5.4.164-rc1 review

This is the start of the stable review cycle for the 5.4.164 release. There are 70 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, 08 Dec 2021 14:55:37 +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-5.4.164-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 5.4.164-rc1

Wei Yongjun <[email protected]>
ipmi: msghandler: Make symbol 'remove_work_wq' static

Tianjia Zhang <[email protected]>
net/tls: Fix authentication failure in CCM mode

Helge Deller <[email protected]>
parisc: Mark cr16 CPU clocksource unstable on all SMP machines

Mordechay Goodstein <[email protected]>
iwlwifi: mvm: retry init flow if failed

Jay Dolan <[email protected]>
serial: 8250_pci: rewrite pericom_do_set_divisor()

Jay Dolan <[email protected]>
serial: 8250_pci: Fix ACCES entries in pci_serial_quirks array

Johan Hovold <[email protected]>
serial: core: fix transmit-buffer reset and memleak

Pierre Gondois <[email protected]>
serial: pl011: Add ACPI SBSA UART match id

Sven Eckelmann <[email protected]>
tty: serial: msm_serial: Deactivate RX DMA for polling support

Joerg Roedel <[email protected]>
x86/64/mm: Map all kernel memory into trampoline_pgd

Feng Tang <[email protected]>
x86/tsc: Disable clocksource watchdog for TSC on qualified platorms

Feng Tang <[email protected]>
x86/tsc: Add a timer to make sure TSC_adjust is always checked

Badhri Jagan Sridharan <[email protected]>
usb: typec: tcpm: Wait in SNK_DEBOUNCED until disconnect

Ole Ernst <[email protected]>
USB: NO_LPM quirk Lenovo Powered USB-C Travel Hub

Mathias Nyman <[email protected]>
xhci: Fix commad ring abort, write all 64 bits to CRCR register.

Maciej W. Rozycki <[email protected]>
vgacon: Propagate console boot parameters before calling `vc_resize'

Helge Deller <[email protected]>
parisc: Fix "make install" on newer debian releases

Helge Deller <[email protected]>
parisc: Fix KBUILD_IMAGE for self-extracting kernel

Qais Yousef <[email protected]>

    sched/uclamp: Fix rq->uclamp_max not set on first enqueue

Like Xu <[email protected]>
KVM: x86/pmu: Fix reserved bits for AMD PerfEvtSeln register

msizanoen1 <[email protected]>
ipv6: fix memory leak in fib6_rule_suppress

Rob Clark <[email protected]>
drm/msm: Do hw_init() before capturing GPU state

Tony Lu <[email protected]>
net/smc: Keep smc_close_final rc during active close

William Kucharski <[email protected]>
net/rds: correct socket tunable error in rds_tcp_tune()

Eric Dumazet <[email protected]>
ipv4: convert fib_num_tclassid_users to atomic_t

Eric Dumazet <[email protected]>

    net: annotate data-races on txq->xmit_lock_owner

Christophe JAILLET <[email protected]>
net: marvell: mvpp2: Fix the computation of shared CPUs

Sven Schuchmann <[email protected]>
net: usb: lan78xx: lan78xx_phy_init(): use PHY_POLL instead of "0" if no IRQ is available

Eiichi Tsukata <[email protected]>
rxrpc: Fix rxrpc_local leak in rxrpc_lookup_peer()

Li Zhijian <[email protected]>
selftests: net: Correct case name

Zhou Qingyang <[email protected]>
net/mlx4_en: Fix an use-after-free bug in mlx4_en_try_alloc_resources()

Arnd Bergmann <[email protected]>
siphash: use _unaligned version by default

Benjamin Poirier <[email protected]>
net: mpls: Fix notifications when deleting a device

Zhou Qingyang <[email protected]>
net: qlogic: qlcnic: Fix a NULL pointer dereference in qlcnic_83xx_add_rings()

Randy Dunlap <[email protected]>
natsemi: xtensa: fix section mismatch warnings

Aaro Koskinen <[email protected]>
i2c: cbus-gpio: set atomic transfer callback

Alain Volmat <[email protected]>
i2c: stm32f7: stop dma transfer in case of NACK

Alain Volmat <[email protected]>
i2c: stm32f7: recover the bus on access timeout

Alain Volmat <[email protected]>
i2c: stm32f7: flush TX FIFO upon transfer errors

Baokun Li <[email protected]>
sata_fsl: fix warning in remove_proc_entry when rmmod sata_fsl

Baokun Li <[email protected]>
sata_fsl: fix UAF in sata_fsl_port_stop when rmmod sata_fsl

Linus Torvalds <[email protected]>
fget: check that the fd still exists after getting a ref to it

Niklas Schnelle <[email protected]>
s390/pci: move pseudo-MMIO to prevent MIO overlap

Xiongfeng Wang <[email protected]>
cpufreq: Fix get_cpu_device() failure in add_cpu_dev_symlink()

Ioanna Alifieraki <[email protected]>
ipmi: Move remove_work to dedicated workqueue

Stanislaw Gruszka <[email protected]>
rt2x00: do not mark device gone on EPROTO errors during start

Masami Hiramatsu <[email protected]>
kprobes: Limit max data_size of the kretprobe instances

Stephen Suryaputra <[email protected]>
vrf: Reset IPCB/IP6CB when processing outbound pkts in vrf dev xmit

Wen Gu <[email protected]>
net/smc: Avoid warning of possible recursive locking

Ian Rogers <[email protected]>
perf report: Fix memory leaks around perf_tip()

Ian Rogers <[email protected]>
perf hist: Fix memory leak of a perf_hpp_fmt

Teng Qi <[email protected]>
net: ethernet: dec: tulip: de4x5: fix possible array overflows in type3_infoblock()

zhangyue <[email protected]>

    net: tulip: de4x5: fix the problem that the array 'lp->phy[8]' may be out of bound

Teng Qi <[email protected]>
ethernet: hisilicon: hns: hns_dsaf_misc: fix a possible array overflow in hns_dsaf_ge_srst_by_port()

Mario Limonciello <[email protected]>
ata: ahci: Add Green Sardine vendor ID as board_ahci_mobile

Mike Christie <[email protected]>
scsi: iscsi: Unblock session then wake up error handler

Manaf Meethalavalappu Pallikunhi <[email protected]>
thermal: core: Reset previous low and high trip during thermal zone init

Wang Yugui <[email protected]>
btrfs: check-integrity: fix a warning on write caching disabled disk

Vasily Gorbik <[email protected]>
s390/setup: avoid using memblock_enforce_memory_limit

Slark Xiao <[email protected]>
platform/x86: thinkpad_acpi: Fix WWAN device disabled issue after S3 deep

liuguoqiang <[email protected]>
net: return correct error code

Zekun Shen <[email protected]>
atlantic: Fix OOB read and write in hw_atl_utils_fw_rpc_wait

Wen Gu <[email protected]>
net/smc: Transfer remaining wait queue entries during fallback

Xing Song <[email protected]>
mac80211: do not access the IV when it was stripped

Julian Braha <[email protected]>
drm/sun4i: fix unmet dependency on RESET_CONTROLLER for PHY_SUN6I_MIPI_DPHY

Andreas Gruenbacher <[email protected]>
gfs2: Fix length of holes reported at end-of-file

Zhang Changzhong <[email protected]>

    can: j1939: j1939_tp_cmd_recv(): check the dst address of TP.CM_BAM

Russell King <[email protected]>
arm64: dts: mcbin: support 2W SFP modules

Geert Uytterhoeven <[email protected]>
of: clk: Make <linux/of_clk.h> self-contained

Benjamin Coddington <[email protected]>
NFSv42: Fix pagecache invalidation after COPY/CLONE

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

Diffstat:

 Makefile                                           |  4 +-
 arch/arm64/boot/dts/marvell/armada-8040-mcbin.dtsi |  3 +
 arch/parisc/Makefile                               |  5 ++
 arch/parisc/install.sh                             |  1 +
 arch/parisc/kernel/time.c                          | 24 ++------
 arch/s390/include/asm/pci_io.h                     |  7 ++-
 arch/s390/kernel/setup.c                           |  3 -
 arch/x86/kernel/tsc.c                              | 28 +++++++--
 arch/x86/kernel/tsc_sync.c                         | 41 +++++++++++++
 arch/x86/kvm/pmu_amd.c                             |  2 +-
 arch/x86/realmode/init.c                           | 12 +++-
 drivers/ata/ahci.c                                 |  1 +
 drivers/ata/sata_fsl.c                             | 20 ++++---
 drivers/char/ipmi/ipmi_msghandler.c                | 13 ++++-
 drivers/cpufreq/cpufreq.c                          |  9 ++-
 drivers/gpu/drm/msm/msm_debugfs.c                  |  1 +
 drivers/gpu/drm/sun4i/Kconfig                      |  1 +
 drivers/i2c/busses/i2c-cbus-gpio.c                 |  5 +-
 drivers/i2c/busses/i2c-stm32f7.c                   | 31 +++++++++-
 .../aquantia/atlantic/hw_atl/hw_atl_utils.c        | 10 ++++
 drivers/net/ethernet/dec/tulip/de4x5.c             | 34 ++++++-----
 drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c |  4 ++
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c    |  2 +-
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c     |  9 ++-
 drivers/net/ethernet/natsemi/xtsonic.c             |  2 +-
 .../net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c    | 10 +++-
 drivers/net/usb/lan78xx.c                          |  2 +-
 drivers/net/vrf.c                                  |  2 +
 drivers/net/wireless/intel/iwlwifi/iwl-drv.c       | 22 ++++---
 drivers/net/wireless/intel/iwlwifi/iwl-drv.h       |  3 +
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 24 +++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  3 +
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |  3 +
 drivers/net/wireless/ralink/rt2x00/rt2x00usb.c     |  3 +
 drivers/platform/x86/thinkpad_acpi.c               | 12 ----
 drivers/scsi/scsi_transport_iscsi.c                |  6 +-
 drivers/thermal/thermal_core.c                     |  2 +
 drivers/tty/serial/8250/8250_pci.c                 | 39 ++++++++-----
 drivers/tty/serial/amba-pl011.c                    |  1 +
 drivers/tty/serial/msm_serial.c                    |  3 +
 drivers/tty/serial/serial_core.c                   | 13 ++++-
 drivers/usb/core/quirks.c                          |  3 +
 drivers/usb/host/xhci-ring.c                       | 21 ++++---
 drivers/usb/typec/tcpm/tcpm.c                      |  4 --
 drivers/video/console/vgacon.c                     | 14 +++--
 fs/btrfs/disk-io.c                                 | 14 ++++-
 fs/file.c                                          |  4 ++
 fs/gfs2/bmap.c                                     |  2 +-
 fs/nfs/nfs42proc.c                                 |  5 +-
 include/linux/kprobes.h                            |  2 +
 include/linux/netdevice.h                          | 19 ++++--
 include/linux/of_clk.h                             |  3 +
 include/linux/siphash.h                            | 14 ++---
 include/net/fib_rules.h                            |  2 +-
 include/net/ip_fib.h                               |  2 +-
 include/net/netns/ipv4.h                           |  2 +-
 kernel/kprobes.c                                   |  3 +
 kernel/sched/core.c                                |  2 +-
 lib/siphash.c                                      | 12 ++--
 net/can/j1939/transport.c                          |  6 ++
 net/core/dev.c                                     |  5 +-
 net/core/fib_rules.c                               |  2 +-
 net/ipv4/devinet.c                                 |  2 +-
 net/ipv4/fib_frontend.c                            |  2 +-
 net/ipv4/fib_rules.c                               |  6 +-
 net/ipv4/fib_semantics.c                           |  4 +-
 net/ipv6/fib6_rules.c                              |  5 +-
 net/mac80211/rx.c                                  |  3 +-
 net/mpls/af_mpls.c                                 | 68 +++++++++++++++++-----
 net/rds/tcp.c                                      |  2 +-
 net/rxrpc/peer_object.c                            | 14 +++--
 net/smc/af_smc.c                                   | 14 +++++
 net/smc/smc_close.c                                |  8 ++-
 net/tls/tls_sw.c                                   |  4 +-
 tools/perf/builtin-report.c                        | 15 +++--
 tools/perf/ui/hist.c                               | 28 ++++-----
 tools/perf/util/hist.h                             |  1 -
 tools/perf/util/util.c                             | 14 ++---
 tools/perf/util/util.h                             |  2 +-
 tools/testing/selftests/net/fcnal-test.sh          |  4 +-
 80 files changed, 532 insertions(+), 225 deletions(-)

From: Benjamin Coddington <[email protected]>

commit 3f015d89a47cd8855cd92f71fff770095bd885a1 upstream.

The mechanism in use to allow the client to see the results of COPY/CLONE is to drop those pages from the pagecache. This forces the client to read those pages once more from the server. However, truncate_pagecache_range() zeros out partial pages instead of dropping them. Let us instead use invalidate_inode_pages2_range() with full-page offsets to ensure the client properly sees the results of COPY/CLONE operations.

---
 fs/nfs/nfs42proc.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -295,8 +295,9 @@ static ssize_t _nfs42_proc_copy(struct f
 			goto out;
 	}
 
-	truncate_pagecache_range(dst_inode, pos_dst,
-				 pos_dst + res->write_res.count);
+	WARN_ON_ONCE(invalidate_inode_pages2_range(dst_inode->i_mapping,
+					pos_dst >> PAGE_SHIFT,
+					(pos_dst + res->write_res.count - 1) >> PAGE_SHIFT));
 
 	status = res->write_res.count;
 out:

From: Geert Uytterhoeven <[email protected]>

commit 5df867145f8adad9e5cdf9d67db1fbc0f71351e9 upstream.

Depending on include order:

    include/linux/of_clk.h:11:45: warning: ‘struct device_node’ declared inside parameter list will not be visible outside of this definition or declaration
     unsigned int of_clk_get_parent_count(struct device_node *np);
						 ^~~~~~~~~~~
    include/linux/of_clk.h:12:43: warning: ‘struct device_node’ declared inside parameter list will not be visible outside of this definition or declaration
     const char *of_clk_get_parent_name(struct device_node *np, int index);
					       ^~~~~~~~~~~
    include/linux/of_clk.h:13:31: warning: ‘struct of_device_id’ declared inside parameter list will not be visible outside of this definition or declaration
     void of_clk_init(const struct of_device_id *matches);
				   ^~~~~~~~~~~~

Fix this by adding forward declarations for struct device_node and struct of_device_id.

---
 include/linux/of_clk.h |    3 +++
 1 file changed, 3 insertions(+)

@@ -6,6 +6,9 @@
 #ifndef __LINUX_OF_CLK_H
 #define __LINUX_OF_CLK_H
 
+struct device_node;
+struct of_device_id;
+
 #if defined(CONFIG_COMMON_CLK) && defined(CONFIG_OF)
 
 unsigned int of_clk_get_parent_count(struct device_node *np);

From: Russell King <[email protected]>

commit 05abc6a5dec2a8c123a50235ecd1ad8d75ffa7b4 upstream.

Allow the SFP cages to be used with 2W SFP modules.

---
 arch/arm64/boot/dts/marvell/armada-8040-mcbin.dtsi |    3 +++
 1 file changed, 3 insertions(+)

@@ -71,6 +71,7 @@
 		tx-fault-gpio  = <&cp1_gpio1 26 GPIO_ACTIVE_HIGH>;
 		pinctrl-names = "default";
 		pinctrl-0 = <&cp1_sfpp0_pins>;
+		maximum-power-milliwatt = <2000>;
 	};
 
 	sfp_eth1: sfp-eth1 {

@@ -83,6 +84,7 @@
 		tx-fault-gpio = <&cp0_gpio2 30 GPIO_ACTIVE_HIGH>;
 		pinctrl-names = "default";
 		pinctrl-0 = <&cp1_sfpp1_pins &cp0_sfpp1_pins>;
+		maximum-power-milliwatt = <2000>;
 	};
 
 	sfp_eth3: sfp-eth3 {

@@ -95,6 +97,7 @@
 		tx-fault-gpio = <&cp0_gpio2 19 GPIO_ACTIVE_HIGH>;
 		pinctrl-names = "default";
 		pinctrl-0 = <&cp0_sfp_1g_pins &cp1_sfp_1g_pins>;
+		maximum-power-milliwatt = <2000>;
 	};
 };
 

From: Zhang Changzhong <[email protected]>

commit 164051a6ab5445bd97f719f50b16db8b32174269 upstream.

The TP.CM_BAM message must be sent to the global address [1], so add a check to drop TP.CM_BAM sent to a non-global address.

Without this patch, the receiver will treat the following packets as normal RTS/CTS transport:
18EC0102#20090002FF002301
18EB0102#0100000000000000
18EB0102#020000FFFFFFFFFF

[1] SAE-J1939-82 2015 A.3.3 Row 1.

---
 net/can/j1939/transport.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -2004,6 +2004,12 @@ static void j1939_tp_cmd_recv(struct j19
 		extd = J1939_ETP;
 		/* fall through */
 	case J1939_TP_CMD_BAM: /* fall through */
+		if (cmd == J1939_TP_CMD_BAM && !j1939_cb_is_broadcast(skcb)) {
+			netdev_err_once(priv->ndev, "%s: BAM to unicast (%02x), ignoring!\n",
+					__func__, skcb->addr.sa);
+			return;
+		}
+		fallthrough;
 	case J1939_TP_CMD_RTS: /* fall through */
 		if (skcb->addr.type != extd)
 			return;

From: Andreas Gruenbacher <[email protected]>

[ Upstream commit f3506eee81d1f700d9ee2d2f4a88fddb669ec032 ]

Fix the length of holes reported at the end of a file: the length is relative to the beginning of the extent, not the seek position which is rounded down to the filesystem block size.

This bug went unnoticed for some time, but is now caught by the following assertion in iomap_iter_done():

WARN_ON_ONCE(iter->iomap.offset + iter->iomap.length <= iter->pos)

---
 fs/gfs2/bmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index aaec3c5b02028..dec5285a02e9d 100644
@@ -940,7 +940,7 @@ static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length,
 		else if (height == ip->i_height)
 			ret = gfs2_hole_size(inode, lblock, len, mp, iomap);
 		else
-			iomap->length = size - pos;
+			iomap->length = size - iomap->offset;
 	} else if (flags & IOMAP_WRITE) {
 		u64 alloc_size;
 

2.33.0

From: Julian Braha <[email protected]>

[ Upstream commit bb162bb2b4394108c8f055d1b115735331205e28 ]

When PHY_SUN6I_MIPI_DPHY is selected, and RESET_CONTROLLER is not selected, Kbuild gives the following warning:

WARNING: unmet direct dependencies detected for PHY_SUN6I_MIPI_DPHY
Depends on [n]: (ARCH_SUNXI [=n] || COMPILE_TEST [=y]) && HAS_IOMEM [=y] && COMMON_CLK [=y] && RESET_CONTROLLER [=n] Selected by [y]:
- DRM_SUN6I_DSI [=y] && HAS_IOMEM [=y] && DRM_SUN4I [=y]

This is because DRM_SUN6I_DSI selects PHY_SUN6I_MIPI_DPHY without selecting or depending on RESET_CONTROLLER, despite PHY_SUN6I_MIPI_DPHY depending on RESET_CONTROLLER.

These unmet dependency bugs were detected by Kismet, a static analysis tool for Kconfig. Please advise if this is not the appropriate solution.

v2:
Fixed indentation to match the rest of the file.

---
 drivers/gpu/drm/sun4i/Kconfig | 1 +
 1 file changed, 1 insertion(+)

index 37e90e42943f6..0e2d304f0d83f 100644
@@ -46,6 +46,7 @@ config DRM_SUN6I_DSI
 	default MACH_SUN8I
 	select CRC_CCITT
 	select DRM_MIPI_DSI
+	select RESET_CONTROLLER
 	select PHY_SUN6I_MIPI_DPHY
 	help
 	  Choose this option if you want have an Allwinner SoC with

2.33.0

From: Xing Song <[email protected]>

[ Upstream commit 77dfc2bc0bb4b8376ecd7a430f27a4a8fff6a5a0 ]

ieee80211_get_keyid() will return false value if IV has been stripped, such as return 0 for IP/ARP frames due to LLC header, and return -EINVAL for disassociation frames due to its length... etc. Don't try to access it if it's not present.

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

index c7e6bf7c22c78..282bf336b15a4 100644
@@ -1918,7 +1918,8 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx)
 		int keyid = rx->sta->ptk_idx;
 		sta_ptk = rcu_dereference(rx->sta->ptk[keyid]);
 
-		if (ieee80211_has_protected(fc)) {
+		if (ieee80211_has_protected(fc) &&
+		    !(status->flag & RX_FLAG_IV_STRIPPED)) {
 			cs = rx->sta->cipher_scheme;
 			keyid = ieee80211_get_keyid(rx->skb, cs);
 

2.33.0

From: Wen Gu <[email protected]>

[ Upstream commit 2153bd1e3d3dbf6a3403572084ef6ed31c53c5f0 ]

The SMC fallback is incomplete currently. There may be some wait queue entries remaining in smc socket->wq, which should be removed to clcsocket->wq during the fallback.

For example, in nginx/wrk benchmark, this issue causes an all-zeros test result:

server: nginx -g 'daemon off;'
client: smc_run wrk -c 1 -t 1 -d 5 http://11.200.15.93/index.html

Running 5s test @ http://11.200.15.93/index.html

     1 threads and 1 connections
     Thread Stats   Avg      Stdev     Max   ± Stdev
     	Latency     0.00us    0.00us   0.00us    -nan%
	Req/Sec     0.00      0.00     0.00      -nan%
	0 requests in 5.00s, 0.00B read
     Requests/sec:      0.00
     Transfer/sec:       0.00B

The reason for this all-zeros result is that when wrk used SMC to replace TCP, it added an eppoll_entry into smc socket->wq and expected to be notified if epoll events like EPOLL_IN/ EPOLL_OUT occurred on the smc socket.

However, once a fallback occurred, wrk switches to use clcsocket. Now it is clcsocket->wq instead of smc socket->wq which will be woken up. The eppoll_entry remaining in smc socket->wq does not work anymore and wrk stops the test.

This patch fixes this issue by removing remaining wait queue entries from smc socket->wq to clcsocket->wq during the fallback.

---
 net/smc/af_smc.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

index 5e1493f8deba7..00d3787387172 100644
@@ -467,12 +467,26 @@ static void smc_link_save_peer_info(struct smc_link *link,
 
 static void smc_switch_to_fallback(struct smc_sock *smc)
 {
+	wait_queue_head_t *smc_wait = sk_sleep(&smc->sk);
+	wait_queue_head_t *clc_wait = sk_sleep(smc->clcsock->sk);
+	unsigned long flags;
+
 	smc->use_fallback = true;
 	if (smc->sk.sk_socket && smc->sk.sk_socket->file) {
 		smc->clcsock->file = smc->sk.sk_socket->file;
 		smc->clcsock->file->private_data = smc->clcsock;
 		smc->clcsock->wq.fasync_list =
 			smc->sk.sk_socket->wq.fasync_list;
+
+		/* There may be some entries remaining in
+		 * smc socket->wq, which should be removed
+		 * to clcsocket->wq during the fallback.
+		 */
+		spin_lock_irqsave(&smc_wait->lock, flags);
+		spin_lock(&clc_wait->lock);
+		list_splice_init(&smc_wait->head, &clc_wait->head);
+		spin_unlock(&clc_wait->lock);
+		spin_unlock_irqrestore(&smc_wait->lock, flags);
 	}
 }
 

2.33.0

From: Zekun Shen <[email protected]>

[ Upstream commit b922f622592af76b57cbc566eaeccda0b31a3496 ]

This bug report shows up when running our research tools. The reports is SOOB read, but it seems SOOB write is also possible a few lines below.

In details, fw.len and sw.len are inputs coming from io. A len over the size of self->rpc triggers SOOB. The patch fixes the bugs by adding sanity checks.

The bugs are triggerable with compromised/malfunctioning devices. They are potentially exploitable given they first leak up to 0xffff bytes and able to overwrite the region later.

The patch is tested with QEMU emulater.
This is NOT tested with a real device.

Attached is the log we found by fuzzing.

BUG: KASAN: slab-out-of-bounds in
	hw_atl_utils_fw_upload_dwords+0x393/0x3c0 [atlantic]
Read of size 4 at addr ffff888016260b08 by task modprobe/213
CPU: 0 PID: 213 Comm: modprobe Not tainted 5.6.0 #1
Call Trace:
 dump_stack+0x76/0xa0
 print_address_description.constprop.0+0x16/0x200
 ? hw_atl_utils_fw_upload_dwords+0x393/0x3c0 [atlantic]
 ? hw_atl_utils_fw_upload_dwords+0x393/0x3c0 [atlantic]
 __kasan_report.cold+0x37/0x7c
 ? aq_hw_read_reg_bit+0x60/0x70 [atlantic]
 ? hw_atl_utils_fw_upload_dwords+0x393/0x3c0 [atlantic]
 kasan_report+0xe/0x20
 hw_atl_utils_fw_upload_dwords+0x393/0x3c0 [atlantic]
 hw_atl_utils_fw_rpc_call+0x95/0x130 [atlantic]
 hw_atl_utils_fw_rpc_wait+0x176/0x210 [atlantic]
 hw_atl_utils_mpi_create+0x229/0x2e0 [atlantic]
 ? hw_atl_utils_fw_rpc_wait+0x210/0x210 [atlantic]
 ? hw_atl_utils_initfw+0x9f/0x1c8 [atlantic]
 hw_atl_utils_initfw+0x12a/0x1c8 [atlantic]
 aq_nic_ndev_register+0x88/0x650 [atlantic]
 ? aq_nic_ndev_init+0x235/0x3c0 [atlantic]
 aq_pci_probe+0x731/0x9b0 [atlantic]
 ? aq_pci_func_init+0xc0/0xc0 [atlantic]
 local_pci_probe+0xd3/0x160
 pci_device_probe+0x23f/0x3e0

---
 .../ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c   | 10 ++++++++++
 1 file changed, 10 insertions(+)

index 873f9865f0d15..b7d70c33459fd 100644
@@ -481,6 +481,11 @@ int hw_atl_utils_fw_rpc_wait(struct aq_hw_s *self,
 			goto err_exit;
 
 		if (fw.len == 0xFFFFU) {
+			if (sw.len > sizeof(self->rpc)) {
+				printk(KERN_INFO "Invalid sw len: %x\n", sw.len);
+				err = -EINVAL;
+				goto err_exit;
+			}
 			err = hw_atl_utils_fw_rpc_call(self, sw.len);
 			if (err < 0)
 				goto err_exit;

@@ -489,6 +494,11 @@ int hw_atl_utils_fw_rpc_wait(struct aq_hw_s *self,
 
 	if (rpc) {
 		if (fw.len) {
+			if (fw.len > sizeof(self->rpc)) {
+				printk(KERN_INFO "Invalid fw len: %x\n", fw.len);
+				err = -EINVAL;
+				goto err_exit;
+			}
 			err =
 			hw_atl_utils_fw_downld_dwords(self,
 						      self->rpc_addr,

2.33.0

From: liuguoqiang <[email protected]>

[ Upstream commit 6def480181f15f6d9ec812bca8cbc62451ba314c ]

When kmemdup called failed and register_net_sysctl return NULL, should return ENOMEM instead of ENOBUFS

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

index 603a3495afa62..4a8ad46397c0e 100644
@@ -2585,7 +2585,7 @@ static int __devinet_sysctl_register(struct net *net, char *dev_name,
 free:
 	kfree(t);
 out:
-	return -ENOBUFS;
+	return -ENOMEM;
 }
 
 static void __devinet_sysctl_unregister(struct net *net,

2.33.0

From: Slark Xiao <[email protected]>

[ Upstream commit 39f53292181081d35174a581a98441de5da22bc9 ]

When WWAN device wake from S3 deep, under thinkpad platform, WWAN would be disabled. This disable status could be checked by command 'nmcli r wwan' or 'rfkill list'.

Issue analysis as below:
When host resume from S3 deep, thinkpad_acpi driver would
call hotkey_resume() function. Finnaly, it will use wan_get_status to check the current status of WWAN device. During this resume progress, wan_get_status would always return off even WWAN boot up completely.
In patch V2, Hans said 'sw_state should be unchanged
after a suspend/resume. It's better to drop the tpacpi_rfk_update_swstate call all together from the resume path'.
And it's confimed by Lenovo that GWAN is no longer
available from WHL generation because the design does not match with current pin control.

---
 drivers/platform/x86/thinkpad_acpi.c | 12 ------------
 1 file changed, 12 deletions(-)

index 3028d9f1ac59c..5d114088c88fb 100644
@@ -1188,15 +1188,6 @@ static int tpacpi_rfk_update_swstate(const struct tpacpi_rfk *tp_rfk)
 	return status;
 }
 
-/* Query FW and update rfkill sw state for all rfkill switches */
-static void tpacpi_rfk_update_swstate_all(void)
-{
-	unsigned int i;
-
-	for (i = 0; i < TPACPI_RFK_SW_MAX; i++)
-		tpacpi_rfk_update_swstate(tpacpi_rfkill_switches[i]);
-}
-
 /*
  * Sync the HW-blocking state of all rfkill switches,
  * do notice it causes the rfkill core to schedule uevents

@@ -3135,9 +3126,6 @@ static void tpacpi_send_radiosw_update(void)
 	if (wlsw == TPACPI_RFK_RADIO_OFF)
 		tpacpi_rfk_update_hwblock_state(true);
 
-	/* Sync sw blocking state */
-	tpacpi_rfk_update_swstate_all();
-
 	/* Sync hw blocking state last if it is hw-unblocked */
 	if (wlsw == TPACPI_RFK_RADIO_ON)
 		tpacpi_rfk_update_hwblock_state(false);

2.33.0

From: Vasily Gorbik <[email protected]>

[ Upstream commit 5dbc4cb4667457b0c53bcd7bff11500b3c362975 ]

There is a difference in how architectures treat "mem=" option. For some that is an amount of online memory, for s390 and x86 this is the limiting max address. Some memblock api like memblock_enforce_memory_limit() take limit argument and explicitly treat it as the size of online memory, and use __find_max_addr to convert it to an actual max address. Current s390 usage:

memblock_enforce_memory_limit(memblock_end_of_DRAM());

yields different results depending on presence of memory holes (offline memory blocks in between online memory). If there are no memory holes limit == max_addr in memblock_enforce_memory_limit() and it does trim online memory and reserved memory regions. With memory holes present it actually does nothing.

Since we already use memblock_remove() explicitly to trim online memory regions to potential limit (think mem=, kdump, addressing limits, etc.) drop the usage of memblock_enforce_memory_limit() altogether. Trimming reserved regions should not be required, since we now use memblock_set_current_limit() to limit allocations and any explicit memory reservations above the limit is an actual problem we should not hide.

---
 arch/s390/kernel/setup.c | 3 ---
 1 file changed, 3 deletions(-)

index f661f176966f5..9a0316a067a11 100644
@@ -841,9 +841,6 @@ static void __init setup_memory(void)
 		storage_key_init_range(reg->base, reg->base + reg->size);
 	}
 	psw_set_key(PAGE_DEFAULT_KEY);
-
-	/* Only cosmetics */
-	memblock_enforce_memory_limit(memblock_end_of_DRAM());
 }
 
 /*

2.33.0

From: Wang Yugui <[email protected]>

[ Upstream commit a91cf0ffbc244792e0b3ecf7d0fddb2f344b461f ]

When a disk has write caching disabled, we skip submission of a bio with flush and sync requests before writing the superblock, since it's not needed. However when the integrity checker is enabled, this results in reports that there are metadata blocks referred by a superblock that were not properly flushed. So don't skip the bio submission only when the integrity checker is enabled for the sake of simplicity, since this is a debug tool and not meant for use in non-debug builds.

fstests/btrfs/220 trigger a check-integrity warning like the following when CONFIG_BTRFS_FS_CHECK_INTEGRITY=y and the disk with WCE=0.

btrfs: attempt to write superblock which references block M @5242880 (sdb2/5242880/0) which is not flushed out of disk's write cache (block flush_gen=1, dev->flush_gen=0)!

  ------------[ cut here ]------------
  WARNING: CPU: 28 PID: 843680 at fs/btrfs/check-integrity.c:2196 btrfsic_process_written_superblock+0x22a/0x2a0 [btrfs]
  CPU: 28 PID: 843680 Comm: umount Not tainted 5.15.0-0.rc5.39.el8.x86_64 #1
  Hardware name: Dell Inc. Precision T7610/0NK70N, BIOS A18 09/11/2019
  RIP: 0010:btrfsic_process_written_superblock+0x22a/0x2a0 [btrfs]
  RSP: 0018:ffffb642afb47940 EFLAGS: 00010246
  RAX: 0000000000000000 RBX: 0000000000000002 RCX: 0000000000000000
  RDX: 00000000ffffffff RSI: ffff8b722fc97d00 RDI: ffff8b722fc97d00
  RBP: ffff8b5601c00000 R08: 0000000000000000 R09: c0000000ffff7fff
  R10: 0000000000000001 R11: ffffb642afb476f8 R12: ffffffffffffffff
  R13: ffffb642afb47974 R14: ffff8b5499254c00 R15: 0000000000000003
  FS:  00007f00a06d4080(0000) GS:ffff8b722fc80000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 00007fff5cff5ff0 CR3: 00000001c0c2a006 CR4: 00000000001706e0
  Call Trace:
   btrfsic_process_written_block+0x2f7/0x850 [btrfs]
   __btrfsic_submit_bio.part.19+0x310/0x330 [btrfs]
   ? bio_associate_blkg_from_css+0xa4/0x2c0
   btrfsic_submit_bio+0x18/0x30 [btrfs]
   write_dev_supers+0x81/0x2a0 [btrfs]
   ? find_get_pages_range_tag+0x219/0x280
   ? pagevec_lookup_range_tag+0x24/0x30
   ? __filemap_fdatawait_range+0x6d/0xf0
   ? __raw_callee_save___native_queued_spin_unlock+0x11/0x1e
   ? find_first_extent_bit+0x9b/0x160 [btrfs]
   ? __raw_callee_save___native_queued_spin_unlock+0x11/0x1e
   write_all_supers+0x1b3/0xa70 [btrfs]
   ? __raw_callee_save___native_queued_spin_unlock+0x11/0x1e
   btrfs_commit_transaction+0x59d/0xac0 [btrfs]
   close_ctree+0x11d/0x339 [btrfs]
   generic_shutdown_super+0x71/0x110
   kill_anon_super+0x14/0x30
   btrfs_kill_super+0x12/0x20 [btrfs]
   deactivate_locked_super+0x31/0x70
   cleanup_mnt+0xb8/0x140
   task_work_run+0x6d/0xb0
   exit_to_user_mode_prepare+0x1f0/0x200
   syscall_exit_to_user_mode+0x12/0x30
   do_syscall_64+0x46/0x80
   entry_SYSCALL_64_after_hwframe+0x44/0xae
  RIP: 0033:0x7f009f711dfb
  RSP: 002b:00007fff5cff7928 EFLAGS: 00000246 ORIG_RAX: 00000000000000a6
  RAX: 0000000000000000 RBX: 000055b68c6c9970 RCX: 00007f009f711dfb
  RDX: 0000000000000001 RSI: 0000000000000000 RDI: 000055b68c6c9b50
  RBP: 0000000000000000 R08: 000055b68c6ca900 R09: 00007f009f795580
  R10: 0000000000000000 R11: 0000000000000246 R12: 000055b68c6c9b50
  R13: 00007f00a04bf184 R14: 0000000000000000 R15: 00000000ffffffff
  ---[ end trace 2c4b82abcef9eec4 ]---
  S-65536(sdb2/65536/1)
   -->
  M-1064960(sdb2/1064960/1)

---
 fs/btrfs/disk-io.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

index 1499531bc1511..f18c6d97932ed 100644
@@ -3636,11 +3636,23 @@ static void btrfs_end_empty_barrier(struct bio *bio)
  */
 static void write_dev_flush(struct btrfs_device *device)
 {
-	struct request_queue *q = bdev_get_queue(device->bdev);
 	struct bio *bio = device->flush_bio;
 
+#ifndef CONFIG_BTRFS_FS_CHECK_INTEGRITY
+	/*
+	 * When a disk has write caching disabled, we skip submission of a bio
+	 * with flush and sync requests before writing the superblock, since
+	 * it's not needed. However when the integrity checker is enabled, this
+	 * results in reports that there are metadata blocks referred by a
+	 * superblock that were not properly flushed. So don't skip the bio
+	 * submission only when the integrity checker is enabled for the sake
+	 * of simplicity, since this is a debug tool and not meant for use in
+	 * non-debug builds.
+	 */
+	struct request_queue *q = bdev_get_queue(device->bdev);
 	if (!test_bit(QUEUE_FLAG_WC, &q->queue_flags))
 		return;
+#endif
 
 	bio_reset(bio);
 	bio->bi_end_io = btrfs_end_empty_barrier;

2.33.0

From: Manaf Meethalavalappu Pallikunhi <[email protected]>

[ Upstream commit 99b63316c39988039965693f5f43d8b4ccb1c86c ]

During the suspend is in process, thermal_zone_device_update bails out thermal zone re-evaluation for any sensor trip violation without setting next valid trip to that sensor. It assumes during resume it will re-evaluate same thermal zone and update trip. But when it is in suspend temperature goes down and on resume path while updating thermal zone if temperature is less than previously violated trip, thermal zone set trip function evaluates the same previous high and previous low trip as new high and low trip. Since there is no change in high/low trip, it bails out from thermal zone set trip API without setting any trip. It leads to a case where sensor high trip or low trip is disabled forever even though thermal zone has a valid high or low trip.

During thermal zone device init, reset thermal zone previous high and low trip. It resolves above mentioned scenario.

---
 drivers/thermal/thermal_core.c | 2 ++
 1 file changed, 2 insertions(+)

index 20eab56b02cb9..f4490b8120176 100644
@@ -460,6 +460,8 @@ static void thermal_zone_device_init(struct thermal_zone_device *tz)
 {
 	struct thermal_instance *pos;
 	tz->temperature = THERMAL_TEMP_INVALID;
+	tz->prev_low_trip = -INT_MAX;
+	tz->prev_high_trip = INT_MAX;
 	list_for_each_entry(pos, &tz->thermal_instances, tz_node)
 		pos->initialized = false;
 }

2.33.0

From: Mike Christie <[email protected]>

[ Upstream commit a0c2f8b6709a9a4af175497ca65f93804f57b248 ]

We can race where iscsi_session_recovery_timedout() has woken up the error handler thread and it's now setting the devices to offline, and session_recovery_timedout()'s call to scsi_target_unblock() is also trying to set the device's state to transport-offline. We can then get a mix of states.

For the case where we can't relogin we want the devices to be in transport-offline so when we have repaired the connection __iscsi_unblock_session() can set the state back to running.

Set the device state then call into libiscsi to wake up the error handler.

---
 drivers/scsi/scsi_transport_iscsi.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

index 6f21cb75d95fd..f6cce0befa7de 100644
@@ -1894,12 +1894,12 @@ static void session_recovery_timedout(struct work_struct *work)
 	}
 	spin_unlock_irqrestore(&session->lock, flags);
 
-	if (session->transport->session_recovery_timedout)
-		session->transport->session_recovery_timedout(session);
-
 	ISCSI_DBG_TRANS_SESSION(session, "Unblocking SCSI target\n");
 	scsi_target_unblock(&session->dev, SDEV_TRANSPORT_OFFLINE);
 	ISCSI_DBG_TRANS_SESSION(session, "Completed unblocking SCSI target\n");
+
+	if (session->transport->session_recovery_timedout)
+		session->transport->session_recovery_timedout(session);
 }
 
 static void __iscsi_unblock_session(struct work_struct *work)

2.33.0

From: Mario Limonciello <[email protected]>

[ Upstream commit 1527f69204fe35f341cb599f1cb01bd02daf4374 ]

AMD requires that the SATA controller be configured for devsleep in order for S0i3 entry to work properly.

commit b1a9585cc396 ("ata: ahci: Enable DEVSLP by default on x86 with SLP_S0") sets up a kernel policy to enable devsleep on Intel mobile platforms that are using s0ix. Add the PCI ID for the SATA controller in Green Sardine platforms to extend this policy by default for AMD based systems using s0i3 as well.

---
 drivers/ata/ahci.c | 1 +
 1 file changed, 1 insertion(+)

index 8beb418ce167b..6f572967b5552 100644
@@ -420,6 +420,7 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	/* AMD */
 	{ PCI_VDEVICE(AMD, 0x7800), board_ahci }, /* AMD Hudson-2 */
 	{ PCI_VDEVICE(AMD, 0x7900), board_ahci }, /* AMD CZ */
+	{ PCI_VDEVICE(AMD, 0x7901), board_ahci_mobile }, /* AMD Green Sardine */
 	/* AMD is using RAID class only for ahci controllers */
 	{ PCI_VENDOR_ID_AMD, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
 	  PCI_CLASS_STORAGE_RAID << 8, 0xffffff, board_ahci },

2.33.0

From: Teng Qi <[email protected]>

[ Upstream commit a66998e0fbf213d47d02813b9679426129d0d114 ]

The if statement:
if (port >= DSAF_GE_NUM)

        return;

limits the value of port less than DSAF_GE_NUM (i.e., 8). However, if the value of port is 6 or 7, an array overflow could occur:
port_rst_off = dsaf_dev->mac_cb[port]->port_rst_off;

because the length of dsaf_dev->mac_cb is DSAF_MAX_PORT_NUM (i.e., 6).

To fix this possible array overflow, we first check port and if it is greater than or equal to DSAF_MAX_PORT_NUM, the function returns.

---
 drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c | 4 ++++
 1 file changed, 4 insertions(+)

index ed3829ae4ef1b..580199fdd0c22 100644
@@ -398,6 +398,10 @@ static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port,
 		return;
 
 	if (!HNS_DSAF_IS_DEBUG(dsaf_dev)) {
+		/* DSAF_MAX_PORT_NUM is 6, but DSAF_GE_NUM is 8.
+		   We need check to prevent array overflow */
+		if (port >= DSAF_MAX_PORT_NUM)
+			return;
 		reg_val_1  = 0x1 << port;
 		port_rst_off = dsaf_dev->mac_cb[port]->port_rst_off;
 		/* there is difference between V1 and V2 in register.*/

2.33.0

From: zhangyue <[email protected]>

[ Upstream commit 61217be886b5f7402843677e4be7e7e83de9cb41 ]

In line 5001, if all id in the array 'lp->phy[8]' is not 0, when the 'for' end, the 'k' is 8.

At this time, the array 'lp->phy[8]' may be out of bound.

---
 drivers/net/ethernet/dec/tulip/de4x5.c | 30 +++++++++++++++-----------
 1 file changed, 17 insertions(+), 13 deletions(-)

index c813e6f2b371e..a80252973171f 100644
@@ -4999,19 +4999,23 @@ mii_get_phy(struct net_device *dev)
 	}
 	if ((j == limit) && (i < DE4X5_MAX_MII)) {
 	    for (k=0; k < DE4X5_MAX_PHY && lp->phy[k].id; k++);
-	    lp->phy[k].addr = i;
-	    lp->phy[k].id = id;
-	    lp->phy[k].spd.reg = GENERIC_REG;      /* ANLPA register         */
-	    lp->phy[k].spd.mask = GENERIC_MASK;    /* 100Mb/s technologies   */
-	    lp->phy[k].spd.value = GENERIC_VALUE;  /* TX & T4, H/F Duplex    */
-	    lp->mii_cnt++;
-	    lp->active++;
-	    printk("%s: Using generic MII device control. If the board doesn't operate,\nplease mail the following dump to the author:\n", dev->name);
-	    j = de4x5_debug;
-	    de4x5_debug |= DEBUG_MII;
-	    de4x5_dbg_mii(dev, k);
-	    de4x5_debug = j;
-	    printk("\n");
+	    if (k < DE4X5_MAX_PHY) {
+		lp->phy[k].addr = i;
+		lp->phy[k].id = id;
+		lp->phy[k].spd.reg = GENERIC_REG;      /* ANLPA register         */
+		lp->phy[k].spd.mask = GENERIC_MASK;    /* 100Mb/s technologies   */
+		lp->phy[k].spd.value = GENERIC_VALUE;  /* TX & T4, H/F Duplex    */
+		lp->mii_cnt++;
+		lp->active++;
+		printk("%s: Using generic MII device control. If the board doesn't operate,\nplease mail the following dump to the author:\n", dev->name);
+		j = de4x5_debug;
+		de4x5_debug |= DEBUG_MII;
+		de4x5_dbg_mii(dev, k);
+		de4x5_debug = j;
+		printk("\n");
+	    } else {
+		goto purgatory;
+	    }
 	}
     }
   purgatory:

2.33.0

From: Teng Qi <[email protected]>

[ Upstream commit 0fa68da72c3be09e06dd833258ee89c33374195f ]

The definition of macro MOTO_SROM_BUG is:
#define MOTO_SROM_BUG (lp->active == 8 && (get_unaligned_le32(
dev->dev_addr) & 0x00ffffff) == 0x3e0008)

and the if statement
if (MOTO_SROM_BUG) lp->active = 0;

using this macro indicates lp->active could be 8. If lp->active is 8 and the second comparison of this macro is false. lp->active will remain 8 in:
lp->phy[lp->active].gep = (*p ? p : NULL); p += (2 * (*p) + 1); lp->phy[lp->active].rst = (*p ? p : NULL); p += (2 * (*p) + 1); lp->phy[lp->active].mc = get_unaligned_le16(p); p += 2; lp->phy[lp->active].ana = get_unaligned_le16(p); p += 2; lp->phy[lp->active].fdx = get_unaligned_le16(p); p += 2; lp->phy[lp->active].ttm = get_unaligned_le16(p); p += 2; lp->phy[lp->active].mci = *p;

However, the length of array lp->phy is 8, so array overflows can occur. To fix these possible array overflows, we first check lp->active and then return -EINVAL if it is greater or equal to ARRAY_SIZE(lp->phy) (i.e. 8).

---
 drivers/net/ethernet/dec/tulip/de4x5.c | 4 ++++
 1 file changed, 4 insertions(+)

index a80252973171f..c97fc0e384ca6 100644
@@ -4708,6 +4708,10 @@ type3_infoblock(struct net_device *dev, u_char count, u_char *p)
         lp->ibn = 3;
         lp->active = *p++;
 	if (MOTO_SROM_BUG) lp->active = 0;
+	/* if (MOTO_SROM_BUG) statement indicates lp->active could
+	 * be 8 (i.e. the size of array lp->phy) */
+	if (WARN_ON(lp->active >= ARRAY_SIZE(lp->phy)))
+		return -EINVAL;
 	lp->phy[lp->active].gep = (*p ? p : NULL); p += (2 * (*p) + 1);
 	lp->phy[lp->active].rst = (*p ? p : NULL); p += (2 * (*p) + 1);
 	lp->phy[lp->active].mc  = get_unaligned_le16(p); p += 2;

2.33.0

From: Ian Rogers <[email protected]>

[ Upstream commit 0ca1f534a776cc7d42f2c33da4732b74ec2790cd ]

perf_hpp__column_unregister() removes an entry from a list but doesn't free the memory causing a memory leak spotted by leak sanitizer.

Add the free while at the same time reducing the scope of the function to static.

---
 tools/perf/ui/hist.c   | 28 ++++++++++++++--------------
 tools/perf/util/hist.h |  1 -
 2 files changed, 14 insertions(+), 15 deletions(-)

index f736755000616..9ae316445f04b 100644
@@ -472,6 +472,18 @@ struct perf_hpp_list perf_hpp_list = {
 #undef __HPP_SORT_ACC_FN
 #undef __HPP_SORT_RAW_FN
 
+static void fmt_free(struct perf_hpp_fmt *fmt)
+{
+	/*
+	 * At this point fmt should be completely
+	 * unhooked, if not it's a bug.
+	 */
+	BUG_ON(!list_empty(&fmt->list));
+	BUG_ON(!list_empty(&fmt->sort_list));
+
+	if (fmt->free)
+		fmt->free(fmt);
+}
 
 void perf_hpp__init(void)
 {

@@ -535,9 +547,10 @@ void perf_hpp_list__prepend_sort_field(struct perf_hpp_list *list,
 	list_add(&format->sort_list, &list->sorts);
 }
 
-void perf_hpp__column_unregister(struct perf_hpp_fmt *format)
+static void perf_hpp__column_unregister(struct perf_hpp_fmt *format)
 {
 	list_del_init(&format->list);
+	fmt_free(format);
 }
 
 void perf_hpp__cancel_cumulate(void)

@@ -609,19 +622,6 @@ void perf_hpp__append_sort_keys(struct perf_hpp_list *list)
 }
 
 
-static void fmt_free(struct perf_hpp_fmt *fmt)
-{
-	/*
-	 * At this point fmt should be completely
-	 * unhooked, if not it's a bug.
-	 */
-	BUG_ON(!list_empty(&fmt->list));
-	BUG_ON(!list_empty(&fmt->sort_list));
-
-	if (fmt->free)
-		fmt->free(fmt);
-}
-
 void perf_hpp__reset_output_field(struct perf_hpp_list *list)
 {
 	struct perf_hpp_fmt *fmt, *tmp;

index 4792731307947..ecce30f086de7 100644
@@ -361,7 +361,6 @@ enum {
 };
 
 void perf_hpp__init(void);
-void perf_hpp__column_unregister(struct perf_hpp_fmt *format);
 void perf_hpp__cancel_cumulate(void);
 void perf_hpp__setup_output_field(struct perf_hpp_list *list);
 void perf_hpp__reset_output_field(struct perf_hpp_list *list);

2.33.0

From: Ian Rogers <[email protected]>

[ Upstream commit d9fc706108c15f8bc2d4ccccf8e50f74830fabd9 ]

perf_tip() may allocate memory or use a literal, this means memory wasn't freed if allocated. Change the API so that literals aren't used.

At the same time add missing frees for system_path. These issues were spotted using leak sanitizer.

---
 tools/perf/builtin-report.c | 15 +++++++++------
 tools/perf/util/util.c      | 14 +++++++-------
 tools/perf/util/util.h      |  2 +-
 3 files changed, 17 insertions(+), 14 deletions(-)

index d3c0b04e2e22b..dc228bdf2bbc2 100644
@@ -569,14 +569,17 @@ static int report__browse_hists(struct report *rep)
 	int ret;
 	struct perf_session *session = rep->session;
 	struct evlist *evlist = session->evlist;
-	const char *help = perf_tip(system_path(TIPDIR));
+	char *help = NULL, *path = NULL;
 
-	if (help == NULL) {
+	path = system_path(TIPDIR);
+	if (perf_tip(&help, path) || help == NULL) {
 		/* fallback for people who don't install perf ;-) */
-		help = perf_tip(DOCDIR);
-		if (help == NULL)
-			help = "Cannot load tips.txt file, please install perf!";
+		free(path);
+		path = system_path(DOCDIR);
+		if (perf_tip(&help, path) || help == NULL)
+			help = strdup("Cannot load tips.txt file, please install perf!");
 	}
+	free(path);
 
 	switch (use_browser) {
 	case 1:

@@ -598,7 +601,7 @@ static int report__browse_hists(struct report *rep)
 		ret = perf_evlist__tty_browse_hists(evlist, rep, help);
 		break;
 	}
-
+	free(help);
 	return ret;
 }
 

index ae56c766eda16..b3c1ae288b478 100644
@@ -343,32 +343,32 @@ fetch_kernel_version(unsigned int *puint, char *str,
 	return 0;
 }
 
-const char *perf_tip(const char *dirpath)
+int perf_tip(char **strp, const char *dirpath)
 {
 	struct strlist *tips;
 	struct str_node *node;
-	char *tip = NULL;
 	struct strlist_config conf = {
 		.dirname = dirpath,
 		.file_only = true,
 	};
+	int ret = 0;
 
+	*strp = NULL;
 	tips = strlist__new("tips.txt", &conf);
 	if (tips == NULL)
-		return errno == ENOENT ? NULL :
-			"Tip: check path of tips.txt or get more memory! ;-p";
+		return -errno;
 
 	if (strlist__nr_entries(tips) == 0)
 		goto out;
 
 	node = strlist__entry(tips, random() % strlist__nr_entries(tips));
-	if (asprintf(&tip, "Tip: %s", node->s) < 0)
-		tip = (char *)"Tip: get more memory! ;-)";
+	if (asprintf(strp, "Tip: %s", node->s) < 0)
+		ret = -ENOMEM;
 
 out:
 	strlist__delete(tips);
 
-	return tip;
+	return ret;
 }
 
 char *perf_exe(char *buf, int len)

index 9969b8b46f7c3..e4a7e1cafc70a 100644
@@ -37,7 +37,7 @@ int fetch_kernel_version(unsigned int *puint,
 #define KVER_FMT	"%d.%d.%d"
 #define KVER_PARAM(x)	KVER_VERSION(x), KVER_PATCHLEVEL(x), KVER_SUBLEVEL(x)
 
-const char *perf_tip(const char *dirpath);
+int perf_tip(char **strp, const char *dirpath);
 
 #ifndef HAVE_SCHED_GETCPU_SUPPORT
 int sched_getcpu(void);

2.33.0

From: Wen Gu <[email protected]>

[ Upstream commit 7a61432dc81375be06b02f0061247d3efbdfce3a ]

Possible recursive locking is detected by lockdep when SMC falls back to TCP. The corresponding warnings are as follows:

============================================
WARNING: possible recursive locking detected
5.16.0-rc1+ #18 Tainted: G E
--------------------------------------------
wrk/1391 is trying to acquire lock:
ffff975246c8e7d8 (&ei->socket.wq.wait){..-.}-{3:3}, at: smc_switch_to_fallback+0x109/0x250 [smc]

but task is already holding lock:
ffff975246c8f918 (&ei->socket.wq.wait){..-.}-{3:3}, at: smc_switch_to_fallback+0xfe/0x250 [smc]

other info that might help us debug this:

  Possible unsafe locking scenario:

        CPU0
        ----

lock(&ei->socket.wq.wait);
lock(&ei->socket.wq.wait);

*** DEADLOCK ***

May be due to missing lock nesting notation

2 locks held by wrk/1391:
#0: ffff975246040130 (sk_lock-AF_SMC){+.+.}-{0:0}, at: smc_connect+0x43/0x150 [smc] #1: ffff975246c8f918 (&ei->socket.wq.wait){..-.}-{3:3}, at: smc_switch_to_fallback+0xfe/0x250 [smc]

 stack backtrace:
 Call Trace:
  <TASK>
  dump_stack_lvl+0x56/0x7b
  __lock_acquire+0x951/0x11f0
  lock_acquire+0x27a/0x320
  ? smc_switch_to_fallback+0x109/0x250 [smc]
  ? smc_switch_to_fallback+0xfe/0x250 [smc]
  _raw_spin_lock_irq+0x3b/0x80
  ? smc_switch_to_fallback+0x109/0x250 [smc]
  smc_switch_to_fallback+0x109/0x250 [smc]
  smc_connect_fallback+0xe/0x30 [smc]
  __smc_connect+0xcf/0x1090 [smc]
  ? mark_held_locks+0x61/0x80
  ? __local_bh_enable_ip+0x77/0xe0
  ? lockdep_hardirqs_on+0xbf/0x130
  ? smc_connect+0x12a/0x150 [smc]
  smc_connect+0x12a/0x150 [smc]
  __sys_connect+0x8a/0xc0
  ? syscall_enter_from_user_mode+0x20/0x70
  __x64_sys_connect+0x16/0x20
  do_syscall_64+0x34/0x90
  entry_SYSCALL_64_after_hwframe+0x44/0xae

The nested locking in smc_switch_to_fallback() is considered to possibly cause a deadlock because smc_wait->lock and clc_wait->lock are the same type of lock. But actually it is safe so far since there is no other place trying to obtain smc_wait->lock when clc_wait->lock is held. So the patch replaces spin_lock() with spin_lock_nested() to avoid false report by lockdep.

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

index 00d3787387172..fa3b20e5f4608 100644
@@ -483,7 +483,7 @@ static void smc_switch_to_fallback(struct smc_sock *smc)
 		 * to clcsocket->wq during the fallback.
 		 */
 		spin_lock_irqsave(&smc_wait->lock, flags);
-		spin_lock(&clc_wait->lock);
+		spin_lock_nested(&clc_wait->lock, SINGLE_DEPTH_NESTING);
 		list_splice_init(&smc_wait->head, &clc_wait->head);
 		spin_unlock(&clc_wait->lock);
 		spin_unlock_irqrestore(&smc_wait->lock, flags);

2.33.0

From: Stephen Suryaputra <[email protected]>

commit ee201011c1e1563c114a55c86eb164b236f18e84 upstream.

IPCB/IP6CB need to be initialized when processing outbound v4 or v6 pkts in the codepath of vrf device xmit function so that leftover garbage doesn't cause futher code that uses the CB to incorrectly process the pkt.

One occasion of the issue might occur when MPLS route uses the vrf device as the outgoing device such as when the route is added using "ip -f mpls route add <label> dev <vrf>" command.

The problems seems to exist since day one. Hence I put the day one commits on the Fixes tags.

---
 drivers/net/vrf.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -221,6 +221,7 @@ static netdev_tx_t vrf_process_v6_outbou
 	/* strip the ethernet header added for pass through VRF device */
 	__skb_pull(skb, skb_network_offset(skb));
 
+	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
 	ret = vrf_ip6_local_out(net, skb->sk, skb);
 	if (unlikely(net_xmit_eval(ret)))
 		dev->stats.tx_errors++;

@@ -304,6 +305,7 @@ static netdev_tx_t vrf_process_v4_outbou
 					       RT_SCOPE_LINK);
 	}
 
+	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
 	ret = vrf_ip_local_out(dev_net(skb_dst(skb)->dev), skb->sk, skb);
 	if (unlikely(net_xmit_eval(ret)))
 		vrf_dev->stats.tx_errors++;

From: Masami Hiramatsu <[email protected]>

commit 6bbfa44116689469267f1a6e3d233b52114139d2 upstream.

The 'kprobe::data_size' is unsigned, thus it can not be negative. But if user sets it enough big number (e.g. (size_t)-8), the result of 'data_size
+ sizeof(struct kretprobe_instance)' becomes smaller than sizeof(struct kretprobe_instance) or zero. In result, the kretprobe_instance are allocated without enough memory, and kretprobe accesses outside of allocated memory.

To avoid this issue, introduce a max limitation of the kretprobe::data_size. 4KB per instance should be OK.

---
 include/linux/kprobes.h |    2 ++
 kernel/kprobes.c        |    3 +++
 2 files changed, 5 insertions(+)

@@ -155,6 +155,8 @@ struct kretprobe {
 	raw_spinlock_t lock;
 };
 
+#define KRETPROBE_MAX_DATA_SIZE	4096
+
 struct kretprobe_instance {
 	struct hlist_node hlist;
 	struct kretprobe *rp;

@@ -2003,6 +2003,9 @@ int register_kretprobe(struct kretprobe
 		}
 	}
 
+	if (rp->data_size > KRETPROBE_MAX_DATA_SIZE)
+		return -E2BIG;
+
 	rp->kp.pre_handler = pre_handler_kretprobe;
 	rp->kp.post_handler = NULL;
 	rp->kp.fault_handler = NULL;

From: Stanislaw Gruszka <[email protected]>

commit ed53ae75693096f1c10b4561edd31a07b631bd72 upstream.

As reported by Exuvo is possible that we have lot's of EPROTO errors during device start i.e. firmware load. But after that device works correctly. Hence marking device gone by few EPROTO errors done by commit e383c70474db ("rt2x00: check number of EPROTO errors") caused regression - Exuvo device stop working after kernel update. To fix disable the check during device start.

---
 drivers/net/wireless/ralink/rt2x00/rt2x00usb.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -25,6 +25,9 @@ static bool rt2x00usb_check_usb_error(st
 	if (status == -ENODEV || status == -ENOENT)
 		return true;
 
+	if (!test_bit(DEVICE_STATE_STARTED, &rt2x00dev->flags))
+		return false;
+
 	if (status == -EPROTO || status == -ETIMEDOUT)
 		rt2x00dev->num_proto_errs++;
 	else

From: Ioanna Alifieraki <[email protected]>

commit 1d49eb91e86e8c1c1614c72e3e958b6b7e2472a9 upstream.

Currently when removing an ipmi_user the removal is deferred as a work on the system's workqueue. Although this guarantees the free operation will occur in non atomic context, it can race with the ipmi_msghandler module removal (see [1]) . In case a remove_user work is scheduled for removal and shortly after ipmi_msghandler module is removed we can end up in a situation where the module is removed fist and when the work is executed the system crashes with :

BUG: unable to handle page fault for address: ffffffffc05c3450

PF: supervisor instruction fetch in kernel mode
PF: error_code(0x0010) - not-present page
because the pages of the module are gone. In cleanup_ipmi() there is no easy way to detect if there are any pending works to flush them before removing the module. This patch creates a separate workqueue and schedules the remove_work works on it. When removing the module the workqueue is drained when destroyed to avoid the race.

[1] https://bugs.launchpad.net/bugs/1950666

Message-Id: <[email protected]>

---
 drivers/char/ipmi/ipmi_msghandler.c |   13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

@@ -220,6 +220,8 @@ struct ipmi_user {
 	struct work_struct remove_work;
 };
 
+struct workqueue_struct *remove_work_wq;
+
 static struct ipmi_user *acquire_ipmi_user(struct ipmi_user *user, int *index)
 	__acquires(user->release_barrier)
 {

@@ -1286,7 +1288,7 @@ static void free_user(struct kref *ref)
 	struct ipmi_user *user = container_of(ref, struct ipmi_user, refcount);
 
 	/* SRCU cleanup must happen in task context. */
-	schedule_work(&user->remove_work);
+	queue_work(remove_work_wq, &user->remove_work);
 }
 
 static void _ipmi_destroy_user(struct ipmi_user *user)

@@ -5161,6 +5163,13 @@ static int ipmi_init_msghandler(void)
 
 	atomic_notifier_chain_register(&panic_notifier_list, &panic_block);
 
+	remove_work_wq = create_singlethread_workqueue("ipmi-msghandler-remove-wq");
+	if (!remove_work_wq) {
+		pr_err("unable to create ipmi-msghandler-remove-wq workqueue");
+		rv = -ENOMEM;
+		goto out;
+	}
+
 	initialized = true;
 
 out:

@@ -5186,6 +5195,8 @@ static void __exit cleanup_ipmi(void)
 	int count;
 
 	if (initialized) {
+		destroy_workqueue(remove_work_wq);
+
 		atomic_notifier_chain_unregister(&panic_notifier_list,
 						 &panic_block);
 

From: Xiongfeng Wang <[email protected]>

commit 2c1b5a84669d2477d8fffe9136e86a2cff591729 upstream.

When I hot added a CPU, I found 'cpufreq' directory was not created below /sys/devices/system/cpu/cpuX/.

It is because get_cpu_device() failed in add_cpu_dev_symlink().

cpufreq_add_dev() is the .add_dev callback of a CPU subsys interface. It will be called when the CPU device registered into the system. The call chain is as follows:

register_cpu()
->device_register()
->device_add()

    ->bus_probe_device()
     ->cpufreq_add_dev()

But only after the CPU device has been registered, we can get the CPU device by get_cpu_device(), otherwise it will return NULL.

Since we already have the CPU device in cpufreq_add_dev(), pass it to add_cpu_dev_symlink().

I noticed that the 'kobj' of the CPU device has been added into the system before cpufreq_add_dev().

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

@@ -995,10 +995,9 @@ static struct kobj_type ktype_cpufreq =
 	.release	= cpufreq_sysfs_release,
 };
 
-static void add_cpu_dev_symlink(struct cpufreq_policy *policy, unsigned int cpu)
+static void add_cpu_dev_symlink(struct cpufreq_policy *policy, unsigned int cpu,
+				struct device *dev)
 {
-	struct device *dev = get_cpu_device(cpu);
-
 	if (unlikely(!dev))
 		return;
 

@@ -1384,7 +1383,7 @@ static int cpufreq_online(unsigned int c
 	if (new_policy) {
 		for_each_cpu(j, policy->related_cpus) {
 			per_cpu(cpufreq_cpu_data, j) = policy;
-			add_cpu_dev_symlink(policy, j);
+			add_cpu_dev_symlink(policy, j, get_cpu_device(j));
 		}
 
 		policy->min_freq_req = kzalloc(2 * sizeof(*policy->min_freq_req),

@@ -1547,7 +1546,7 @@ static int cpufreq_add_dev(struct device
 	/* Create sysfs link on CPU registration */
 	policy = per_cpu(cpufreq_cpu_data, cpu);
 	if (policy)
-		add_cpu_dev_symlink(policy, cpu);
+		add_cpu_dev_symlink(policy, cpu, dev);
 
 	return 0;
 }

From: Niklas Schnelle <[email protected]>

commit 52d04d408185b7aa47628d2339c28ec70074e0ae upstream.

When running without MIO support, with pci=nomio or for devices which are not MIO-capable the zPCI subsystem generates pseudo-MMIO addresses to allow access to PCI BARs via MMIO based Linux APIs even though the platform uses function handles and BAR numbers.

This is done by stashing an index into our global IOMAP array which contains the function handle in the 16 most significant bits of the addresses returned by ioremap() always setting the most significant bit.

On the other hand the MIO addresses assigned by the platform for use, while requiring special instructions, allow PCI access with virtually mapped physical addresses. Now the problem is that these MIO addresses and our own pseudo-MMIO addresses may overlap, while functionally this would not be a problem by itself this overlap is detected by common code as both address types are added as resources in the iomem_resource tree. This leads to the overlapping resource claim of either the MIO capable or non-MIO capable devices with being rejected.

Since PCI is tightly coupled to the use of the iomem_resource tree, see for example the code for request_mem_region(), we can't reasonably get rid of the overlap being detected by keeping our pseudo-MMIO addresses out of the iomem_resource tree.

Instead let's move the range used by our own pseudo-MMIO addresses by starting at (1UL << 62) and only using addresses below (1UL << 63) thus avoiding the range currently used for MIO addresses.

---
 arch/s390/include/asm/pci_io.h |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

@@ -14,12 +14,13 @@
 
 /* I/O Map */
 #define ZPCI_IOMAP_SHIFT		48
-#define ZPCI_IOMAP_ADDR_BASE		0x8000000000000000UL
+#define ZPCI_IOMAP_ADDR_SHIFT		62
+#define ZPCI_IOMAP_ADDR_BASE		(1UL << ZPCI_IOMAP_ADDR_SHIFT)
 #define ZPCI_IOMAP_ADDR_OFF_MASK	((1UL << ZPCI_IOMAP_SHIFT) - 1)
 #define ZPCI_IOMAP_MAX_ENTRIES							\
-	((ULONG_MAX - ZPCI_IOMAP_ADDR_BASE + 1) / (1UL << ZPCI_IOMAP_SHIFT))
+	(1UL << (ZPCI_IOMAP_ADDR_SHIFT - ZPCI_IOMAP_SHIFT))
 #define ZPCI_IOMAP_ADDR_IDX_MASK						\
-	(~ZPCI_IOMAP_ADDR_OFF_MASK - ZPCI_IOMAP_ADDR_BASE)
+	((ZPCI_IOMAP_ADDR_BASE - 1) & ~ZPCI_IOMAP_ADDR_OFF_MASK)
 
 struct zpci_iomap_entry {
 	u32 fh;

From: Linus Torvalds <[email protected]>

commit 054aa8d439b9185d4f5eb9a90282d1ce74772969 upstream.

Jann Horn points out that there is another possible race wrt Unix domain socket garbage collection, somewhat reminiscent of the one fixed in commit cbcf01128d0a ("af_unix: fix garbage collect vs MSG_PEEK").

See the extended comment about the garbage collection requirements added to unix_peek_fds() by that commit for details.

The race comes from how we can locklessly look up a file descriptor just as it is in the process of being closed, and with the right artificial timing (Jann added a few strategic 'mdelay(500)' calls to do that), the Unix domain socket garbage collector could see the reference count decrement of the close() happen before fget() took its reference to the file and the file was attached onto a new file descriptor.

This is all (intentionally) correct on the 'struct file *' side, with RCU lookups and lockless reference counting very much part of the design. Getting that reference count out of order isn't a problem per se.

But the garbage collector can get confused by seeing this situation of having seen a file not having any remaining external references and then seeing it being attached to an fd.

In commit cbcf01128d0a ("af_unix: fix garbage collect vs MSG_PEEK") the fix was to serialize the file descriptor install with the garbage collector by taking and releasing the unix_gc_lock.

That's not really an option here, but since this all happens when we are in the process of looking up a file descriptor, we can instead simply just re-check that the file hasn't been closed in the meantime, and just re-do the lookup if we raced with a concurrent close() of the same file descriptor.

---
 fs/file.c |    4 ++++
 1 file changed, 4 insertions(+)

@@ -723,6 +723,10 @@ loop:
 			file = NULL;
 		else if (!get_file_rcu_many(file, refs))
 			goto loop;
+		else if (__fcheck_files(files, fd) != file) {
+			fput_many(file, refs);
+			goto loop;
+		}
 	}
 	rcu_read_unlock();
 

From: Baokun Li <[email protected]>

commit 6c8ad7e8cf29eb55836e7a0215f967746ab2b504 upstream.

When the rmmod sata_fsl.ko command is executed in the PPC64 GNU/Linux, a bug is reported:
==================================================================

 BUG: Unable to handle kernel data access on read at 0x80000800805b502c
 Oops: Kernel access of bad area, sig: 11 [#1]
 NIP [c0000000000388a4] .ioread32+0x4/0x20
 LR [80000000000c6034] .sata_fsl_port_stop+0x44/0xe0 [sata_fsl]
 Call Trace:
  .free_irq+0x1c/0x4e0 (unreliable)
  .ata_host_stop+0x74/0xd0 [libata]
  .release_nodes+0x330/0x3f0
  .device_release_driver_internal+0x178/0x2c0
  .driver_detach+0x64/0xd0
  .bus_remove_driver+0x70/0xf0
  .driver_unregister+0x38/0x80
  .platform_driver_unregister+0x14/0x30
  .fsl_sata_driver_exit+0x18/0xa20 [sata_fsl]
  .__se_sys_delete_module+0x1ec/0x2d0
  .system_call_exception+0xfc/0x1f0
  system_call_common+0xf8/0x200
 ==================================================================

The triggering of the BUG is shown in the following stack:

driver_detach
device_release_driver_internal

    __device_release_driver
      drv->remove(dev) --> platform_drv_remove/platform_remove
        drv->remove(dev) --> sata_fsl_remove
          iounmap(host_priv->hcr_base);			<---- unmap
          kfree(host_priv);                             <---- free
      devres_release_all
        release_nodes
          dr->node.release(dev, dr->data) --> ata_host_stop
            ap->ops->port_stop(ap) --> sata_fsl_port_stop
                ioread32(hcr_base + HCONTROL)           <---- UAF
            host->ops->host_stop(host)

The iounmap(host_priv->hcr_base) and kfree(host_priv) functions should not be executed in drv->remove. These functions should be executed in host_stop after port_stop. Therefore, we move these functions to the new function sata_fsl_host_stop and bind the new function to host_stop.

---
 drivers/ata/sata_fsl.c |   12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

@@ -1394,6 +1394,14 @@ static int sata_fsl_init_controller(stru
 	return 0;
 }
 
+static void sata_fsl_host_stop(struct ata_host *host)
+{
+        struct sata_fsl_host_priv *host_priv = host->private_data;
+
+        iounmap(host_priv->hcr_base);
+        kfree(host_priv);
+}
+
 /*
  * scsi mid-layer and libata interface structures
  */

@@ -1426,6 +1434,8 @@ static struct ata_port_operations sata_f
 	.port_start = sata_fsl_port_start,
 	.port_stop = sata_fsl_port_stop,
 
+	.host_stop      = sata_fsl_host_stop,
+
 	.pmp_attach = sata_fsl_pmp_attach,
 	.pmp_detach = sata_fsl_pmp_detach,
 };

@@ -1558,8 +1568,6 @@ static int sata_fsl_remove(struct platfo
 	ata_host_detach(host);
 
 	irq_dispose_mapping(host_priv->irq);
-	iounmap(host_priv->hcr_base);
-	kfree(host_priv);
 
 	return 0;
 }

From: Baokun Li <[email protected]>

commit 6f48394cf1f3e8486591ad98c11cdadb8f1ef2ad upstream.

Trying to remove the fsl-sata module in the PPC64 GNU/Linux leads to the following warning:

 ------------[ cut here ]------------
 remove_proc_entry: removing non-empty directory 'irq/69',
   leaking at least 'fsl-sata[ff0221000.sata]'
 WARNING: CPU: 3 PID: 1048 at fs/proc/generic.c:722
   .remove_proc_entry+0x20c/0x220
 IRQMASK: 0
 NIP [c00000000033826c] .remove_proc_entry+0x20c/0x220
 LR [c000000000338268] .remove_proc_entry+0x208/0x220
 Call Trace:
  .remove_proc_entry+0x208/0x220 (unreliable)
  .unregister_irq_proc+0x104/0x140
  .free_desc+0x44/0xb0
  .irq_free_descs+0x9c/0xf0
  .irq_dispose_mapping+0x64/0xa0
  .sata_fsl_remove+0x58/0xa0 [sata_fsl]
  .platform_drv_remove+0x40/0x90
  .device_release_driver_internal+0x160/0x2c0
  .driver_detach+0x64/0xd0
  .bus_remove_driver+0x70/0xf0
  .driver_unregister+0x38/0x80
  .platform_driver_unregister+0x14/0x30
  .fsl_sata_driver_exit+0x18/0xa20 [sata_fsl]
 ---[ end trace 0ea876d4076908f5 ]---

The driver creates the mapping by calling irq_of_parse_and_map(), so it also has to dispose the mapping. But the easy way out is to simply use platform_get_irq() instead of irq_of_parse_map(). Also we should adapt return value checking and propagate error values.

In this case the mapping is not managed by the device but by the of core, so the device has not to dispose the mapping.

---
 drivers/ata/sata_fsl.c |    8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

@@ -1490,9 +1490,9 @@ static int sata_fsl_probe(struct platfor
 	host_priv->ssr_base = ssr_base;
 	host_priv->csr_base = csr_base;
 
-	irq = irq_of_parse_and_map(ofdev->dev.of_node, 0);
-	if (!irq) {
-		dev_err(&ofdev->dev, "invalid irq from platform\n");
+	irq = platform_get_irq(ofdev, 0);
+	if (irq < 0) {
+		retval = irq;
 		goto error_exit_with_cleanup;
 	}
 	host_priv->irq = irq;

@@ -1567,8 +1567,6 @@ static int sata_fsl_remove(struct platfo
 
 	ata_host_detach(host);
 
-	irq_dispose_mapping(host_priv->irq);
-
 	return 0;
 }
 

From: Alain Volmat <[email protected]>

commit 0c21d02ca469574d2082379db52d1a27b99eed0c upstream.

While handling an error during transfer (ex: NACK), it could happen that the driver has already written data into TXDR before the transfer get stopped.
This commit add TXDR Flush after end of transfer in case of error to avoid sending a wrong data on any other slave upon next transfer.

---
 drivers/i2c/busses/i2c-stm32f7.c |   20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

@@ -1586,6 +1586,16 @@ static int stm32f7_i2c_xfer(struct i2c_a
 	time_left = wait_for_completion_timeout(&i2c_dev->complete,
 						i2c_dev->adap.timeout);
 	ret = f7_msg->result;
+	if (ret) {
+		/*
+		 * It is possible that some unsent data have already been
+		 * written into TXDR. To avoid sending old data in a
+		 * further transfer, flush TXDR in case of any error
+		 */
+		writel_relaxed(STM32F7_I2C_ISR_TXE,
+			       i2c_dev->base + STM32F7_I2C_ISR);
+		goto pm_free;
+	}
 
 	if (!time_left) {
 		dev_dbg(i2c_dev->dev, "Access to slave 0x%x timed out\n",

@@ -1634,8 +1644,16 @@ static int stm32f7_i2c_smbus_xfer(struct
 	timeout = wait_for_completion_timeout(&i2c_dev->complete,
 					      i2c_dev->adap.timeout);
 	ret = f7_msg->result;
-	if (ret)
+	if (ret) {
+		/*
+		 * It is possible that some unsent data have already been
+		 * written into TXDR. To avoid sending old data in a
+		 * further transfer, flush TXDR in case of any error
+		 */
+		writel_relaxed(STM32F7_I2C_ISR_TXE,
+			       i2c_dev->base + STM32F7_I2C_ISR);
 		goto pm_free;
+	}
 
 	if (!timeout) {
 		dev_dbg(dev, "Access to slave 0x%x timed out\n", f7_msg->addr);

From: Alain Volmat <[email protected]>

commit b933d1faf8fa30d16171bcff404e39c41b2a7c84 upstream.

When getting an access timeout, ensure that the bus is in a proper state prior to returning the error.

---
 drivers/i2c/busses/i2c-stm32f7.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -1602,6 +1602,7 @@ static int stm32f7_i2c_xfer(struct i2c_a
 			i2c_dev->msg->addr);
 		if (i2c_dev->use_dma)
 			dmaengine_terminate_all(dma->chan_using);
+		stm32f7_i2c_wait_free_bus(i2c_dev);
 		ret = -ETIMEDOUT;
 	}
 

@@ -1659,6 +1660,7 @@ static int stm32f7_i2c_smbus_xfer(struct
 		dev_dbg(dev, "Access to slave 0x%x timed out\n", f7_msg->addr);
 		if (i2c_dev->use_dma)
 			dmaengine_terminate_all(dma->chan_using);
+		stm32f7_i2c_wait_free_bus(i2c_dev);
 		ret = -ETIMEDOUT;
 		goto pm_free;
 	}

From: Alain Volmat <[email protected]>

commit 31b90a95ccbbb4b628578ac17e3b3cc8eeacfe31 upstream.

In case of receiving a NACK, the dma transfer should be stopped to avoid feeding data into the FIFO.
Also ensure to properly return the proper error code and avoid waiting for the end of the dma completion in case of error happening during the transmission.

---
 drivers/i2c/busses/i2c-stm32f7.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

@@ -1394,6 +1394,7 @@ static irqreturn_t stm32f7_i2c_isr_event
 {
 	struct stm32f7_i2c_dev *i2c_dev = data;
 	struct stm32f7_i2c_msg *f7_msg = &i2c_dev->f7_msg;
+	struct stm32_i2c_dma *dma = i2c_dev->dma;
 	void __iomem *base = i2c_dev->base;
 	u32 status, mask;
 	int ret = IRQ_HANDLED;

@@ -1418,6 +1419,10 @@ static irqreturn_t stm32f7_i2c_isr_event
 	if (status & STM32F7_I2C_ISR_NACKF) {
 		dev_dbg(i2c_dev->dev, "<%s>: Receive NACK\n", __func__);
 		writel_relaxed(STM32F7_I2C_ICR_NACKCF, base + STM32F7_I2C_ICR);
+		if (i2c_dev->use_dma) {
+			stm32f7_i2c_disable_dma_req(i2c_dev);
+			dmaengine_terminate_all(dma->chan_using);
+		}
 		f7_msg->result = -ENXIO;
 	}
 

@@ -1433,7 +1438,7 @@ static irqreturn_t stm32f7_i2c_isr_event
 		/* Clear STOP flag */
 		writel_relaxed(STM32F7_I2C_ICR_STOPCF, base + STM32F7_I2C_ICR);
 
-		if (i2c_dev->use_dma) {
+		if (i2c_dev->use_dma && !f7_msg->result) {
 			ret = IRQ_WAKE_THREAD;
 		} else {
 			i2c_dev->master_mode = false;

@@ -1446,7 +1451,7 @@ static irqreturn_t stm32f7_i2c_isr_event
 		if (f7_msg->stop) {
 			mask = STM32F7_I2C_CR2_STOP;
 			stm32f7_i2c_set_bits(base + STM32F7_I2C_CR2, mask);
-		} else if (i2c_dev->use_dma) {
+		} else if (i2c_dev->use_dma && !f7_msg->result) {
 			ret = IRQ_WAKE_THREAD;
 		} else if (f7_msg->smbus) {
 			stm32f7_i2c_smbus_rep_start(i2c_dev);

From: Aaro Koskinen <[email protected]>

commit b12764695c3fcade145890b67f82f8b139174cc7 upstream.

CBUS transfers have always been atomic, but after commit 63b96983a5dd ("i2c: core: introduce callbacks for atomic transfers") we started to see warnings during e.g. poweroff as the atomic callback is not explicitly set. Fix that.

Fixes the following WARNING seen during Nokia N810 power down:

[  786.570617] reboot: Power down
[  786.573913] ------------[ cut here ]------------
[  786.578826] WARNING: CPU: 0 PID: 672 at drivers/i2c/i2c-core.h:40 i2c_smbus_xfer+0x100/0x110
[  786.587799] No atomic I2C transfer handler for 'i2c-2'

---
 drivers/i2c/busses/i2c-cbus-gpio.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -195,8 +195,9 @@ static u32 cbus_i2c_func(struct i2c_adap
 }
 
 static const struct i2c_algorithm cbus_i2c_algo = {
-	.smbus_xfer	= cbus_i2c_smbus_xfer,
-	.functionality	= cbus_i2c_func,
+	.smbus_xfer		= cbus_i2c_smbus_xfer,
+	.smbus_xfer_atomic	= cbus_i2c_smbus_xfer,
+	.functionality		= cbus_i2c_func,
 };
 
 static int cbus_i2c_remove(struct platform_device *pdev)

From: Randy Dunlap <[email protected]>

commit b0f38e15979fa8851e88e8aa371367f264e7b6e9 upstream.

Fix section mismatch warnings in xtsonic. The first one appears to be bogus and after fixing the second one, the first one is gone.

WARNING: modpost: vmlinux.o(.text+0x529adc): Section mismatch in reference from the function sonic_get_stats() to the function .init.text:set_reset_devices() The function sonic_get_stats() references
the function __init set_reset_devices().
This is often because sonic_get_stats lacks a __init
annotation or the annotation of set_reset_devices is wrong.

WARNING: modpost: vmlinux.o(.text+0x529b3b): Section mismatch in reference from the function xtsonic_probe() to the function .init.text:sonic_probe1() The function xtsonic_probe() references
the function __init sonic_probe1().
This is often because xtsonic_probe lacks a __init
annotation or the annotation of sonic_probe1 is wrong.

---
 drivers/net/ethernet/natsemi/xtsonic.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -120,7 +120,7 @@ static const struct net_device_ops xtson
 	.ndo_set_mac_address	= eth_mac_addr,
 };
 
-static int __init sonic_probe1(struct net_device *dev)
+static int sonic_probe1(struct net_device *dev)
 {
 	unsigned int silicon_revision;
 	struct sonic_local *lp = netdev_priv(dev);

From: Zhou Qingyang <[email protected]>

commit e2dabc4f7e7b60299c20a36d6a7b24ed9bf8e572 upstream.

In qlcnic_83xx_add_rings(), the indirect function of ahw->hw_ops->alloc_mbx_args will be called to allocate memory for cmd.req.arg, and there is a dereference of it in qlcnic_83xx_add_rings(), which could lead to a NULL pointer dereference on failure of the indirect function like qlcnic_83xx_alloc_mbx_args().

Fix this bug by adding a check of alloc_mbx_args(), this patch imitates the logic of mbx_cmd()'s failure handling.

This bug was found by a static analyzer. The analysis employs differential checking to identify inconsistent security operations (e.g., checks or kfrees) between two code paths and confirms that the inconsistent operations are not recovered in the current function or the callers, so they constitute bugs.

Note that, as a bug found by static analysis, it can be a false positive or hard to trigger. Multiple researchers have cross-reviewed the bug.

Builds with CONFIG_QLCNIC=m show no new warnings, and our static analyzer no longer warns about this code.

---
 drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

@@ -1079,8 +1079,14 @@ static int qlcnic_83xx_add_rings(struct
 	sds_mbx_size = sizeof(struct qlcnic_sds_mbx);
 	context_id = recv_ctx->context_id;
 	num_sds = adapter->drv_sds_rings - QLCNIC_MAX_SDS_RINGS;
-	ahw->hw_ops->alloc_mbx_args(&cmd, adapter,
-				    QLCNIC_CMD_ADD_RCV_RINGS);
+	err = ahw->hw_ops->alloc_mbx_args(&cmd, adapter,
+					QLCNIC_CMD_ADD_RCV_RINGS);
+	if (err) {
+		dev_err(&adapter->pdev->dev,
+			"Failed to alloc mbx args %d\n", err);
+		return err;
+	}
+
 	cmd.req.arg[1] = 0 | (num_sds << 8) | (context_id << 16);
 
 	/* set up status rings, mbx 2-81 */

From: Benjamin Poirier <[email protected]>

commit 7d4741eacdefa5f0475431645b56baf00784df1f upstream.

There are various problems related to netlink notifications for mpls route changes in response to interfaces being deleted:
* delete interface of only nexthop
DELROUTE notification is missing RTA_OIF attribute
* delete interface of non-last nexthop
NEWROUTE notification is missing entirely
* delete interface of last nexthop
DELROUTE notification is missing nexthop

All of these problems stem from the fact that existing routes are modified in-place before sending a notification. Restructure mpls_ifdown() to avoid changing the route in the DELROUTE cases and to create a copy in the NEWROUTE case.

---
 net/mpls/af_mpls.c |   68 ++++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 52 insertions(+), 16 deletions(-)

@@ -1491,22 +1491,52 @@ static void mpls_dev_destroy_rcu(struct
 	kfree(mdev);
 }
 
-static void mpls_ifdown(struct net_device *dev, int event)
+static int mpls_ifdown(struct net_device *dev, int event)
 {
 	struct mpls_route __rcu **platform_label;
 	struct net *net = dev_net(dev);
-	u8 alive, deleted;
 	unsigned index;
 
 	platform_label = rtnl_dereference(net->mpls.platform_label);
 	for (index = 0; index < net->mpls.platform_labels; index++) {
 		struct mpls_route *rt = rtnl_dereference(platform_label[index]);
+		bool nh_del = false;
+		u8 alive = 0;
 
 		if (!rt)
 			continue;
 
-		alive = 0;
-		deleted = 0;
+		if (event == NETDEV_UNREGISTER) {
+			u8 deleted = 0;
+
+			for_nexthops(rt) {
+				struct net_device *nh_dev =
+					rtnl_dereference(nh->nh_dev);
+
+				if (!nh_dev || nh_dev == dev)
+					deleted++;
+				if (nh_dev == dev)
+					nh_del = true;
+			} endfor_nexthops(rt);
+
+			/* if there are no more nexthops, delete the route */
+			if (deleted == rt->rt_nhn) {
+				mpls_route_update(net, index, NULL, NULL);
+				continue;
+			}
+
+			if (nh_del) {
+				size_t size = sizeof(*rt) + rt->rt_nhn *
+					rt->rt_nh_size;
+				struct mpls_route *orig = rt;
+
+				rt = kmalloc(size, GFP_KERNEL);
+				if (!rt)
+					return -ENOMEM;
+				memcpy(rt, orig, size);
+			}
+		}
+
 		change_nexthops(rt) {
 			unsigned int nh_flags = nh->nh_flags;
 

@@ -1530,16 +1560,15 @@ static void mpls_ifdown(struct net_devic
 next:
 			if (!(nh_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN)))
 				alive++;
-			if (!rtnl_dereference(nh->nh_dev))
-				deleted++;
 		} endfor_nexthops(rt);
 
 		WRITE_ONCE(rt->rt_nhn_alive, alive);
 
-		/* if there are no more nexthops, delete the route */
-		if (event == NETDEV_UNREGISTER && deleted == rt->rt_nhn)
-			mpls_route_update(net, index, NULL, NULL);
+		if (nh_del)
+			mpls_route_update(net, index, rt, NULL);
 	}
+
+	return 0;
 }
 
 static void mpls_ifup(struct net_device *dev, unsigned int flags)

@@ -1607,8 +1636,12 @@ static int mpls_dev_notify(struct notifi
 		return NOTIFY_OK;
 
 	switch (event) {
+		int err;
+
 	case NETDEV_DOWN:
-		mpls_ifdown(dev, event);
+		err = mpls_ifdown(dev, event);
+		if (err)
+			return notifier_from_errno(err);
 		break;
 	case NETDEV_UP:
 		flags = dev_get_flags(dev);

@@ -1619,13 +1652,18 @@ static int mpls_dev_notify(struct notifi
 		break;
 	case NETDEV_CHANGE:
 		flags = dev_get_flags(dev);
-		if (flags & (IFF_RUNNING | IFF_LOWER_UP))
+		if (flags & (IFF_RUNNING | IFF_LOWER_UP)) {
 			mpls_ifup(dev, RTNH_F_DEAD | RTNH_F_LINKDOWN);
-		else
-			mpls_ifdown(dev, event);
+		} else {
+			err = mpls_ifdown(dev, event);
+			if (err)
+				return notifier_from_errno(err);
+		}
 		break;
 	case NETDEV_UNREGISTER:
-		mpls_ifdown(dev, event);
+		err = mpls_ifdown(dev, event);
+		if (err)
+			return notifier_from_errno(err);
 		mdev = mpls_dev_get(dev);
 		if (mdev) {
 			mpls_dev_sysctl_unregister(dev, mdev);

@@ -1636,8 +1674,6 @@ static int mpls_dev_notify(struct notifi
 	case NETDEV_CHANGENAME:
 		mdev = mpls_dev_get(dev);
 		if (mdev) {
-			int err;
-
 			mpls_dev_sysctl_unregister(dev, mdev);
 			err = mpls_dev_sysctl_register(dev, mdev);
 			if (err)

From: Arnd Bergmann <[email protected]>

commit f7e5b9bfa6c8820407b64eabc1f29c9a87e8993d upstream.

On ARM v6 and later, we define CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS because the ordinary load/store instructions (ldr, ldrh, ldrb) can tolerate any misalignment of the memory address. However, load/store double and load/store multiple instructions (ldrd, ldm) may still only be used on memory addresses that are 32-bit aligned, and so we have to use the CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS macro with care, or we may end up with a severe performance hit due to alignment traps that require fixups by the kernel. Testing shows that this currently happens with clang-13 but not gcc-11. In theory, any compiler version can produce this bug or other problems, as we are dealing with undefined behavior in C99 even on architectures that support this in hardware, see also https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100363.

Fortunately, the get_unaligned() accessors do the right thing: when building for ARMv6 or later, the compiler will emit unaligned accesses using the ordinary load/store instructions (but avoid the ones that require 32-bit alignment). When building for older ARM, those accessors will emit the appropriate sequence of ldrb/mov/orr instructions. And on architectures that can truly tolerate any kind of misalignment, the get_unaligned() accessors resolve to the leXX_to_cpup accessors that operate on aligned addresses.

Since the compiler will in fact emit ldrd or ldm instructions when building this code for ARM v6 or later, the solution is to use the unaligned accessors unconditionally on architectures where this is known to be fast. The _aligned version of the hash function is however still needed to get the best performance on architectures that cannot do any unaligned access in hardware.

This new version avoids the undefined behavior and should produce the fastest hash on all architectures we support.

---
 include/linux/siphash.h |   14 ++++----------
 lib/siphash.c           |   12 ++++++------
 2 files changed, 10 insertions(+), 16 deletions(-)

@@ -27,9 +27,7 @@ static inline bool siphash_key_is_zero(c
 }
 
 u64 __siphash_aligned(const void *data, size_t len, const siphash_key_t *key);
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u64 __siphash_unaligned(const void *data, size_t len, const siphash_key_t *key);
-#endif
 
 u64 siphash_1u64(const u64 a, const siphash_key_t *key);
 u64 siphash_2u64(const u64 a, const u64 b, const siphash_key_t *key);

@@ -82,10 +80,9 @@ static inline u64 ___siphash_aligned(con
 static inline u64 siphash(const void *data, size_t len,
 			  const siphash_key_t *key)
 {
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
-	if (!IS_ALIGNED((unsigned long)data, SIPHASH_ALIGNMENT))
+	if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) ||
+	    !IS_ALIGNED((unsigned long)data, SIPHASH_ALIGNMENT))
 		return __siphash_unaligned(data, len, key);
-#endif
 	return ___siphash_aligned(data, len, key);
 }
 

@@ -96,10 +93,8 @@ typedef struct {
 
 u32 __hsiphash_aligned(const void *data, size_t len,
 		       const hsiphash_key_t *key);
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_unaligned(const void *data, size_t len,
 			 const hsiphash_key_t *key);
-#endif
 
 u32 hsiphash_1u32(const u32 a, const hsiphash_key_t *key);
 u32 hsiphash_2u32(const u32 a, const u32 b, const hsiphash_key_t *key);

@@ -135,10 +130,9 @@ static inline u32 ___hsiphash_aligned(co
 static inline u32 hsiphash(const void *data, size_t len,
 			   const hsiphash_key_t *key)
 {
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
-	if (!IS_ALIGNED((unsigned long)data, HSIPHASH_ALIGNMENT))
+	if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) ||
+	    !IS_ALIGNED((unsigned long)data, HSIPHASH_ALIGNMENT))
 		return __hsiphash_unaligned(data, len, key);
-#endif
 	return ___hsiphash_aligned(data, len, key);
 }
 

@@ -49,6 +49,7 @@
 	SIPROUND; \
 	return (v0 ^ v1) ^ (v2 ^ v3);
 
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u64 __siphash_aligned(const void *data, size_t len, const siphash_key_t *key)
 {
 	const u8 *end = data + len - (len % sizeof(u64));

@@ -80,8 +81,8 @@ u64 __siphash_aligned(const void *data,
 	POSTAMBLE
 }
 EXPORT_SYMBOL(__siphash_aligned);
+#endif
 
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u64 __siphash_unaligned(const void *data, size_t len, const siphash_key_t *key)
 {
 	const u8 *end = data + len - (len % sizeof(u64));

@@ -113,7 +114,6 @@ u64 __siphash_unaligned(const void *data
 	POSTAMBLE
 }
 EXPORT_SYMBOL(__siphash_unaligned);
-#endif
 
 /**
  * siphash_1u64 - compute 64-bit siphash PRF value of a u64

@@ -250,6 +250,7 @@ EXPORT_SYMBOL(siphash_3u32);
 	HSIPROUND; \
 	return (v0 ^ v1) ^ (v2 ^ v3);
 
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_aligned(const void *data, size_t len, const hsiphash_key_t *key)
 {
 	const u8 *end = data + len - (len % sizeof(u64));

@@ -280,8 +281,8 @@ u32 __hsiphash_aligned(const void *data,
 	HPOSTAMBLE
 }
 EXPORT_SYMBOL(__hsiphash_aligned);
+#endif
 
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_unaligned(const void *data, size_t len,
 			 const hsiphash_key_t *key)
 {

@@ -313,7 +314,6 @@ u32 __hsiphash_unaligned(const void *dat
 	HPOSTAMBLE
 }
 EXPORT_SYMBOL(__hsiphash_unaligned);
-#endif
 
 /**
  * hsiphash_1u32 - compute 64-bit hsiphash PRF value of a u32

@@ -418,6 +418,7 @@ EXPORT_SYMBOL(hsiphash_4u32);
 	HSIPROUND; \
 	return v1 ^ v3;
 
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_aligned(const void *data, size_t len, const hsiphash_key_t *key)
 {
 	const u8 *end = data + len - (len % sizeof(u32));

@@ -438,8 +439,8 @@ u32 __hsiphash_aligned(const void *data,
 	HPOSTAMBLE
 }
 EXPORT_SYMBOL(__hsiphash_aligned);
+#endif
 
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_unaligned(const void *data, size_t len,
 			 const hsiphash_key_t *key)
 {

@@ -461,7 +462,6 @@ u32 __hsiphash_unaligned(const void *dat
 	HPOSTAMBLE
 }
 EXPORT_SYMBOL(__hsiphash_unaligned);
-#endif
 
 /**
  * hsiphash_1u32 - compute 32-bit hsiphash PRF value of a u32

From: Zhou Qingyang <[email protected]>

commit addad7643142f500080417dd7272f49b7a185570 upstream.

In mlx4_en_try_alloc_resources(), mlx4_en_copy_priv() is called and tmp->tx_cq will be freed on the error path of mlx4_en_copy_priv(). After that mlx4_en_alloc_resources() is called and there is a dereference of &tmp->tx_cq[t][i] in mlx4_en_alloc_resources(), which could lead to a use after free problem on failure of mlx4_en_copy_priv().

Fix this bug by adding a check of mlx4_en_copy_priv()

This bug was found by a static analyzer. The analysis employs differential checking to identify inconsistent security operations (e.g., checks or kfrees) between two code paths and confirms that the inconsistent operations are not recovered in the current function or the callers, so they constitute bugs.

Note that, as a bug found by static analysis, it can be a false positive or hard to trigger. Multiple researchers have cross-reviewed the bug.

Builds with CONFIG_MLX4_EN=m show no new warnings, and our static analyzer no longer warns about this code.

---
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

@@ -2281,9 +2281,14 @@ int mlx4_en_try_alloc_resources(struct m
 				bool carry_xdp_prog)
 {
 	struct bpf_prog *xdp_prog;
-	int i, t;
+	int i, t, ret;
 
-	mlx4_en_copy_priv(tmp, priv, prof);
+	ret = mlx4_en_copy_priv(tmp, priv, prof);
+	if (ret) {
+		en_warn(priv, "%s: mlx4_en_copy_priv() failed, return\n",
+			__func__);
+		return ret;
+	}
 
 	if (mlx4_en_alloc_resources(tmp)) {
 		en_warn(priv,

From: Li Zhijian <[email protected]>

commit a05431b22be819d75db72ca3d44381d18a37b092 upstream.

ipv6_addr_bind/ipv4_addr_bind are function names. Previously, bind test would not be run by default due to the wrong case names

---
 tools/testing/selftests/net/fcnal-test.sh |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -3450,8 +3450,8 @@ EOF
 ################################################################################
 # main
 
-TESTS_IPV4="ipv4_ping ipv4_tcp ipv4_udp ipv4_addr_bind ipv4_runtime ipv4_netfilter"
-TESTS_IPV6="ipv6_ping ipv6_tcp ipv6_udp ipv6_addr_bind ipv6_runtime ipv6_netfilter"
+TESTS_IPV4="ipv4_ping ipv4_tcp ipv4_udp ipv4_bind ipv4_runtime ipv4_netfilter"
+TESTS_IPV6="ipv6_ping ipv6_tcp ipv6_udp ipv6_bind ipv6_runtime ipv6_netfilter"
 TESTS_OTHER="use_cases"
 
 PAUSE_ON_FAIL=no

From: Eiichi Tsukata <[email protected]>

commit beacff50edbd6c9659a6f15fc7f6126909fade29 upstream.

Need to call rxrpc_put_local() for peer candidate before kfree() as it holds a ref to rxrpc_local.

[DH: v2: Changed to abstract the peer freeing code out into a function]

---
 net/rxrpc/peer_object.c |   14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

@@ -297,6 +297,12 @@ static struct rxrpc_peer *rxrpc_create_p
 	return peer;
 }
 
+static void rxrpc_free_peer(struct rxrpc_peer *peer)
+{
+	rxrpc_put_local(peer->local);
+	kfree_rcu(peer, rcu);
+}
+
 /*
  * Set up a new incoming peer.  There shouldn't be any other matching peers
  * since we've already done a search in the list from the non-reentrant context

@@ -363,7 +369,7 @@ struct rxrpc_peer *rxrpc_lookup_peer(str
 		spin_unlock_bh(&rxnet->peer_hash_lock);
 
 		if (peer)
-			kfree(candidate);
+			rxrpc_free_peer(candidate);
 		else
 			peer = candidate;
 	}

@@ -418,8 +424,7 @@ static void __rxrpc_put_peer(struct rxrp
 	list_del_init(&peer->keepalive_link);
 	spin_unlock_bh(&rxnet->peer_hash_lock);
 
-	rxrpc_put_local(peer->local);
-	kfree_rcu(peer, rcu);
+	rxrpc_free_peer(peer);
 }
 
 /*

@@ -455,8 +460,7 @@ void rxrpc_put_peer_locked(struct rxrpc_
 	if (n == 0) {
 		hash_del_rcu(&peer->hash_link);
 		list_del_init(&peer->keepalive_link);
-		rxrpc_put_local(peer->local);
-		kfree_rcu(peer, rcu);
+		rxrpc_free_peer(peer);
 	}
 }
 

From: Sven Schuchmann <[email protected]>

commit 817b653160db9852d5a0498a31f047e18ce27e5b upstream.

On most systems request for IRQ 0 will fail, phylib will print an error message and fall back to polling. To fix this set the phydev->irq to PHY_POLL if no IRQ is available.

---
 drivers/net/usb/lan78xx.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -2136,7 +2136,7 @@ static int lan78xx_phy_init(struct lan78
 	if (dev->domain_data.phyirq > 0)
 		phydev->irq = dev->domain_data.phyirq;
 	else
-		phydev->irq = 0;
+		phydev->irq = PHY_POLL;
 	netdev_dbg(dev->net, "phydev->irq = %d\n", phydev->irq);
 
 	/* set to AUTOMDIX */

From: Christophe JAILLET <[email protected]>

commit b83f5ac7d922e69a109261f5f940eebbd4e514c4 upstream.

'bitmap_fill()' fills a bitmap one 'long' at a time. It is likely that an exact number of bits is expected.

Use 'bitmap_set()' instead in order not to set unexpected bits.

---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -5773,7 +5773,7 @@ static int mvpp2_probe(struct platform_d
 
 	shared = num_present_cpus() - priv->nthreads;
 	if (shared > 0)
-		bitmap_fill(&priv->lock_map,
+		bitmap_set(&priv->lock_map, 0,
 			    min_t(int, shared, MVPP2_MAX_THREADS));
 
 	for (i = 0; i < MVPP2_MAX_THREADS; i++) {

From: Eric Dumazet <[email protected]>

commit 7a10d8c810cfad3e79372d7d1c77899d86cd6662 upstream.

syzbot found that __dev_queue_xmit() is reading txq->xmit_lock_owner without annotations.

No serious issue there, let's document what is happening there.

BUG: KCSAN: data-race in __dev_queue_xmit / __dev_queue_xmit

write to 0xffff888139d09484 of 4 bytes by interrupt on cpu 0:

 __netif_tx_unlock include/linux/netdevice.h:4437 [inline]
 __dev_queue_xmit+0x948/0xf70 net/core/dev.c:4229
 dev_queue_xmit_accel+0x19/0x20 net/core/dev.c:4265
 macvlan_queue_xmit drivers/net/macvlan.c:543 [inline]
 macvlan_start_xmit+0x2b3/0x3d0 drivers/net/macvlan.c:567
 __netdev_start_xmit include/linux/netdevice.h:4987 [inline]
 netdev_start_xmit include/linux/netdevice.h:5001 [inline]
 xmit_one+0x105/0x2f0 net/core/dev.c:3590
 dev_hard_start_xmit+0x72/0x120 net/core/dev.c:3606
 sch_direct_xmit+0x1b2/0x7c0 net/sched/sch_generic.c:342
 __dev_xmit_skb+0x83d/0x1370 net/core/dev.c:3817
 __dev_queue_xmit+0x590/0xf70 net/core/dev.c:4194
 dev_queue_xmit+0x13/0x20 net/core/dev.c:4259
 neigh_hh_output include/net/neighbour.h:511 [inline]
 neigh_output include/net/neighbour.h:525 [inline]
 ip6_finish_output2+0x995/0xbb0 net/ipv6/ip6_output.c:126
 __ip6_finish_output net/ipv6/ip6_output.c:191 [inline]
 ip6_finish_output+0x444/0x4c0 net/ipv6/ip6_output.c:201
 NF_HOOK_COND include/linux/netfilter.h:296 [inline]
 ip6_output+0x10e/0x210 net/ipv6/ip6_output.c:224
 dst_output include/net/dst.h:450 [inline]
 NF_HOOK include/linux/netfilter.h:307 [inline]
 ndisc_send_skb+0x486/0x610 net/ipv6/ndisc.c:508
 ndisc_send_rs+0x3b0/0x3e0 net/ipv6/ndisc.c:702
 addrconf_rs_timer+0x370/0x540 net/ipv6/addrconf.c:3898
 call_timer_fn+0x2e/0x240 kernel/time/timer.c:1421
 expire_timers+0x116/0x240 kernel/time/timer.c:1466
 __run_timers+0x368/0x410 kernel/time/timer.c:1734
 run_timer_softirq+0x2e/0x60 kernel/time/timer.c:1747
 __do_softirq+0x158/0x2de kernel/softirq.c:558
 __irq_exit_rcu kernel/softirq.c:636 [inline]
 irq_exit_rcu+0x37/0x70 kernel/softirq.c:648
 sysvec_apic_timer_interrupt+0x3e/0xb0 arch/x86/kernel/apic/apic.c:1097
 asm_sysvec_apic_timer_interrupt+0x12/0x20

read to 0xffff888139d09484 of 4 bytes by interrupt on cpu 1:

 __dev_queue_xmit+0x5e3/0xf70 net/core/dev.c:4213
 dev_queue_xmit_accel+0x19/0x20 net/core/dev.c:4265
 macvlan_queue_xmit drivers/net/macvlan.c:543 [inline]
 macvlan_start_xmit+0x2b3/0x3d0 drivers/net/macvlan.c:567
 __netdev_start_xmit include/linux/netdevice.h:4987 [inline]
 netdev_start_xmit include/linux/netdevice.h:5001 [inline]
 xmit_one+0x105/0x2f0 net/core/dev.c:3590
 dev_hard_start_xmit+0x72/0x120 net/core/dev.c:3606
 sch_direct_xmit+0x1b2/0x7c0 net/sched/sch_generic.c:342
 __dev_xmit_skb+0x83d/0x1370 net/core/dev.c:3817
 __dev_queue_xmit+0x590/0xf70 net/core/dev.c:4194
 dev_queue_xmit+0x13/0x20 net/core/dev.c:4259
 neigh_resolve_output+0x3db/0x410 net/core/neighbour.c:1523
 neigh_output include/net/neighbour.h:527 [inline]
 ip6_finish_output2+0x9be/0xbb0 net/ipv6/ip6_output.c:126
 __ip6_finish_output net/ipv6/ip6_output.c:191 [inline]
 ip6_finish_output+0x444/0x4c0 net/ipv6/ip6_output.c:201
 NF_HOOK_COND include/linux/netfilter.h:296 [inline]
 ip6_output+0x10e/0x210 net/ipv6/ip6_output.c:224
 dst_output include/net/dst.h:450 [inline]
 NF_HOOK include/linux/netfilter.h:307 [inline]
 ndisc_send_skb+0x486/0x610 net/ipv6/ndisc.c:508
 ndisc_send_rs+0x3b0/0x3e0 net/ipv6/ndisc.c:702
 addrconf_rs_timer+0x370/0x540 net/ipv6/addrconf.c:3898
 call_timer_fn+0x2e/0x240 kernel/time/timer.c:1421
 expire_timers+0x116/0x240 kernel/time/timer.c:1466
 __run_timers+0x368/0x410 kernel/time/timer.c:1734
 run_timer_softirq+0x2e/0x60 kernel/time/timer.c:1747
 __do_softirq+0x158/0x2de kernel/softirq.c:558
 __irq_exit_rcu kernel/softirq.c:636 [inline]
 irq_exit_rcu+0x37/0x70 kernel/softirq.c:648
 sysvec_apic_timer_interrupt+0x8d/0xb0 arch/x86/kernel/apic/apic.c:1097
 asm_sysvec_apic_timer_interrupt+0x12/0x20
 kcsan_setup_watchpoint+0x94/0x420 kernel/kcsan/core.c:443
 folio_test_anon include/linux/page-flags.h:581 [inline]
 PageAnon include/linux/page-flags.h:586 [inline]
 zap_pte_range+0x5ac/0x10e0 mm/memory.c:1347
 zap_pmd_range mm/memory.c:1467 [inline]
 zap_pud_range mm/memory.c:1496 [inline]
 zap_p4d_range mm/memory.c:1517 [inline]
 unmap_page_range+0x2dc/0x3d0 mm/memory.c:1538
 unmap_single_vma+0x157/0x210 mm/memory.c:1583
 unmap_vmas+0xd0/0x180 mm/memory.c:1615
 exit_mmap+0x23d/0x470 mm/mmap.c:3170
 __mmput+0x27/0x1b0 kernel/fork.c:1113
 mmput+0x3d/0x50 kernel/fork.c:1134
 exit_mm+0xdb/0x170 kernel/exit.c:507
 do_exit+0x608/0x17a0 kernel/exit.c:819
 do_group_exit+0xce/0x180 kernel/exit.c:929
 get_signal+0xfc3/0x1550 kernel/signal.c:2852
 arch_do_signal_or_restart+0x8c/0x2e0 arch/x86/kernel/signal.c:868
 handle_signal_work kernel/entry/common.c:148 [inline]
 exit_to_user_mode_loop kernel/entry/common.c:172 [inline]
 exit_to_user_mode_prepare+0x113/0x190 kernel/entry/common.c:207
 __syscall_exit_to_user_mode_work kernel/entry/common.c:289 [inline]
 syscall_exit_to_user_mode+0x20/0x40 kernel/entry/common.c:300
 do_syscall_64+0x50/0xd0 arch/x86/entry/common.c:86
 entry_SYSCALL_64_after_hwframe+0x44/0xae

value changed: 0x00000000 -> 0xffffffff

Reported by Kernel Concurrency Sanitizer on:

CPU: 1 PID: 28712 Comm: syz-executor.0 Tainted: G        W         5.16.0-rc1-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011

---
 include/linux/netdevice.h |   19 +++++++++++++------
 net/core/dev.c            |    5 ++++-
 2 files changed, 17 insertions(+), 7 deletions(-)

@@ -3943,7 +3943,8 @@ static inline u32 netif_msg_init(int deb
 static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
 {
 	spin_lock(&txq->_xmit_lock);
-	txq->xmit_lock_owner = cpu;
+	/* Pairs with READ_ONCE() in __dev_queue_xmit() */
+	WRITE_ONCE(txq->xmit_lock_owner, cpu);
 }
 
 static inline bool __netif_tx_acquire(struct netdev_queue *txq)

@@ -3960,26 +3961,32 @@ static inline void __netif_tx_release(st
 static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
 {
 	spin_lock_bh(&txq->_xmit_lock);
-	txq->xmit_lock_owner = smp_processor_id();
+	/* Pairs with READ_ONCE() in __dev_queue_xmit() */
+	WRITE_ONCE(txq->xmit_lock_owner, smp_processor_id());
 }
 
 static inline bool __netif_tx_trylock(struct netdev_queue *txq)
 {
 	bool ok = spin_trylock(&txq->_xmit_lock);
-	if (likely(ok))
-		txq->xmit_lock_owner = smp_processor_id();
+
+	if (likely(ok)) {
+		/* Pairs with READ_ONCE() in __dev_queue_xmit() */
+		WRITE_ONCE(txq->xmit_lock_owner, smp_processor_id());
+	}
 	return ok;
 }
 
 static inline void __netif_tx_unlock(struct netdev_queue *txq)
 {
-	txq->xmit_lock_owner = -1;
+	/* Pairs with READ_ONCE() in __dev_queue_xmit() */
+	WRITE_ONCE(txq->xmit_lock_owner, -1);
 	spin_unlock(&txq->_xmit_lock);
 }
 
 static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
 {
-	txq->xmit_lock_owner = -1;
+	/* Pairs with READ_ONCE() in __dev_queue_xmit() */
+	WRITE_ONCE(txq->xmit_lock_owner, -1);
 	spin_unlock_bh(&txq->_xmit_lock);
 }
 

@@ -3766,7 +3766,10 @@ static int __dev_queue_xmit(struct sk_bu
 	if (dev->flags & IFF_UP) {
 		int cpu = smp_processor_id(); /* ok because BHs are off */
 
-		if (txq->xmit_lock_owner != cpu) {
+		/* Other cpus might concurrently change txq->xmit_lock_owner
+		 * to -1 or to their cpu id, but not to our id.
+		 */
+		if (READ_ONCE(txq->xmit_lock_owner) != cpu) {
 			if (dev_xmit_recursion())
 				goto recursion_alert;
 

From: Eric Dumazet <[email protected]>

commit 213f5f8f31f10aa1e83187ae20fb7fa4e626b724 upstream.

Before commit faa041a40b9f ("ipv4: Create cleanup helper for fib_nh") changes to net->ipv4.fib_num_tclassid_users were protected by RTNL.

After the change, this is no longer the case, as free_fib_info_rcu() runs after rcu grace period, without rtnl being held.

---
 include/net/ip_fib.h     |    2 +-
 include/net/netns/ipv4.h |    2 +-
 net/ipv4/fib_frontend.c  |    2 +-
 net/ipv4/fib_rules.c     |    4 ++--
 net/ipv4/fib_semantics.c |    4 ++--
 5 files changed, 7 insertions(+), 7 deletions(-)

@@ -412,7 +412,7 @@ int fib_validate_source(struct sk_buff *
 #ifdef CONFIG_IP_ROUTE_CLASSID
 static inline int fib_num_tclassid_users(struct net *net)
 {
-	return net->ipv4.fib_num_tclassid_users;
+	return atomic_read(&net->ipv4.fib_num_tclassid_users);
 }
 #else
 static inline int fib_num_tclassid_users(struct net *net)

@@ -61,7 +61,7 @@ struct netns_ipv4 {
 #endif
 	bool			fib_has_custom_local_routes;
 #ifdef CONFIG_IP_ROUTE_CLASSID
-	int			fib_num_tclassid_users;
+	atomic_t		fib_num_tclassid_users;
 #endif
 	struct hlist_head	*fib_table_hash;
 	bool			fib_offload_disabled;

@@ -1588,7 +1588,7 @@ static int __net_init fib_net_init(struc
 	int error;
 
 #ifdef CONFIG_IP_ROUTE_CLASSID
-	net->ipv4.fib_num_tclassid_users = 0;
+	atomic_set(&net->ipv4.fib_num_tclassid_users, 0);
 #endif
 	error = ip_fib_net_init(net);
 	if (error < 0)

@@ -258,7 +258,7 @@ static int fib4_rule_configure(struct fi
 	if (tb[FRA_FLOW]) {
 		rule4->tclassid = nla_get_u32(tb[FRA_FLOW]);
 		if (rule4->tclassid)
-			net->ipv4.fib_num_tclassid_users++;
+			atomic_inc(&net->ipv4.fib_num_tclassid_users);
 	}
 #endif
 

@@ -290,7 +290,7 @@ static int fib4_rule_delete(struct fib_r
 
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	if (((struct fib4_rule *)rule)->tclassid)
-		net->ipv4.fib_num_tclassid_users--;
+		atomic_dec(&net->ipv4.fib_num_tclassid_users);
 #endif
 	net->ipv4.fib_has_custom_rules = true;
 

@@ -222,7 +222,7 @@ void fib_nh_release(struct net *net, str
 {
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	if (fib_nh->nh_tclassid)
-		net->ipv4.fib_num_tclassid_users--;
+		atomic_dec(&net->ipv4.fib_num_tclassid_users);
 #endif
 	fib_nh_common_release(&fib_nh->nh_common);
 }

@@ -624,7 +624,7 @@ int fib_nh_init(struct net *net, struct
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	nh->nh_tclassid = cfg->fc_flow;
 	if (nh->nh_tclassid)
-		net->ipv4.fib_num_tclassid_users++;
+		atomic_inc(&net->ipv4.fib_num_tclassid_users);
 #endif
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 	nh->fib_nh_weight = nh_weight;

From: William Kucharski <[email protected]>

commit 19f36edf14bcdb783aef3af8217df96f76a8ce34 upstream.

Correct an error where setting /proc/sys/net/rds/tcp/rds_tcp_rcvbuf would instead modify the socket's sk_sndbuf and would leave sk_rcvbuf untouched.

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

@@ -510,7 +510,7 @@ void rds_tcp_tune(struct socket *sock)
 		sk->sk_userlocks |= SOCK_SNDBUF_LOCK;
 	}
 	if (rtn->rcvbuf_size > 0) {
-		sk->sk_sndbuf = rtn->rcvbuf_size;
+		sk->sk_rcvbuf = rtn->rcvbuf_size;
 		sk->sk_userlocks |= SOCK_RCVBUF_LOCK;
 	}
 	release_sock(sk);

From: Tony Lu <[email protected]>

commit 00e158fb91dfaff3f94746f260d11f1a4853506e upstream.

When smc_close_final() returns error, the return code overwrites by kernel_sock_shutdown() in smc_close_active(). The return code of smc_close_final() is more important than kernel_sock_shutdown(), and it will pass to userspace directly.

Fix it by keeping both return codes, if smc_close_final() raises an error, return it or kernel_sock_shutdown()'s.

---
 net/smc/smc_close.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

@@ -183,6 +183,7 @@ int smc_close_active(struct smc_sock *sm
 	int old_state;
 	long timeout;
 	int rc = 0;
+	int rc1 = 0;
 
 	timeout = current->flags & PF_EXITING ?
 		  0 : sock_flag(sk, SOCK_LINGER) ?

@@ -222,8 +223,11 @@ again:
 			/* actively shutdown clcsock before peer close it,
 			 * prevent peer from entering TIME_WAIT state.
 			 */
-			if (smc->clcsock && smc->clcsock->sk)
-				rc = kernel_sock_shutdown(smc->clcsock, SHUT_RDWR);
+			if (smc->clcsock && smc->clcsock->sk) {
+				rc1 = kernel_sock_shutdown(smc->clcsock,
+							   SHUT_RDWR);
+				rc = rc ? rc : rc1;
+			}
 		} else {
 			/* peer event has changed the state */
 			goto again;

From: Rob Clark <[email protected]>

commit e4840d537c2c6b1189d4de16ee0f4820e069dcea upstream.

In particular, we need to ensure all the necessary blocks are switched to 64b mode (a5xx+) otherwise the high bits of the address of the BO to snapshot state into will be ignored, resulting in:

*** gpu fault: ttbr0=0000000000000000 iova=0000000000012000 dir=READ type=TRANSLATION source=CP (0,0,0,0) platform 506a000.gmu: [drm:a6xx_gmu_set_oob] *ERROR* Timeout waiting for GMU OOB set BOOT_SLUMBER: 0x0

---
 drivers/gpu/drm/msm/msm_debugfs.c |    1 +
 1 file changed, 1 insertion(+)

@@ -77,6 +77,7 @@ static int msm_gpu_open(struct inode *in
 		goto free_priv;
 
 	pm_runtime_get_sync(&gpu->pdev->dev);
+	msm_gpu_hw_init(gpu);
 	show_priv->state = gpu->funcs->gpu_state_get(gpu);
 	pm_runtime_put_sync(&gpu->pdev->dev);
 

From: msizanoen1 <[email protected]>

commit cdef485217d30382f3bf6448c54b4401648fe3f1 upstream.

The kernel leaks memory when a fib rule is present in IPv6 nftables firewall rules and a suppress_prefix rule is present in the IPv6 routing rules (used by certain tools such as wg-quick). In such scenarios, every incoming packet will leak an allocation in ip6_dst_cache slab cache.

After some hours of bpftrace-ing and source code reading, I tracked down the issue to ca7a03c41753 ("ipv6: do not free rt if FIB_LOOKUP_NOREF is set on suppress rule").

The problem with that change is that the generic args->flags always have FIB_LOOKUP_NOREF set[1][2] but the IPv6-specific flag RT6_LOOKUP_F_DST_NOREF might not be, leading to fib6_rule_suppress not decreasing the refcount when needed.

How to reproduce:
- Add the following nftables rule to a prerouting chain:
meta nfproto ipv6 fib saddr . mark . iif oif missing drop
This can be done with:
sudo nft create table inet test
sudo nft create chain inet test test_chain '{ type filter hook prerouting priority filter + 10; policy accept; }' sudo nft add rule inet test test_chain meta nfproto ipv6 fib saddr . mark . iif oif missing drop
- Run:
sudo ip -6 rule add table main suppress_prefixlength 0
- Watch sudo slabtop -o | grep ip6_dst_cache to see memory usage increase
with every incoming ipv6 packet.

This patch exposes the protocol-specific flags to the protocol specific suppress function, and check the protocol-specific flags argument for RT6_LOOKUP_F_DST_NOREF instead of the generic FIB_LOOKUP_NOREF when decreasing the refcount, like this.

[1]: https://github.com/torvalds/linux/blob/ca7a03c4175366a92cee0ccc4fec0038c3266e26/net/ipv6/fib6_rules.c#L71
[2]: https://github.com/torvalds/linux/blob/ca7a03c4175366a92cee0ccc4fec0038c3266e26/net/ipv6/fib6_rules.c#L99

---
 include/net/fib_rules.h |    2 +-
 net/core/fib_rules.c    |    2 +-
 net/ipv4/fib_rules.c    |    2 +-
 net/ipv6/fib6_rules.c   |    5 ++---
 4 files changed, 5 insertions(+), 6 deletions(-)

@@ -68,7 +68,7 @@ struct fib_rules_ops {
 	int			(*action)(struct fib_rule *,
 					  struct flowi *, int,
 					  struct fib_lookup_arg *);
-	bool			(*suppress)(struct fib_rule *,
+	bool			(*suppress)(struct fib_rule *, int,
 					    struct fib_lookup_arg *);
 	int			(*match)(struct fib_rule *,
 					 struct flowi *, int);

@@ -300,7 +300,7 @@ jumped:
 		else
 			err = ops->action(rule, fl, flags, arg);
 
-		if (!err && ops->suppress && ops->suppress(rule, arg))
+		if (!err && ops->suppress && ops->suppress(rule, flags, arg))
 			continue;
 
 		if (err != -EAGAIN) {

@@ -137,7 +137,7 @@ static int fib4_rule_action(struct fib_r
 	return err;
 }
 
-static bool fib4_rule_suppress(struct fib_rule *rule, struct fib_lookup_arg *arg)
+static bool fib4_rule_suppress(struct fib_rule *rule, int flags, struct fib_lookup_arg *arg)
 {
 	struct fib_result *result = (struct fib_result *) arg->result;
 	struct net_device *dev = NULL;

@@ -260,7 +260,7 @@ static int fib6_rule_action(struct fib_r
 	return __fib6_rule_action(rule, flp, flags, arg);
 }
 
-static bool fib6_rule_suppress(struct fib_rule *rule, struct fib_lookup_arg *arg)
+static bool fib6_rule_suppress(struct fib_rule *rule, int flags, struct fib_lookup_arg *arg)
 {
 	struct fib6_result *res = arg->result;
 	struct rt6_info *rt = res->rt6;

@@ -287,8 +287,7 @@ static bool fib6_rule_suppress(struct fi
 	return false;
 
 suppress_route:
-	if (!(arg->flags & FIB_LOOKUP_NOREF))
-		ip6_rt_put(rt);
+	ip6_rt_put_flags(rt, flags);
 	return true;
 }
 

From: Like Xu <[email protected]>

[ Upstream commit cb1d220da0faa5ca0deb93449aff953f0c2cce6d ]

If we run the following perf command in an AMD Milan guest:

  perf stat \
  -e cpu/event=0x1d0/ \
  -e cpu/event=0x1c7/ \
  -e cpu/umask=0x1f,event=0x18e/ \
  -e cpu/umask=0x7,event=0x18e/ \
  -e cpu/umask=0x18,event=0x18e/ \
  ./workload

dmesg will report a #GP warning from an unchecked MSR access error on MSR_F15H_PERF_CTLx.

This is because according to APM (Revision: 4.03) Figure 13-7, the bits [35:32] of AMD PerfEvtSeln register is a part of the event select encoding, which extends the EVENT_SELECT field from 8 bits to 12 bits.

Opportunistically update pmu->reserved_bits for reserved bit 19.

Message-Id: <[email protected]>

---
 arch/x86/kvm/pmu_amd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index c8388389a3b05..d9990951fd0ac 100644
@@ -266,7 +266,7 @@ static void amd_pmu_refresh(struct kvm_vcpu *vcpu)
 		pmu->nr_arch_gp_counters = AMD64_NUM_COUNTERS;
 
 	pmu->counter_bitmask[KVM_PMC_GP] = ((u64)1 << 48) - 1;
-	pmu->reserved_bits = 0xffffffff00200000ull;
+	pmu->reserved_bits = 0xfffffff000280000ull;
 	pmu->version = 1;
 	/* not applicable to AMD; but clean them to prevent any fall out */
 	pmu->counter_bitmask[KVM_PMC_FIXED] = 0;

2.33.0

From: Qais Yousef <[email protected]>

[ Upstream commit 315c4f884800c45cb6bd8c90422fad554a8b9588 ]

Commit d81ae8aac85c ("sched/uclamp: Fix initialization of struct uclamp_rq") introduced a bug where uclamp_max of the rq is not reset to match the woken up task's uclamp_max when the rq is idle.

The code was relying on rq->uclamp_max initialized to zero, so on first enqueue

	static inline void uclamp_rq_inc_id(struct rq *rq, struct task_struct *p,
					    enum uclamp_id clamp_id)
	{
		...

		if (uc_se->value > READ_ONCE(uc_rq->value))
			WRITE_ONCE(uc_rq->value, uc_se->value);
	}

was actually resetting it. But since commit d81ae8aac85c changed the default to 1024, this no longer works. And since rq->uclamp_flags is also initialized to 0, neither above code path nor uclamp_idle_reset() update the rq->uclamp_max on first wake up from idle.

This is only visible from first wake up(s) until the first dequeue to idle after enabling the static key. And it only matters if the uclamp_max of this task is < 1024 since only then its uclamp_max will be effectively ignored.

Fix it by properly initializing rq->uclamp_flags = UCLAMP_FLAG_IDLE to ensure uclamp_idle_reset() is called which then will update the rq uclamp_max value as expected.

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

index f8ca0738d729e..5befdecefe947 100644
@@ -1337,7 +1337,7 @@ static void __init init_uclamp_rq(struct rq *rq)
 		};
 	}
 
-	rq->uclamp_flags = 0;
+	rq->uclamp_flags = UCLAMP_FLAG_IDLE;
 }
 
 static void __init init_uclamp(void)

2.33.0

From: Helge Deller <[email protected]>

commit 1d7c29b77725d05faff6754d2f5e7c147aedcf93 upstream.

Default KBUILD_IMAGE to $(boot)/bzImage if a self-extracting (CONFIG_PARISC_SELF_EXTRACT=y) kernel is to be built. This fixes the bindeb-pkg make target.

---
 arch/parisc/Makefile |    5 +++++
 1 file changed, 5 insertions(+)

@@ -17,7 +17,12 @@
 # Mike Shaver, Helge Deller and Martin K. Petersen
 #
 
+ifdef CONFIG_PARISC_SELF_EXTRACT
+boot := arch/parisc/boot
+KBUILD_IMAGE := $(boot)/bzImage
+else
 KBUILD_IMAGE := vmlinuz
+endif
 
 NM		= sh $(srctree)/arch/parisc/nm
 CHECKFLAGS	+= -D__hppa__=1

From: Helge Deller <[email protected]>

commit 0f9fee4cdebfbe695c297e5b603a275e2557c1cc upstream.

On newer debian releases the debian-provided "installkernel" script is installed in /usr/sbin. Fix the kernel install.sh script to look for the script in this directory as well.

---
 arch/parisc/install.sh |    1 +
 1 file changed, 1 insertion(+)

@@ -39,6 +39,7 @@ verify "$3"
 if [ -n "${INSTALLKERNEL}" ]; then
   if [ -x ~/bin/${INSTALLKERNEL} ]; then exec ~/bin/${INSTALLKERNEL} "$@"; fi
   if [ -x /sbin/${INSTALLKERNEL} ]; then exec /sbin/${INSTALLKERNEL} "$@"; fi
+  if [ -x /usr/sbin/${INSTALLKERNEL} ]; then exec /usr/sbin/${INSTALLKERNEL} "$@"; fi
 fi
 
 # Default install

From: Maciej W. Rozycki <[email protected]>

commit 3dfac26e2ef29ff2abc2a75aa4cd48fce25a2c4b upstream.

Fix a division by zero in `vgacon_resize' with a backtrace like:

vgacon_resize
vc_do_resize
vgacon_init
do_bind_con_driver
do_unbind_con_driver
fbcon_fb_unbind
do_unregister_framebuffer
do_register_framebuffer
register_framebuffer
__drm_fb_helper_initial_config_and_unlock
drm_helper_hpd_irq_event
dw_hdmi_irq
irq_thread
kthread

caused by `c->vc_cell_height' not having been initialized. This has only started to trigger with commit 860dafa90259 ("vt: Fix character height handling with VT_RESIZEX"), however the ultimate offender is commit 50ec42edd978 ("[PATCH] Detaching fbcon: fix vgacon to allow retaking of the console").

Said commit has added a call to vc_resize' whenever vgacon_init' is called with the init' argument set to 0, which did not happen before. And the call is made before a key vgacon boot parameter retrieved in vgacon_startup' has been propagated in vgacon_init' for vc_resize' to use to the console structure being worked on. Previously the parameter was c->vc_font.height' and now it is c->vc_cell_height'.

In this particular scenario the registration of fbcon has failed and vt resorts to vgacon. Now fbcon does have initialized c->vc_font.height' somehow, unlike c->vc_cell_height', which is why this code did not crash before, but either way the boot parameters should have been copied to the console structure ahead of the call to vc_resize' rather than afterwards, so that first the call has a chance to use them and second they do not change the console structure to something possibly different from what was used by vc_resize'.

Move the propagation of the vgacon boot parameters ahead of the call to `vc_resize' then. Adjust the comment accordingly.

---
 drivers/video/console/vgacon.c |   14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

@@ -370,11 +370,17 @@ static void vgacon_init(struct vc_data *
 	struct uni_pagedir *p;
 
 	/*
-	 * We cannot be loaded as a module, therefore init is always 1,
-	 * but vgacon_init can be called more than once, and init will
-	 * not be 1.
+	 * We cannot be loaded as a module, therefore init will be 1
+	 * if we are the default console, however if we are a fallback
+	 * console, for example if fbcon has failed registration, then
+	 * init will be 0, so we need to make sure our boot parameters
+	 * have been copied to the console structure for vgacon_resize
+	 * ultimately called by vc_resize.  Any subsequent calls to
+	 * vgacon_init init will have init set to 0 too.
 	 */
 	c->vc_can_do_color = vga_can_do_color;
+	c->vc_scan_lines = vga_scan_lines;
+	c->vc_font.height = c->vc_cell_height = vga_video_font_height;
 
 	/* set dimensions manually if init != 0 since vc_resize() will fail */
 	if (init) {

@@ -383,8 +389,6 @@ static void vgacon_init(struct vc_data *
 	} else
 		vc_resize(c, vga_video_num_columns, vga_video_num_lines);
 
-	c->vc_scan_lines = vga_scan_lines;
-	c->vc_font.height = c->vc_cell_height = vga_video_font_height;
 	c->vc_complement_mask = 0x7700;
 	if (vga_512_chars)
 		c->vc_hi_font_mask = 0x0800;

From: Mathias Nyman <[email protected]>

commit 09f736aa95476631227d2dc0e6b9aeee1ad7ed58 upstream.

Turns out some xHC controllers require all 64 bits in the CRCR register to be written to execute a command abort.

The lower 32 bits containing the command abort bit is written first. In case the command ring stops before we write the upper 32 bits then hardware may use these upper bits to set the commnd ring dequeue pointer.

Solve this by making sure the upper 32 bits contain a valid command ring dequeue pointer.

The original patch that only wrote the first 32 to stop the ring went to stable, so this fix should go there as well.

---
 drivers/usb/host/xhci-ring.c |   21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)

@@ -339,7 +339,9 @@ static void xhci_handle_stopped_cmd_ring
 /* Must be called with xhci->lock held, releases and aquires lock back */
 static int xhci_abort_cmd_ring(struct xhci_hcd *xhci, unsigned long flags)
 {
-	u32 temp_32;
+	struct xhci_segment *new_seg	= xhci->cmd_ring->deq_seg;
+	union xhci_trb *new_deq		= xhci->cmd_ring->dequeue;
+	u64 crcr;
 	int ret;
 
 	xhci_dbg(xhci, "Abort command ring\n");

@@ -348,13 +350,18 @@ static int xhci_abort_cmd_ring(struct xh
 
 	/*
 	 * The control bits like command stop, abort are located in lower
-	 * dword of the command ring control register. Limit the write
-	 * to the lower dword to avoid corrupting the command ring pointer
-	 * in case if the command ring is stopped by the time upper dword
-	 * is written.
+	 * dword of the command ring control register.
+	 * Some controllers require all 64 bits to be written to abort the ring.
+	 * Make sure the upper dword is valid, pointing to the next command,
+	 * avoiding corrupting the command ring pointer in case the command ring
+	 * is stopped by the time the upper dword is written.
 	 */
-	temp_32 = readl(&xhci->op_regs->cmd_ring);
-	writel(temp_32 | CMD_RING_ABORT, &xhci->op_regs->cmd_ring);
+	next_trb(xhci, NULL, &new_seg, &new_deq);
+	if (trb_is_link(new_deq))
+		next_trb(xhci, NULL, &new_seg, &new_deq);
+
+	crcr = xhci_trb_virt_to_dma(new_seg, new_deq);
+	xhci_write_64(xhci, crcr | CMD_RING_ABORT, &xhci->op_regs->cmd_ring);
 
 	/* Section 4.6.1.2 of xHCI 1.0 spec says software should also time the
 	 * completion of the Command Abort operation. If CRR is not negated in 5

From: Ole Ernst <[email protected]>

commit d2a004037c3c6afd36d40c384d2905f47cd51c57 upstream.

This is another branded 8153 device that doesn't work well with LPM:
r8152 2-2.1:1.0 enp0s13f0u2u1: Stop submitting intr, status -71

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 Powered USB-C Travel Hub (4X90S92381, RTL8153 GigE) */
+	{ USB_DEVICE(0x17ef, 0x721e), .driver_info = USB_QUIRK_NO_LPM },
+
 	/* Lenovo ThinkCenter A630Z TI024Gen3 usb-audio */
 	{ USB_DEVICE(0x17ef, 0xa012), .driver_info =
 			USB_QUIRK_DISCONNECT_SUSPEND },

From: Badhri Jagan Sridharan <[email protected]>

commit fbcd13df1e78eb2ba83a3c160eefe2d6f574beaf upstream.

Stub from the spec:
"4.5.2.2.4.2 Exiting from AttachWait.SNK State
A Sink shall transition to Unattached.SNK when the state of both the CC1 and CC2 pins is SNK.Open for at least tPDDebounce. A DRP shall transition to Unattached.SRC when the state of both the CC1 and CC2 pins is SNK.Open for at least tPDDebounce."

This change makes TCPM to wait in SNK_DEBOUNCED state until CC1 and CC2 pins is SNK.Open for at least tPDDebounce. Previously, TCPM resets the port if vbus is not present in PD_T_PS_SOURCE_ON. This causes TCPM to loop continuously when connected to a faulty power source that does not present vbus. Waiting in SNK_DEBOUNCED also ensures that TCPM is adherant to "4.5.2.2.4.2 Exiting from AttachWait.SNK State" requirements.

[ 6169.280751] CC1: 0 -> 0, CC2: 0 -> 5 [state TOGGLING, polarity 0, connected]
[ 6169.280759] state change TOGGLING -> SNK_ATTACH_WAIT [rev2 NONE_AMS]
[ 6169.280771] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @ 170 ms [rev2 NONE_AMS]
[ 6169.282427] CC1: 0 -> 0, CC2: 5 -> 5 [state SNK_ATTACH_WAIT, polarity 0, connected]
[ 6169.450825] state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED [delayed 170 ms]
[ 6169.450834] pending state change SNK_DEBOUNCED -> PORT_RESET @ 480 ms [rev2 NONE_AMS]
[ 6169.930892] state change SNK_DEBOUNCED -> PORT_RESET [delayed 480 ms]
[ 6169.931296] disable vbus discharge ret:0
[ 6169.931301] Setting usb_comm capable false
[ 6169.932783] Setting voltage/current limit 0 mV 0 mA
[ 6169.932802] polarity 0
[ 6169.933706] Requesting mux state 0, usb-role 0, orientation 0
[ 6169.936689] cc:=0
[ 6169.936812] pending state change PORT_RESET -> PORT_RESET_WAIT_OFF @ 100 ms [rev2 NONE_AMS]
[ 6169.937157] CC1: 0 -> 0, CC2: 5 -> 0 [state PORT_RESET, polarity 0, disconnected]
[ 6170.036880] state change PORT_RESET -> PORT_RESET_WAIT_OFF [delayed 100 ms]
[ 6170.036890] state change PORT_RESET_WAIT_OFF -> SNK_UNATTACHED [rev2 NONE_AMS]
[ 6170.036896] Start toggling
[ 6170.041412] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected]
[ 6170.042973] CC1: 0 -> 0, CC2: 0 -> 5 [state TOGGLING, polarity 0, connected]
[ 6170.042976] state change TOGGLING -> SNK_ATTACH_WAIT [rev2 NONE_AMS]
[ 6170.042981] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @ 170 ms [rev2 NONE_AMS]
[ 6170.213014] state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED [delayed 170 ms]
[ 6170.213019] pending state change SNK_DEBOUNCED -> PORT_RESET @ 480 ms [rev2 NONE_AMS]
[ 6170.693068] state change SNK_DEBOUNCED -> PORT_RESET [delayed 480 ms]
[ 6170.693304] disable vbus discharge ret:0
[ 6170.693308] Setting usb_comm capable false
[ 6170.695193] Setting voltage/current limit 0 mV 0 mA
[ 6170.695210] polarity 0
[ 6170.695990] Requesting mux state 0, usb-role 0, orientation 0
[ 6170.701896] cc:=0
[ 6170.702181] pending state change PORT_RESET -> PORT_RESET_WAIT_OFF @ 100 ms [rev2 NONE_AMS]
[ 6170.703343] CC1: 0 -> 0, CC2: 5 -> 0 [state PORT_RESET, polarity 0, disconnected]

---
 drivers/usb/typec/tcpm/tcpm.c |    4 ----
 1 file changed, 4 deletions(-)

@@ -3118,11 +3118,7 @@ static void run_state_machine(struct tcp
 				       tcpm_try_src(port) ? SRC_TRY
 							  : SNK_ATTACHED,
 				       0);
-		else
-			/* Wait for VBUS, but not forever */
-			tcpm_set_state(port, PORT_RESET, PD_T_PS_SOURCE_ON);
 		break;
-
 	case SRC_TRY:
 		port->try_src_count++;
 		tcpm_set_cc(port, tcpm_rp_cc(port));

From: Feng Tang <[email protected]>

commit c7719e79347803b8e3b6b50da8c6db410a3012b5 upstream.

The TSC_ADJUST register is checked every time a CPU enters idle state, but Thomas Gleixner mentioned there is still a caveat that a system won't enter idle [1], either because it's too busy or configured purposely to not enter idle.

Setup a periodic timer (every 10 minutes) to make sure the check is happening on a regular base.

[1] https://lore.kernel.org/lkml/[email protected]/

---
 arch/x86/kernel/tsc_sync.c |   41 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

@@ -30,6 +30,7 @@ struct tsc_adjust {
 };
 
 static DEFINE_PER_CPU(struct tsc_adjust, tsc_adjust);
+static struct timer_list tsc_sync_check_timer;
 
 /*
  * TSC's on different sockets may be reset asynchronously.

@@ -77,6 +78,46 @@ void tsc_verify_tsc_adjust(bool resume)
 	}
 }
 
+/*
+ * Normally the tsc_sync will be checked every time system enters idle
+ * state, but there is still caveat that a system won't enter idle,
+ * either because it's too busy or configured purposely to not enter
+ * idle.
+ *
+ * So setup a periodic timer (every 10 minutes) to make sure the check
+ * is always on.
+ */
+
+#define SYNC_CHECK_INTERVAL		(HZ * 600)
+
+static void tsc_sync_check_timer_fn(struct timer_list *unused)
+{
+	int next_cpu;
+
+	tsc_verify_tsc_adjust(false);
+
+	/* Run the check for all onlined CPUs in turn */
+	next_cpu = cpumask_next(raw_smp_processor_id(), cpu_online_mask);
+	if (next_cpu >= nr_cpu_ids)
+		next_cpu = cpumask_first(cpu_online_mask);
+
+	tsc_sync_check_timer.expires += SYNC_CHECK_INTERVAL;
+	add_timer_on(&tsc_sync_check_timer, next_cpu);
+}
+
+static int __init start_sync_check_timer(void)
+{
+	if (!cpu_feature_enabled(X86_FEATURE_TSC_ADJUST) || tsc_clocksource_reliable)
+		return 0;
+
+	timer_setup(&tsc_sync_check_timer, tsc_sync_check_timer_fn, 0);
+	tsc_sync_check_timer.expires = jiffies + SYNC_CHECK_INTERVAL;
+	add_timer(&tsc_sync_check_timer);
+
+	return 0;
+}
+late_initcall(start_sync_check_timer);
+
 static void tsc_sanitize_first_cpu(struct tsc_adjust *cur, s64 bootval,
 				   unsigned int cpu, bool bootcpu)
 {

From: Feng Tang <[email protected]>

commit b50db7095fe002fa3e16605546cba66bf1b68a3e upstream.

There are cases that the TSC clocksource is wrongly judged as unstable by the clocksource watchdog mechanism which tries to validate the TSC against HPET, PM_TIMER or jiffies. While there is hardly a general reliable way to check the validity of a watchdog, Thomas Gleixner proposed [1]:

"I'm inclined to lift that requirement when the CPU has:

1) X86_FEATURE_CONSTANT_TSC
2) X86_FEATURE_NONSTOP_TSC
3) X86_FEATURE_NONSTOP_TSC_S3
4) X86_FEATURE_TSC_ADJUST
5) At max. 4 sockets

After two decades of horrors we're finally at a point where TSC seems to be halfway reliable and less abused by BIOS tinkerers. TSC_ADJUST was really key as we can now detect even small modifications reliably and the important point is that we can cure them as well (not pretty but better than all other options)."

As feature #3 X86_FEATURE_NONSTOP_TSC_S3 only exists on several generations of Atom processorz, and is always coupled with X86_FEATURE_CONSTANT_TSC and X86_FEATURE_NONSTOP_TSC, skip checking it, and also be more defensive to use maximal 2 sockets.

The check is done inside tsc_init() before registering 'tsc-early' and 'tsc' clocksources, as there were cases that both of them had been wrongly judged as unreliable.

For more background of tsc/watchdog, there is a good summary in [2]

[tglx} Update vs. jiffies:

On systems where the only remaining clocksource aside of TSC is jiffies there is no way to make this work because that creates a circular dependency. Jiffies accuracy depends on not missing a periodic timer interrupt, which is not guaranteed. That could be detected by TSC, but as TSC is not trusted this cannot be compensated. The consequence is a circulus vitiosus which results in shutting down TSC and falling back to the jiffies clocksource which is even more unreliable.

[1]. https://lore.kernel.org/lkml/[email protected]/
[2]. https://lore.kernel.org/lkml/[email protected]/

[ tglx: Refine comment and amend changelog ]

---
 arch/x86/kernel/tsc.c |   28 ++++++++++++++++++++++++----
 1 file changed, 24 insertions(+), 4 deletions(-)

@@ -1162,6 +1162,12 @@ void mark_tsc_unstable(char *reason)
 
 EXPORT_SYMBOL_GPL(mark_tsc_unstable);
 
+static void __init tsc_disable_clocksource_watchdog(void)
+{
+	clocksource_tsc_early.flags &= ~CLOCK_SOURCE_MUST_VERIFY;
+	clocksource_tsc.flags &= ~CLOCK_SOURCE_MUST_VERIFY;
+}
+
 static void __init check_system_tsc_reliable(void)
 {
 #if defined(CONFIG_MGEODEGX1) || defined(CONFIG_MGEODE_LX) || defined(CONFIG_X86_GENERIC)

@@ -1178,6 +1184,23 @@ static void __init check_system_tsc_reli
 #endif
 	if (boot_cpu_has(X86_FEATURE_TSC_RELIABLE))
 		tsc_clocksource_reliable = 1;
+
+	/*
+	 * Disable the clocksource watchdog when the system has:
+	 *  - TSC running at constant frequency
+	 *  - TSC which does not stop in C-States
+	 *  - the TSC_ADJUST register which allows to detect even minimal
+	 *    modifications
+	 *  - not more than two sockets. As the number of sockets cannot be
+	 *    evaluated at the early boot stage where this has to be
+	 *    invoked, check the number of online memory nodes as a
+	 *    fallback solution which is an reasonable estimate.
+	 */
+	if (boot_cpu_has(X86_FEATURE_CONSTANT_TSC) &&
+	    boot_cpu_has(X86_FEATURE_NONSTOP_TSC) &&
+	    boot_cpu_has(X86_FEATURE_TSC_ADJUST) &&
+	    nr_online_nodes <= 2)
+		tsc_disable_clocksource_watchdog();
 }
 
 /*

@@ -1369,9 +1392,6 @@ static int __init init_tsc_clocksource(v
 	if (tsc_unstable)
 		goto unreg;
 
-	if (tsc_clocksource_reliable || no_tsc_watchdog)
-		clocksource_tsc.flags &= ~CLOCK_SOURCE_MUST_VERIFY;
-
 	if (boot_cpu_has(X86_FEATURE_NONSTOP_TSC_S3))
 		clocksource_tsc.flags |= CLOCK_SOURCE_SUSPEND_NONSTOP;
 

@@ -1506,7 +1526,7 @@ void __init tsc_init(void)
 	}
 
 	if (tsc_clocksource_reliable || no_tsc_watchdog)
-		clocksource_tsc_early.flags &= ~CLOCK_SOURCE_MUST_VERIFY;
+		tsc_disable_clocksource_watchdog();
 
 	clocksource_register_khz(&clocksource_tsc_early, tsc_khz);
 	detect_art();

From: Joerg Roedel <[email protected]>

commit 51523ed1c26758de1af7e58730a656875f72f783 upstream.

The trampoline_pgd only maps the 0xfffffff000000000-0xffffffffffffffff range of kernel memory (with 4-level paging). This range contains the kernel's text+data+bss mappings and the module mapping space but not the direct mapping and the vmalloc area.

This is enough to get the application processors out of real-mode, but for code that switches back to real-mode the trampoline_pgd is missing important parts of the address space. For example, consider this code from arch/x86/kernel/reboot.c, function machine_real_restart() for a 64-bit kernel:

  #ifdef CONFIG_X86_32
  	load_cr3(initial_page_table);
  #else
  	write_cr3(real_mode_header->trampoline_pgd);

  	/* Exiting long mode will fail if CR4.PCIDE is set. */
  	if (boot_cpu_has(X86_FEATURE_PCID))
  		cr4_clear_bits(X86_CR4_PCIDE);
  #endif

  	/* Jump to the identity-mapped low memory code */
  #ifdef CONFIG_X86_32
  	asm volatile("jmpl *%0" : :
  		     "rm" (real_mode_header->machine_real_restart_asm),
  		     "a" (type));
  #else
  	asm volatile("ljmpl *%0" : :
  		     "m" (real_mode_header->machine_real_restart_asm),
  		     "D" (type));
  #endif

The code switches to the trampoline_pgd, which unmaps the direct mapping and also the kernel stack. The call to cr4_clear_bits() will find no stack and crash the machine. The real_mode_header pointer below points into the direct mapping, and dereferencing it also causes a crash.

The reason this does not crash always is only that kernel mappings are global and the CR3 switch does not flush those mappings. But if theses mappings are not in the TLB already, the above code will crash before it can jump to the real-mode stub.

Extend the trampoline_pgd to contain all kernel mappings to prevent these crashes and to make code which runs on this page-table more robust.