x64_sys_exit_group at ffffffff9fOe5994 13 [ffffa7ce8Of3bf4O] do_syscall_64 at ffffffff9f9d4373
14 [ffffa7ce8Of3bfsO] entry_SYSCALL_64_after_hwframe at ffffffff9fa0007c

This issue dues to flush_work(), which is used to wait for smc_connect_work() to finish in smc_release(). Once lots of smc_connect_work() was pending or all executing work dangling, smc_release() has to block until one worker comes to free, which is equivalent to wait another smc_connnect_work() to finish.

In order to fix this, There are two changes:

1. For those idle smc_connect_work(), cancel it from the workqueue; for executing smc_connect_work(), waiting for it to finish. For that purpose, replace flush_work() with cancel_work_sync().

2. Since smc_connect() hold a reference for passive closing, if smc_connect_work() has been cancelled, release the reference.

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

index fa3b20e5f4608..06684ac346abd 100644
@@ -183,7 +183,9 @@ static int smc_release(struct socket *sock)
 	/* cleanup for a dangling non-blocking connect */
 	if (smc->connect_nonblock && sk->sk_state == SMC_INIT)
 		tcp_abort(smc->clcsock->sk, ECONNABORTED);
-	flush_work(&smc->connect_work);
+
+	if (cancel_work_sync(&smc->connect_work))
+		sock_put(&smc->sk); /* sock_hold in smc_connect for passive closing */
 
 	if (sk->sk_state == SMC_LISTEN)
 		/* smc_close_non_accepted() is called and acquires

2.33.0

From: Florian Fainelli <[email protected]>

[ Upstream commit 8b8e6e782456f1ce02a7ae914bbd5b1053f0b034 ]

The descriptor list is a shared resource across all of the transmit queues, and the locking mechanism used today only protects concurrency across a given transmit queue between the transmit and reclaiming. This creates an opportunity for the SYSTEMPORT hardware to work on corrupted descriptors if we have multiple producers at once which is the case when using multiple transmit queues.

This was particularly noticeable when using multiple flows/transmit queues and it showed up in interesting ways in that UDP packets would get a correct UDP header checksum being calculated over an incorrect packet length. Similarly TCP packets would get an equally correct checksum computed by the hardware over an incorrect packet length.

The SYSTEMPORT hardware maintains an internal descriptor list that it re-arranges when the driver produces a new descriptor anytime it writes to the WRITE_PORT_{HI,LO} registers, there is however some delay in the hardware to re-organize its descriptors and it is possible that concurrent TX queues eventually break this internal allocation scheme to the point where the length/status part of the descriptor gets used for an incorrect data buffer.

The fix is to impose a global serialization for all TX queues in the short section where we are writing to the WRITE_PORT_{HI,LO} registers which solves the corruption even with multiple concurrent TX queues being used.

---
 drivers/net/ethernet/broadcom/bcmsysport.c | 5 ++++-
 drivers/net/ethernet/broadcom/bcmsysport.h | 1 +
 2 files changed, 5 insertions(+), 1 deletion(-)

index 470d12e308814..5a2094a281e15 100644
@@ -1277,11 +1277,11 @@ static netdev_tx_t bcm_sysport_xmit(struct sk_buff *skb,
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	struct device *kdev = &priv->pdev->dev;
 	struct bcm_sysport_tx_ring *ring;
+	unsigned long flags, desc_flags;
 	struct bcm_sysport_cb *cb;
 	struct netdev_queue *txq;
 	u32 len_status, addr_lo;
 	unsigned int skb_len;
-	unsigned long flags;
 	dma_addr_t mapping;
 	u16 queue;
 	int ret;

@@ -1339,8 +1339,10 @@ static netdev_tx_t bcm_sysport_xmit(struct sk_buff *skb,
 	ring->desc_count--;
 
 	/* Ports are latched, so write upper address first */
+	spin_lock_irqsave(&priv->desc_lock, desc_flags);
 	tdma_writel(priv, len_status, TDMA_WRITE_PORT_HI(ring->index));
 	tdma_writel(priv, addr_lo, TDMA_WRITE_PORT_LO(ring->index));
+	spin_unlock_irqrestore(&priv->desc_lock, desc_flags);
 
 	/* Check ring space and update SW control flow */
 	if (ring->desc_count == 0)

@@ -1970,6 +1972,7 @@ static int bcm_sysport_open(struct net_device *dev)
 	}
 
 	/* Initialize both hardware and software ring */
+	spin_lock_init(&priv->desc_lock);
 	for (i = 0; i < dev->num_tx_queues; i++) {
 		ret = bcm_sysport_init_tx_ring(priv, i);
 		if (ret) {

index 6d80735fbc7f4..57336ca3f4277 100644
@@ -742,6 +742,7 @@ struct bcm_sysport_priv {
 	int			wol_irq;
 
 	/* Transmit rings */
+	spinlock_t		desc_lock;
 	struct bcm_sysport_tx_ring *tx_rings;
 
 	/* Receive queue */

2.33.0

From: Eric Dumazet <[email protected]>

[ Upstream commit e28587cc491ef0f3c51258fdc87fbc386b1d4c59 ]

ipip6_dev_free is sit dev->priv_destructor, already called by register_netdevice() if something goes wrong.

Alternative would be to make ipip6_dev_free() robust against multiple invocations, but other drivers do not implement this strategy.

syzbot reported:

dst_release underflow

WARNING: CPU: 0 PID: 5059 at net/core/dst.c:173 dst_release+0xd8/0xe0 net/core/dst.c:173
Modules linked in:
CPU: 1 PID: 5059 Comm: syz-executor.4 Not tainted 5.16.0-rc5-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:dst_release+0xd8/0xe0 net/core/dst.c:173
Code: 4c 89 f2 89 d9 31 c0 5b 41 5e 5d e9 da d5 44 f9 e8 1d 90 5f f9 c6 05 87 48 c6 05 01 48 c7 c7 80 44 99 8b 31 c0 e8 e8 67 29 f9 <0f> 0b eb 85 0f 1f 40 00 53 48 89 fb e8 f7 8f 5f f9 48 83 c3 a8 48
RSP: 0018:ffffc9000aa5faa0 EFLAGS: 00010246
RAX: d6894a925dd15a00 RBX: 00000000ffffffff RCX: 0000000000040000
RDX: ffffc90005e19000 RSI: 000000000003ffff RDI: 0000000000040000
RBP: 0000000000000000 R08: ffffffff816a1f42 R09: ffffed1017344f2c
R10: ffffed1017344f2c R11: 0000000000000000 R12: 0000607f462b1358
R13: 1ffffffff1bfd305 R14: ffffe8ffffcb1358 R15: dffffc0000000000
FS:  00007f66c71a2700(0000) GS:ffff8880b9a00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f88aaed5058 CR3: 0000000023e0f000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 dst_cache_destroy+0x107/0x1e0 net/core/dst_cache.c:160
 ipip6_dev_free net/ipv6/sit.c:1414 [inline]
 sit_init_net+0x229/0x550 net/ipv6/sit.c:1936
 ops_init+0x313/0x430 net/core/net_namespace.c:140
 setup_net+0x35b/0x9d0 net/core/net_namespace.c:326
 copy_net_ns+0x359/0x5c0 net/core/net_namespace.c:470
 create_new_namespaces+0x4ce/0xa00 kernel/nsproxy.c:110
 unshare_nsproxy_namespaces+0x11e/0x180 kernel/nsproxy.c:226
 ksys_unshare+0x57d/0xb50 kernel/fork.c:3075
 __do_sys_unshare kernel/fork.c:3146 [inline]
 __se_sys_unshare kernel/fork.c:3144 [inline]
 __x64_sys_unshare+0x34/0x40 kernel/fork.c:3144
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f66c882ce99
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f66c71a2168 EFLAGS: 00000246 ORIG_RAX: 0000000000000110
RAX: ffffffffffffffda RBX: 00007f66c893ff60 RCX: 00007f66c882ce99
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000048040200
RBP: 00007f66c8886ff1 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007fff6634832f R14: 00007f66c71a2300 R15: 0000000000022000
 </TASK>

---
 net/ipv6/sit.c | 1 -
 1 file changed, 1 deletion(-)

index 7f9cae4c49e7e..16e75a996b749 100644
@@ -1876,7 +1876,6 @@ static int __net_init sit_init_net(struct net *net)
 	return 0;
 
 err_reg_dev:
-	ipip6_dev_free(sitn->fb_tunnel_dev);
 	free_netdev(sitn->fb_tunnel_dev);
 err_alloc_dev:
 	return err;

2.33.0

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

[ Upstream commit f08adf5add9a071160c68bb2a61d697f39ab0758 ]

Szymon rightly pointed out that the previous check for the endpoint direction in bRequestType was not looking at only the bit involved, but rather the whole value. Normally this is ok, but for some request types, bits other than bit 8 could be set and the check for the endpoint length could not stall correctly.

Fix that up by only checking the single bit.

---
 drivers/usb/gadget/composite.c    | 6 +++---
 drivers/usb/gadget/legacy/dbgp.c  | 6 +++---
 drivers/usb/gadget/legacy/inode.c | 6 +++---
 3 files changed, 9 insertions(+), 9 deletions(-)

index d2980e30f3417..c5acf5c39fb18 100644
@@ -1649,14 +1649,14 @@ composite_setup(struct usb_gadget *gadget, const struct usb_ctrlrequest *ctrl)
 	u8				endp;
 
 	if (w_length > USB_COMP_EP0_BUFSIZ) {
-		if (ctrl->bRequestType == USB_DIR_OUT) {
-			goto done;
-		} else {
+		if (ctrl->bRequestType & USB_DIR_IN) {
 			/* Cast away the const, we are going to overwrite on purpose. */
 			__le16 *temp = (__le16 *)&ctrl->wLength;
 
 			*temp = cpu_to_le16(USB_COMP_EP0_BUFSIZ);
 			w_length = USB_COMP_EP0_BUFSIZ;
+		} else {
+			goto done;
 		}
 	}
 

index 355bc7dab9d5f..6bcbad3825802 100644
@@ -346,14 +346,14 @@ static int dbgp_setup(struct usb_gadget *gadget,
 	u16 len = 0;
 
 	if (length > DBGP_REQ_LEN) {
-		if (ctrl->bRequestType == USB_DIR_OUT) {
-			return err;
-		} else {
+		if (ctrl->bRequestType & USB_DIR_IN) {
 			/* Cast away the const, we are going to overwrite on purpose. */
 			__le16 *temp = (__le16 *)&ctrl->wLength;
 
 			*temp = cpu_to_le16(DBGP_REQ_LEN);
 			length = DBGP_REQ_LEN;
+		} else {
+			return err;
 		}
 	}
 

index f0aff79f544c3..5f1e15172403e 100644
@@ -1336,14 +1336,14 @@ gadgetfs_setup (struct usb_gadget *gadget, const struct usb_ctrlrequest *ctrl)
 	u16				w_length = le16_to_cpu(ctrl->wLength);
 
 	if (w_length > RBUF_SIZE) {
-		if (ctrl->bRequestType == USB_DIR_OUT) {
-			return value;
-		} else {
+		if (ctrl->bRequestType & USB_DIR_IN) {
 			/* Cast away the const, we are going to overwrite on purpose. */
 			__le16 *temp = (__le16 *)&ctrl->wLength;
 
 			*temp = cpu_to_le16(RBUF_SIZE);
 			w_length = RBUF_SIZE;
+		} else {
+			return value;
 		}
 	}
 

2.34.1

From: Jimmy Wang <[email protected]>

commit 0ad3bd562bb91853b9f42bda145b5db6255aee90 upstream.

This device doesn't work well with LPM, losing connectivity intermittently. Disable LPM to resolve the issue.

---
 drivers/usb/core/quirks.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -435,6 +435,9 @@ static const struct usb_device_id usb_qu
 	{ USB_DEVICE(0x1532, 0x0116), .driver_info =
 			USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL },
 
+	/* Lenovo USB-C to Ethernet Adapter RTL8153-04 */
+	{ USB_DEVICE(0x17ef, 0x720c), .driver_info = USB_QUIRK_NO_LPM },
+
 	/* Lenovo Powered USB-C Travel Hub (4X90S92381, RTL8153 GigE) */
 	{ USB_DEVICE(0x17ef, 0x721e), .driver_info = USB_QUIRK_NO_LPM },
 

From: Thomas Gleixner <[email protected]>

commit 94185adbfad56815c2c8401e16d81bdb74a79201 upstream.

PCI_MSIX_FLAGS_MASKALL is set in the MSI-X control register at MSI-X interrupt setup time. It's cleared on success, but the error handling path only clears the PCI_MSIX_FLAGS_ENABLE bit.

That's incorrect as the reset state of the PCI_MSIX_FLAGS_MASKALL bit is zero. That can be observed via lspci:

Capabilities: [b0] MSI-X: Enable- Count=67 Masked+

Clear the bit in the error path to restore the reset state.

---
 drivers/pci/msi.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -878,7 +878,7 @@ out_free:
 	free_msi_irqs(dev);
 
 out_disable:
-	pci_msix_clear_and_set_ctrl(dev, PCI_MSIX_FLAGS_ENABLE, 0);
+	pci_msix_clear_and_set_ctrl(dev, PCI_MSIX_FLAGS_MASKALL | PCI_MSIX_FLAGS_ENABLE, 0);
 
 	return ret;
 }

From: Stefan Roese <[email protected]>

commit 83dbf898a2d45289be875deb580e93050ba67529 upstream.

Masking all unused MSI-X entries is done to ensure that a crash kernel starts from a clean slate, which correponds to the reset state of the device as defined in the PCI-E specificion 3.0 and later:

Vector Control for MSI-X Table Entries
--------------------------------------

"00: Mask bit: When this bit is set, the function is prohibited from
sending a message using this MSI-X Table entry.
...
This bit’s state after reset is 1 (entry is masked)."

A Marvell NVME device fails to deliver MSI interrupts after trying to enable MSI-X interrupts due to that masking. It seems to take the MSI-X mask bits into account even when MSI-X is disabled.

While not specification compliant, this can be cured by moving the masking into the success path, so that the MSI-X table entries stay in device reset state when the MSI-X setup fails.

[ tglx: Move it into the success path, add comment and amend changelog ]

---
 drivers/pci/msi.c |   13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

@@ -827,9 +827,6 @@ static int msix_capability_init(struct p
 		goto out_disable;
 	}
 
-	/* Ensure that all table entries are masked. */
-	msix_mask_all(base, tsize);
-
 	ret = msix_setup_entries(dev, base, entries, nvec, affd);
 	if (ret)
 		goto out_disable;

@@ -852,6 +849,16 @@ static int msix_capability_init(struct p
 	/* Set MSI-X enabled bits and unmask the function */
 	pci_intx_for_msi(dev, 0);
 	dev->msix_enabled = 1;
+
+	/*
+	 * Ensure that all table entries are masked to prevent
+	 * stale entries from firing in a crash kernel.
+	 *
+	 * Done late to deal with a broken Marvell NVME device
+	 * which takes the MSI-X mask bits into account even
+	 * when MSI-X is disabled, which prevents MSI delivery.
+	 */
+	msix_mask_all(base, tsize);
 	pci_msix_clear_and_set_ctrl(dev, PCI_MSIX_FLAGS_MASKALL, 0);
 
 	pcibios_free_irq(dev);

From: Nehal Bakulchandra Shah <[email protected]>

commit f886d4fbb7c97b8f5f447c92d2dab99c841803c0 upstream.

AMD's Yellow Carp platform has few more XHCI controllers, enable the runtime power management support for the same.

---
 drivers/usb/host/xhci-pci.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

@@ -65,6 +65,8 @@
 #define PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_4		0x161e
 #define PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_5		0x15d6
 #define PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_6		0x15d7
+#define PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_7		0x161c
+#define PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_8		0x161f
 
 #define PCI_DEVICE_ID_ASMEDIA_1042_XHCI			0x1042
 #define PCI_DEVICE_ID_ASMEDIA_1042A_XHCI		0x1142

@@ -303,7 +305,9 @@ static void xhci_pci_quirks(struct devic
 	    pdev->device == PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_3 ||
 	    pdev->device == PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_4 ||
 	    pdev->device == PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_5 ||
-	    pdev->device == PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_6))
+	    pdev->device == PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_6 ||
+	    pdev->device == PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_7 ||
+	    pdev->device == PCI_DEVICE_ID_AMD_YELLOW_CARP_XHCI_8))
 		xhci->quirks |= XHCI_DEFAULT_PM_RUNTIME_ALLOW;
 
 	if (xhci->quirks & XHCI_RESET_ON_RESUME)

From: Johan Hovold <[email protected]>

commit 83b67041f3eaf33f98a075249aa7f4c7617c2f85 upstream.

When generalising GPIO support and adding support for CP2102N, the GPIO registration for some CP2105 devices accidentally broke. Specifically, when all the pins of a port are in "modem" mode, and thus unavailable for GPIO use, the GPIO chip would now be registered without having initialised the number of GPIO lines. This would in turn be rejected by gpiolib and some errors messages would be printed (but importantly probe would still succeed).

Fix this by initialising the number of GPIO lines before registering the GPIO chip.

Note that as for the other device types, and as when all CP2105 pins are muxed for LED function, the GPIO chip is registered also when no pins are available for GPIO use.

---
 drivers/usb/serial/cp210x.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

@@ -1552,6 +1552,8 @@ static int cp2105_gpioconf_init(struct u
 
 	/*  2 banks of GPIO - One for the pins taken from each serial port */
 	if (intf_num == 0) {
+		priv->gc.ngpio = 2;
+
 		if (mode.eci == CP210X_PIN_MODE_MODEM) {
 			/* mark all GPIOs of this interface as reserved */
 			priv->gpio_altfunc = 0xff;

@@ -1562,8 +1564,9 @@ static int cp2105_gpioconf_init(struct u
 		priv->gpio_pushpull = (u8)((le16_to_cpu(config.gpio_mode) &
 						CP210X_ECI_GPIO_MODE_MASK) >>
 						CP210X_ECI_GPIO_MODE_OFFSET);
-		priv->gc.ngpio = 2;
 	} else if (intf_num == 1) {
+		priv->gc.ngpio = 3;
+
 		if (mode.sci == CP210X_PIN_MODE_MODEM) {
 			/* mark all GPIOs of this interface as reserved */
 			priv->gpio_altfunc = 0xff;

@@ -1574,7 +1577,6 @@ static int cp2105_gpioconf_init(struct u
 		priv->gpio_pushpull = (u8)((le16_to_cpu(config.gpio_mode) &
 						CP210X_SCI_GPIO_MODE_MASK) >>
 						CP210X_SCI_GPIO_MODE_OFFSET);
-		priv->gc.ngpio = 3;
 	} else {
 		return -ENODEV;
 	}

From: Daniele Palmas <[email protected]>

commit 2b503c8598d1b232e7fc7526bce9326d92331541 upstream.

Add the following Telit FN990 compositions:

0x1070: tty, adb, rmnet, tty, tty, tty, tty 0x1071: tty, adb, mbim, tty, tty, tty, tty 0x1072: rndis, tty, adb, tty, tty, tty, tty 0x1073: tty, adb, ecm, tty, tty, tty, tty

---
 drivers/usb/serial/option.c |    8 ++++++++
 1 file changed, 8 insertions(+)

@@ -1219,6 +1219,14 @@ static const struct usb_device_id option
 	  .driver_info = NCTRL(2) | RSVD(3) },
 	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x1063, 0xff),	/* Telit LN920 (ECM) */
 	  .driver_info = NCTRL(0) | RSVD(1) },
+	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x1070, 0xff),	/* Telit FN990 (rmnet) */
+	  .driver_info = NCTRL(0) | RSVD(1) | RSVD(2) },
+	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x1071, 0xff),	/* Telit FN990 (MBIM) */
+	  .driver_info = NCTRL(0) | RSVD(1) },
+	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x1072, 0xff),	/* Telit FN990 (RNDIS) */
+	  .driver_info = NCTRL(2) | RSVD(3) },
+	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x1073, 0xff),	/* Telit FN990 (ECM) */
+	  .driver_info = NCTRL(0) | RSVD(1) },
 	{ USB_DEVICE(TELIT_VENDOR_ID, TELIT_PRODUCT_ME910),
 	  .driver_info = NCTRL(0) | RSVD(1) | RSVD(3) },
 	{ USB_DEVICE(TELIT_VENDOR_ID, TELIT_PRODUCT_ME910_DUAL_MODEM),

From: Yu Liao <[email protected]>

commit 4e8c11b6b3f0b6a283e898344f154641eda94266 upstream.

Even after commit e1d7ba873555 ("time: Always make sure wall_to_monotonic isn't positive") it is still possible to make wall_to_monotonic positive by running the following code:

    int main(void)
    {
        struct timespec time;

        clock_gettime(CLOCK_MONOTONIC, &time);
        time.tv_nsec = 0;
        clock_settime(CLOCK_REALTIME, &time);
        return 0;
    }

The reason is that the second parameter of timespec64_compare(), ts_delta, may be unnormalized because the delta is calculated with an open coded substraction which causes the comparison of tv_sec to yield the wrong result:

wall_to_monotonic = { .tv_sec = -10, .tv_nsec = 900000000 }
ts_delta = { .tv_sec = -9, .tv_nsec = -900000000 }

That makes timespec64_compare() claim that wall_to_monotonic < ts_delta, but actually the result should be wall_to_monotonic > ts_delta.

After normalization, the result of timespec64_compare() is correct because the tv_sec comparison is not longer misleading:

wall_to_monotonic = { .tv_sec = -10, .tv_nsec = 900000000 }
ts_delta = { .tv_sec = -10, .tv_nsec = 100000000 }

Use timespec64_sub() to ensure that ts_delta is normalized, which fixes the issue.

---
 kernel/time/timekeeping.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

@@ -1236,8 +1236,7 @@ int do_settimeofday64(const struct times
 	timekeeping_forward_now(tk);
 
 	xt = tk_xtime(tk);
-	ts_delta.tv_sec = ts->tv_sec - xt.tv_sec;
-	ts_delta.tv_nsec = ts->tv_nsec - xt.tv_nsec;
+	ts_delta = timespec64_sub(*ts, xt);
 
 	if (timespec64_compare(&tk->wall_to_monotonic, &ts_delta) > 0) {
 		ret = -EINVAL;

From: George Kennedy <[email protected]>

commit 5da5231bb47864e5dd6c6731151e98b6ee498827 upstream.

Avoid data corruption by rejecting pass-through commands where T_LENGTH is zero (No data is transferred) and the dma direction is not DMA_NONE.

---
 drivers/ata/libata-scsi.c |   15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

@@ -3164,8 +3164,19 @@ static unsigned int ata_scsi_pass_thru(s
 		goto invalid_fld;
 	}
 
-	if (ata_is_ncq(tf->protocol) && (cdb[2 + cdb_offset] & 0x3) == 0)
-		tf->protocol = ATA_PROT_NCQ_NODATA;
+	if ((cdb[2 + cdb_offset] & 0x3) == 0) {
+		/*
+		 * When T_LENGTH is zero (No data is transferred), dir should
+		 * be DMA_NONE.
+		 */
+		if (scmd->sc_data_direction != DMA_NONE) {
+			fp = 2 + cdb_offset;
+			goto invalid_fld;
+		}
+
+		if (ata_is_ncq(tf->protocol))
+			tf->protocol = ATA_PROT_NCQ_NODATA;
+	}
 
 	/* enable LBA */
 	tf->flags |= ATA_TFLAG_LBA;

From: Le Ma <[email protected]>

commit f3a8076eb28cae1553958c629aecec479394bbe2 upstream.

should count on GC IP base address

---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -2906,8 +2906,8 @@ static void gfx_v9_0_init_pg(struct amdg
 			      AMD_PG_SUPPORT_CP |
 			      AMD_PG_SUPPORT_GDS |
 			      AMD_PG_SUPPORT_RLC_SMU_HS)) {
-		WREG32(mmRLC_JUMP_TABLE_RESTORE,
-		       adev->gfx.rlc.cp_table_gpu_addr >> 8);
+		WREG32_SOC15(GC, 0, mmRLC_JUMP_TABLE_RESTORE,
+			     adev->gfx.rlc.cp_table_gpu_addr >> 8);
 		gfx_v9_0_init_gfx_power_gating(adev);
 	}
 }

From: Johannes Berg <[email protected]>

commit 768c0b19b50665e337c96858aa2b7928d6dcf756 upstream.

Before attempting to parse an extended element, verify that the extended element ID is present.

---
 net/mac80211/util.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -1227,6 +1227,8 @@ _ieee802_11_parse_elems_crc(const u8 *st
 				elems->max_idle_period_ie = (void *)pos;
 			break;
 		case WLAN_EID_EXTENSION:
+			if (!elen)
+				break;
 			if (pos[0] == WLAN_EID_EXT_HE_MU_EDCA &&
 			    elen >= (sizeof(*elems->mu_edca_param_set) + 1)) {
 				elems->mu_edca_param_set = (void *)&pos[1];

From: Nathan Chancellor <[email protected]>

commit 6a953dc4dbd1c7057fb765a24f37a5e953c85fb0 upstream.

A new warning in clang points out when macro expansion might result in a GNU C statement expression. There is an instance of this in the mwifiex driver:

drivers/net/wireless/marvell/mwifiex/cmdevt.c:217:34: warning: '}' and ')' tokens terminating statement expression appear in different macro expansion contexts [-Wcompound-token-split-by-macro]

        host_cmd->seq_num = cpu_to_le16(HostCmd_SET_SEQ_NO_BSS_INFO
                                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/net/wireless/marvell/mwifiex/fw.h:519:46: note: expanded from
macro 'HostCmd_SET_SEQ_NO_BSS_INFO'
        (((type) & 0x000f) << 12);                  }
                                                    ^

This does not appear to be a real issue. Removing the braces and replacing them with parentheses will fix the warning and not change the meaning of the code.

---
 drivers/net/wireless/marvell/mwifiex/cmdevt.c |    4 ++--
 drivers/net/wireless/marvell/mwifiex/fw.h     |    8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

@@ -322,9 +322,9 @@ static int mwifiex_dnld_sleep_confirm_cm
 
 	adapter->seq_num++;
 	sleep_cfm_buf->seq_num =
-		cpu_to_le16((HostCmd_SET_SEQ_NO_BSS_INFO
+		cpu_to_le16(HostCmd_SET_SEQ_NO_BSS_INFO
 					(adapter->seq_num, priv->bss_num,
-					 priv->bss_type)));
+					 priv->bss_type));
 
 	mwifiex_dbg(adapter, CMD,
 		    "cmd: DNLD_CMD: %#x, act %#x, len %d, seqno %#x\n",

@@ -512,10 +512,10 @@ enum mwifiex_channel_flags {
 
 #define RF_ANTENNA_AUTO                 0xFFFF
 
-#define HostCmd_SET_SEQ_NO_BSS_INFO(seq, num, type) {   \
-	(((seq) & 0x00ff) |                             \
-	 (((num) & 0x000f) << 8)) |                     \
-	(((type) & 0x000f) << 12);                  }
+#define HostCmd_SET_SEQ_NO_BSS_INFO(seq, num, type) \
+	((((seq) & 0x00ff) |                        \
+	 (((num) & 0x000f) << 8)) |                 \
+	(((type) & 0x000f) << 12))
 
 #define HostCmd_GET_SEQ_NO(seq)       \
 	((seq) & HostCmd_SEQ_NUM_MASK)

From: Nathan Chancellor <[email protected]>

commit a02dcde595f7cbd240ccd64de96034ad91cffc40 upstream.

A new warning in clang points out a few places in this driver where a bitwise OR is being used with boolean types:

drivers/input/touchscreen.c:81:17: warning: use of bitwise '|' with boolean operands [-Wbitwise-instead-of-logical]
data_present = touchscreen_get_prop_u32(dev, "touchscreen-min-x",
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

This use of a bitwise OR is intentional, as bitwise operations do not short circuit, which allows all the calls to touchscreen_get_prop_u32() to happen so that the last parameter is initialized while coalescing the results of the calls to make a decision after they are all evaluated.

To make this clearer to the compiler, use the '|=' operator to assign the result of each touchscreen_get_prop_u32() call to data_present, which keeps the meaning of the code the same but makes it obvious that every one of these calls is expected to happen.

---
 drivers/input/touchscreen/of_touchscreen.c |   18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

@@ -81,8 +81,8 @@ void touchscreen_parse_properties(struct
 		       touchscreen_get_prop_u32(dev, "touchscreen-size-x",
 						input_abs_get_max(input,
 								  axis) + 1,
-						&maximum) |
-		       touchscreen_get_prop_u32(dev, "touchscreen-fuzz-x",
+						&maximum);
+	data_present |= touchscreen_get_prop_u32(dev, "touchscreen-fuzz-x",
 						input_abs_get_fuzz(input, axis),
 						&fuzz);
 	if (data_present)

@@ -95,8 +95,8 @@ void touchscreen_parse_properties(struct
 		       touchscreen_get_prop_u32(dev, "touchscreen-size-y",
 						input_abs_get_max(input,
 								  axis) + 1,
-						&maximum) |
-		       touchscreen_get_prop_u32(dev, "touchscreen-fuzz-y",
+						&maximum);
+	data_present |= touchscreen_get_prop_u32(dev, "touchscreen-fuzz-y",
 						input_abs_get_fuzz(input, axis),
 						&fuzz);
 	if (data_present)

@@ -106,11 +106,11 @@ void touchscreen_parse_properties(struct
 	data_present = touchscreen_get_prop_u32(dev,
 						"touchscreen-max-pressure",
 						input_abs_get_max(input, axis),
-						&maximum) |
-		       touchscreen_get_prop_u32(dev,
-						"touchscreen-fuzz-pressure",
-						input_abs_get_fuzz(input, axis),
-						&fuzz);
+						&maximum);
+	data_present |= touchscreen_get_prop_u32(dev,
+						 "touchscreen-fuzz-pressure",
+						 input_abs_get_fuzz(input, axis),
+						 &fuzz);
 	if (data_present)
 		touchscreen_set_params(input, axis, 0, maximum, fuzz);
 

From: Fabio Estevam <[email protected]>

commit 737e65c7956795b3553781fb7bc82fce1c39503f upstream.

According to the i.MX6ULL Reference Manual, pad CSI_DATA07 may have the ESAI_TX0 functionality, not ESAI_T0.

Also, NXP's i.MX Config Tools 10.0 generates dtsi with the MX6ULL_PAD_CSI_DATA07__ESAI_TX0 naming, so fix it accordingly.

There are no devicetree users in mainline that use the old name, so just remove the old entry.

---
 arch/arm/boot/dts/imx6ull-pinfunc.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -82,6 +82,6 @@
 #define MX6ULL_PAD_CSI_DATA04__ESAI_TX_FS                         0x01F4 0x0480 0x0000 0x9 0x0
 #define MX6ULL_PAD_CSI_DATA05__ESAI_TX_CLK                        0x01F8 0x0484 0x0000 0x9 0x0
 #define MX6ULL_PAD_CSI_DATA06__ESAI_TX5_RX0                       0x01FC 0x0488 0x0000 0x9 0x0
-#define MX6ULL_PAD_CSI_DATA07__ESAI_T0                            0x0200 0x048C 0x0000 0x9 0x0
+#define MX6ULL_PAD_CSI_DATA07__ESAI_TX0                           0x0200 0x048C 0x0000 0x9 0x0
 
 #endif /* __DTS_IMX6ULL_PINFUNC_H */

From: Magnus Karlsson <[email protected]>

commit bd0687c18e635b63233dc87f38058cd728802ab4 upstream.

Do not sleep in poll() when the need_wakeup flag is set. When this flag is set, the application needs to explicitly wake up the driver with a syscall (poll, recvmsg, sendmsg, etc.) to guarantee that Rx and/or Tx processing will be processed promptly. But the current code in poll(), sleeps first then wakes up the driver. This means that no driver processing will occur (baring any interrupts) until the timeout has expired.

Fix this by checking the need_wakeup flag first and if set, wake the driver and return to the application. Only if need_wakeup is not set should the process sleep if there is a timeout set in the poll() call.

---
 net/xdp/xsk.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -434,8 +434,6 @@ static __poll_t xsk_poll(struct file *fi
 	struct xdp_sock *xs = xdp_sk(sk);
 	struct xdp_umem *umem;
 
-	sock_poll_wait(file, sock, wait);
-
 	if (unlikely(!xsk_is_bound(xs)))
 		return mask;
 

@@ -447,6 +445,8 @@ static __poll_t xsk_poll(struct file *fi
 		else
 			/* Poll needs to drive Tx also in copy mode */
 			__xsk_sendmsg(sk);
+	} else {
+		sock_poll_wait(file, sock, wait);
 	}
 
 	if (xs->rx && !xskq_empty_desc(xs->rx))

From: Pavel Skripkin <[email protected]>

commit 44870a9e7a3c24acbb3f888b2a7cc22c9bdf7e7f upstream.

Syzbot reported, that mxl111sf_ctrl_msg() uses uninitialized mutex. The problem was in wrong mutex_init() location.

Previous mutex_init(&state->msg_lock) call was in ->init() function, but dvb_usbv2_init() has this order of calls:

	dvb_usbv2_init()
	  dvb_usbv2_adapter_init()
	    dvb_usbv2_adapter_frontend_init()
	      props->frontend_attach()

	  props->init()

Since mxl111sf_* devices call mxl111sf_ctrl_msg() in ->frontend_attach() internally we need to initialize state->msg_lock before frontend_attach(). To achieve it, ->probe() call added to all mxl111sf_* devices, which will simply initiaize mutex.

---
 drivers/media/usb/dvb-usb-v2/mxl111sf.c |   16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

@@ -931,8 +931,6 @@ static int mxl111sf_init(struct dvb_usb_
 		  .len = sizeof(eeprom), .buf = eeprom },
 	};
 
-	mutex_init(&state->msg_lock);
-
 	ret = get_chip_info(state);
 	if (mxl_fail(ret))
 		pr_err("failed to get chip info during probe");

@@ -1074,6 +1072,14 @@ static int mxl111sf_get_stream_config_dv
 	return 0;
 }
 
+static int mxl111sf_probe(struct dvb_usb_device *dev)
+{
+	struct mxl111sf_state *state = d_to_priv(dev);
+
+	mutex_init(&state->msg_lock);
+	return 0;
+}
+
 static struct dvb_usb_device_properties mxl111sf_props_dvbt = {
 	.driver_name = KBUILD_MODNAME,
 	.owner = THIS_MODULE,

@@ -1083,6 +1089,7 @@ static struct dvb_usb_device_properties
 	.generic_bulk_ctrl_endpoint = 0x02,
 	.generic_bulk_ctrl_endpoint_response = 0x81,
 
+	.probe             = mxl111sf_probe,
 	.i2c_algo          = &mxl111sf_i2c_algo,
 	.frontend_attach   = mxl111sf_frontend_attach_dvbt,
 	.tuner_attach      = mxl111sf_attach_tuner,

@@ -1124,6 +1131,7 @@ static struct dvb_usb_device_properties
 	.generic_bulk_ctrl_endpoint = 0x02,
 	.generic_bulk_ctrl_endpoint_response = 0x81,
 
+	.probe             = mxl111sf_probe,
 	.i2c_algo          = &mxl111sf_i2c_algo,
 	.frontend_attach   = mxl111sf_frontend_attach_atsc,
 	.tuner_attach      = mxl111sf_attach_tuner,

@@ -1165,6 +1173,7 @@ static struct dvb_usb_device_properties
 	.generic_bulk_ctrl_endpoint = 0x02,
 	.generic_bulk_ctrl_endpoint_response = 0x81,
 
+	.probe             = mxl111sf_probe,
 	.i2c_algo          = &mxl111sf_i2c_algo,
 	.frontend_attach   = mxl111sf_frontend_attach_mh,
 	.tuner_attach      = mxl111sf_attach_tuner,

@@ -1233,6 +1242,7 @@ static struct dvb_usb_device_properties
 	.generic_bulk_ctrl_endpoint = 0x02,
 	.generic_bulk_ctrl_endpoint_response = 0x81,
 
+	.probe             = mxl111sf_probe,
 	.i2c_algo          = &mxl111sf_i2c_algo,
 	.frontend_attach   = mxl111sf_frontend_attach_atsc_mh,
 	.tuner_attach      = mxl111sf_attach_tuner,

@@ -1311,6 +1321,7 @@ static struct dvb_usb_device_properties
 	.generic_bulk_ctrl_endpoint = 0x02,
 	.generic_bulk_ctrl_endpoint_response = 0x81,
 
+	.probe             = mxl111sf_probe,
 	.i2c_algo          = &mxl111sf_i2c_algo,
 	.frontend_attach   = mxl111sf_frontend_attach_mercury,
 	.tuner_attach      = mxl111sf_attach_tuner,

@@ -1381,6 +1392,7 @@ static struct dvb_usb_device_properties
 	.generic_bulk_ctrl_endpoint = 0x02,
 	.generic_bulk_ctrl_endpoint_response = 0x81,
 
+	.probe             = mxl111sf_probe,
 	.i2c_algo          = &mxl111sf_i2c_algo,
 	.frontend_attach   = mxl111sf_frontend_attach_mercury_mh,
 	.tuner_attach      = mxl111sf_attach_tuner,

From: Miklos Szeredi <[email protected]>

commit bda9a71980e083699a0360963c0135657b73f47a upstream.

Add missing inode lock annotatation; found by syzbot.

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

@@ -1032,7 +1032,7 @@ int fuse_reverse_inval_entry(struct supe
 	if (!parent)
 		return -ENOENT;
 
-	inode_lock(parent);
+	inode_lock_nested(parent, I_MUTEX_PARENT);
 	if (!S_ISDIR(parent->i_mode))
 		goto unlock;
 

From: Miklos Szeredi <[email protected]>

commit 1f5573cfe7a7056e80a92c7a037a3e69f3a13d1c upstream.

Syzbot triggered the following warning in ovl_workdir_create() -> ovl_create_real():

	if (!err && WARN_ON(!newdentry->d_inode)) {

The reason is that the cgroup2 filesystem returns from mkdir without instantiating the new dentry.

Weird filesystems such as this will be rejected by overlayfs at a later stage during setup, but to prevent such a warning, call ovl_mkdir_real() directly from ovl_workdir_create() and reject this case early.

---
 fs/overlayfs/dir.c       |    3 +--
 fs/overlayfs/overlayfs.h |    1 +
 fs/overlayfs/super.c     |   12 ++++++++----
 3 files changed, 10 insertions(+), 6 deletions(-)

@@ -113,8 +113,7 @@ kill_whiteout:
 	goto out;
 }
 
-static int ovl_mkdir_real(struct inode *dir, struct dentry **newdentry,
-			  umode_t mode)
+int ovl_mkdir_real(struct inode *dir, struct dentry **newdentry, umode_t mode)
 {
 	int err;
 	struct dentry *d, *dentry = *newdentry;

@@ -409,6 +409,7 @@ struct ovl_cattr {
 
 #define OVL_CATTR(m) (&(struct ovl_cattr) { .mode = (m) })
 
+int ovl_mkdir_real(struct inode *dir, struct dentry **newdentry, umode_t mode);
 struct dentry *ovl_create_real(struct inode *dir, struct dentry *newdentry,
 			       struct ovl_cattr *attr);
 int ovl_cleanup(struct inode *dir, struct dentry *dentry);

@@ -650,10 +650,14 @@ retry:
 			goto retry;
 		}
 
-		work = ovl_create_real(dir, work, OVL_CATTR(attr.ia_mode));
-		err = PTR_ERR(work);
-		if (IS_ERR(work))
-			goto out_err;
+		err = ovl_mkdir_real(dir, &work, attr.ia_mode);
+		if (err)
+			goto out_dput;
+
+		/* Weird filesystem returning with hashed negative (kernfs)? */
+		err = -EINVAL;
+		if (d_really_is_negative(work))
+			goto out_dput;
 
 		/*
 		 * Try to remove POSIX ACL xattrs from workdir.  We are good if:

From: George Kennedy <[email protected]>

commit e0a2c28da11e2c2b963fc01d50acbf03045ac732 upstream.

In resp_mode_select() sanity check the block descriptor len to avoid UAF.

BUG: KASAN: use-after-free in resp_mode_select+0xa4c/0xb40 drivers/scsi/scsi_debug.c:2509
Read of size 1 at addr ffff888026670f50 by task scsicmd/15032

CPU: 1 PID: 15032 Comm: scsicmd Not tainted 5.15.0-01d0625 #15
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
Call Trace:
 <TASK>
 dump_stack_lvl+0x89/0xb5 lib/dump_stack.c:107
 print_address_description.constprop.9+0x28/0x160 mm/kasan/report.c:257
 kasan_report.cold.14+0x7d/0x117 mm/kasan/report.c:443
 __asan_report_load1_noabort+0x14/0x20 mm/kasan/report_generic.c:306
 resp_mode_select+0xa4c/0xb40 drivers/scsi/scsi_debug.c:2509
 schedule_resp+0x4af/0x1a10 drivers/scsi/scsi_debug.c:5483
 scsi_debug_queuecommand+0x8c9/0x1e70 drivers/scsi/scsi_debug.c:7537
 scsi_queue_rq+0x16b4/0x2d10 drivers/scsi/scsi_lib.c:1521
 blk_mq_dispatch_rq_list+0xb9b/0x2700 block/blk-mq.c:1640
 __blk_mq_sched_dispatch_requests+0x28f/0x590 block/blk-mq-sched.c:325
 blk_mq_sched_dispatch_requests+0x105/0x190 block/blk-mq-sched.c:358
 __blk_mq_run_hw_queue+0xe5/0x150 block/blk-mq.c:1762
 __blk_mq_delay_run_hw_queue+0x4f8/0x5c0 block/blk-mq.c:1839
 blk_mq_run_hw_queue+0x18d/0x350 block/blk-mq.c:1891
 blk_mq_sched_insert_request+0x3db/0x4e0 block/blk-mq-sched.c:474
 blk_execute_rq_nowait+0x16b/0x1c0 block/blk-exec.c:63
 sg_common_write.isra.18+0xeb3/0x2000 drivers/scsi/sg.c:837
 sg_new_write.isra.19+0x570/0x8c0 drivers/scsi/sg.c:775
 sg_ioctl_common+0x14d6/0x2710 drivers/scsi/sg.c:941
 sg_ioctl+0xa2/0x180 drivers/scsi/sg.c:1166
 __x64_sys_ioctl+0x19d/0x220 fs/ioctl.c:52
 do_syscall_64+0x3a/0x80 arch/x86/entry/common.c:50
 entry_SYSCALL_64_after_hwframe+0x44/0xae arch/x86/entry/entry_64.S:113

---
 drivers/scsi/scsi_debug.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -2296,11 +2296,11 @@ static int resp_mode_select(struct scsi_
 			    __func__, param_len, res);
 	md_len = mselect6 ? (arr[0] + 1) : (get_unaligned_be16(arr + 0) + 2);
 	bd_len = mselect6 ? arr[3] : get_unaligned_be16(arr + 6);
-	if (md_len > 2) {
+	off = bd_len + (mselect6 ? 4 : 8);
+	if (md_len > 2 || off >= res) {
 		mk_sense_invalid_fld(scp, SDEB_IN_DATA, 0, -1);
 		return check_condition_result;
 	}
-	off = bd_len + (mselect6 ? 4 : 8);
 	mpage = arr[off] & 0x3f;
 	ps = !!(arr[off] & 0x80);
 	if (ps) {

From: Paul E. McKenney <[email protected]>

commit 2431774f04d1050292054c763070021bade7b151 upstream.

This commit marks accesses to the rcu_state.n_force_qs. These data races are hard to make happen, but syzkaller was equal to the task.

---
 kernel/rcu/tree.c |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

@@ -1602,7 +1602,7 @@ static void rcu_gp_fqs(bool first_time)
 	struct rcu_node *rnp = rcu_get_root();
 
 	WRITE_ONCE(rcu_state.gp_activity, jiffies);
-	rcu_state.n_force_qs++;
+	WRITE_ONCE(rcu_state.n_force_qs, rcu_state.n_force_qs + 1);
 	if (first_time) {
 		/* Collect dyntick-idle snapshots. */
 		force_qs_rnp(dyntick_save_progress_counter);

@@ -2207,7 +2207,7 @@ static void rcu_do_batch(struct rcu_data
 	/* Reset ->qlen_last_fqs_check trigger if enough CBs have drained. */
 	if (count == 0 && rdp->qlen_last_fqs_check != 0) {
 		rdp->qlen_last_fqs_check = 0;
-		rdp->n_force_qs_snap = rcu_state.n_force_qs;
+		rdp->n_force_qs_snap = READ_ONCE(rcu_state.n_force_qs);
 	} else if (count < rdp->qlen_last_fqs_check - qhimark)
 		rdp->qlen_last_fqs_check = count;
 

@@ -2535,10 +2535,10 @@ static void __call_rcu_core(struct rcu_d
 		} else {
 			/* Give the grace period a kick. */
 			rdp->blimit = DEFAULT_MAX_RCU_BLIMIT;
-			if (rcu_state.n_force_qs == rdp->n_force_qs_snap &&
+			if (READ_ONCE(rcu_state.n_force_qs) == rdp->n_force_qs_snap &&
 			    rcu_segcblist_first_pend_cb(&rdp->cblist) != head)
 				rcu_force_quiescent_state();
-			rdp->n_force_qs_snap = rcu_state.n_force_qs;
+			rdp->n_force_qs_snap = READ_ONCE(rcu_state.n_force_qs);
 			rdp->qlen_last_fqs_check = rcu_segcblist_n_cbs(&rdp->cblist);
 		}
 	}

@@ -3029,7 +3029,7 @@ int rcutree_prepare_cpu(unsigned int cpu
 	/* Set up local state, ensuring consistent view of global state. */
 	raw_spin_lock_irqsave_rcu_node(rnp, flags);
 	rdp->qlen_last_fqs_check = 0;
-	rdp->n_force_qs_snap = rcu_state.n_force_qs;
+	rdp->n_force_qs_snap = READ_ONCE(rcu_state.n_force_qs);
 	rdp->blimit = blimit;
 	if (rcu_segcblist_empty(&rdp->cblist) && /* No early-boot CBs? */
 	    !rcu_segcblist_is_offloaded(&rdp->cblist))

From: Felix Fietkau <[email protected]>

commit 73111efacd3c6d9e644acca1d132566932be8af0 upstream.

Some drivers that do their own sequence number allocation (e.g. ath9k) rely on being able to modify params->ssn on starting tx ampdu sessions. This was broken by a change that modified it to use sta->tid_seq[tid] instead.

---
 net/mac80211/agg-tx.c   |    4 ++--
 net/mac80211/sta_info.h |    1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

@@ -481,8 +481,7 @@ static void ieee80211_send_addba_with_ti
 
 	/* send AddBA request */
 	ieee80211_send_addba_request(sdata, sta->sta.addr, tid,
-				     tid_tx->dialog_token,
-				     sta->tid_seq[tid] >> 4,
+				     tid_tx->dialog_token, tid_tx->ssn,
 				     buf_size, tid_tx->timeout);
 }
 

@@ -522,6 +521,7 @@ void ieee80211_tx_ba_session_handle_star
 
 	params.ssn = sta->tid_seq[tid] >> 4;
 	ret = drv_ampdu_action(local, sdata, &params);
+	tid_tx->ssn = params.ssn;
 	if (ret) {
 		ht_dbg(sdata,
 		       "BA request denied - HW unavailable for %pM tid %d\n",

@@ -180,6 +180,7 @@ struct tid_ampdu_tx {
 	u8 stop_initiator;
 	bool tx_stop;
 	u16 buf_size;
+	u16 ssn;
 
 	u16 failed_bar_ssn;
 	bool bar_pending;

From: Leon Romanovsky <[email protected]>

commit d086a1c65aabb5a4e1edc580ca583e2964c62b44 upstream.

The access of tcf_tunnel_info() produces the following splat, so fix it by dereferencing the tcf_tunnel_key_params pointer with marker that internal tcfa_liock is held.

=============================
WARNING: suspicious RCU usage
5.9.0+ #1 Not tainted
-----------------------------
include/net/tc_act/tc_tunnel_key.h:59 suspicious rcu_dereference_protected() usage! other info that might help us debug this:

rcu_scheduler_active = 2, debug_locks = 1
1 lock held by tc/34839:
#0: ffff88828572c2a0 (&p->tcfa_lock){+...}-{2:2}, at: tc_setup_flow_action+0xb3/0x48b5

 stack backtrace:
 CPU: 1 PID: 34839 Comm: tc Not tainted 5.9.0+ #1
 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
 Call Trace:
  dump_stack+0x9a/0xd0
  tc_setup_flow_action+0x14cb/0x48b5
  fl_hw_replace_filter+0x347/0x690 [cls_flower]
  fl_change+0x2bad/0x4875 [cls_flower]
  tc_new_tfilter+0xf6f/0x1ba0
  rtnetlink_rcv_msg+0x5f2/0x870
  netlink_rcv_skb+0x124/0x350
  netlink_unicast+0x433/0x700
  netlink_sendmsg+0x6f1/0xbd0
  sock_sendmsg+0xb0/0xe0
  ____sys_sendmsg+0x4fa/0x6d0
  ___sys_sendmsg+0x12e/0x1b0
  __sys_sendmsg+0xa4/0x120
  do_syscall_64+0x2d/0x40
  entry_SYSCALL_64_after_hwframe+0x44/0xa9
 RIP: 0033:0x7f1f8cd4fe57
 Code: 0c 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 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 89 74 24 10
 RSP: 002b:00007ffdc1e193b8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f1f8cd4fe57
 RDX: 0000000000000000 RSI: 00007ffdc1e19420 RDI: 0000000000000003
 RBP: 000000005f85aafa R08: 0000000000000001 R09: 00007ffdc1e1936c
 R10: 000000000040522d R11: 0000000000000246 R12: 0000000000000001
 R13: 0000000000000000 R14: 00007ffdc1e1d6f0 R15: 0000000000482420

---
 include/net/tc_act/tc_tunnel_key.h |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

@@ -52,7 +52,10 @@ static inline struct ip_tunnel_info *tcf
 {
 #ifdef CONFIG_NET_CLS_ACT
 	struct tcf_tunnel_key *t = to_tunnel_key(a);
-	struct tcf_tunnel_key_params *params = rtnl_dereference(t->params);
+	struct tcf_tunnel_key_params *params;
+
+	params = rcu_dereference_protected(t->params,
+					   lockdep_is_held(&a->tcfa_lock));
 
 	return &params->tcft_enc_metadata->u.tun_info;
 #else

From: Magnus Karlsson <[email protected]>

commit 0706a78f31c4217ca144f630063ec9561a21548d upstream.

This reverts commit bd0687c18e635b63233dc87f38058cd728802ab4.

This patch causes a Tx only workload to go to sleep even when it does not have to, leading to misserable performance in skb mode. It fixed one rare problem but created a much worse one, so this need to be reverted while I try to craft a proper solution to the original problem.