Skip to contentSkip to conversation

5.15.126-rc1 review

PATCH In review Greg Kroah-Hartman 10 participants 121 messages Aug 9, 2023 – Aug 11, 2023

Subsystems (5)
lore
Share & tools
lore

Greg Kroah-Hartman

author maintainer
#0 lore raw

cover 5.15.126-rc1 review

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

    Linux 5.15.126-rc1

Johan Hovold <[email protected]>
PM: sleep: wakeirq: fix wake irq arming

Chunfeng Yun <[email protected]>
PM / wakeirq: support enabling wake-up irq after runtime_suspend called

Johan Hovold <[email protected]>
soundwire: fix enumeration completion

Pierre-Louis Bossart <[email protected]>
soundwire: bus: pm_runtime_request_resume on peripheral attachment

Sean Christopherson <[email protected]>
selftests/rseq: Play nice with binaries statically linked against glibc 2.35+

Michael Jeanson <[email protected]>
selftests/rseq: check if libc rseq support is registered

Alexander Stein <[email protected]>
drm/imx/ipuv3: Fix front porch adjustment upon hactive aligning

Thomas Zimmermann <[email protected]>
drm/fsl-dcu: Use drm_plane_helper_destroy()

Aneesh Kumar K.V <[email protected]>
powerpc/mm/altmap: Fix altmap boundary check

Christophe JAILLET <[email protected]>
mtd: rawnand: fsl_upm: Fix an off-by one test in fun_exec_op()

Johan Jonker <[email protected]>
mtd: rawnand: rockchip: Align hwecc vs. raw page helper layouts

Johan Jonker <[email protected]>
mtd: rawnand: rockchip: fix oobfree offset and description

Roger Quadros <[email protected]>
mtd: rawnand: omap_elm: Fix incorrect type in assignment

Jan Kara <[email protected]>
ext2: Drop fragment support

Jan Kara <[email protected]>
fs: Protect reconfiguration of sb read-write from racing writes

Alan Stern <[email protected]>
net: usbnet: Fix WARNING in usbnet_start_xmit/usb_submit_urb

Sungwoo Kim <[email protected]>
Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb

Prince Kumar Maurya <[email protected]>
fs/sysv: Null check to prevent null-ptr-deref bug

Tetsuo Handa <[email protected]>
fs/ntfs3: Use __GFP_NOWARN allocation at ntfs_load_attr_list()

Linus Torvalds <[email protected]>
file: reinstate f_pos locking optimization for regular files

Hou Tao <[email protected]>
bpf, cpumap: Make sure kthread is running before map update returns

Guchun Chen <[email protected]>
drm/ttm: check null pointer before accessing when swapping

Aleksa Sarai <[email protected]>
open: make RESOLVE_CACHED correctly test for O_TMPFILE

Jiri Olsa <[email protected]>
bpf: Disable preemption in bpf_event_output

Ilya Dryomov <[email protected]>
rbd: prevent busy loop when requesting exclusive lock

Paul Fertser <[email protected]>
wifi: mt76: mt7615: do not advertise 5 GHz on first phy of MT7615D (DBDC)

Laszlo Ersek <[email protected]>
net: tap_open(): set sk_uid from current_fsuid()

Laszlo Ersek <[email protected]>
net: tun_chr_open(): set sk_uid from current_fsuid()

Dinh Nguyen <[email protected]>
arm64: dts: stratix10: fix incorrect I2C property for SCL signal

Arseniy Krasnov <[email protected]>
mtd: rawnand: meson: fix OOB available bytes for ECC

Olivier Maignial <[email protected]>
mtd: spinand: toshiba: Fix ecc_get_status

Sungjong Seo <[email protected]>
exfat: release s_lock before calling dir_emit()

gaoming <[email protected]>
exfat: use kvmalloc_array/kvfree instead of kmalloc_array/kfree

Krzysztof Kozlowski <[email protected]>
firmware: arm_scmi: Drop OF node reference in the transport channel setup

Xiubo Li <[email protected]>
ceph: defer stopping mdsc delayed_work

Ross Maynard <[email protected]>

    USB: zaurus: Add ID for A-300/B-500/C-700

Ilya Dryomov <[email protected]>
libceph: fix potential hang in ceph_osdc_notify()

Michael Kelley <[email protected]>
scsi: storvsc: Limit max_sectors for virtual Fibre Channel devices

Steffen Maier <[email protected]>
scsi: zfcp: Defer fc_rport blocking until after ADISC response

Eric Dumazet <[email protected]>
tcp_metrics: fix data-race in tcpm_suck_dst() vs fastopen

Eric Dumazet <[email protected]>

    tcp_metrics: annotate data-races around tm->tcpm_net

Eric Dumazet <[email protected]>

    tcp_metrics: annotate data-races around tm->tcpm_vals[]

Eric Dumazet <[email protected]>

    tcp_metrics: annotate data-races around tm->tcpm_lock

Eric Dumazet <[email protected]>

    tcp_metrics: annotate data-races around tm->tcpm_stamp

Eric Dumazet <[email protected]>
tcp_metrics: fix addr_same() helper

Jonas Gorski <[email protected]>
prestera: fix fallback to previous version on same major version

Jianbo Liu <[email protected]>
net/mlx5: fs_core: Skip the FTs in the same FS_TYPE_PRIO_CHAINS fs_prio

Jianbo Liu <[email protected]>
net/mlx5: fs_core: Make find_closest_ft more generic

Benjamin Poirier <[email protected]>
vxlan: Fix nexthop hash size

Yue Haibing <[email protected]>
ip6mr: Fix skb_under_panic in ip6mr_cache_report()

Alexandra Winter <[email protected]>
s390/qeth: Don't call dev_close/dev_open (DOWN/UP)

Lin Ma <[email protected]>
net: dcb: choose correct policy to parse DCB_ATTR_BCN

Mark Brown <[email protected]>
net: netsec: Ignore 'phy-mode' on SynQuacer in DT mode

Yuanjun Gong <[email protected]>
net: korina: handle clk prepare error in korina_probe()

Dan Carpenter <[email protected]>
net: ll_temac: fix error checking of irq_of_parse_and_map()

Yang Yingliang <[email protected]>
net: ll_temac: Switch to use dev_err_probe() helper

Tomas Glozar <[email protected]>
bpf: sockmap: Remove preempt_disable in sock_map_sk_acquire

valis <[email protected]>
net/sched: cls_route: No longer copy tcf_result on update to avoid use-after-free

valis <[email protected]>
net/sched: cls_fw: No longer copy tcf_result on update to avoid use-after-free

valis <[email protected]>
net/sched: cls_u32: No longer copy tcf_result on update to avoid use-after-free

Hou Tao <[email protected]>
bpf, cpumap: Handle skb as well when clean up ptr_ring

Kuniyuki Iwashima <[email protected]>
net/sched: taprio: Limit TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME to INT_MAX.

Eric Dumazet <[email protected]>
net: add missing data-race annotation for sk_ll_usec

Eric Dumazet <[email protected]>

    net: add missing data-race annotations around sk->sk_peek_off

Eric Dumazet <[email protected]>

    net: add missing READ_ONCE(sk->sk_rcvbuf) annotation

Eric Dumazet <[email protected]>

    net: add missing READ_ONCE(sk->sk_sndbuf) annotation

Eric Dumazet <[email protected]>

    net: add missing READ_ONCE(sk->sk_rcvlowat) annotation

Eric Dumazet <[email protected]>

    net: annotate data-races around sk->sk_max_pacing_rate

Konstantin Khorenko <[email protected]>
qed: Fix scheduling in a tasklet while getting stats

Prabhakar Kushwaha <[email protected]>
qed: Fix kernel-doc warnings

Chengfeng Ye <[email protected]>

    mISDN: hfcpci: Fix potential deadlock on &hc->lock

Jamal Hadi Salim <[email protected]>
net: sched: cls_u32: Fix match key mis-addressing

Georg Müller <[email protected]>
perf test uprobe_from_different_cu: Skip if there is no gcc

Yuanjun Gong <[email protected]>
net: dsa: fix value check in bcm_sf2_sw_probe()

Lin Ma <[email protected]>
rtnetlink: let rtnl_bridge_setlink checks IFLA_BRIDGE_MODE length

Lin Ma <[email protected]>
bpf: Add length check for SK_DIAG_BPF_STORAGE_REQ_MAP_FD parsing

Yuanjun Gong <[email protected]>
net/mlx5e: fix return value check in mlx5e_ipsec_remove_trailer()

Zhengchao Shao <[email protected]>
net/mlx5: DR, fix memory leak in mlx5dr_cmd_create_reformat_ctx

Ilan Peer <[email protected]>
wifi: cfg80211: Fix return value in scan logic

Heiko Carstens <[email protected]>
KVM: s390: fix sthyi error handling

[email protected] <[email protected]>
word-at-a-time: use the same return type for has_zero regardless of endianness

Cristian Marussi <[email protected]>
firmware: arm_scmi: Fix chan_free cleanup on SMC

Hugo Villeneuve <[email protected]>
arm64: dts: imx8mn-var-som: add missing pull-up for onboard PHY reset pinmux

Robin Murphy <[email protected]>
iommu/arm-smmu-v3: Document nesting-related errata

Robin Murphy <[email protected]>
iommu/arm-smmu-v3: Add explicit feature for nesting

Robin Murphy <[email protected]>

    iommu/arm-smmu-v3: Document MMU-700 erratum 2812531

Robin Murphy <[email protected]>

    iommu/arm-smmu-v3: Work around MMU-600 erratum 1076982

Suzuki K Poulose <[email protected]>
arm64: errata: Add detection for TRBE write to out-of-range

Suzuki K Poulose <[email protected]>
arm64: errata: Add workaround for TSB flush failures

Shay Drory <[email protected]>
net/mlx5: Free irqs only on shutdown callback

Peter Zijlstra <[email protected]>
perf: Fix function pointer case

Jens Axboe <[email protected]>
io_uring: gate iowait schedule on having pending requests

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

Diffstat:

 Documentation/arm64/silicon-errata.rst             |  12 +
 Makefile                                           |   4 +-
 arch/arm64/Kconfig                                 |  74 ++
 .../boot/dts/altera/socfpga_stratix10_socdk.dts    |   2 +-
 .../dts/altera/socfpga_stratix10_socdk_nand.dts    |   2 +-
 arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi  |   2 +-
 arch/arm64/include/asm/barrier.h                   |  16 +-
 arch/arm64/kernel/cpu_errata.c                     |  39 +
 arch/arm64/tools/cpucaps                           |   2 +
 arch/powerpc/include/asm/word-at-a-time.h          |   2 +-
 arch/powerpc/mm/init_64.c                          |   3 +-
 arch/s390/kernel/sthyi.c                           |   6 +-
 arch/s390/kvm/intercept.c                          |   9 +-
 drivers/base/power/power.h                         |   8 +-
 drivers/base/power/runtime.c                       |   6 +-
 drivers/base/power/wakeirq.c                       | 111 ++-
 drivers/block/rbd.c                                |  28 +-
 drivers/firmware/arm_scmi/mailbox.c                |   4 +-
 drivers/firmware/arm_scmi/smc.c                    |  21 +-
 drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c        |   8 +-
 drivers/gpu/drm/imx/ipuv3-crtc.c                   |   2 +-
 drivers/gpu/drm/ttm/ttm_bo.c                       |   3 +-
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c        |  50 ++
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h        |   8 +
 drivers/isdn/hardware/mISDN/hfcpci.c               |  10 +-
 drivers/mtd/nand/raw/fsl_upm.c                     |   2 +-
 drivers/mtd/nand/raw/meson_nand.c                  |   3 +-
 drivers/mtd/nand/raw/omap_elm.c                    |  24 +-
 drivers/mtd/nand/raw/rockchip-nand-controller.c    |  45 +-
 drivers/mtd/nand/spi/toshiba.c                     |   4 +-
 drivers/net/dsa/bcm_sf2.c                          |   8 +-
 drivers/net/ethernet/korina.c                      |   3 +-
 .../net/ethernet/marvell/prestera/prestera_pci.c   |   3 +-
 .../mellanox/mlx5/core/en_accel/ipsec_rxtx.c       |   4 +-
 drivers/net/ethernet/mellanox/mlx5/core/eq.c       |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  | 105 ++-
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  |  29 +
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  |   5 +-
 drivers/net/ethernet/qlogic/qed/qed.h              |   9 +-
 drivers/net/ethernet/qlogic/qed/qed_cxt.h          | 138 +--
 drivers/net/ethernet/qlogic/qed/qed_dev_api.h      | 361 ++++----
 drivers/net/ethernet/qlogic/qed/qed_fcoe.c         |  19 +-
 drivers/net/ethernet/qlogic/qed/qed_fcoe.h         |  17 +-
 drivers/net/ethernet/qlogic/qed/qed_hsi.h          | 922 +++++++++++----------
 drivers/net/ethernet/qlogic/qed/qed_hw.c           |  26 +-
 drivers/net/ethernet/qlogic/qed/qed_hw.h           | 214 ++---
 drivers/net/ethernet/qlogic/qed/qed_init_ops.h     |  58 +-
 drivers/net/ethernet/qlogic/qed/qed_int.h          | 274 +++---
 drivers/net/ethernet/qlogic/qed/qed_iscsi.c        |  19 +-
 drivers/net/ethernet/qlogic/qed/qed_iscsi.h        |  17 +-
 drivers/net/ethernet/qlogic/qed/qed_l2.c           |  19 +-
 drivers/net/ethernet/qlogic/qed/qed_l2.h           | 158 ++--
 drivers/net/ethernet/qlogic/qed/qed_ll2.h          | 130 +--
 drivers/net/ethernet/qlogic/qed/qed_main.c         |   6 +-
 drivers/net/ethernet/qlogic/qed/qed_mcp.h          | 757 +++++++++--------
 drivers/net/ethernet/qlogic/qed/qed_selftest.h     |  30 +-
 drivers/net/ethernet/qlogic/qed/qed_sp.h           | 215 +++--
 drivers/net/ethernet/qlogic/qed/qed_sriov.h        |  99 ++-
 drivers/net/ethernet/qlogic/qed/qed_vf.h           | 301 ++++---
 drivers/net/ethernet/qlogic/qede/qede_main.c       |   5 +-
 drivers/net/ethernet/socionext/netsec.c            |  11 +
 drivers/net/ethernet/xilinx/ll_temac_main.c        |  16 +-
 drivers/net/tap.c                                  |   2 +-
 drivers/net/tun.c                                  |   2 +-
 drivers/net/usb/cdc_ether.c                        |  21 +
 drivers/net/usb/usbnet.c                           |   6 +
 drivers/net/usb/zaurus.c                           |  21 +
 drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c |   6 +-
 drivers/s390/net/qeth_core.h                       |   1 -
 drivers/s390/net/qeth_core_main.c                  |   2 -
 drivers/s390/net/qeth_l2_main.c                    |   9 +-
 drivers/s390/net/qeth_l3_main.c                    |   8 +-
 drivers/s390/scsi/zfcp_fc.c                        |   6 +-
 drivers/scsi/storvsc_drv.c                         |   4 +
 drivers/soundwire/bus.c                            |  20 +-
 fs/ceph/mds_client.c                               |   4 +-
 fs/ceph/mds_client.h                               |   5 +
 fs/ceph/super.c                                    |  10 +
 fs/exfat/balloc.c                                  |   6 +-
 fs/exfat/dir.c                                     |  27 +-
 fs/ext2/ext2.h                                     |  12 -
 fs/ext2/super.c                                    |  23 +-
 fs/file.c                                          |  18 +-
 fs/ntfs3/attrlist.c                                |   4 +-
 fs/open.c                                          |   2 +-
 fs/super.c                                         |  11 +-
 fs/sysv/itree.c                                    |   4 +
 include/asm-generic/word-at-a-time.h               |   2 +-
 include/linux/pm_wakeirq.h                         |   9 +-
 include/linux/qed/qed_chain.h                      |  97 ++-
 include/linux/qed/qed_if.h                         | 255 +++---
 include/linux/qed/qed_iscsi_if.h                   |   2 +-
 include/linux/qed/qed_ll2_if.h                     |  42 +-
 include/linux/qed/qed_nvmetcp_if.h                 |  17 +
 include/net/vxlan.h                                |   4 +-
 io_uring/io_uring.c                                |  23 +-
 kernel/bpf/cpumap.c                                |  35 +-
 kernel/events/core.c                               |   8 +-
 kernel/trace/bpf_trace.c                           |   6 +-
 net/bluetooth/l2cap_sock.c                         |   2 +
 net/ceph/osd_client.c                              |  20 +-
 net/core/bpf_sk_storage.c                          |   5 +-
 net/core/rtnetlink.c                               |   8 +-
 net/core/sock.c                                    |  21 +-
 net/core/sock_map.c                                |   2 -
 net/dcb/dcbnl.c                                    |   2 +-
 net/ipv4/tcp_metrics.c                             |  70 +-
 net/ipv6/ip6mr.c                                   |   2 +-
 net/sched/cls_fw.c                                 |   1 -
 net/sched/cls_route.c                              |   1 -
 net/sched/cls_u32.c                                |  57 +-
 net/sched/sch_taprio.c                             |  15 +-
 net/unix/af_unix.c                                 |   2 +-
 net/wireless/scan.c                                |   2 +-
 .../tests/shell/test_uprobe_from_different_cu.sh   |   8 +-
 tools/testing/selftests/rseq/rseq.c                |  31 +-
 117 files changed, 3227 insertions(+), 2247 deletions(-)
author maintainer
#1 lore raw

PATCH 1/92 io_uring: gate iowait schedule on having pending requests

From: Jens Axboe <[email protected]>

Commit 7b72d661f1f2f950ab8c12de7e2bc48bdac8ed69 upstream.

A previous commit made all cqring waits marked as iowait, as a way to improve performance for short schedules with pending IO. However, for use cases that have a special reaper thread that does nothing but wait on events on the ring, this causes a cosmetic issue where we know have one core marked as being "busy" with 100% iowait.

While this isn't a grave issue, it is confusing to users. Rather than always mark us as being in iowait, gate setting of current->in_iowait to 1 by whether or not the waiting task has pending requests.

---
 io_uring/io_uring.c |   23 +++++++++++++++++------
 1 file changed, 17 insertions(+), 6 deletions(-)
io_uring/io_uring.c +17 -6
@@ -7802,12 +7802,21 @@ static int io_run_task_work_sig(void)
 	return -EINTR;
 }
 
+static bool current_pending_io(void)
+{
+	struct io_uring_task *tctx = current->io_uring;
+
+	if (!tctx)
+		return false;
+	return percpu_counter_read_positive(&tctx->inflight);
+}
+
 /* when returns >0, the caller should retry */
 static inline int io_cqring_wait_schedule(struct io_ring_ctx *ctx,
 					  struct io_wait_queue *iowq,
 					  ktime_t *timeout)
 {
-	int token, ret;
+	int io_wait, ret;
 
 	/* make sure we run task_work before checking for signals */
 	ret = io_run_task_work_sig();
@@ -7818,15 +7827,17 @@ static inline int io_cqring_wait_schedul
 		return 1;
 
 	/*
-	 * Use io_schedule_prepare/finish, so cpufreq can take into account
-	 * that the task is waiting for IO - turns out to be important for low
-	 * QD IO.
+	 * Mark us as being in io_wait if we have pending requests, so cpufreq
+	 * can take into account that the task is waiting for IO - turns out
+	 * to be important for low QD IO.
 	 */
-	token = io_schedule_prepare();
+	io_wait = current->in_iowait;
+	if (current_pending_io())
+		current->in_iowait = 1;
 	ret = 1;
 	if (!schedule_hrtimeout(timeout, HRTIMER_MODE_ABS))
 		ret = -ETIME;
-	io_schedule_finish(token);
+	current->in_iowait = io_wait;
 	return ret;
 }
 
author maintainer
#2 lore raw

PATCH 2/92 perf: Fix function pointer case

In reply to Greg Kroah-Hartman

From: Peter Zijlstra <[email protected]>

commit 1af6239d1d3e61d33fd2f0ba53d3d1a67cc50574 upstream.

With the advent of CFI it is no longer acceptible to cast function pointers.

The robot complains thusly:

kernel-events-core.c:warning:cast-from-int-(-)(struct-perf_cpu_pmu_context-)-to-remote_function_f-(aka-int-(-)(void-)-)-converts-to-incompatible-function-type

---
 kernel/events/core.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
kernel/events/core.c +6 -2
@@ -1224,6 +1224,11 @@ static int perf_mux_hrtimer_restart(stru
 	return 0;
 }
 
+static int perf_mux_hrtimer_restart_ipi(void *arg)
+{
+	return perf_mux_hrtimer_restart(arg);
+}
+
 void perf_pmu_disable(struct pmu *pmu)
 {
 	int *count = this_cpu_ptr(pmu->pmu_disable_count);
@@ -11137,8 +11142,7 @@ perf_event_mux_interval_ms_store(struct
 		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
 		cpuctx->hrtimer_interval = ns_to_ktime(NSEC_PER_MSEC * timer);
 
-		cpu_function_call(cpu,
-			(remote_function_f)perf_mux_hrtimer_restart, cpuctx);
+		cpu_function_call(cpu, perf_mux_hrtimer_restart_ipi, cpuctx);
 	}
 	cpus_read_unlock();
 	mutex_unlock(&mux_interval_mutex);
author maintainer
#3 lore raw

PATCH 3/92 net/mlx5: Free irqs only on shutdown callback

In reply to Greg Kroah-Hartman

From: Shay Drory <[email protected]>

commit 9c2d08010963a61a171e8cb2852d3ce015b60cb4 upstream.

Whenever a shutdown is invoked, free irqs only and keep mlx5_irq synthetic wrapper intact in order to avoid use-after-free on system shutdown.

for example:
==================================================================

BUG: KASAN: use-after-free in _find_first_bit+0x66/0x80
Read of size 8 at addr ffff88823fc0d318 by task kworker/u192:0/13608

CPU: 25 PID: 13608 Comm: kworker/u192:0 Tainted: G    B   W  O  6.1.21-cloudflare-kasan-2023.3.21 #1
Hardware name: GIGABYTE R162-R2-GEN0/MZ12-HD2-CD, BIOS R14 05/03/2021
Workqueue: mlx5e mlx5e_tx_timeout_work [mlx5_core]
Call Trace:
  <TASK>
  dump_stack_lvl+0x34/0x48
  print_report+0x170/0x473
  ? _find_first_bit+0x66/0x80
  kasan_report+0xad/0x130
  ? _find_first_bit+0x66/0x80
  _find_first_bit+0x66/0x80
  mlx5e_open_channels+0x3c5/0x3a10 [mlx5_core]
  ? console_unlock+0x2fa/0x430
  ? _raw_spin_lock_irqsave+0x8d/0xf0
  ? _raw_spin_unlock_irqrestore+0x42/0x80
  ? preempt_count_add+0x7d/0x150
  ? __wake_up_klogd.part.0+0x7d/0xc0
  ? vprintk_emit+0xfe/0x2c0
  ? mlx5e_trigger_napi_sched+0x40/0x40 [mlx5_core]
  ? dev_attr_show.cold+0x35/0x35
  ? devlink_health_do_dump.part.0+0x174/0x340
  ? devlink_health_report+0x504/0x810
  ? mlx5e_reporter_tx_timeout+0x29d/0x3a0 [mlx5_core]
  ? mlx5e_tx_timeout_work+0x17c/0x230 [mlx5_core]
  ? process_one_work+0x680/0x1050
  mlx5e_safe_switch_params+0x156/0x220 [mlx5_core]
  ? mlx5e_switch_priv_channels+0x310/0x310 [mlx5_core]
  ? mlx5_eq_poll_irq_disabled+0xb6/0x100 [mlx5_core]
  mlx5e_tx_reporter_timeout_recover+0x123/0x240 [mlx5_core]
  ? __mutex_unlock_slowpath.constprop.0+0x2b0/0x2b0
  devlink_health_reporter_recover+0xa6/0x1f0
  devlink_health_report+0x2f7/0x810
  ? vsnprintf+0x854/0x15e0
  mlx5e_reporter_tx_timeout+0x29d/0x3a0 [mlx5_core]
  ? mlx5e_reporter_tx_err_cqe+0x1a0/0x1a0 [mlx5_core]
  ? mlx5e_tx_reporter_timeout_dump+0x50/0x50 [mlx5_core]
  ? mlx5e_tx_reporter_dump_sq+0x260/0x260 [mlx5_core]
  ? newidle_balance+0x9b7/0xe30
  ? psi_group_change+0x6a7/0xb80
  ? mutex_lock+0x96/0xf0
  ? __mutex_lock_slowpath+0x10/0x10
  mlx5e_tx_timeout_work+0x17c/0x230 [mlx5_core]
  process_one_work+0x680/0x1050
  worker_thread+0x5a0/0xeb0
  ? process_one_work+0x1050/0x1050
  kthread+0x2a2/0x340
  ? kthread_complete_and_exit+0x20/0x20
  ret_from_fork+0x22/0x30
  </TASK>

Freed by task 1:
  kasan_save_stack+0x23/0x50
  kasan_set_track+0x21/0x30
  kasan_save_free_info+0x2a/0x40
  ____kasan_slab_free+0x169/0x1d0
  slab_free_freelist_hook+0xd2/0x190
  __kmem_cache_free+0x1a1/0x2f0
  irq_pool_free+0x138/0x200 [mlx5_core]
  mlx5_irq_table_destroy+0xf6/0x170 [mlx5_core]
  mlx5_core_eq_free_irqs+0x74/0xf0 [mlx5_core]
  shutdown+0x194/0x1aa [mlx5_core]
  pci_device_shutdown+0x75/0x120
  device_shutdown+0x35c/0x620
  kernel_restart+0x60/0xa0
  __do_sys_reboot+0x1cb/0x2c0
  do_syscall_64+0x3b/0x90
  entry_SYSCALL_64_after_hwframe+0x4b/0xb5

The buggy address belongs to the object at ffff88823fc0d300
  which belongs to the cache kmalloc-192 of size 192

The buggy address is located 24 bytes inside of
192-byte region [ffff88823fc0d300, ffff88823fc0d3c0)

The buggy address belongs to the physical page:

page:0000000010139587 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x23fc0c
head:0000000010139587 order:1 compound_mapcount:0 compound_pincount:0
flags: 0x2ffff800010200(slab|head|node=0|zone=2|lastcpupid=0x1ffff)

raw: 002ffff800010200 0000000000000000 dead000000000122 ffff88810004ca00
raw: 0000000000000000 0000000000200020 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
  ffff88823fc0d200: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
  ffff88823fc0d280: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
 >ffff88823fc0d300: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                             ^
  ffff88823fc0d380: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
  ffff88823fc0d400: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
==================================================================
general protection fault, probably for non-canonical address
0xdffffc005c40d7ac: 0000 [#1] PREEMPT SMP KASAN NOPTI
KASAN: probably user-memory-access in range [0x00000002e206bd60-0x00000002e206bd67]
CPU: 25 PID: 13608 Comm: kworker/u192:0 Tainted: G    B   W  O  6.1.21-cloudflare-kasan-2023.3.21 #1
Hardware name: GIGABYTE R162-R2-GEN0/MZ12-HD2-CD, BIOS R14 05/03/2021
Workqueue: mlx5e mlx5e_tx_timeout_work [mlx5_core]
RIP: 0010:__alloc_pages+0x141/0x5c0
Call Trace:
  <TASK>
  ? sysvec_apic_timer_interrupt+0xa0/0xc0
  ? asm_sysvec_apic_timer_interrupt+0x16/0x20
  ? __alloc_pages_slowpath.constprop.0+0x1ec0/0x1ec0
  ? _raw_spin_unlock_irqrestore+0x3d/0x80
  __kmalloc_large_node+0x80/0x120
  ? kvmalloc_node+0x4e/0x170
  __kmalloc_node+0xd4/0x150
  kvmalloc_node+0x4e/0x170
  mlx5e_open_channels+0x631/0x3a10 [mlx5_core]
  ? console_unlock+0x2fa/0x430
  ? _raw_spin_lock_irqsave+0x8d/0xf0
  ? _raw_spin_unlock_irqrestore+0x42/0x80
  ? preempt_count_add+0x7d/0x150
  ? __wake_up_klogd.part.0+0x7d/0xc0
  ? vprintk_emit+0xfe/0x2c0
  ? mlx5e_trigger_napi_sched+0x40/0x40 [mlx5_core]
  ? dev_attr_show.cold+0x35/0x35
  ? devlink_health_do_dump.part.0+0x174/0x340
  ? devlink_health_report+0x504/0x810
  ? mlx5e_reporter_tx_timeout+0x29d/0x3a0 [mlx5_core]
  ? mlx5e_tx_timeout_work+0x17c/0x230 [mlx5_core]
  ? process_one_work+0x680/0x1050
  mlx5e_safe_switch_params+0x156/0x220 [mlx5_core]
  ? mlx5e_switch_priv_channels+0x310/0x310 [mlx5_core]
  ? mlx5_eq_poll_irq_disabled+0xb6/0x100 [mlx5_core]
  mlx5e_tx_reporter_timeout_recover+0x123/0x240 [mlx5_core]
  ? __mutex_unlock_slowpath.constprop.0+0x2b0/0x2b0
  devlink_health_reporter_recover+0xa6/0x1f0
  devlink_health_report+0x2f7/0x810
  ? vsnprintf+0x854/0x15e0
  mlx5e_reporter_tx_timeout+0x29d/0x3a0 [mlx5_core]
  ? mlx5e_reporter_tx_err_cqe+0x1a0/0x1a0 [mlx5_core]
  ? mlx5e_tx_reporter_timeout_dump+0x50/0x50 [mlx5_core]
  ? mlx5e_tx_reporter_dump_sq+0x260/0x260 [mlx5_core]
  ? newidle_balance+0x9b7/0xe30
  ? psi_group_change+0x6a7/0xb80
  ? mutex_lock+0x96/0xf0
  ? __mutex_lock_slowpath+0x10/0x10
  mlx5e_tx_timeout_work+0x17c/0x230 [mlx5_core]
  process_one_work+0x680/0x1050
  worker_thread+0x5a0/0xeb0
  ? process_one_work+0x1050/0x1050
  kthread+0x2a2/0x340
  ? kthread_complete_and_exit+0x20/0x20
  ret_from_fork+0x22/0x30
  </TASK>
---[ end trace 0000000000000000  ]---
RIP: 0010:__alloc_pages+0x141/0x5c0
Code: e0 39 a3 96 89 e9 b8 22 01 32 01 83 e1 0f 48 89 fa 01 c9 48 c1 ea
03 d3 f8 83 e0 03 89 44 24 6c 48 b8 00 00 00 00 00 fc ff df <80> 3c 02
00 0f 85 fc 03 00 00 89 e8 4a 8b 14 f5 e0 39 a3 96 4c 89
RSP: 0018:ffff888251f0f438 EFLAGS: 00010202
RAX: dffffc0000000000 RBX: 1ffff1104a3e1e8b RCX: 0000000000000000
RDX: 000000005c40d7ac RSI: 0000000000000003 RDI: 00000002e206bd60
RBP: 0000000000052dc0 R08: ffff8882b0044218 R09: ffff8882b0045e8a
R10: fffffbfff300fefc R11: ffff888167af4000 R12: 0000000000000003
R13: 0000000000000000 R14: 00000000696c7070 R15: ffff8882373f4380
FS:  0000000000000000(0000) GS:ffff88bf2be80000(0000)
knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00005641d031eee8 CR3: 0000002e7ca14000 CR4: 0000000000350ee0
Kernel panic - not syncing: Fatal exception
Kernel Offset: 0x11000000 from 0xffffffff81000000 (relocation range:
0xffffffff80000000-0xffffffffbfffffff)
---[ end Kernel panic - not syncing: Fatal exception  ]---]

[hardik: Refer to the irqn member of the mlx5_irq struct, instead of
the msi_map, since we don't have upstream v6.4 commit 235a25fe28de ("net/mlx5: Modify struct mlx5_irq to use struct msi_map")].
[hardik: Refer to the pf_pool member of the mlx5_irq_table struct,
instead of pcif_pool, since we don't have upstream v6.4 commit 8bebfd767909 ("net/mlx5: Improve naming of pci function vectors")].

---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c       |    2 -
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h |    1 
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  |   29 +++++++++++++++++++++
 3 files changed, 31 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/eq.c +1 -1
@@ -981,7 +981,7 @@ void mlx5_core_eq_free_irqs(struct mlx5_
 	mutex_lock(&table->lock); /* sync with create/destroy_async_eq */
 	if (!mlx5_core_is_sf(dev))
 		clear_rmap(dev);
-	mlx5_irq_table_destroy(dev);
+	mlx5_irq_table_free_irqs(dev);
 	mutex_unlock(&table->lock);
 }
 
drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h +1 -0
@@ -16,6 +16,7 @@ int mlx5_irq_table_init(struct mlx5_core
 void mlx5_irq_table_cleanup(struct mlx5_core_dev *dev);
 int mlx5_irq_table_create(struct mlx5_core_dev *dev);
 void mlx5_irq_table_destroy(struct mlx5_core_dev *dev);
+void mlx5_irq_table_free_irqs(struct mlx5_core_dev *dev);
 int mlx5_irq_table_get_num_comp(struct mlx5_irq_table *table);
 int mlx5_irq_table_get_sfs_vec(struct mlx5_irq_table *table);
 struct mlx5_irq_table *mlx5_irq_table_get(struct mlx5_core_dev *dev);
drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c +29 -0
@@ -550,6 +550,24 @@ static void irq_pools_destroy(struct mlx
 	irq_pool_free(table->pf_pool);
 }
 
+static void mlx5_irq_pool_free_irqs(struct mlx5_irq_pool *pool)
+{
+	struct mlx5_irq *irq;
+	unsigned long index;
+
+	xa_for_each(&pool->irqs, index, irq)
+		free_irq(irq->irqn, &irq->nh);
+}
+
+static void mlx5_irq_pools_free_irqs(struct mlx5_irq_table *table)
+{
+	if (table->sf_ctrl_pool) {
+		mlx5_irq_pool_free_irqs(table->sf_comp_pool);
+		mlx5_irq_pool_free_irqs(table->sf_ctrl_pool);
+	}
+	mlx5_irq_pool_free_irqs(table->pf_pool);
+}
+
 /* irq_table API */
 
 int mlx5_irq_table_init(struct mlx5_core_dev *dev)
@@ -630,6 +648,17 @@ void mlx5_irq_table_destroy(struct mlx5_
 	pci_free_irq_vectors(dev->pdev);
 }
 
+void mlx5_irq_table_free_irqs(struct mlx5_core_dev *dev)
+{
+	struct mlx5_irq_table *table = dev->priv.irq_table;
+
+	if (mlx5_core_is_sf(dev))
+		return;
+
+	mlx5_irq_pools_free_irqs(table);
+	pci_free_irq_vectors(dev->pdev);
+}
+
 int mlx5_irq_table_get_sfs_vec(struct mlx5_irq_table *table)
 {
 	if (table->sf_comp_pool)
author maintainer
#4 lore raw

PATCH 4/92 arm64: errata: Add workaround for TSB flush failures

In reply to Greg Kroah-Hartman

From: Suzuki K Poulose <[email protected]>

commit fa82d0b4b833790ac4572377fb777dcea24a9d69 upstream

Arm Neoverse-N2 (#2067961) and Cortex-A710 (#2054223) suffers from errata, where a TSB (trace synchronization barrier) fails to flush the trace data completely, when executed from a trace prohibited region. In Linux we always execute it after we have moved the PE to trace prohibited region. So, we can apply the workaround every time a TSB is executed.

The work around is to issue two TSB consecutively.

NOTE: This errata is defined as LOCAL_CPU_ERRATUM, implying that a late CPU could be blocked from booting if it is the first CPU that requires the workaround. This is because we do not allow setting a cpu_hwcaps after the SMP boot. The other alternative is to use "this_cpu_has_cap()" instead of the faster system wide check, which may be a bit of an overhead, given we may have to do this in nvhe KVM host before a guest entry.

---
 Documentation/arm64/silicon-errata.rst |    4 ++++
 arch/arm64/Kconfig                     |   33 +++++++++++++++++++++++++++++++++
 arch/arm64/include/asm/barrier.h       |   16 +++++++++++++++-
 arch/arm64/kernel/cpu_errata.c         |   19 +++++++++++++++++++
 arch/arm64/tools/cpucaps               |    1 +
 5 files changed, 72 insertions(+), 1 deletion(-)
Documentation/arm64/silicon-errata.rst +4 -0
@@ -104,6 +104,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A710     | #2119858        | ARM64_ERRATUM_2119858       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Cortex-A710     | #2054223        | ARM64_ERRATUM_2054223       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-N1     | #1188873,1418040| ARM64_ERRATUM_1418040       |
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-N1     | #1349291        | N/A                         |
@@ -112,6 +114,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-N2     | #2139208        | ARM64_ERRATUM_2139208       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Neoverse-N2     | #2067961        | ARM64_ERRATUM_2067961       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | MMU-500         | #841119,826419  | N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
 +----------------+-----------------+-----------------+-----------------------------+
arch/arm64/Kconfig +33 -0
@@ -774,6 +774,39 @@ config ARM64_ERRATUM_2139208
 
 	  If unsure, say Y.
 
+config ARM64_WORKAROUND_TSB_FLUSH_FAILURE
+	bool
+
+config ARM64_ERRATUM_2054223
+	bool "Cortex-A710: 2054223: workaround TSB instruction failing to flush trace"
+	default y
+	select ARM64_WORKAROUND_TSB_FLUSH_FAILURE
+	help
+	  Enable workaround for ARM Cortex-A710 erratum 2054223
+
+	  Affected cores may fail to flush the trace data on a TSB instruction, when
+	  the PE is in trace prohibited state. This will cause losing a few bytes
+	  of the trace cached.
+
+	  Workaround is to issue two TSB consecutively on affected cores.
+
+	  If unsure, say Y.
+
+config ARM64_ERRATUM_2067961
+	bool "Neoverse-N2: 2067961: workaround TSB instruction failing to flush trace"
+	default y
+	select ARM64_WORKAROUND_TSB_FLUSH_FAILURE
+	help
+	  Enable workaround for ARM Neoverse-N2 erratum 2067961
+
+	  Affected cores may fail to flush the trace data on a TSB instruction, when
+	  the PE is in trace prohibited state. This will cause losing a few bytes
+	  of the trace cached.
+
+	  Workaround is to issue two TSB consecutively on affected cores.
+
+	  If unsure, say Y.
+
 config CAVIUM_ERRATUM_22375
 	bool "Cavium erratum 22375, 24313"
 	default y
arch/arm64/include/asm/barrier.h +15 -1
@@ -23,7 +23,7 @@
 #define dsb(opt)	asm volatile("dsb " #opt : : : "memory")
 
 #define psb_csync()	asm volatile("hint #17" : : : "memory")
-#define tsb_csync()	asm volatile("hint #18" : : : "memory")
+#define __tsb_csync()	asm volatile("hint #18" : : : "memory")
 #define csdb()		asm volatile("hint #20" : : : "memory")
 
 #ifdef CONFIG_ARM64_PSEUDO_NMI
@@ -46,6 +46,20 @@
 #define dma_rmb()	dmb(oshld)
 #define dma_wmb()	dmb(oshst)
 
+
+#define tsb_csync()								\
+	do {									\
+		/*								\
+		 * CPUs affected by Arm Erratum 2054223 or 2067961 needs	\
+		 * another TSB to ensure the trace is flushed. The barriers	\
+		 * don't have to be strictly back to back, as long as the	\
+		 * CPU is in trace prohibited state.				\
+		 */								\
+		if (cpus_have_final_cap(ARM64_WORKAROUND_TSB_FLUSH_FAILURE))	\
+			__tsb_csync();						\
+		__tsb_csync();							\
+	} while (0)
+
 /*
  * Generate a mask for array_index__nospec() that is ~0UL when 0 <= idx < sz
  * and 0 otherwise.
arch/arm64/kernel/cpu_errata.c +19 -0
@@ -375,6 +375,18 @@ static const struct midr_range trbe_over
 };
 #endif	/* CONFIG_ARM64_WORKAROUND_TRBE_OVERWRITE_FILL_MODE */
 
+#ifdef CONFIG_ARM64_WORKAROUND_TSB_FLUSH_FAILURE
+static const struct midr_range tsb_flush_fail_cpus[] = {
+#ifdef CONFIG_ARM64_ERRATUM_2067961
+	MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N2),
+#endif
+#ifdef CONFIG_ARM64_ERRATUM_2054223
+	MIDR_ALL_VERSIONS(MIDR_CORTEX_A710),
+#endif
+	{},
+};
+#endif	/* CONFIG_ARM64_WORKAROUND_TSB_FLUSH_FAILURE */
+
 const struct arm64_cpu_capabilities arm64_errata[] = {
 #ifdef CONFIG_ARM64_WORKAROUND_CLEAN_CACHE
 	{
@@ -607,6 +619,13 @@ const struct arm64_cpu_capabilities arm6
 		CAP_MIDR_RANGE_LIST(trbe_overwrite_fill_mode_cpus),
 	},
 #endif
+#ifdef CONFIG_ARM64_WORKAROUND_TSB_FLUSH_FAILURE
+	{
+		.desc = "ARM erratum 2067961 or 2054223",
+		.capability = ARM64_WORKAROUND_TSB_FLUSH_FAILURE,
+		ERRATA_MIDR_RANGE_LIST(tsb_flush_fail_cpus),
+	},
+#endif
 	{
 	}
 };
arch/arm64/tools/cpucaps +1 -0
@@ -57,6 +57,7 @@ WORKAROUND_1542419
 WORKAROUND_1742098
 WORKAROUND_2457168
 WORKAROUND_TRBE_OVERWRITE_FILL_MODE
+WORKAROUND_TSB_FLUSH_FAILURE
 WORKAROUND_CAVIUM_23154
 WORKAROUND_CAVIUM_27456
 WORKAROUND_CAVIUM_30115
author maintainer
#5 lore raw

PATCH 5/92 arm64: errata: Add detection for TRBE write to out-of-range

In reply to Greg Kroah-Hartman

From: Suzuki K Poulose <[email protected]>

commit 8d81b2a38ddfc4b03662d2359765648c8b4cc73c upstream

Arm Neoverse-N2 and Cortex-A710 cores are affected by an erratum where the trbe, under some circumstances, might write upto 64bytes to an address after the Limit as programmed by the TRBLIMITR_EL1.LIMIT. This might -
- Corrupt a page in the ring buffer, which may corrupt trace from a previous session, consumed by userspace.
- Hit the guard page at the end of the vmalloc area and raise a fault.

To keep the handling simpler, we always leave the last page from the range, which TRBE is allowed to write. This can be achieved by ensuring that we always have more than a PAGE worth space in the range, while calculating the LIMIT for TRBE. And then the LIMIT pointer can be adjusted to leave the PAGE (TRBLIMITR.LIMIT -= PAGE_SIZE), out of the TRBE range while enabling it. This makes sure that the TRBE will only write to an area within its allowed limit (i.e, [head-head+size]) and we do not have to handle address faults within the driver.

---
 Documentation/arm64/silicon-errata.rst |    4 +++
 arch/arm64/Kconfig                     |   41 +++++++++++++++++++++++++++++++++
 arch/arm64/kernel/cpu_errata.c         |   20 ++++++++++++++++
 arch/arm64/tools/cpucaps               |    1 
 4 files changed, 66 insertions(+)
Documentation/arm64/silicon-errata.rst +4 -0
@@ -106,6 +106,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A710     | #2054223        | ARM64_ERRATUM_2054223       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Cortex-A710     | #2224489        | ARM64_ERRATUM_2224489       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-N1     | #1188873,1418040| ARM64_ERRATUM_1418040       |
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-N1     | #1349291        | N/A                         |
@@ -116,6 +118,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-N2     | #2067961        | ARM64_ERRATUM_2067961       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Neoverse-N2     | #2253138        | ARM64_ERRATUM_2253138       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | MMU-500         | #841119,826419  | N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
 +----------------+-----------------+-----------------+-----------------------------+
arch/arm64/Kconfig +41 -0

47 lines not on this page · raw message

arch/arm64/kernel/cpu_errata.c +20 -0
@@ -387,6 +387,18 @@ static const struct midr_range tsb_flush
 };
 #endif	/* CONFIG_ARM64_WORKAROUND_TSB_FLUSH_FAILURE */
 
+#ifdef CONFIG_ARM64_WORKAROUND_TRBE_WRITE_OUT_OF_RANGE
+static struct midr_range trbe_write_out_of_range_cpus[] = {
+#ifdef CONFIG_ARM64_ERRATUM_2253138
+	MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N2),
+#endif
+#ifdef CONFIG_ARM64_ERRATUM_2224489
+	MIDR_ALL_VERSIONS(MIDR_CORTEX_A710),
+#endif
+	{},
+};
+#endif /* CONFIG_ARM64_WORKAROUND_TRBE_WRITE_OUT_OF_RANGE */
+
 const struct arm64_cpu_capabilities arm64_errata[] = {
 #ifdef CONFIG_ARM64_WORKAROUND_CLEAN_CACHE
 	{
@@ -626,6 +638,14 @@ const struct arm64_cpu_capabilities arm6
 		ERRATA_MIDR_RANGE_LIST(tsb_flush_fail_cpus),
 	},
 #endif
+#ifdef CONFIG_ARM64_WORKAROUND_TRBE_WRITE_OUT_OF_RANGE
+	{
+		.desc = "ARM erratum 2253138 or 2224489",
+		.capability = ARM64_WORKAROUND_TRBE_WRITE_OUT_OF_RANGE,
+		.type = ARM64_CPUCAP_WEAK_LOCAL_CPU_FEATURE,
+		CAP_MIDR_RANGE_LIST(trbe_write_out_of_range_cpus),
+	},
+#endif
 	{
 	}
 };
arch/arm64/tools/cpucaps +1 -0
@@ -58,6 +58,7 @@ WORKAROUND_1742098
 WORKAROUND_2457168
 WORKAROUND_TRBE_OVERWRITE_FILL_MODE
 WORKAROUND_TSB_FLUSH_FAILURE
+WORKAROUND_TRBE_WRITE_OUT_OF_RANGE
 WORKAROUND_CAVIUM_23154
 WORKAROUND_CAVIUM_27456
 WORKAROUND_CAVIUM_30115
author maintainer
#6 lore raw

PATCH 6/92 iommu/arm-smmu-v3: Work around MMU-600 erratum 1076982

In reply to Greg Kroah-Hartman

From: Robin Murphy <[email protected]>

commit f322e8af35c7f23a8c08b595c38d6c855b2d836f upstream

MMU-600 versions prior to r1p0 fail to correctly generate a WFE wakeup event when the command queue transitions fom full to non-full. We can easily work around this by simply hiding the SEV capability such that we fall back to polling for space in the queue - since MMU-600 implements MSIs we wouldn't expect to need SEV for sync completion either, so this should have little to no impact.

---
 Documentation/arm64/silicon-errata.rst      |    2 +
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c |   29 ++++++++++++++++++++++++++++
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h |    6 +++++
 3 files changed, 37 insertions(+)
Documentation/arm64/silicon-errata.rst +2 -0
@@ -122,6 +122,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | MMU-500         | #841119,826419  | N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | MMU-600         | #1076982        | N/A                         |
++----------------+-----------------+-----------------+-----------------------------+
 +----------------+-----------------+-----------------+-----------------------------+
 | Broadcom       | Brahma-B53      | N/A             | ARM64_ERRATUM_845719        |
 +----------------+-----------------+-----------------+-----------------------------+
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +29 -0
@@ -3459,6 +3459,33 @@ static int arm_smmu_device_reset(struct
 	return 0;
 }
 
+#define IIDR_IMPLEMENTER_ARM		0x43b
+#define IIDR_PRODUCTID_ARM_MMU_600	0x483
+
+static void arm_smmu_device_iidr_probe(struct arm_smmu_device *smmu)
+{
+	u32 reg;
+	unsigned int implementer, productid, variant, revision;
+
+	reg = readl_relaxed(smmu->base + ARM_SMMU_IIDR);
+	implementer = FIELD_GET(IIDR_IMPLEMENTER, reg);
+	productid = FIELD_GET(IIDR_PRODUCTID, reg);
+	variant = FIELD_GET(IIDR_VARIANT, reg);
+	revision = FIELD_GET(IIDR_REVISION, reg);
+
+	switch (implementer) {
+	case IIDR_IMPLEMENTER_ARM:
+		switch (productid) {
+		case IIDR_PRODUCTID_ARM_MMU_600:
+			/* Arm erratum 1076982 */
+			if (variant == 0 && revision <= 2)
+				smmu->features &= ~ARM_SMMU_FEAT_SEV;
+			break;
+		}
+		break;
+	}
+}
+
 static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 {
 	u32 reg;
@@ -3664,6 +3691,8 @@ static int arm_smmu_device_hw_probe(stru
 
 	smmu->ias = max(smmu->ias, smmu->oas);
 
+	arm_smmu_device_iidr_probe(smmu);
+
 	if (arm_smmu_sva_supported(smmu))
 		smmu->features |= ARM_SMMU_FEAT_SVA;
 
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h +6 -0
@@ -69,6 +69,12 @@
 #define IDR5_VAX			GENMASK(11, 10)
 #define IDR5_VAX_52_BIT			1
 
+#define ARM_SMMU_IIDR			0x18
+#define IIDR_PRODUCTID			GENMASK(31, 20)
+#define IIDR_VARIANT			GENMASK(19, 16)
+#define IIDR_REVISION			GENMASK(15, 12)
+#define IIDR_IMPLEMENTER		GENMASK(11, 0)
+
 #define ARM_SMMU_CR0			0x20
 #define CR0_ATSCHK			(1 << 4)
 #define CR0_CMDQEN			(1 << 3)
author maintainer
#7 lore raw

PATCH 7/92 iommu/arm-smmu-v3: Document MMU-700 erratum 2812531

In reply to Greg Kroah-Hartman

From: Robin Murphy <[email protected]>

commit 309a15cb16bb075da1c99d46fb457db6a1a2669e upstream

To work around MMU-700 erratum 2812531 we need to ensure that certain sequences of commands cannot be issued without an intervening sync. In practice this falls out of our current command-batching machinery anyway - each batch only contains a single type of invalidation command, and ends with a sync. The only exception is when a batch is sufficiently large to need issuing across multiple command queue slots, wherein the earlier slots will not contain a sync and thus may in theory interleave with another batch being issued in parallel to create an affected sequence across the slot boundary.

Since MMU-700 supports range invalidate commands and thus we will prefer to use them (which also happens to avoid conditions for other errata), I'm not entirely sure it's even possible for a single high-level invalidate call to generate a batch of more than 63 commands, but for the sake of robustness and documentation, wire up an option to enforce that a sync is always inserted for every slot issued.

The other aspect is that the relative order of DVM commands cannot be controlled, so DVM cannot be used. Again that is already the status quo, but since we have at least defined ARM_SMMU_FEAT_BTM, we can explicitly disable it for documentation purposes even if it's not wired up anywhere yet.

---
 Documentation/arm64/silicon-errata.rst      |    2 ++
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c |   12 ++++++++++++
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h |    1 +
 3 files changed, 15 insertions(+)
Documentation/arm64/silicon-errata.rst +2 -0
@@ -124,6 +124,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | MMU-600         | #1076982        | N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | MMU-700         | #2812531        | N/A                         |
++----------------+-----------------+-----------------+-----------------------------+
 +----------------+-----------------+-----------------+-----------------------------+
 | Broadcom       | Brahma-B53      | N/A             | ARM64_ERRATUM_845719        |
 +----------------+-----------------+-----------------+-----------------------------+
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +12 -0
@@ -897,6 +897,12 @@ static void arm_smmu_cmdq_batch_add(stru
 				    struct arm_smmu_cmdq_batch *cmds,
 				    struct arm_smmu_cmdq_ent *cmd)
 {
+	if (cmds->num == CMDQ_BATCH_ENTRIES - 1 &&
+	    (smmu->options & ARM_SMMU_OPT_CMDQ_FORCE_SYNC)) {
+		arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, true);
+		cmds->num = 0;
+	}
+
 	if (cmds->num == CMDQ_BATCH_ENTRIES) {
 		arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, false);
 		cmds->num = 0;
@@ -3461,6 +3467,7 @@ static int arm_smmu_device_reset(struct
 
 #define IIDR_IMPLEMENTER_ARM		0x43b
 #define IIDR_PRODUCTID_ARM_MMU_600	0x483
+#define IIDR_PRODUCTID_ARM_MMU_700	0x487
 
 static void arm_smmu_device_iidr_probe(struct arm_smmu_device *smmu)
 {
@@ -3481,6 +3488,11 @@ static void arm_smmu_device_iidr_probe(s
 			if (variant == 0 && revision <= 2)
 				smmu->features &= ~ARM_SMMU_FEAT_SEV;
 			break;
+		case IIDR_PRODUCTID_ARM_MMU_700:
+			/* Arm erratum 2812531 */
+			smmu->features &= ~ARM_SMMU_FEAT_BTM;
+			smmu->options |= ARM_SMMU_OPT_CMDQ_FORCE_SYNC;
+			break;
 		}
 		break;
 	}
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h +1 -0
@@ -651,6 +651,7 @@ struct arm_smmu_device {
 #define ARM_SMMU_OPT_SKIP_PREFETCH	(1 << 0)
 #define ARM_SMMU_OPT_PAGE0_REGS_ONLY	(1 << 1)
 #define ARM_SMMU_OPT_MSIPOLL		(1 << 2)
+#define ARM_SMMU_OPT_CMDQ_FORCE_SYNC	(1 << 3)
 	u32				options;
 
 	struct arm_smmu_cmdq		cmdq;
author maintainer
#8 lore raw

PATCH 8/92 iommu/arm-smmu-v3: Add explicit feature for nesting

In reply to Greg Kroah-Hartman

From: Robin Murphy <[email protected]>

commit 1d9777b9f3d55b4b6faf186ba4f1d6fb560c0523 upstream

In certain cases we may want to refuse to allow nested translation even when both stages are implemented, so let's add an explicit feature for nesting support which we can control in its own right. For now this merely serves as documentation, but it means a nice convenient check will be ready and waiting for the future nesting code.

---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c |    4 ++++
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h |    1 +
 2 files changed, 5 insertions(+)
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +4 -0
@@ -3703,6 +3703,10 @@ static int arm_smmu_device_hw_probe(stru
 
 	smmu->ias = max(smmu->ias, smmu->oas);
 
+	if ((smmu->features & ARM_SMMU_FEAT_TRANS_S1) &&
+	    (smmu->features & ARM_SMMU_FEAT_TRANS_S2))
+		smmu->features |= ARM_SMMU_FEAT_NESTING;
+
 	arm_smmu_device_iidr_probe(smmu);
 
 	if (arm_smmu_sva_supported(smmu))
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h +1 -0
@@ -646,6 +646,7 @@ struct arm_smmu_device {
 #define ARM_SMMU_FEAT_BTM		(1 << 16)
 #define ARM_SMMU_FEAT_SVA		(1 << 17)
 #define ARM_SMMU_FEAT_E2H		(1 << 18)
+#define ARM_SMMU_FEAT_NESTING		(1 << 19)
 	u32				features;
 
 #define ARM_SMMU_OPT_SKIP_PREFETCH	(1 << 0)
author maintainer
#9 lore raw

PATCH 9/92 iommu/arm-smmu-v3: Document nesting-related errata

In reply to Greg Kroah-Hartman

From: Robin Murphy <[email protected]>

commit 0bfbfc526c70606bf0fad302e4821087cbecfaf4 upstream

Both MMU-600 and MMU-700 have similar errata around TLB invalidation while both stages of translation are active, which will need some consideration once nesting support is implemented. For now, though, it's very easy to make our implicit lack of nesting support explicit for those cases, so they're less likely to be missed in future.

---
 Documentation/arm64/silicon-errata.rst      |    4 ++--
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c |    5 +++++
 2 files changed, 7 insertions(+), 2 deletions(-)
Documentation/arm64/silicon-errata.rst +2 -2
@@ -122,9 +122,9 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | MMU-500         | #841119,826419  | N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
-| ARM            | MMU-600         | #1076982        | N/A                         |
+| ARM            | MMU-600         | #1076982,1209401| N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
-| ARM            | MMU-700         | #2812531        | N/A                         |
+| ARM            | MMU-700         | #2268618,2812531| N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
 +----------------+-----------------+-----------------+-----------------------------+
 | Broadcom       | Brahma-B53      | N/A             | ARM64_ERRATUM_845719        |
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +5 -0
@@ -3487,11 +3487,16 @@ static void arm_smmu_device_iidr_probe(s
 			/* Arm erratum 1076982 */
 			if (variant == 0 && revision <= 2)
 				smmu->features &= ~ARM_SMMU_FEAT_SEV;
+			/* Arm erratum 1209401 */
+			if (variant < 2)
+				smmu->features &= ~ARM_SMMU_FEAT_NESTING;
 			break;
 		case IIDR_PRODUCTID_ARM_MMU_700:
 			/* Arm erratum 2812531 */
 			smmu->features &= ~ARM_SMMU_FEAT_BTM;
 			smmu->options |= ARM_SMMU_OPT_CMDQ_FORCE_SYNC;
+			/* Arm errata 2268618, 2812531 */
+			smmu->features &= ~ARM_SMMU_FEAT_NESTING;
 			break;
 		}
 		break;
author maintainer
#10 lore raw

PATCH 10/92 arm64: dts: imx8mn-var-som: add missing pull-up for onboard PHY reset pinmux

In reply to Greg Kroah-Hartman

From: Hugo Villeneuve <[email protected]>

[ Upstream commit 253be5b53c2792fb4384f8005b05421e6f040ee3 ]

For SOMs with an onboard PHY, the RESET_N pull-up resistor is currently deactivated in the pinmux configuration. When the pinmux code selects the GPIO function for this pin, with a default direction of input, this prevents the RESET_N pin from being taken to the proper 3.3V level (deasserted), and this results in the PHY being not detected since it is held in reset.

Taken from RESET_N pin description in ADIN13000 datasheet:
This pin requires a 1K pull-up resistor to AVDD_3P3.

Activate the pull-up resistor to fix the issue.

Fixes ade0176dd8a0 ("arm64: dts: imx8mn-var-som: Add Variscite VAR-SOM-MX8MN System on Module")
---
 arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi +1 -1
index d053ef302fb82..faafefe562e4b 100644
@@ -351,7 +351,7 @@
 			MX8MN_IOMUXC_ENET_RXC_ENET1_RGMII_RXC		0x91
 			MX8MN_IOMUXC_ENET_RX_CTL_ENET1_RGMII_RX_CTL	0x91
 			MX8MN_IOMUXC_ENET_TX_CTL_ENET1_RGMII_TX_CTL	0x1f
-			MX8MN_IOMUXC_GPIO1_IO09_GPIO1_IO9		0x19
+			MX8MN_IOMUXC_GPIO1_IO09_GPIO1_IO9		0x159
 		>;
 	};
 
Signature · 1 line
2.40.1
author maintainer
#11 lore raw

PATCH 11/92 firmware: arm_scmi: Fix chan_free cleanup on SMC

In reply to Greg Kroah-Hartman

From: Cristian Marussi <[email protected]>

[ Upstream commit d1ff11d7ad8704f8d615f6446041c221b2d2ec4d ]

SCMI transport based on SMC can optionally use an additional IRQ to signal message completion. The associated interrupt handler is currently allocated using devres but on shutdown the core SCMI stack will call .chan_free() well before any managed cleanup is invoked by devres. As a consequence, the arrival of a late reply to an in-flight pending transaction could still trigger the interrupt handler well after the SCMI core has cleaned up the channels, with unpleasant results.

Inhibit further message processing on the IRQ path by explicitly freeing the IRQ inside .chan_free() callback itself.

Fixes dd820ee21d5e ("firmware: arm_scmi: Augment SMC/HVC to allow optional interrupt")
---
 drivers/firmware/arm_scmi/smc.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)
drivers/firmware/arm_scmi/smc.c +11 -6
index 4effecc3bb463..f529004f1922e 100644
@@ -21,6 +21,7 @@
 /**
  * struct scmi_smc - Structure representing a SCMI smc transport
  *
+ * @irq: An optional IRQ for completion
  * @cinfo: SCMI channel info
  * @shmem: Transmit/Receive shared memory area
  * @shmem_lock: Lock to protect access to Tx/Rx shared memory area
@@ -30,6 +31,7 @@
  */
 
 struct scmi_smc {
+	int irq;
 	struct scmi_chan_info *cinfo;
 	struct scmi_shared_mem __iomem *shmem;
 	struct mutex shmem_lock;
@@ -66,7 +68,7 @@ static int smc_chan_setup(struct scmi_chan_info *cinfo, struct device *dev,
 	struct resource res;
 	struct device_node *np;
 	u32 func_id;
-	int ret, irq;
+	int ret;
 
 	if (!tx)
 		return -ENODEV;
@@ -102,11 +104,10 @@ static int smc_chan_setup(struct scmi_chan_info *cinfo, struct device *dev,
 	 * completion of a message is signaled by an interrupt rather than by
 	 * the return of the SMC call.
 	 */
-	irq = of_irq_get_byname(cdev->of_node, "a2p");
-	if (irq > 0) {
-		ret = devm_request_irq(dev, irq, smc_msg_done_isr,
-				       IRQF_NO_SUSPEND,
-				       dev_name(dev), scmi_info);
+	scmi_info->irq = of_irq_get_byname(cdev->of_node, "a2p");
+	if (scmi_info->irq > 0) {
+		ret = request_irq(scmi_info->irq, smc_msg_done_isr,
+				  IRQF_NO_SUSPEND, dev_name(dev), scmi_info);
 		if (ret) {
 			dev_err(dev, "failed to setup SCMI smc irq\n");
 			return ret;
@@ -128,6 +129,10 @@ static int smc_chan_free(int id, void *p, void *data)
 	struct scmi_chan_info *cinfo = p;
 	struct scmi_smc *scmi_info = cinfo->transport_info;
 
+	/* Ignore any possible further reception on the IRQ path */
+	if (scmi_info->irq > 0)
+		free_irq(scmi_info->irq, scmi_info);
+
 	cinfo->transport_info = NULL;
 	scmi_info->cinfo = NULL;
 
Signature · 1 line
2.40.1
#12 lore raw

Re: [PATCH 5.15 11/92] firmware: arm_scmi: Fix chan_free cleanup on SMC

Quoted 38 lines · Greg Kroah-Hartman

From: Cristian Marussi <[email protected]>

[ Upstream commit d1ff11d7ad8704f8d615f6446041c221b2d2ec4d ]

SCMI transport based on SMC can optionally use an additional IRQ to signal message completion. The associated interrupt handler is currently allocated using devres but on shutdown the core SCMI stack will call .chan_free() well before any managed cleanup is invoked by devres. As a consequence, the arrival of a late reply to an in-flight pending transaction could still trigger the interrupt handler well after the SCMI core has cleaned up the channels, with unpleasant results.

Inhibit further message processing on the IRQ path by explicitly freeing the IRQ inside .chan_free() callback itself.

Fixes dd820ee21d5e ("firmware: arm_scmi: Augment SMC/HVC to allow optional interrupt")
---
  drivers/firmware/arm_scmi/smc.c | 17 +++++++++++------
  1 file changed, 11 insertions(+), 6 deletions(-)
drivers/firmware/arm_scmi/smc.c +1 -0
index 4effecc3bb463..f529004f1922e 100644
@@ -21,6 +21,7 @@
  /**
   * struct scmi_smc - Structure representing a SCMI smc transport
   *
+ * @irq: An optional IRQ for completion
   * @cinfo: SCMI channel info
   * @shmem: Transmit/Receive shared memory area
   * @shmem_lock: Lock to protect access to Tx/Rx shared memory area

drivers/firmware/arm_scmi/smc.c line 34

Quoted 4 lines
drivers/firmware/arm_scmi/smc.c +1 -0
@@ -30,6 +31,7 @@
   */
 
  struct scmi_smc {
+	int irq;

For this backport to apply as-is and not define a duplicate "int irq" field we need to take in f716cbd33f038af87824c30e165b3b70e4c6be1e ("firmware: arm_scmi: Make smc transport use common completions") which did remove the "int irq" from struct scmi_smc.

Alternatively, we can just omit this hunk adding the "int irq" member from the back port.

This is a 5.15 stable kernel problem only because f716cbd33f038af87824c30e165b3b70e4c6be1e is in v5.18 and newer.

Signature · 1 line
Florian
#13 lore raw

Re: [PATCH 5.15 11/92] firmware: arm_scmi: Fix chan_free cleanup on SMC

Quoted 48 lines · Florian Fainelli
Quoted 38 lines · Greg Kroah-Hartman

From: Cristian Marussi <[email protected]>

[ Upstream commit d1ff11d7ad8704f8d615f6446041c221b2d2ec4d ]

SCMI transport based on SMC can optionally use an additional IRQ to signal message completion. The associated interrupt handler is currently allocated using devres but on shutdown the core SCMI stack will call .chan_free() well before any managed cleanup is invoked by devres. As a consequence, the arrival of a late reply to an in-flight pending transaction could still trigger the interrupt handler well after the SCMI core has cleaned up the channels, with unpleasant results.

Inhibit further message processing on the IRQ path by explicitly freeing the IRQ inside .chan_free() callback itself.

Fixes dd820ee21d5e ("firmware: arm_scmi: Augment SMC/HVC to allow optional interrupt")
---
  drivers/firmware/arm_scmi/smc.c | 17 +++++++++++------
  1 file changed, 11 insertions(+), 6 deletions(-)
drivers/firmware/arm_scmi/smc.c +1 -0
index 4effecc3bb463..f529004f1922e 100644
@@ -21,6 +21,7 @@
  /**
   * struct scmi_smc - Structure representing a SCMI smc transport
   *
+ * @irq: An optional IRQ for completion
   * @cinfo: SCMI channel info
   * @shmem: Transmit/Receive shared memory area
   * @shmem_lock: Lock to protect access to Tx/Rx shared memory area

drivers/firmware/arm_scmi/smc.c line 33

Quoted 3 lines
drivers/firmware/arm_scmi/smc.c +1 -0
@@ -30,6 +31,7 @@
   */
  struct scmi_smc {
+	int irq;

For this backport to apply as-is and not define a duplicate "int irq" field we need to take in f716cbd33f038af87824c30e165b3b70e4c6be1e ("firmware:
arm_scmi: Make smc transport use common completions") which did remove the "int irq" from struct scmi_smc.

Quoted 3 lines

Alternatively, we can just omit this hunk adding the "int irq" member from the back port.

There is a bit disconnect in the communication here. I didn't see Greg was not cc-ed on the earlier email to Sasha[1]. The request to drop it was also made here[2].

Quoted 2 lines

This is a 5.15 stable kernel problem only because f716cbd33f038af87824c30e165b3b70e4c6be1e is in v5.18 and newer.

Signature · 6 lines
Regards,
Sudeep

[1] https://marc.info/?l=linux-stable-commits&m=169159392424103&w=2
(couldn't find a lore link for the above)
[2] https://lore.kernel.org/all/ZNIdrd+SQ0KjYWKA@e120937-lin/
author maintainer
#14 lore raw

PATCH 12/92 word-at-a-time: use the same return type for has_zero regardless of endianness

In reply to Greg Kroah-Hartman

From: [email protected] <[email protected]>

[ Upstream commit 79e8328e5acbe691bbde029a52c89d70dcbc22f3 ]

Compiling big-endian targets with Clang produces the diagnostic:

fs/namei.c:2173:13: warning: use of bitwise '|' with boolean operands [-Wbitwise-instead-of-logical]

	} while (!(has_zero(a, &adata, &constants) | has_zero(b, &bdata, &constants)));
	          ~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
                                               ||

fs/namei.c:2173:13: note: cast one or both operands to int to silence this warning

It appears that when has_zero was introduced, two definitions were produced with different signatures (in particular different return types).

Looking at the usage in hash_name() in fs/namei.c, I suspect that has_zero() is meant to be invoked twice per while loop iteration; using logical-or would not update bdata when a did not have zeros. So I think it's preferred to always return an unsigned long rather than a bool than update the while loop in hash_name() to use a logical-or rather than bitwise-or.

[ Also changed powerpc version to do the same - Linus ]

---
 arch/powerpc/include/asm/word-at-a-time.h | 2 +-
 include/asm-generic/word-at-a-time.h      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/include/asm/word-at-a-time.h +1 -1
index f3f4710d4ff52..99129b0cd8b8a 100644
@@ -34,7 +34,7 @@ static inline long find_zero(unsigned long mask)
 	return leading_zero_bits >> 3;
 }
 
-static inline bool has_zero(unsigned long val, unsigned long *data, const struct word_at_a_time *c)
+static inline unsigned long has_zero(unsigned long val, unsigned long *data, const struct word_at_a_time *c)
 {
 	unsigned long rhs = val | c->low_bits;
 	*data = rhs;
include/asm-generic/word-at-a-time.h +1 -1
index 20c93f08c9933..95a1d214108a5 100644
@@ -38,7 +38,7 @@ static inline long find_zero(unsigned long mask)
 	return (mask >> 8) ? byte : byte + 1;
 }
 
-static inline bool has_zero(unsigned long val, unsigned long *data, const struct word_at_a_time *c)
+static inline unsigned long has_zero(unsigned long val, unsigned long *data, const struct word_at_a_time *c)
 {
 	unsigned long rhs = val | c->low_bits;
 	*data = rhs;
Signature · 1 line
2.40.1
author maintainer
#15 lore raw

PATCH 13/92 KVM: s390: fix sthyi error handling

In reply to Greg Kroah-Hartman

From: Heiko Carstens <[email protected]>

[ Upstream commit 0c02cc576eac161601927b41634f80bfd55bfa9e ]

Commit 9fb6c9b3fea1 ("s390/sthyi: add cache to store hypervisor info") added cache handling for store hypervisor info. This also changed the possible return code for sthyi_fill().

Instead of only returning a condition code like the sthyi instruction would do, it can now also return a negative error value (-ENOMEM). handle_styhi() was not changed accordingly. In case of an error, the negative error value would incorrectly injected into the guest PSW.

Add proper error handling to prevent this, and update the comment which describes the possible return values of sthyi_fill().

Fixes 9fb6c9b3fea1 ("s390/sthyi: add cache to store hypervisor info")
---
 arch/s390/kernel/sthyi.c  | 6 +++---
 arch/s390/kvm/intercept.c | 9 ++++++---
 2 files changed, 9 insertions(+), 6 deletions(-)
arch/s390/kernel/sthyi.c +3 -3
index 4d141e2c132e5..2ea7f208f0e73 100644
@@ -459,9 +459,9 @@ static int sthyi_update_cache(u64 *rc)
  *
  * Fills the destination with system information returned by the STHYI
  * instruction. The data is generated by emulation or execution of STHYI,
- * if available. The return value is the condition code that would be
- * returned, the rc parameter is the return code which is passed in
- * register R2 + 1.
+ * if available. The return value is either a negative error value or
+ * the condition code that would be returned, the rc parameter is the
+ * return code which is passed in register R2 + 1.
  */
 int sthyi_fill(void *dst, u64 *rc)
 {
arch/s390/kvm/intercept.c +6 -3
index aeb0e0865e890..458b42b50b8cb 100644
@@ -389,8 +389,8 @@ static int handle_partial_execution(struct kvm_vcpu *vcpu)
  */
 int handle_sthyi(struct kvm_vcpu *vcpu)
 {
-	int reg1, reg2, r = 0;
-	u64 code, addr, cc = 0, rc = 0;
+	int reg1, reg2, cc = 0, r = 0;
+	u64 code, addr, rc = 0;
 	struct sthyi_sctns *sctns = NULL;
 
 	if (!test_kvm_facility(vcpu->kvm, 74))
@@ -421,7 +421,10 @@ int handle_sthyi(struct kvm_vcpu *vcpu)
 		return -ENOMEM;
 
 	cc = sthyi_fill(sctns, &rc);
-
+	if (cc < 0) {
+		free_page((unsigned long)sctns);
+		return cc;
+	}
 out:
 	if (!cc) {
 		if (kvm_s390_pv_cpu_is_protected(vcpu)) {
Signature · 1 line
2.40.1
author maintainer
#16 lore raw

PATCH 14/92 wifi: cfg80211: Fix return value in scan logic

In reply to Greg Kroah-Hartman

From: Ilan Peer <[email protected]>

[ Upstream commit fd7f08d92fcd7cc3eca0dd6c853f722a4c6176df ]

The reporter noticed a warning when running iwlwifi:

WARNING: CPU: 8 PID: 659 at mm/page_alloc.c:4453 __alloc_pages+0x329/0x340

As cfg80211_parse_colocated_ap() is not expected to return a negative value return 0 and not a negative value if cfg80211_calc_short_ssid() fails.

---
 net/wireless/scan.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/wireless/scan.c +1 -1
index a565476809f02..c7192d7bcbd76 100644
@@ -641,7 +641,7 @@ static int cfg80211_parse_colocated_ap(const struct cfg80211_bss_ies *ies,
 
 	ret = cfg80211_calc_short_ssid(ies, &ssid_elem, &s_ssid_tmp);
 	if (ret)
-		return ret;
+		return 0;
 
 	/* RNR IE may contain more than one NEIGHBOR_AP_INFO */
 	while (pos + sizeof(*ap_info) <= end) {
Signature · 1 line
2.40.1
author maintainer
#17 lore raw

PATCH 15/92 net/mlx5: DR, fix memory leak in mlx5dr_cmd_create_reformat_ctx

In reply to Greg Kroah-Hartman

From: Zhengchao Shao <[email protected]>

[ Upstream commit 5dd77585dd9d0e03dd1bceb95f0269a7eaf6b936 ]

when mlx5_cmd_exec failed in mlx5dr_cmd_create_reformat_ctx, the memory pointed by 'in' is not released, which will cause memory leak. Move memory release after mlx5_cmd_exec.

Fixes 1d9186476e12 ("net/mlx5: DR, Add direct rule command utilities")
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c +3 -2
index fcf705ce421f3..aa003a75946bb 100644
@@ -528,11 +528,12 @@ int mlx5dr_cmd_create_reformat_ctx(struct mlx5_core_dev *mdev,
 
 	err = mlx5_cmd_exec(mdev, in, inlen, out, sizeof(out));
 	if (err)
-		return err;
+		goto err_free_in;
 
 	*reformat_id = MLX5_GET(alloc_packet_reformat_context_out, out, packet_reformat_id);
-	kvfree(in);
 
+err_free_in:
+	kvfree(in);
 	return err;
 }
 
Signature · 1 line
2.40.1
author maintainer
#18 lore raw

PATCH 16/92 net/mlx5e: fix return value check in mlx5e_ipsec_remove_trailer()

In reply to Greg Kroah-Hartman

From: Yuanjun Gong <[email protected]>

[ Upstream commit e5bcb7564d3bd0c88613c76963c5349be9c511c5 ]

mlx5e_ipsec_remove_trailer() should return an error code if function pskb_trim() returns an unexpected value.

Fixes 2ac9cfe78223 ("net/mlx5e: IPSec, Add Innova IPSec offload TX data path")
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c +3 -1
index b56fea142c246..4590d19c25cf7 100644
@@ -121,7 +121,9 @@ static int mlx5e_ipsec_remove_trailer(struct sk_buff *skb, struct xfrm_state *x)
 
 	trailer_len = alen + plen + 2;
 
-	pskb_trim(skb, skb->len - trailer_len);
+	ret = pskb_trim(skb, skb->len - trailer_len);
+	if (unlikely(ret))
+		return ret;
 	if (skb->protocol == htons(ETH_P_IP)) {
 		ipv4hdr->tot_len = htons(ntohs(ipv4hdr->tot_len) - trailer_len);
 		ip_send_check(ipv4hdr);
Signature · 1 line
2.40.1
author maintainer
#19 lore raw

PATCH 17/92 bpf: Add length check for SK_DIAG_BPF_STORAGE_REQ_MAP_FD parsing

In reply to Greg Kroah-Hartman

From: Lin Ma <[email protected]>

[ Upstream commit bcc29b7f5af6797702c2306a7aacb831fc5ce9cb ]

The nla_for_each_nested parsing in function bpf_sk_storage_diag_alloc does not check the length of the nested attribute. This can lead to an out-of-attribute read and allow a malformed nlattr (e.g., length 0) to be viewed as a 4 byte integer.

This patch adds an additional check when the nlattr is getting counted. This makes sure the latter nla_get_u32 can access the attributes with the correct length.

---
 net/core/bpf_sk_storage.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
net/core/bpf_sk_storage.c +4 -1
index 910ca41cb9e67..4953abee79fea 100644
@@ -521,8 +521,11 @@ bpf_sk_storage_diag_alloc(const struct nlattr *nla_stgs)
 		return ERR_PTR(-EPERM);
 
 	nla_for_each_nested(nla, nla_stgs, rem) {
-		if (nla_type(nla) == SK_DIAG_BPF_STORAGE_REQ_MAP_FD)
+		if (nla_type(nla) == SK_DIAG_BPF_STORAGE_REQ_MAP_FD) {
+			if (nla_len(nla) != sizeof(u32))
+				return ERR_PTR(-EINVAL);
 			nr_maps++;
+		}
 	}
 
 	diag = kzalloc(struct_size(diag, maps, nr_maps), GFP_KERNEL);
Signature · 1 line
2.40.1
author maintainer
#20 lore raw

PATCH 18/92 rtnetlink: let rtnl_bridge_setlink checks IFLA_BRIDGE_MODE length

In reply to Greg Kroah-Hartman

From: Lin Ma <[email protected]>

[ Upstream commit d73ef2d69c0dba5f5a1cb9600045c873bab1fb7f ]

There are totally 9 ndo_bridge_setlink handlers in the current kernel, which are 1) bnxt_bridge_setlink, 2) be_ndo_bridge_setlink 3) i40e_ndo_bridge_setlink 4) ice_bridge_setlink 5) ixgbe_ndo_bridge_setlink 6) mlx5e_bridge_setlink 7) nfp_net_bridge_setlink 8) qeth_l2_bridge_setlink 9) br_setlink.

By investigating the code, we find that 1-7 parse and use nlattr IFLA_BRIDGE_MODE but 3 and 4 forget to do the nla_len check. This can lead to an out-of-attribute read and allow a malformed nlattr (e.g., length 0) to be viewed as a 2 byte integer.

To avoid such issues, also for other ndo_bridge_setlink handlers in the future. This patch adds the nla_len check in rtnl_bridge_setlink and does an early error return if length mismatches. To make it works, the break is removed from the parsing for IFLA_BRIDGE_FLAGS to make sure this nla_for_each_nested iterates every attribute.

Fixes b1edc14a3fbf ("ice: Implement ice_bridge_getlink and ice_bridge_setlink")
Fixes 51616018dd1b ("i40e: Add support for getlink, setlink ndo ops")
---
 net/core/rtnetlink.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
net/core/rtnetlink.c +6 -2
index 49766446797c1..b055e196f5306 100644
@@ -4919,13 +4919,17 @@ static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC);
 	if (br_spec) {
 		nla_for_each_nested(attr, br_spec, rem) {
-			if (nla_type(attr) == IFLA_BRIDGE_FLAGS) {
+			if (nla_type(attr) == IFLA_BRIDGE_FLAGS && !have_flags) {
 				if (nla_len(attr) < sizeof(flags))
 					return -EINVAL;
 
 				have_flags = true;
 				flags = nla_get_u16(attr);
-				break;
+			}
+
+			if (nla_type(attr) == IFLA_BRIDGE_MODE) {
+				if (nla_len(attr) < sizeof(u16))
+					return -EINVAL;
 			}
 		}
 	}
Signature · 1 line
2.40.1
author maintainer
#21 lore raw

PATCH 19/92 net: dsa: fix value check in bcm_sf2_sw_probe()

In reply to Greg Kroah-Hartman

From: Yuanjun Gong <[email protected]>

[ Upstream commit dadc5b86cc9459581f37fe755b431adc399ea393 ]

in bcm_sf2_sw_probe(), check the return value of clk_prepare_enable() and return the error code if clk_prepare_enable() returns an unexpected value.

Fixes e9ec5c3bd238 ("net: dsa: bcm_sf2: request and handle clocks")
---
 drivers/net/dsa/bcm_sf2.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/net/dsa/bcm_sf2.c +6 -2
index d76b2377d66ef..773d751ef169f 100644
@@ -1422,7 +1422,9 @@ static int bcm_sf2_sw_probe(struct platform_device *pdev)
 	if (IS_ERR(priv->clk))
 		return PTR_ERR(priv->clk);
 
-	clk_prepare_enable(priv->clk);
+	ret = clk_prepare_enable(priv->clk);
+	if (ret)
+		return ret;
 
 	priv->clk_mdiv = devm_clk_get_optional(&pdev->dev, "sw_switch_mdiv");
 	if (IS_ERR(priv->clk_mdiv)) {
@@ -1430,7 +1432,9 @@ static int bcm_sf2_sw_probe(struct platform_device *pdev)
 		goto out_clk;
 	}
 
-	clk_prepare_enable(priv->clk_mdiv);
+	ret = clk_prepare_enable(priv->clk_mdiv);
+	if (ret)
+		goto out_clk;
 
 	ret = bcm_sf2_sw_rst(priv);
 	if (ret) {
Signature · 1 line
2.40.1
author maintainer
#22 lore raw

PATCH 20/92 perf test uprobe_from_different_cu: Skip if there is no gcc

In reply to Greg Kroah-Hartman

From: Georg Müller <[email protected]>

[ Upstream commit 98ce8e4a9dcfb448b30a2d7a16190f4a00382377 ]

Without gcc, the test will fail.

On cleanup, ignore probe removal errors. Otherwise, in case of an error adding the probe, the temporary directory is not removed.

Fixes 56cbeacf14353057 ("perf probe: Add test for regression introduced by switch to die_get_decl_file()")
Cc Adrian Hunter <[email protected]>
Cc Alexander Shishkin <[email protected]>
Cc Georg Müller <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Mark Rutland <[email protected]>
Cc Masami Hiramatsu <[email protected]>
Cc Namhyung Kim <[email protected]>
Cc Peter Zijlstra <[email protected]>
---
 tools/perf/tests/shell/test_uprobe_from_different_cu.sh | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
tools/perf/tests/shell/test_uprobe_from_different_cu.sh +7 -1
index 00d2e0e2e0c28..319f36ebb9a40 100644
@@ -4,6 +4,12 @@
 
 set -e
 
+# skip if there's no gcc
+if ! [ -x "$(command -v gcc)" ]; then
+        echo "failed: no gcc compiler"
+        exit 2
+fi
+
 temp_dir=$(mktemp -d /tmp/perf-uprobe-different-cu-sh.XXXXXXXXXX)
 
 cleanup()
@@ -11,7 +17,7 @@ cleanup()
 	trap - EXIT TERM INT
 	if [[ "${temp_dir}" =~ ^/tmp/perf-uprobe-different-cu-sh.*$ ]]; then
 		echo "--- Cleaning up ---"
-		perf probe -x ${temp_dir}/testfile -d foo
+		perf probe -x ${temp_dir}/testfile -d foo || true
 		rm -f "${temp_dir}/"*
 		rmdir "${temp_dir}"
 	fi
Signature · 1 line
2.40.1
author maintainer
#23 lore raw

PATCH 21/92 net: sched: cls_u32: Fix match key mis-addressing

In reply to Greg Kroah-Hartman

From: Jamal Hadi Salim <[email protected]>

[ Upstream commit e68409db995380d1badacba41ff24996bd396171 ]

A match entry is uniquely identified with an "address" or "path" in the form of: hashtable ID(12b):bucketid(8b):nodeid(12b).

When creating table match entries all of hash table id, bucket id and node (match entry id) are needed to be either specified by the user or reasonable in-kernel defaults are used. The in-kernel default for a table id is 0x800(omnipresent root table); for bucketid it is 0x0. Prior to this fix there was none for a nodeid i.e. the code assumed that the user passed the correct nodeid and if the user passes a nodeid of 0 (as Mingi Cho did) then that is what was used. But nodeid of 0 is reserved for identifying the table. This is not a problem until we dump. The dump code notices that the nodeid is zero and assumes it is referencing a table and therefore references table struct tc_u_hnode instead of what was created i.e match entry struct tc_u_knode.

Ming does an equivalent of:
tc filter add dev dummy0 parent 10: prio 1 handle 0x1000 \
protocol ip u32 match ip src 10.0.0.1/32 classid 10:1 action ok

Essentially specifying a table id 0, bucketid 1 and nodeid of zero Tableid 0 is remapped to the default of 0x800.
Bucketid 1 is ignored and defaults to 0x00.
Nodeid was assumed to be what Ming passed - 0x000

dumping before fix shows:
~$ tc filter ls dev dummy0 parent 10:
filter protocol ip pref 1 u32 chain 0
filter protocol ip pref 1 u32 chain 0 fh 800: ht divisor 1 filter protocol ip pref 1 u32 chain 0 fh 800: ht divisor -30591

Note that the last line reports a table instead of a match entry (you can tell this because it says "ht divisor...").
As a result of reporting the wrong data type (misinterpretting of struct tc_u_knode as being struct tc_u_hnode) the divisor is reported with value of -30591. Ming identified this as part of the heap address (physmap_base is 0xffff8880 (-30591 - 1)).

The fix is to ensure that when table entry matches are added and no nodeid is specified (i.e nodeid == 0) then we get the next available nodeid from the table's pool.

After the fix, this is what the dump shows:

$ tc filter ls dev dummy0 parent 10:
filter protocol ip pref 1 u32 chain 0
filter protocol ip pref 1 u32 chain 0 fh 800: ht divisor 1
filter protocol ip pref 1 u32 chain 0 fh 800::800 order 2048 key ht 800 bkt 0 flowid 10:1 not_in_hw
  match 0a000001/ffffffff at 12
	action order 1: gact action pass
	 random type none pass val 0
	 index 1 ref 1 bind 1
---
 net/sched/cls_u32.c | 56 ++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 50 insertions(+), 6 deletions(-)
net/sched/cls_u32.c +50 -6
index 0025fa837e857..e4e38aa75111e 100644
@@ -999,18 +999,62 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 		return -EINVAL;
 	}
 
+	/* At this point, we need to derive the new handle that will be used to
+	 * uniquely map the identity of this table match entry. The
+	 * identity of the entry that we need to construct is 32 bits made of:
+	 *     htid(12b):bucketid(8b):node/entryid(12b)
+	 *
+	 * At this point _we have the table(ht)_ in which we will insert this
+	 * entry. We carry the table's id in variable "htid".
+	 * Note that earlier code picked the ht selection either by a) the user
+	 * providing the htid specified via TCA_U32_HASH attribute or b) when
+	 * no such attribute is passed then the root ht, is default to at ID
+	 * 0x[800][00][000]. Rule: the root table has a single bucket with ID 0.
+	 * If OTOH the user passed us the htid, they may also pass a bucketid of
+	 * choice. 0 is fine. For example a user htid is 0x[600][01][000] it is
+	 * indicating hash bucketid of 1. Rule: the entry/node ID _cannot_ be
+	 * passed via the htid, so even if it was non-zero it will be ignored.
+	 *
+	 * We may also have a handle, if the user passed one. The handle also
+	 * carries the same addressing of htid(12b):bucketid(8b):node/entryid(12b).
+	 * Rule: the bucketid on the handle is ignored even if one was passed;
+	 * rather the value on "htid" is always assumed to be the bucketid.
+	 */
 	if (handle) {
+		/* Rule: The htid from handle and tableid from htid must match */
 		if (TC_U32_HTID(handle) && TC_U32_HTID(handle ^ htid)) {
 			NL_SET_ERR_MSG_MOD(extack, "Handle specified hash table address mismatch");
 			return -EINVAL;
 		}
-		handle = htid | TC_U32_NODE(handle);
-		err = idr_alloc_u32(&ht->handle_idr, NULL, &handle, handle,
-				    GFP_KERNEL);
-		if (err)
-			return err;
-	} else
+		/* Ok, so far we have a valid htid(12b):bucketid(8b) but we
+		 * need to finalize the table entry identification with the last
+		 * part - the node/entryid(12b)). Rule: Nodeid _cannot be 0_ for
+		 * entries. Rule: nodeid of 0 is reserved only for tables(see
+		 * earlier code which processes TC_U32_DIVISOR attribute).
+		 * Rule: The nodeid can only be derived from the handle (and not
+		 * htid).
+		 * Rule: if the handle specified zero for the node id example
+		 * 0x60000000, then pick a new nodeid from the pool of IDs
+		 * this hash table has been allocating from.
+		 * If OTOH it is specified (i.e for example the user passed a
+		 * handle such as 0x60000123), then we use it generate our final
+		 * handle which is used to uniquely identify the match entry.
+		 */
+		if (!TC_U32_NODE(handle)) {
+			handle = gen_new_kid(ht, htid);
+		} else {
+			handle = htid | TC_U32_NODE(handle);
+			err = idr_alloc_u32(&ht->handle_idr, NULL, &handle,
+					    handle, GFP_KERNEL);
+			if (err)
+				return err;
+		}
+	} else {
+		/* The user did not give us a handle; lets just generate one
+		 * from the table's pool of nodeids.
+		 */
 		handle = gen_new_kid(ht, htid);
+	}
 
 	if (tb[TCA_U32_SEL] == NULL) {
 		NL_SET_ERR_MSG_MOD(extack, "Selector not specified");
Signature · 1 line
2.40.1
author maintainer
#24 lore raw

PATCH 22/92 mISDN: hfcpci: Fix potential deadlock on &hc->lock

In reply to Greg Kroah-Hartman

From: Chengfeng Ye <[email protected]>

[ Upstream commit 56c6be35fcbed54279df0a2c9e60480a61841d6f ]

As &hc->lock is acquired by both timer _hfcpci_softirq() and hardirq hfcpci_int(), the timer should disable irq before lock acquisition otherwise deadlock could happen if the timmer is preemtped by the hadr irq.

Possible deadlock scenario:
hfcpci_softirq() (timer)

    -> _hfcpci_softirq()
    -> spin_lock(&hc->lock);
        <irq interruption>
        -> hfcpci_int()
        -> spin_lock(&hc->lock); (deadlock here)

This flaw was found by an experimental static analysis tool I am developing for irq-related deadlock.

The tentative patch fixes the potential deadlock by spin_lock_irq() in timer.

---
 drivers/isdn/hardware/mISDN/hfcpci.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/isdn/hardware/mISDN/hfcpci.c +5 -5
index eba58b99cd29d..d6cf01c32a33d 100644
@@ -839,7 +839,7 @@ hfcpci_fill_fifo(struct bchannel *bch)
 		*z1t = cpu_to_le16(new_z1);	/* now send data */
 		if (bch->tx_idx < bch->tx_skb->len)
 			return;
-		dev_kfree_skb(bch->tx_skb);
+		dev_kfree_skb_any(bch->tx_skb);
 		if (get_next_bframe(bch))
 			goto next_t_frame;
 		return;
@@ -895,7 +895,7 @@ hfcpci_fill_fifo(struct bchannel *bch)
 	}
 	bz->za[new_f1].z1 = cpu_to_le16(new_z1);	/* for next buffer */
 	bz->f1 = new_f1;	/* next frame */
-	dev_kfree_skb(bch->tx_skb);
+	dev_kfree_skb_any(bch->tx_skb);
 	get_next_bframe(bch);
 }
 
@@ -1119,7 +1119,7 @@ tx_birq(struct bchannel *bch)
 	if (bch->tx_skb && bch->tx_idx < bch->tx_skb->len)
 		hfcpci_fill_fifo(bch);
 	else {
-		dev_kfree_skb(bch->tx_skb);
+		dev_kfree_skb_any(bch->tx_skb);
 		if (get_next_bframe(bch))
 			hfcpci_fill_fifo(bch);
 	}
@@ -2277,7 +2277,7 @@ _hfcpci_softirq(struct device *dev, void *unused)
 		return 0;
 
 	if (hc->hw.int_m2 & HFCPCI_IRQ_ENABLE) {
-		spin_lock(&hc->lock);
+		spin_lock_irq(&hc->lock);
 		bch = Sel_BCS(hc, hc->hw.bswapped ? 2 : 1);
 		if (bch && bch->state == ISDN_P_B_RAW) { /* B1 rx&tx */
 			main_rec_hfcpci(bch);
@@ -2288,7 +2288,7 @@ _hfcpci_softirq(struct device *dev, void *unused)
 			main_rec_hfcpci(bch);
 			tx_birq(bch);
 		}
-		spin_unlock(&hc->lock);
+		spin_unlock_irq(&hc->lock);
 	}
 	return 0;
 }
Signature · 1 line
2.40.1
author maintainer
#25 lore raw

PATCH 23/92 qed: Fix kernel-doc warnings

In reply to Greg Kroah-Hartman

From: Prabhakar Kushwaha <[email protected]>

[ Upstream commit 19198e4ec97dc9d173b458a75ace3c3ca55c2d85 ]

This patch fixes all the qed and qede kernel-doc warnings according to the guidelines that are described in Documentation/doc-guide/kernel-doc.rst.

Stable-dep-of: e346e231b42b ("qed: Fix scheduling in a tasklet while getting stats")

---
 drivers/net/ethernet/qlogic/qed/qed.h         |   9 +-
 drivers/net/ethernet/qlogic/qed/qed_cxt.h     | 138 +--
 drivers/net/ethernet/qlogic/qed/qed_dev_api.h | 343 ++++---
 drivers/net/ethernet/qlogic/qed/qed_hsi.h     | 956 +++++++++---------
 drivers/net/ethernet/qlogic/qed/qed_hw.h      | 222 ++--
 .../net/ethernet/qlogic/qed/qed_init_ops.h    |  58 +-
 drivers/net/ethernet/qlogic/qed/qed_int.h     | 284 +++---
 drivers/net/ethernet/qlogic/qed/qed_iscsi.h   |   9 +-
 drivers/net/ethernet/qlogic/qed/qed_l2.h      | 134 +--
 drivers/net/ethernet/qlogic/qed/qed_ll2.h     | 130 +--
 drivers/net/ethernet/qlogic/qed/qed_mcp.h     | 757 +++++++-------
 .../net/ethernet/qlogic/qed/qed_selftest.h    |  30 +-
 drivers/net/ethernet/qlogic/qed/qed_sp.h      | 215 ++--
 drivers/net/ethernet/qlogic/qed/qed_sriov.h   |  99 +-
 drivers/net/ethernet/qlogic/qed/qed_vf.h      | 301 +++---
 drivers/net/ethernet/qlogic/qede/qede_main.c  |   5 +-
 include/linux/qed/qed_chain.h                 |  97 +-
 include/linux/qed/qed_if.h                    | 255 ++---
 include/linux/qed/qed_iscsi_if.h              |   2 +-
 include/linux/qed/qed_ll2_if.h                |  42 +-
 include/linux/qed/qed_nvmetcp_if.h            |  17 +
 21 files changed, 2186 insertions(+), 1917 deletions(-)
drivers/net/ethernet/qlogic/qed/qed.h +5 -4
index d58e021614cd0..b656408b9d700 100644
@@ -877,12 +877,13 @@ u32 qed_get_hsi_def_val(struct qed_dev *cdev, enum qed_hsi_def_type type);
 
 
 /**
- * @brief qed_concrete_to_sw_fid - get the sw function id from
- *        the concrete value.
+ * qed_concrete_to_sw_fid(): Get the sw function id from
+ *                           the concrete value.
  *
- * @param concrete_fid
+ * @cdev: Qed dev pointer.
+ * @concrete_fid: Concrete fid.
  *
- * @return inline u8
+ * Return: inline u8.
  */
 static inline u8 qed_concrete_to_sw_fid(struct qed_dev *cdev,
 					u32 concrete_fid)
drivers/net/ethernet/qlogic/qed/qed_cxt.h +75 -63

241 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_dev_api.h +185 -158

573 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_hsi.h +499 -457

1,404 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_hw.h +125 -97

347 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_init_ops.h +26 -32

105 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_int.h +154 -130

313 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_iscsi.h +6 -3
index dab7a5d09f874..dec2b00259d42 100644

Lines of this file are not shown. View the raw message.

drivers/net/ethernet/qlogic/qed/qed_l2.h +68 -66

134 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_ll2.h +66 -64

130 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_mcp.h +399 -358

757 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_selftest.h +15 -15
index e27dd9a4547e8..7a3bd749e1e4c 100644

Lines of this file are not shown. View the raw message.

drivers/net/ethernet/qlogic/qed/qed_sp.h +105 -110

215 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_sriov.h +55 -44

99 lines not on this page · raw message

drivers/net/ethernet/qlogic/qed/qed_vf.h +169 -132

301 lines not on this page · raw message

drivers/net/ethernet/qlogic/qede/qede_main.c +4 -1
index 6c22bfc16ee6b..fee47c8eeff49 100644

Lines of this file are not shown. View the raw message.

include/linux/qed/qed_chain.h +52 -45

97 lines not on this page · raw message

include/linux/qed/qed_if.h +139 -116

255 lines not on this page · raw message

include/linux/qed/qed_iscsi_if.h +1 -1
index 04180d9af560e..494cdc3cd840b 100644

Lines of this file are not shown. View the raw message.

include/linux/qed/qed_ll2_if.h +21 -21

42 lines not on this page · raw message

include/linux/qed/qed_nvmetcp_if.h +17 -0
index 14671bc19ed11..1d51df347560d 100644

Lines of this file are not shown. View the raw message.

The rest of this diff is not shown. View the raw message.

Signature · 1 line
2.40.1

Part of the diff is not shown here. View the raw message.

author maintainer
#26 lore raw

PATCH 24/92 qed: Fix scheduling in a tasklet while getting stats

In reply to Greg Kroah-Hartman

From: Konstantin Khorenko <[email protected]>

[ Upstream commit e346e231b42bcae6822a6326acfb7b741e9e6026 ]

Here we've got to a situation when tasklet called usleep_range() in PTT acquire logic, thus welcome to the "scheduling while atomic" BUG().

  BUG: scheduling while atomic: swapper/24/0/0x00000100

   [<ffffffffb41c6199>] schedule+0x29/0x70
   [<ffffffffb41c5512>] schedule_hrtimeout_range_clock+0xb2/0x150
   [<ffffffffb41c55c3>] schedule_hrtimeout_range+0x13/0x20
   [<ffffffffb41c3bcf>] usleep_range+0x4f/0x70
   [<ffffffffc08d3e58>] qed_ptt_acquire+0x38/0x100 [qed]
   [<ffffffffc08eac48>] _qed_get_vport_stats+0x458/0x580 [qed]
   [<ffffffffc08ead8c>] qed_get_vport_stats+0x1c/0xd0 [qed]
   [<ffffffffc08dffd3>] qed_get_protocol_stats+0x93/0x100 [qed]
                        qed_mcp_send_protocol_stats
            case MFW_DRV_MSG_GET_LAN_STATS:
            case MFW_DRV_MSG_GET_FCOE_STATS:
            case MFW_DRV_MSG_GET_ISCSI_STATS:
            case MFW_DRV_MSG_GET_RDMA_STATS:
   [<ffffffffc08e36d8>] qed_mcp_handle_events+0x2d8/0x890 [qed]
                        qed_int_assertion
                        qed_int_attentions
   [<ffffffffc08d9490>] qed_int_sp_dpc+0xa50/0xdc0 [qed]
   [<ffffffffb3aa7623>] tasklet_action+0x83/0x140
   [<ffffffffb41d9125>] __do_softirq+0x125/0x2bb
   [<ffffffffb41d560c>] call_softirq+0x1c/0x30
   [<ffffffffb3a30645>] do_softirq+0x65/0xa0
   [<ffffffffb3aa78d5>] irq_exit+0x105/0x110
   [<ffffffffb41d8996>] do_IRQ+0x56/0xf0

Fix this by making caller to provide the context whether it could be in atomic context flow or not when getting stats from QED driver. QED driver based on the context provided decide to schedule out or not when acquiring the PTT BAR window.

We faced the BUG_ON() while getting vport stats, but according to the code same issue could happen for fcoe and iscsi statistics as well, so fixing them too.

Fixes 6c75424612a7 ("qed: Add support for NCSI statistics.")
Fixes 1e128c81290a ("qed: Add support for hardware offloaded FCoE.")
Fixes 2f2b2614e893 ("qed: Provide iSCSI statistics to management")
Cc Sudarsana Kalluru <[email protected]>
Cc David Miller <[email protected]>
Cc Manish Chopra <[email protected]>
---
 drivers/net/ethernet/qlogic/qed/qed_dev_api.h | 16 ++++++++++++
 drivers/net/ethernet/qlogic/qed/qed_fcoe.c    | 19 ++++++++++----
 drivers/net/ethernet/qlogic/qed/qed_fcoe.h    | 17 ++++++++++--
 drivers/net/ethernet/qlogic/qed/qed_hw.c      | 26 ++++++++++++++++---
 drivers/net/ethernet/qlogic/qed/qed_iscsi.c   | 19 ++++++++++----
 drivers/net/ethernet/qlogic/qed/qed_iscsi.h   |  8 ++++--
 drivers/net/ethernet/qlogic/qed/qed_l2.c      | 19 ++++++++++----
 drivers/net/ethernet/qlogic/qed/qed_l2.h      | 24 +++++++++++++++++
 drivers/net/ethernet/qlogic/qed/qed_main.c    |  6 ++---
 9 files changed, 128 insertions(+), 26 deletions(-)
drivers/net/ethernet/qlogic/qed/qed_dev_api.h +16 -0
index f0a825b985a4b..a0a766a1723cc 100644
@@ -194,6 +194,22 @@ void qed_hw_remove(struct qed_dev *cdev);
  */
 struct qed_ptt *qed_ptt_acquire(struct qed_hwfn *p_hwfn);
 
+/**
+ * qed_ptt_acquire_context(): Allocate a PTT window honoring the context
+ *			      atomicy.
+ *
+ * @p_hwfn: HW device data.
+ * @is_atomic: Hint from the caller - if the func can sleep or not.
+ *
+ * Context: The function should not sleep in case is_atomic == true.
+ * Return: struct qed_ptt.
+ *
+ * Should be called at the entry point to the driver
+ * (at the beginning of an exported function).
+ */
+struct qed_ptt *qed_ptt_acquire_context(struct qed_hwfn *p_hwfn,
+					bool is_atomic);
+
 /**
  * qed_ptt_release(): Release PTT Window.
  *
drivers/net/ethernet/qlogic/qed/qed_fcoe.c +14 -5
index b768f0698170e..0c55249b3a358 100644
@@ -694,13 +694,14 @@ static void _qed_fcoe_get_pstats(struct qed_hwfn *p_hwfn,
 }
 
 static int qed_fcoe_get_stats(struct qed_hwfn *p_hwfn,
-			      struct qed_fcoe_stats *p_stats)
+			      struct qed_fcoe_stats *p_stats,
+			      bool is_atomic)
 {
 	struct qed_ptt *p_ptt;
 
 	memset(p_stats, 0, sizeof(*p_stats));
 
-	p_ptt = qed_ptt_acquire(p_hwfn);
+	p_ptt = qed_ptt_acquire_context(p_hwfn, is_atomic);
 
 	if (!p_ptt) {
 		DP_ERR(p_hwfn, "Failed to acquire ptt\n");
@@ -974,19 +975,27 @@ static int qed_fcoe_destroy_conn(struct qed_dev *cdev,
 					QED_SPQ_MODE_EBLOCK, NULL);
 }
 
+static int qed_fcoe_stats_context(struct qed_dev *cdev,
+				  struct qed_fcoe_stats *stats,
+				  bool is_atomic)
+{
+	return qed_fcoe_get_stats(QED_AFFIN_HWFN(cdev), stats, is_atomic);
+}
+
 static int qed_fcoe_stats(struct qed_dev *cdev, struct qed_fcoe_stats *stats)
 {
-	return qed_fcoe_get_stats(QED_AFFIN_HWFN(cdev), stats);
+	return qed_fcoe_stats_context(cdev, stats, false);
 }
 
 void qed_get_protocol_stats_fcoe(struct qed_dev *cdev,
-				 struct qed_mcp_fcoe_stats *stats)
+				 struct qed_mcp_fcoe_stats *stats,
+				 bool is_atomic)
 {
 	struct qed_fcoe_stats proto_stats;
 
 	/* Retrieve FW statistics */
 	memset(&proto_stats, 0, sizeof(proto_stats));
-	if (qed_fcoe_stats(cdev, &proto_stats)) {
+	if (qed_fcoe_stats_context(cdev, &proto_stats, is_atomic)) {
 		DP_VERBOSE(cdev, QED_MSG_STORAGE,
 			   "Failed to collect FCoE statistics\n");
 		return;
drivers/net/ethernet/qlogic/qed/qed_fcoe.h +15 -2
index 19c85adf4ceb1..214e8299ecb4e 100644
@@ -28,8 +28,20 @@ int qed_fcoe_alloc(struct qed_hwfn *p_hwfn);
 void qed_fcoe_setup(struct qed_hwfn *p_hwfn);
 
 void qed_fcoe_free(struct qed_hwfn *p_hwfn);
+/**
+ * qed_get_protocol_stats_fcoe(): Fills provided statistics
+ *				  struct with statistics.
+ *
+ * @cdev: Qed dev pointer.
+ * @stats: Points to struct that will be filled with statistics.
+ * @is_atomic: Hint from the caller - if the func can sleep or not.
+ *
+ * Context: The function should not sleep in case is_atomic == true.
+ * Return: Void.
+ */
 void qed_get_protocol_stats_fcoe(struct qed_dev *cdev,
-				 struct qed_mcp_fcoe_stats *stats);
+				 struct qed_mcp_fcoe_stats *stats,
+				 bool is_atomic);
 #else /* CONFIG_QED_FCOE */
 static inline int qed_fcoe_alloc(struct qed_hwfn *p_hwfn)
 {
@@ -40,7 +52,8 @@ static inline void qed_fcoe_setup(struct qed_hwfn *p_hwfn) {}
 static inline void qed_fcoe_free(struct qed_hwfn *p_hwfn) {}
 
 static inline void qed_get_protocol_stats_fcoe(struct qed_dev *cdev,
-					       struct qed_mcp_fcoe_stats *stats)
+					       struct qed_mcp_fcoe_stats *stats,
+					       bool is_atomic)
 {
 }
 #endif /* CONFIG_QED_FCOE */
drivers/net/ethernet/qlogic/qed/qed_hw.c +22 -4
index 554f30b0cfd5e..6263f847b6b92 100644
@@ -23,7 +23,10 @@
 #include "qed_reg_addr.h"
 #include "qed_sriov.h"
 
-#define QED_BAR_ACQUIRE_TIMEOUT 1000
+#define QED_BAR_ACQUIRE_TIMEOUT_USLEEP_CNT	1000
+#define QED_BAR_ACQUIRE_TIMEOUT_USLEEP		1000
+#define QED_BAR_ACQUIRE_TIMEOUT_UDELAY_CNT	100000
+#define QED_BAR_ACQUIRE_TIMEOUT_UDELAY		10
 
 /* Invalid values */
 #define QED_BAR_INVALID_OFFSET          (cpu_to_le32(-1))
@@ -84,12 +87,22 @@ void qed_ptt_pool_free(struct qed_hwfn *p_hwfn)
 }
 
 struct qed_ptt *qed_ptt_acquire(struct qed_hwfn *p_hwfn)
+{
+	return qed_ptt_acquire_context(p_hwfn, false);
+}
+
+struct qed_ptt *qed_ptt_acquire_context(struct qed_hwfn *p_hwfn, bool is_atomic)
 {
 	struct qed_ptt *p_ptt;
-	unsigned int i;
+	unsigned int i, count;
+
+	if (is_atomic)
+		count = QED_BAR_ACQUIRE_TIMEOUT_UDELAY_CNT;
+	else
+		count = QED_BAR_ACQUIRE_TIMEOUT_USLEEP_CNT;
 
 	/* Take the free PTT from the list */
-	for (i = 0; i < QED_BAR_ACQUIRE_TIMEOUT; i++) {
+	for (i = 0; i < count; i++) {
 		spin_lock_bh(&p_hwfn->p_ptt_pool->lock);
 
 		if (!list_empty(&p_hwfn->p_ptt_pool->free_list)) {
@@ -105,7 +118,12 @@ struct qed_ptt *qed_ptt_acquire(struct qed_hwfn *p_hwfn)
 		}
 
 		spin_unlock_bh(&p_hwfn->p_ptt_pool->lock);
-		usleep_range(1000, 2000);
+
+		if (is_atomic)
+			udelay(QED_BAR_ACQUIRE_TIMEOUT_UDELAY);
+		else
+			usleep_range(QED_BAR_ACQUIRE_TIMEOUT_USLEEP,
+				     QED_BAR_ACQUIRE_TIMEOUT_USLEEP * 2);
 	}
 
 	DP_NOTICE(p_hwfn, "PTT acquire timeout - failed to allocate PTT\n");
drivers/net/ethernet/qlogic/qed/qed_iscsi.c +14 -5
index db926d8b30334..f111391772778 100644
@@ -1000,13 +1000,14 @@ static void _qed_iscsi_get_pstats(struct qed_hwfn *p_hwfn,
 }
 
 static int qed_iscsi_get_stats(struct qed_hwfn *p_hwfn,
-			       struct qed_iscsi_stats *stats)
+			       struct qed_iscsi_stats *stats,
+			       bool is_atomic)
 {
 	struct qed_ptt *p_ptt;
 
 	memset(stats, 0, sizeof(*stats));
 
-	p_ptt = qed_ptt_acquire(p_hwfn);
+	p_ptt = qed_ptt_acquire_context(p_hwfn, is_atomic);
 	if (!p_ptt) {
 		DP_ERR(p_hwfn, "Failed to acquire ptt\n");
 		return -EAGAIN;
@@ -1337,9 +1338,16 @@ static int qed_iscsi_destroy_conn(struct qed_dev *cdev,
 					   QED_SPQ_MODE_EBLOCK, NULL);
 }
 
+static int qed_iscsi_stats_context(struct qed_dev *cdev,
+				   struct qed_iscsi_stats *stats,
+				   bool is_atomic)
+{
+	return qed_iscsi_get_stats(QED_AFFIN_HWFN(cdev), stats, is_atomic);
+}
+
 static int qed_iscsi_stats(struct qed_dev *cdev, struct qed_iscsi_stats *stats)
 {
-	return qed_iscsi_get_stats(QED_AFFIN_HWFN(cdev), stats);
+	return qed_iscsi_stats_context(cdev, stats, false);
 }
 
 static int qed_iscsi_change_mac(struct qed_dev *cdev,
@@ -1359,13 +1367,14 @@ static int qed_iscsi_change_mac(struct qed_dev *cdev,
 }
 
 void qed_get_protocol_stats_iscsi(struct qed_dev *cdev,
-				  struct qed_mcp_iscsi_stats *stats)
+				  struct qed_mcp_iscsi_stats *stats,
+				  bool is_atomic)
 {
 	struct qed_iscsi_stats proto_stats;
 
 	/* Retrieve FW statistics */
 	memset(&proto_stats, 0, sizeof(proto_stats));
-	if (qed_iscsi_stats(cdev, &proto_stats)) {
+	if (qed_iscsi_stats_context(cdev, &proto_stats, is_atomic)) {
 		DP_VERBOSE(cdev, QED_MSG_STORAGE,
 			   "Failed to collect ISCSI statistics\n");
 		return;
drivers/net/ethernet/qlogic/qed/qed_iscsi.h +6 -2
index dec2b00259d42..974cb8d26608c 100644
@@ -39,11 +39,14 @@ void qed_iscsi_free(struct qed_hwfn *p_hwfn);
  *
  * @cdev: Qed dev pointer.
  * @stats: Points to struct that will be filled with statistics.
+ * @is_atomic: Hint from the caller - if the func can sleep or not.
  *
+ * Context: The function should not sleep in case is_atomic == true.
  * Return: Void.
  */
 void qed_get_protocol_stats_iscsi(struct qed_dev *cdev,
-				  struct qed_mcp_iscsi_stats *stats);
+				  struct qed_mcp_iscsi_stats *stats,
+				  bool is_atomic);
 #else /* IS_ENABLED(CONFIG_QED_ISCSI) */
 static inline int qed_iscsi_alloc(struct qed_hwfn *p_hwfn)
 {
@@ -56,7 +59,8 @@ static inline void qed_iscsi_free(struct qed_hwfn *p_hwfn) {}
 
 static inline void
 qed_get_protocol_stats_iscsi(struct qed_dev *cdev,
-			     struct qed_mcp_iscsi_stats *stats) {}
+			     struct qed_mcp_iscsi_stats *stats,
+			     bool is_atomic) {}
 #endif /* IS_ENABLED(CONFIG_QED_ISCSI) */
 
 #endif
drivers/net/ethernet/qlogic/qed/qed_l2.c +14 -5
index bc17bc36d346e..6ffa6425a75a5 100644
@@ -1863,7 +1863,8 @@ static void __qed_get_vport_stats(struct qed_hwfn *p_hwfn,
 }
 
 static void _qed_get_vport_stats(struct qed_dev *cdev,
-				 struct qed_eth_stats *stats)
+				 struct qed_eth_stats *stats,
+				 bool is_atomic)
 {
 	u8 fw_vport = 0;
 	int i;
@@ -1872,10 +1873,11 @@ static void _qed_get_vport_stats(struct qed_dev *cdev,
 
 	for_each_hwfn(cdev, i) {
 		struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
-		struct qed_ptt *p_ptt = IS_PF(cdev) ? qed_ptt_acquire(p_hwfn)
-						    :  NULL;
+		struct qed_ptt *p_ptt;
 		bool b_get_port_stats;
 
+		p_ptt = IS_PF(cdev) ? qed_ptt_acquire_context(p_hwfn, is_atomic)
+				    : NULL;
 		if (IS_PF(cdev)) {
 			/* The main vport index is relative first */
 			if (qed_fw_vport(p_hwfn, 0, &fw_vport)) {
@@ -1900,6 +1902,13 @@ static void _qed_get_vport_stats(struct qed_dev *cdev,
 }
 
 void qed_get_vport_stats(struct qed_dev *cdev, struct qed_eth_stats *stats)
+{
+	qed_get_vport_stats_context(cdev, stats, false);
+}
+
+void qed_get_vport_stats_context(struct qed_dev *cdev,
+				 struct qed_eth_stats *stats,
+				 bool is_atomic)
 {
 	u32 i;
 
@@ -1908,7 +1917,7 @@ void qed_get_vport_stats(struct qed_dev *cdev, struct qed_eth_stats *stats)
 		return;
 	}
 
-	_qed_get_vport_stats(cdev, stats);
+	_qed_get_vport_stats(cdev, stats, is_atomic);
 
 	if (!cdev->reset_stats)
 		return;
@@ -1960,7 +1969,7 @@ void qed_reset_vport_stats(struct qed_dev *cdev)
 	if (!cdev->reset_stats) {
 		DP_INFO(cdev, "Reset stats not allocated\n");
 	} else {
-		_qed_get_vport_stats(cdev, cdev->reset_stats);
+		_qed_get_vport_stats(cdev, cdev->reset_stats, false);
 		cdev->reset_stats->common.link_change_count = 0;
 	}
 }
drivers/net/ethernet/qlogic/qed/qed_l2.h +24 -0
index 2ab7f3f0cf6c9..602a12a348b2e 100644
@@ -250,8 +250,32 @@ qed_sp_eth_rx_queues_update(struct qed_hwfn *p_hwfn,
 			    enum spq_mode comp_mode,
 			    struct qed_spq_comp_cb *p_comp_data);
 
+/**
+ * qed_get_vport_stats(): Fills provided statistics
+ *			  struct with statistics.
+ *
+ * @cdev: Qed dev pointer.
+ * @stats: Points to struct that will be filled with statistics.
+ *
+ * Return: Void.
+ */
 void qed_get_vport_stats(struct qed_dev *cdev, struct qed_eth_stats *stats);
 
+/**
+ * qed_get_vport_stats_context(): Fills provided statistics
+ *				  struct with statistics.
+ *
+ * @cdev: Qed dev pointer.
+ * @stats: Points to struct that will be filled with statistics.
+ * @is_atomic: Hint from the caller - if the func can sleep or not.
+ *
+ * Context: The function should not sleep in case is_atomic == true.
+ * Return: Void.
+ */
+void qed_get_vport_stats_context(struct qed_dev *cdev,
+				 struct qed_eth_stats *stats,
+				 bool is_atomic);
+
 void qed_reset_vport_stats(struct qed_dev *cdev);
 
 /**
drivers/net/ethernet/qlogic/qed/qed_main.c +3 -3
index d10e1cd6d2ba9..26700b0b4b370 100644
@@ -3054,7 +3054,7 @@ void qed_get_protocol_stats(struct qed_dev *cdev,
 
 	switch (type) {
 	case QED_MCP_LAN_STATS:
-		qed_get_vport_stats(cdev, &eth_stats);
+		qed_get_vport_stats_context(cdev, &eth_stats, true);
 		stats->lan_stats.ucast_rx_pkts =
 					eth_stats.common.rx_ucast_pkts;
 		stats->lan_stats.ucast_tx_pkts =
@@ -3062,10 +3062,10 @@ void qed_get_protocol_stats(struct qed_dev *cdev,
 		stats->lan_stats.fcs_err = -1;
 		break;
 	case QED_MCP_FCOE_STATS:
-		qed_get_protocol_stats_fcoe(cdev, &stats->fcoe_stats);
+		qed_get_protocol_stats_fcoe(cdev, &stats->fcoe_stats, true);
 		break;
 	case QED_MCP_ISCSI_STATS:
-		qed_get_protocol_stats_iscsi(cdev, &stats->iscsi_stats);
+		qed_get_protocol_stats_iscsi(cdev, &stats->iscsi_stats, true);
 		break;
 	default:
 		DP_VERBOSE(cdev, QED_MSG_SP,
Signature · 1 line
2.40.1
author maintainer
#27 lore raw

PATCH 25/92 net: annotate data-races around sk->sk_max_pacing_rate

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit ea7f45ef77b39e72244d282e47f6cb1ef4135cd2 ]

sk_getsockopt() runs locklessly. This means sk->sk_max_pacing_rate can be read while other threads are changing its value.

Fixes 62748f32d501 ("net: introduce SO_MAX_PACING_RATE")
---
 net/core/sock.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
net/core/sock.c +6 -3
index cf1e437ae4875..d392efa0d9551 100644
@@ -1297,7 +1297,8 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
 			cmpxchg(&sk->sk_pacing_status,
 				SK_PACING_NONE,
 				SK_PACING_NEEDED);
-		sk->sk_max_pacing_rate = ulval;
+		/* Pairs with READ_ONCE() from sk_getsockopt() */
+		WRITE_ONCE(sk->sk_max_pacing_rate, ulval);
 		sk->sk_pacing_rate = min(sk->sk_pacing_rate, ulval);
 		break;
 		}
@@ -1680,12 +1681,14 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 #endif
 
 	case SO_MAX_PACING_RATE:
+		/* The READ_ONCE() pair with the WRITE_ONCE() in sk_setsockopt() */
 		if (sizeof(v.ulval) != sizeof(v.val) && len >= sizeof(v.ulval)) {
 			lv = sizeof(v.ulval);
-			v.ulval = sk->sk_max_pacing_rate;
+			v.ulval = READ_ONCE(sk->sk_max_pacing_rate);
 		} else {
 			/* 32bit version */
-			v.val = min_t(unsigned long, sk->sk_max_pacing_rate, ~0U);
+			v.val = min_t(unsigned long, ~0U,
+				      READ_ONCE(sk->sk_max_pacing_rate));
 		}
 		break;
 
Signature · 1 line
2.40.1
author maintainer
#28 lore raw

PATCH 26/92 net: add missing READ_ONCE(sk->sk_rcvlowat) annotation

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit e6d12bdb435d23ff6c1890c852d85408a2f496ee ]

In a prior commit, I forgot to change sk_getsockopt() when reading sk->sk_rcvlowat locklessly.

Fixes eac66402d1c3 ("net: annotate sk->sk_rcvlowat lockless reads")
---
 net/core/sock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/core/sock.c +1 -1
index d392efa0d9551..ffd566d52eec3 100644
@@ -1549,7 +1549,7 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 		break;
 
 	case SO_RCVLOWAT:
-		v.val = sk->sk_rcvlowat;
+		v.val = READ_ONCE(sk->sk_rcvlowat);
 		break;
 
 	case SO_SNDLOWAT:
Signature · 1 line
2.40.1
author maintainer
#29 lore raw

PATCH 27/92 net: add missing READ_ONCE(sk->sk_sndbuf) annotation

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit 74bc084327c643499474ba75df485607da37dd6e ]

In a prior commit, I forgot to change sk_getsockopt() when reading sk->sk_sndbuf locklessly.

Fixes e292f05e0df7 ("tcp: annotate sk->sk_sndbuf lockless reads")
---
 net/core/sock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/core/sock.c +1 -1
index ffd566d52eec3..0e480ac6c4de5 100644
@@ -1456,7 +1456,7 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 		break;
 
 	case SO_SNDBUF:
-		v.val = sk->sk_sndbuf;
+		v.val = READ_ONCE(sk->sk_sndbuf);
 		break;
 
 	case SO_RCVBUF:
Signature · 1 line
2.40.1
author maintainer
#30 lore raw

PATCH 28/92 net: add missing READ_ONCE(sk->sk_rcvbuf) annotation

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit b4b553253091cafe9ec38994acf42795e073bef5 ]

In a prior commit, I forgot to change sk_getsockopt() when reading sk->sk_rcvbuf locklessly.

Fixes ebb3b78db7bf ("tcp: annotate sk->sk_rcvbuf lockless reads")
---
 net/core/sock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/core/sock.c +1 -1
index 0e480ac6c4de5..59abfb2c92266 100644
@@ -1460,7 +1460,7 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 		break;
 
 	case SO_RCVBUF:
-		v.val = sk->sk_rcvbuf;
+		v.val = READ_ONCE(sk->sk_rcvbuf);
 		break;
 
 	case SO_REUSEADDR:
Signature · 1 line
2.40.1
author maintainer
#31 lore raw

PATCH 29/92 net: add missing data-race annotations around sk->sk_peek_off

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit 11695c6e966b0ec7ed1d16777d294cef865a5c91 ]

sk_getsockopt() runs locklessly, thus we need to annotate the read of sk->sk_peek_off.

While we are at it, add corresponding annotations to sk_set_peek_off() and unix_set_peek_off().

Fixes b9bb53f3836f ("sock: convert sk_peek_offset functions to WRITE_ONCE")
Cc Willem de Bruijn <[email protected]>
---
 net/core/sock.c    | 4 ++--
 net/unix/af_unix.c | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)
net/core/sock.c +2 -2
index 59abfb2c92266..0cca1ccf14990 100644
@@ -1643,7 +1643,7 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 		if (!sock->ops->set_peek_off)
 			return -EOPNOTSUPP;
 
-		v.val = sk->sk_peek_off;
+		v.val = READ_ONCE(sk->sk_peek_off);
 		break;
 	case SO_NOFCS:
 		v.val = sock_flag(sk, SOCK_NOFCS);
@@ -2901,7 +2901,7 @@ EXPORT_SYMBOL(__sk_mem_reclaim);
 
 int sk_set_peek_off(struct sock *sk, int val)
 {
-	sk->sk_peek_off = val;
+	WRITE_ONCE(sk->sk_peek_off, val);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(sk_set_peek_off);
net/unix/af_unix.c +1 -1
index d326540e4938c..7a076d5017d1c 100644
@@ -717,7 +717,7 @@ static int unix_set_peek_off(struct sock *sk, int val)
 	if (mutex_lock_interruptible(&u->iolock))
 		return -EINTR;
 
-	sk->sk_peek_off = val;
+	WRITE_ONCE(sk->sk_peek_off, val);
 	mutex_unlock(&u->iolock);
 
 	return 0;
Signature · 1 line
2.40.1
author maintainer
#32 lore raw

PATCH 30/92 net: add missing data-race annotation for sk_ll_usec

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit e5f0d2dd3c2faa671711dac6d3ff3cef307bcfe3 ]

In a prior commit I forgot that sk_getsockopt() reads sk->sk_ll_usec without holding a lock.

Fixes 0dbffbb5335a ("net: annotate data race around sk_ll_usec")
---
 net/core/sock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/core/sock.c +1 -1
index 0cca1ccf14990..1f9401d757cbb 100644
@@ -1673,7 +1673,7 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	case SO_BUSY_POLL:
-		v.val = sk->sk_ll_usec;
+		v.val = READ_ONCE(sk->sk_ll_usec);
 		break;
 	case SO_PREFER_BUSY_POLL:
 		v.val = READ_ONCE(sk->sk_prefer_busy_poll);
Signature · 1 line
2.40.1
author maintainer
#33 lore raw

PATCH 31/92 net/sched: taprio: Limit TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME to INT_MAX.

In reply to Greg Kroah-Hartman

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit e739718444f7bf2fa3d70d101761ad83056ca628 ]

syzkaller found zero division error [0] in div_s64_rem() called from get_cycle_time_elapsed(), where sched->cycle_time is the divisor.

We have tests in parse_taprio_schedule() so that cycle_time will never be 0, and actually cycle_time is not 0 in get_cycle_time_elapsed().

The problem is that the types of divisor are different; cycle_time is s64, but the argument of div_s64_rem() is s32.

syzkaller fed this input and 0x100000000 is cast to s32 to be 0.

@TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME={0xc, 0x8, 0x100000000}

We use s64 for cycle_time to cast it to ktime_t, so let's keep it and set max for cycle_time.

While at it, we prevent overflow in setup_txtime() and add another test in parse_taprio_schedule() to check if cycle_time overflows.

Also, we add a new tdc test case for this issue.

[0]:
divide error: 0000 [#1] PREEMPT SMP KASAN NOPTI

CPU: 1 PID: 103 Comm: kworker/1:3 Not tainted 6.5.0-rc1-00330-g60cc1f7d0605 #3
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
Workqueue: ipv6_addrconf addrconf_dad_work
RIP: 0010:div_s64_rem include/linux/math64.h:42 [inline]
RIP: 0010:get_cycle_time_elapsed net/sched/sch_taprio.c:223 [inline]
RIP: 0010:find_entry_to_transmit+0x252/0x7e0 net/sched/sch_taprio.c:344
Code: 3c 02 00 0f 85 5e 05 00 00 48 8b 4c 24 08 4d 8b bd 40 01 00 00 48 8b 7c 24 48 48 89 c8 4c 29 f8 48 63 f7 48 99 48 89 74 24 70 <48> f7 fe 48 29 d1 48 8d 04 0f 49 89 cc 48 89 44 24 20 49 8d 85 10
RSP: 0018:ffffc90000acf260 EFLAGS: 00010206
RAX: 177450e0347560cf RBX: 0000000000000000 RCX: 177450e0347560cf
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000100000000
RBP: 0000000000000056 R08: 0000000000000000 R09: ffffed10020a0934
R10: ffff8880105049a7 R11: ffff88806cf3a520 R12: ffff888010504800
R13: ffff88800c00d800 R14: ffff8880105049a0 R15: 0000000000000000
FS:  0000000000000000(0000) GS:ffff88806cf00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f0edf84f0e8 CR3: 000000000d73c002 CR4: 0000000000770ee0
PKRU: 55555554
Call Trace:
 <TASK>
 get_packet_txtime net/sched/sch_taprio.c:508 [inline]
 taprio_enqueue_one+0x900/0xff0 net/sched/sch_taprio.c:577
 taprio_enqueue+0x378/0xae0 net/sched/sch_taprio.c:658
 dev_qdisc_enqueue+0x46/0x170 net/core/dev.c:3732
 __dev_xmit_skb net/core/dev.c:3821 [inline]
 __dev_queue_xmit+0x1b2f/0x3000 net/core/dev.c:4169
 dev_queue_xmit include/linux/netdevice.h:3088 [inline]
 neigh_resolve_output net/core/neighbour.c:1552 [inline]
 neigh_resolve_output+0x4a7/0x780 net/core/neighbour.c:1532
 neigh_output include/net/neighbour.h:544 [inline]
 ip6_finish_output2+0x924/0x17d0 net/ipv6/ip6_output.c:135
 __ip6_finish_output+0x620/0xaa0 net/ipv6/ip6_output.c:196
 ip6_finish_output net/ipv6/ip6_output.c:207 [inline]
 NF_HOOK_COND include/linux/netfilter.h:292 [inline]
 ip6_output+0x206/0x410 net/ipv6/ip6_output.c:228
 dst_output include/net/dst.h:458 [inline]
 NF_HOOK.constprop.0+0xea/0x260 include/linux/netfilter.h:303
 ndisc_send_skb+0x872/0xe80 net/ipv6/ndisc.c:508
 ndisc_send_ns+0xb5/0x130 net/ipv6/ndisc.c:666
 addrconf_dad_work+0xc14/0x13f0 net/ipv6/addrconf.c:4175
 process_one_work+0x92c/0x13a0 kernel/workqueue.c:2597
 worker_thread+0x60f/0x1240 kernel/workqueue.c:2748
 kthread+0x2fe/0x3f0 kernel/kthread.c:389
 ret_from_fork+0x2c/0x50 arch/x86/entry/entry_64.S:308
 </TASK>
Modules linked in:
---
 net/sched/sch_taprio.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)
net/sched/sch_taprio.c +13 -2
index e203deacc9533..e40b4425eb6b5 100644
@@ -780,6 +780,11 @@ static const struct nla_policy entry_policy[TCA_TAPRIO_SCHED_ENTRY_MAX + 1] = {
 	[TCA_TAPRIO_SCHED_ENTRY_INTERVAL]  = { .type = NLA_U32 },
 };
 
+static struct netlink_range_validation_signed taprio_cycle_time_range = {
+	.min = 0,
+	.max = INT_MAX,
+};
+
 static const struct nla_policy taprio_policy[TCA_TAPRIO_ATTR_MAX + 1] = {
 	[TCA_TAPRIO_ATTR_PRIOMAP]	       = {
 		.len = sizeof(struct tc_mqprio_qopt)
@@ -788,7 +793,8 @@ static const struct nla_policy taprio_policy[TCA_TAPRIO_ATTR_MAX + 1] = {
 	[TCA_TAPRIO_ATTR_SCHED_BASE_TIME]            = { .type = NLA_S64 },
 	[TCA_TAPRIO_ATTR_SCHED_SINGLE_ENTRY]         = { .type = NLA_NESTED },
 	[TCA_TAPRIO_ATTR_SCHED_CLOCKID]              = { .type = NLA_S32 },
-	[TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME]           = { .type = NLA_S64 },
+	[TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME]           =
+		NLA_POLICY_FULL_RANGE_SIGNED(NLA_S64, &taprio_cycle_time_range),
 	[TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME_EXTENSION] = { .type = NLA_S64 },
 	[TCA_TAPRIO_ATTR_FLAGS]                      = { .type = NLA_U32 },
 	[TCA_TAPRIO_ATTR_TXTIME_DELAY]		     = { .type = NLA_U32 },
@@ -923,6 +929,11 @@ static int parse_taprio_schedule(struct taprio_sched *q, struct nlattr **tb,
 			return -EINVAL;
 		}
 
+		if (cycle < 0 || cycle > INT_MAX) {
+			NL_SET_ERR_MSG(extack, "'cycle_time' is too big");
+			return -EINVAL;
+		}
+
 		new->cycle_time = cycle;
 	}
 
@@ -1127,7 +1138,7 @@ static void setup_txtime(struct taprio_sched *q,
 			 struct sched_gate_list *sched, ktime_t base)
 {
 	struct sched_entry *entry;
-	u32 interval = 0;
+	u64 interval = 0;
 
 	list_for_each_entry(entry, &sched->entries, list) {
 		entry->next_txtime = ktime_add_ns(base, interval);
Signature · 1 line
2.40.1
author maintainer
#34 lore raw

PATCH 32/92 bpf, cpumap: Handle skb as well when clean up ptr_ring

In reply to Greg Kroah-Hartman

From: Hou Tao <[email protected]>

[ Upstream commit 7c62b75cd1a792e14b037fa4f61f9b18914e7de1 ]

The following warning was reported when running xdp_redirect_cpu with both skb-mode and stress-mode enabled:

  ------------[ cut here ]------------
  Incorrect XDP memory type (-2128176192) usage
  WARNING: CPU: 7 PID: 1442 at net/core/xdp.c:405
  Modules linked in:
  CPU: 7 PID: 1442 Comm: kworker/7:0 Tainted: G  6.5.0-rc2+ #1
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
  Workqueue: events __cpu_map_entry_free
  RIP: 0010:__xdp_return+0x1e4/0x4a0
  ......
  Call Trace:
   <TASK>
   ? show_regs+0x65/0x70
   ? __warn+0xa5/0x240
   ? __xdp_return+0x1e4/0x4a0
   ......
   xdp_return_frame+0x4d/0x150
   __cpu_map_entry_free+0xf9/0x230
   process_one_work+0x6b0/0xb80
   worker_thread+0x96/0x720
   kthread+0x1a5/0x1f0
   ret_from_fork+0x3a/0x70
   ret_from_fork_asm+0x1b/0x30
   </TASK>

The reason for the warning is twofold. One is due to the kthread cpu_map_kthread_run() is stopped prematurely. Another one is __cpu_map_ring_cleanup() doesn't handle skb mode and treats skbs in ptr_ring as XDP frames.

Prematurely-stopped kthread will be fixed by the preceding patch and ptr_ring will be empty when __cpu_map_ring_cleanup() is called. But as the comments in __cpu_map_ring_cleanup() said, handling and freeing skbs in ptr_ring as well to "catch any broken behaviour gracefully".

---
 kernel/bpf/cpumap.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)
kernel/bpf/cpumap.c +10 -4
index db6221773e43f..17f52efba1c80 100644
@@ -133,11 +133,17 @@ static void __cpu_map_ring_cleanup(struct ptr_ring *ring)
 	 * invoked cpu_map_kthread_stop(). Catch any broken behaviour
 	 * gracefully and warn once.
 	 */
-	struct xdp_frame *xdpf;
+	void *ptr;
 
-	while ((xdpf = ptr_ring_consume(ring)))
-		if (WARN_ON_ONCE(xdpf))
-			xdp_return_frame(xdpf);
+	while ((ptr = ptr_ring_consume(ring))) {
+		WARN_ON_ONCE(1);
+		if (unlikely(__ptr_test_bit(0, &ptr))) {
+			__ptr_clear_bit(0, &ptr);
+			kfree_skb(ptr);
+			continue;
+		}
+		xdp_return_frame(ptr);
+	}
 }
 
 static void put_cpu_map_entry(struct bpf_cpu_map_entry *rcpu)
Signature · 1 line
2.40.1
author maintainer
#35 lore raw

PATCH 33/92 net/sched: cls_u32: No longer copy tcf_result on update to avoid use-after-free

In reply to Greg Kroah-Hartman

From: valis <[email protected]>

[ Upstream commit 3044b16e7c6fe5d24b1cdbcf1bd0a9d92d1ebd81 ]

When u32_change() is called on an existing filter, the whole tcf_result struct is always copied into the new instance of the filter.

This causes a problem when updating a filter bound to a class, as tcf_unbind_filter() is always called on the old instance in the success path, decreasing filter_cnt of the still referenced class and allowing it to be deleted, leading to a use-after-free.

Fix this by no longer copying the tcf_result struct from the old filter.

---
 net/sched/cls_u32.c | 1 -
 1 file changed, 1 deletion(-)
net/sched/cls_u32.c +0 -1
index e4e38aa75111e..17edcf1d1c3b6 100644
@@ -812,7 +812,6 @@ static struct tc_u_knode *u32_init_knode(struct net *net, struct tcf_proto *tp,
 
 	new->ifindex = n->ifindex;
 	new->fshift = n->fshift;
-	new->res = n->res;
 	new->flags = n->flags;
 	RCU_INIT_POINTER(new->ht_down, ht);
 
Signature · 1 line
2.40.1
author maintainer
#36 lore raw

PATCH 34/92 net/sched: cls_fw: No longer copy tcf_result on update to avoid use-after-free

In reply to Greg Kroah-Hartman

From: valis <[email protected]>

[ Upstream commit 76e42ae831991c828cffa8c37736ebfb831ad5ec ]

When fw_change() is called on an existing filter, the whole tcf_result struct is always copied into the new instance of the filter.

This causes a problem when updating a filter bound to a class, as tcf_unbind_filter() is always called on the old instance in the success path, decreasing filter_cnt of the still referenced class and allowing it to be deleted, leading to a use-after-free.

Fix this by no longer copying the tcf_result struct from the old filter.

---
 net/sched/cls_fw.c | 1 -
 1 file changed, 1 deletion(-)
net/sched/cls_fw.c +0 -1
index ea52c320f67c4..a2f53aee39097 100644
@@ -265,7 +265,6 @@ static int fw_change(struct net *net, struct sk_buff *in_skb,
 			return -ENOBUFS;
 
 		fnew->id = f->id;
-		fnew->res = f->res;
 		fnew->ifindex = f->ifindex;
 		fnew->tp = f->tp;
 
Signature · 1 line
2.40.1
author maintainer
#37 lore raw

PATCH 35/92 net/sched: cls_route: No longer copy tcf_result on update to avoid use-after-free

In reply to Greg Kroah-Hartman

From: valis <[email protected]>

[ Upstream commit b80b829e9e2c1b3f7aae34855e04d8f6ecaf13c8 ]

When route4_change() is called on an existing filter, the whole tcf_result struct is always copied into the new instance of the filter.

This causes a problem when updating a filter bound to a class, as tcf_unbind_filter() is always called on the old instance in the success path, decreasing filter_cnt of the still referenced class and allowing it to be deleted, leading to a use-after-free.

Fix this by no longer copying the tcf_result struct from the old filter.

---
 net/sched/cls_route.c | 1 -
 1 file changed, 1 deletion(-)
net/sched/cls_route.c +0 -1
index 48712bc51bda7..194468d0355a1 100644
@@ -511,7 +511,6 @@ static int route4_change(struct net *net, struct sk_buff *in_skb,
 	if (fold) {
 		f->id = fold->id;
 		f->iif = fold->iif;
-		f->res = fold->res;
 		f->handle = fold->handle;
 
 		f->tp = fold->tp;
Signature · 1 line
2.40.1
author maintainer
#38 lore raw

PATCH 36/92 bpf: sockmap: Remove preempt_disable in sock_map_sk_acquire

In reply to Greg Kroah-Hartman

From: Tomas Glozar <[email protected]>

[ Upstream commit 13d2618b48f15966d1adfe1ff6a1985f5eef40ba ]

Disabling preemption in sock_map_sk_acquire conflicts with GFP_ATOMIC allocation later in sk_psock_init_link on PREEMPT_RT kernels, since GFP_ATOMIC might sleep on RT (see bpf: Make BPF and PREEMPT_RT co-exist patchset notes for details).

This causes calling bpf_map_update_elem on BPF_MAP_TYPE_SOCKMAP maps to BUG (sleeping function called from invalid context) on RT kernels.

preempt_disable was introduced together with lock_sk and rcu_read_lock in commit 99ba2b5aba24e ("bpf: sockhash, disallow bpf_tcp_close and update in parallel"), probably to match disabled migration of BPF programs, and is no longer necessary.

Remove preempt_disable to fix BUG in sock_map_update_common on RT.

---
 net/core/sock_map.c | 2 --
 1 file changed, 2 deletions(-)
net/core/sock_map.c +0 -2
index 86b4e8909ad1e..85d3c62bdfa6a 100644
@@ -117,7 +117,6 @@ static void sock_map_sk_acquire(struct sock *sk)
 	__acquires(&sk->sk_lock.slock)
 {
 	lock_sock(sk);
-	preempt_disable();
 	rcu_read_lock();
 }
 
@@ -125,7 +124,6 @@ static void sock_map_sk_release(struct sock *sk)
 	__releases(&sk->sk_lock.slock)
 {
 	rcu_read_unlock();
-	preempt_enable();
 	release_sock(sk);
 }
 
Signature · 1 line
2.40.1
author maintainer
#39 lore raw

PATCH 37/92 net: ll_temac: Switch to use dev_err_probe() helper

In reply to Greg Kroah-Hartman

From: Yang Yingliang <[email protected]>

[ Upstream commit 75ae8c284c00dc3584b7c173f6fcf96ee15bd02c ]

dev_err() can be replace with dev_err_probe() which will check if error code is -EPROBE_DEFER.

Stable-dep-of: ef45e8400f5b ("net: ll_temac: fix error checking of irq_of_parse_and_map()")

---
 drivers/net/ethernet/xilinx/ll_temac_main.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)
drivers/net/ethernet/xilinx/ll_temac_main.c +6 -10
index 2ab29efa6b6e4..303de9293fc71 100644
@@ -1556,16 +1556,12 @@ static int temac_probe(struct platform_device *pdev)
 	}
 
 	/* Error handle returned DMA RX and TX interrupts */
-	if (lp->rx_irq < 0) {
-		if (lp->rx_irq != -EPROBE_DEFER)
-			dev_err(&pdev->dev, "could not get DMA RX irq\n");
-		return lp->rx_irq;
-	}
-	if (lp->tx_irq < 0) {
-		if (lp->tx_irq != -EPROBE_DEFER)
-			dev_err(&pdev->dev, "could not get DMA TX irq\n");
-		return lp->tx_irq;
-	}
+	if (lp->rx_irq < 0)
+		return dev_err_probe(&pdev->dev, lp->rx_irq,
+				     "could not get DMA RX irq\n");
+	if (lp->tx_irq < 0)
+		return dev_err_probe(&pdev->dev, lp->tx_irq,
+				     "could not get DMA TX irq\n");
 
 	if (temac_np) {
 		/* Retrieve the MAC address */
Signature · 1 line
2.40.1
author maintainer
#40 lore raw

PATCH 38/92 net: ll_temac: fix error checking of irq_of_parse_and_map()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit ef45e8400f5bb66b03cc949f76c80e2a118447de ]

Most kernel functions return negative error codes but some irq functions return zero on error. In this code irq_of_parse_and_map(), returns zero and platform_get_irq() returns negative error codes. We need to handle both cases appropriately.

---
 drivers/net/ethernet/xilinx/ll_temac_main.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
drivers/net/ethernet/xilinx/ll_temac_main.c +8 -4
index 303de9293fc71..b4db50c9e7038 100644
@@ -1556,12 +1556,16 @@ static int temac_probe(struct platform_device *pdev)
 	}
 
 	/* Error handle returned DMA RX and TX interrupts */
-	if (lp->rx_irq < 0)
-		return dev_err_probe(&pdev->dev, lp->rx_irq,
+	if (lp->rx_irq <= 0) {
+		rc = lp->rx_irq ?: -EINVAL;
+		return dev_err_probe(&pdev->dev, rc,
 				     "could not get DMA RX irq\n");
-	if (lp->tx_irq < 0)
-		return dev_err_probe(&pdev->dev, lp->tx_irq,
+	}
+	if (lp->tx_irq <= 0) {
+		rc = lp->tx_irq ?: -EINVAL;
+		return dev_err_probe(&pdev->dev, rc,
 				     "could not get DMA TX irq\n");
+	}
 
 	if (temac_np) {
 		/* Retrieve the MAC address */
Signature · 1 line
2.40.1
author maintainer
#41 lore raw

PATCH 39/92 net: korina: handle clk prepare error in korina_probe()

In reply to Greg Kroah-Hartman

From: Yuanjun Gong <[email protected]>

[ Upstream commit 0b6291ad1940c403734312d0e453e8dac9148f69 ]

in korina_probe(), the return value of clk_prepare_enable() should be checked since it might fail. we can use devm_clk_get_optional_enabled() instead of devm_clk_get_optional() and clk_prepare_enable() to automatically handle the error.

Fixes e4cd854ec487 ("net: korina: Get mdio input clock via common clock framework")
---
 drivers/net/ethernet/korina.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/ethernet/korina.c +1 -2
index df9a8eefa007a..916e85039b610 100644
@@ -1301,11 +1301,10 @@ static int korina_probe(struct platform_device *pdev)
 	else if (of_get_ethdev_address(pdev->dev.of_node, dev) < 0)
 		eth_hw_addr_random(dev);
 
-	clk = devm_clk_get_optional(&pdev->dev, "mdioclk");
+	clk = devm_clk_get_optional_enabled(&pdev->dev, "mdioclk");
 	if (IS_ERR(clk))
 		return PTR_ERR(clk);
 	if (clk) {
-		clk_prepare_enable(clk);
 		lp->mii_clock_freq = clk_get_rate(clk);
 	} else {
 		lp->mii_clock_freq = 200000000; /* max possible input clk */
Signature · 1 line
2.40.1
author maintainer
#42 lore raw

PATCH 40/92 net: netsec: Ignore phy-mode on SynQuacer in DT mode

In reply to Greg Kroah-Hartman

From: Mark Brown <[email protected]>

[ Upstream commit f3bb7759a924713bc54d15f6d0d70733b5935fad ]

As documented in acd7aaf51b20 ("netsec: ignore 'phy-mode' device property on ACPI systems") the SocioNext SynQuacer platform ships with firmware defining the PHY mode as RGMII even though the physical configuration of the PHY is for TX and RX delays. Since bbc4d71d63549bc ("net: phy: realtek: fix rtl8211e rx/tx delay config") this has caused misconfiguration of the PHY, rendering the network unusable.

This was worked around for ACPI by ignoring the phy-mode property but the system is also used with DT. For DT instead if we're running on a SynQuacer force a working PHY mode, as well as the standard EDK2 firmware with DT there are also some of these systems that use u-boot and might not initialise the PHY if not netbooting. Newer firmware imagaes for at least EDK2 are available from Linaro so print a warning when doing this.

---
 drivers/net/ethernet/socionext/netsec.c | 11 +++++++++++
 1 file changed, 11 insertions(+)
drivers/net/ethernet/socionext/netsec.c +11 -0
index 6b8013fb17c38..eb59e8abe6915 100644
@@ -1851,6 +1851,17 @@ static int netsec_of_probe(struct platform_device *pdev,
 		return err;
 	}
 
+	/*
+	 * SynQuacer is physically configured with TX and RX delays
+	 * but the standard firmware claimed otherwise for a long
+	 * time, ignore it.
+	 */
+	if (of_machine_is_compatible("socionext,developer-box") &&
+	    priv->phy_interface != PHY_INTERFACE_MODE_RGMII_ID) {
+		dev_warn(&pdev->dev, "Outdated firmware reports incorrect PHY mode, overriding\n");
+		priv->phy_interface = PHY_INTERFACE_MODE_RGMII_ID;
+	}
+
 	priv->phy_np = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
 	if (!priv->phy_np) {
 		dev_err(&pdev->dev, "missing required property 'phy-handle'\n");
Signature · 1 line
2.40.1
author maintainer
#43 lore raw

PATCH 41/92 net: dcb: choose correct policy to parse DCB_ATTR_BCN

In reply to Greg Kroah-Hartman

From: Lin Ma <[email protected]>

[ Upstream commit 31d49ba033095f6e8158c60f69714a500922e0c3 ]

The dcbnl_bcn_setcfg uses erroneous policy to parse tb[DCB_ATTR_BCN], which is introduced in commit 859ee3c43812 ("DCB: Add support for DCB BCN"). Please see the comment in below code

static int dcbnl_bcn_setcfg(...)
{
...
ret = nla_parse_nested_deprecated(..., dcbnl_pfc_up_nest, .. ) // !!! dcbnl_pfc_up_nest for attributes
// DCB_PFC_UP_ATTR_0 to DCB_PFC_UP_ATTR_ALL in enum dcbnl_pfc_up_attrs
...
for (i = DCB_BCN_ATTR_RP_0; i <= DCB_BCN_ATTR_RP_7; i++) { // !!! DCB_BCN_ATTR_RP_0 to DCB_BCN_ATTR_RP_7 in enum dcbnl_bcn_attrs

    ...
    value_byte = nla_get_u8(data[i]);
    ...
  }
  ...
  for (i = DCB_BCN_ATTR_BCNA_0; i <= DCB_BCN_ATTR_RI; i++) {
  // !!! DCB_BCN_ATTR_BCNA_0 to DCB_BCN_ATTR_RI in enum dcbnl_bcn_attrs
  ...
    value_int = nla_get_u32(data[i]);

...
}
...
}

That is, the nla_parse_nested_deprecated uses dcbnl_pfc_up_nest attributes to parse nlattr defined in dcbnl_pfc_up_attrs. But the following access code fetch each nlattr as dcbnl_bcn_attrs attributes. By looking up the associated nla_policy for dcbnl_bcn_attrs. We can find the beginning part of these two policies are "same".

static const struct nla_policy dcbnl_pfc_up_nest[...] = {
        [DCB_PFC_UP_ATTR_0]   = {.type = NLA_U8},
        [DCB_PFC_UP_ATTR_1]   = {.type = NLA_U8},
        [DCB_PFC_UP_ATTR_2]   = {.type = NLA_U8},
        [DCB_PFC_UP_ATTR_3]   = {.type = NLA_U8},
        [DCB_PFC_UP_ATTR_4]   = {.type = NLA_U8},
        [DCB_PFC_UP_ATTR_5]   = {.type = NLA_U8},
        [DCB_PFC_UP_ATTR_6]   = {.type = NLA_U8},
        [DCB_PFC_UP_ATTR_7]   = {.type = NLA_U8},
        [DCB_PFC_UP_ATTR_ALL] = {.type = NLA_FLAG},
};
static const struct nla_policy dcbnl_bcn_nest[...] = {
        [DCB_BCN_ATTR_RP_0]         = {.type = NLA_U8},
        [DCB_BCN_ATTR_RP_1]         = {.type = NLA_U8},
        [DCB_BCN_ATTR_RP_2]         = {.type = NLA_U8},
        [DCB_BCN_ATTR_RP_3]         = {.type = NLA_U8},
        [DCB_BCN_ATTR_RP_4]         = {.type = NLA_U8},
        [DCB_BCN_ATTR_RP_5]         = {.type = NLA_U8},
        [DCB_BCN_ATTR_RP_6]         = {.type = NLA_U8},
        [DCB_BCN_ATTR_RP_7]         = {.type = NLA_U8},
        [DCB_BCN_ATTR_RP_ALL]       = {.type = NLA_FLAG},
        // from here is somewhat different
        [DCB_BCN_ATTR_BCNA_0]       = {.type = NLA_U32},
        ...
        [DCB_BCN_ATTR_ALL]          = {.type = NLA_FLAG},
};

Therefore, the current code is buggy and this nla_parse_nested_deprecated could overflow the dcbnl_pfc_up_nest and use the adjacent nla_policy to parse attributes from DCB_BCN_ATTR_BCNA_0.

Hence use the correct policy dcbnl_bcn_nest to parse the nested tb[DCB_ATTR_BCN] TLV.

---
 net/dcb/dcbnl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/dcb/dcbnl.c +1 -1
index dc4fb699b56c3..d2981e89d3638 100644
@@ -946,7 +946,7 @@ static int dcbnl_bcn_setcfg(struct net_device *netdev, struct nlmsghdr *nlh,
 		return -EOPNOTSUPP;
 
 	ret = nla_parse_nested_deprecated(data, DCB_BCN_ATTR_MAX,
-					  tb[DCB_ATTR_BCN], dcbnl_pfc_up_nest,
+					  tb[DCB_ATTR_BCN], dcbnl_bcn_nest,
 					  NULL);
 	if (ret)
 		return ret;
Signature · 1 line
2.40.1
author maintainer
#44 lore raw

PATCH 42/92 s390/qeth: Dont call dev_close/dev_open (DOWN/UP)

In reply to Greg Kroah-Hartman

From: Alexandra Winter <[email protected]>

[ Upstream commit 1cfef80d4c2b2c599189f36f36320b205d9447d9 ]

dev_close() and dev_open() are issued to change the interface state to DOWN or UP (dev->flags IFF_UP). When the netdev is set DOWN it loses e.g its Ipv6 addresses and routes. We don't want this in cases of device recovery (triggered by hardware or software) or when the qeth device is set offline.

Setting a qeth device offline or online and device recovery actions call netif_device_detach() and/or netif_device_attach(). That will reset or set the LOWER_UP indication i.e. change the dev->state Bit __LINK_STATE_PRESENT. That is enough to e.g. cause bond failovers, and still preserves the interface settings that are handled by the network stack.

Don't call dev_open() nor dev_close() from the qeth device driver. Let the network stack handle this.

Fixes d4560150cb47 ("s390/qeth: call dev_close() during recovery")
---
 drivers/s390/net/qeth_core.h      | 1 -
 drivers/s390/net/qeth_core_main.c | 2 --
 drivers/s390/net/qeth_l2_main.c   | 9 ++++++---
 drivers/s390/net/qeth_l3_main.c   | 8 +++++---
 4 files changed, 11 insertions(+), 9 deletions(-)
drivers/s390/net/qeth_core.h +0 -1
index a5aa0bdc61d69..e8c360879883b 100644
@@ -717,7 +717,6 @@ struct qeth_card_info {
 	u16 chid;
 	u8 ids_valid:1; /* cssid,iid,chid */
 	u8 dev_addr_is_registered:1;
-	u8 open_when_online:1;
 	u8 promisc_mode:1;
 	u8 use_v1_blkt:1;
 	u8 is_vm_nic:1;
drivers/s390/net/qeth_core_main.c +0 -2
index e9807d2996a9d..62e7576bff536 100644
@@ -5459,8 +5459,6 @@ int qeth_set_offline(struct qeth_card *card, const struct qeth_discipline *disc,
 	qeth_clear_ipacmd_list(card);
 
 	rtnl_lock();
-	card->info.open_when_online = card->dev->flags & IFF_UP;
-	dev_close(card->dev);
 	netif_device_detach(card->dev);
 	netif_carrier_off(card->dev);
 	rtnl_unlock();
drivers/s390/net/qeth_l2_main.c +6 -3
index d694e3ff80865..7cdf3274cf964 100644
@@ -2373,9 +2373,12 @@ static int qeth_l2_set_online(struct qeth_card *card, bool carrier_ok)
 		qeth_enable_hw_features(dev);
 		qeth_l2_enable_brport_features(card);
 
-		if (card->info.open_when_online) {
-			card->info.open_when_online = 0;
-			dev_open(dev, NULL);
+		if (netif_running(dev)) {
+			local_bh_disable();
+			napi_schedule(&card->napi);
+			/* kick-start the NAPI softirq: */
+			local_bh_enable();
+			qeth_l2_set_rx_mode(dev);
 		}
 		rtnl_unlock();
 	}
drivers/s390/net/qeth_l3_main.c +5 -3
index 6fd3e288f0595..93f55c7348026 100644
@@ -2029,9 +2029,11 @@ static int qeth_l3_set_online(struct qeth_card *card, bool carrier_ok)
 		netif_device_attach(dev);
 		qeth_enable_hw_features(dev);
 
-		if (card->info.open_when_online) {
-			card->info.open_when_online = 0;
-			dev_open(dev, NULL);
+		if (netif_running(dev)) {
+			local_bh_disable();
+			napi_schedule(&card->napi);
+			/* kick-start the NAPI softirq: */
+			local_bh_enable();
 		}
 		rtnl_unlock();
 	}
Signature · 1 line
2.40.1
author maintainer
#45 lore raw

PATCH 43/92 ip6mr: Fix skb_under_panic in ip6mr_cache_report()

In reply to Greg Kroah-Hartman

From: Yue Haibing <[email protected]>

[ Upstream commit 30e0191b16e8a58e4620fa3e2839ddc7b9d4281c ]

skbuff: skb_under_panic: text:ffffffff88771f69 len:56 put:-4
head:ffff88805f86a800 data:ffff887f5f86a850 tail:0x88 end:0x2c0 dev:pim6reg

 ------------[ cut here ]------------
 kernel BUG at net/core/skbuff.c:192!
 invalid opcode: 0000 [#1] PREEMPT SMP KASAN
 CPU: 2 PID: 22968 Comm: kworker/2:11 Not tainted 6.5.0-rc3-00044-g0a8db05b571a #236
 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
 Workqueue: ipv6_addrconf addrconf_dad_work
 RIP: 0010:skb_panic+0x152/0x1d0
 Call Trace:
  <TASK>
  skb_push+0xc4/0xe0
  ip6mr_cache_report+0xd69/0x19b0
  reg_vif_xmit+0x406/0x690
  dev_hard_start_xmit+0x17e/0x6e0
  __dev_queue_xmit+0x2d6a/0x3d20
  vlan_dev_hard_start_xmit+0x3ab/0x5c0
  dev_hard_start_xmit+0x17e/0x6e0
  __dev_queue_xmit+0x2d6a/0x3d20
  neigh_connected_output+0x3ed/0x570
  ip6_finish_output2+0x5b5/0x1950
  ip6_finish_output+0x693/0x11c0
  ip6_output+0x24b/0x880
  NF_HOOK.constprop.0+0xfd/0x530
  ndisc_send_skb+0x9db/0x1400
  ndisc_send_rs+0x12a/0x6c0
  addrconf_dad_completed+0x3c9/0xea0
  addrconf_dad_work+0x849/0x1420
  process_one_work+0xa22/0x16e0
  worker_thread+0x679/0x10c0
  ret_from_fork+0x28/0x60
  ret_from_fork_asm+0x11/0x20

When setup a vlan device on dev pim6reg, DAD ns packet may sent on reg_vif_xmit(). reg_vif_xmit()
ip6mr_cache_report()
skb_push(skb, -skb_network_offset(pkt));//skb_network_offset(pkt) is 4
And skb_push declared as:

	void *skb_push(struct sk_buff *skb, unsigned int len);
		skb->data -= len;
		//0xffff88805f86a84c - 0xfffffffc = 0xffff887f5f86a850

skb->data is set to 0xffff887f5f86a850, which is invalid mem addr, lead to skb_push() fails.

Fixes 14fb64e1f449 ("[IPV6] MROUTE: Support PIM-SM (SSM).")
---
 net/ipv6/ip6mr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/ip6mr.c +1 -1
index 91f1c5f56d5fa..ee094645c7cea 100644
@@ -1068,7 +1068,7 @@ static int ip6mr_cache_report(struct mr_table *mrt, struct sk_buff *pkt,
 		   And all this only to mangle msg->im6_msgtype and
 		   to set msg->im6_mbz to "mbz" :-)
 		 */
-		skb_push(skb, -skb_network_offset(pkt));
+		__skb_pull(skb, skb_network_offset(pkt));
 
 		skb_push(skb, sizeof(*msg));
 		skb_reset_transport_header(skb);
Signature · 1 line
2.40.1
author maintainer
#46 lore raw

PATCH 44/92 vxlan: Fix nexthop hash size

In reply to Greg Kroah-Hartman

From: Benjamin Poirier <[email protected]>

[ Upstream commit 0756384fb1bd38adb2ebcfd1307422f433a1d772 ]

The nexthop code expects a 31 bit hash, such as what is returned by fib_multipath_hash() and rt6_multipath_hash(). Passing the 32 bit hash returned by skb_get_hash() can lead to problems related to the fact that 'int hash' is a negative number when the MSB is set.

In the case of hash threshold nexthop groups, nexthop_select_path_hthr() will disproportionately select the first nexthop group entry. In the case of resilient nexthop groups, nexthop_select_path_res() may do an out of bounds access in nh_buckets[], for example:

    hash = -912054133
    num_nh_buckets = 2
    bucket_index = 65535

which leads to the following panic:

BUG: unable to handle page fault for address: ffffc900025910c8
PGD 100000067 P4D 100000067 PUD 10026b067 PMD 0
Oops: 0002 [#1] PREEMPT SMP KASAN NOPTI
CPU: 4 PID: 856 Comm: kworker/4:3 Not tainted 6.5.0-rc2+ #34
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-debian-1.16.2-1 04/01/2014
Workqueue: ipv6_addrconf addrconf_dad_work
RIP: 0010:nexthop_select_path+0x197/0xbf0
Code: c1 e4 05 be 08 00 00 00 4c 8b 35 a4 14 7e 01 4e 8d 6c 25 00 4a 8d 7c 25 08 48 01 dd e8 c2 25 15 ff 49 8d 7d 08 e8 39 13 15 ff <4d> 89 75 08 48 89 ef e8 7d 12 15 ff 48 8b 5d 00 e8 14 55 2f 00 85
RSP: 0018:ffff88810c36f260 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 00000000002000c0 RCX: ffffffffaf02dd77
RDX: dffffc0000000000 RSI: 0000000000000008 RDI: ffffc900025910c8
RBP: ffffc900025910c0 R08: 0000000000000001 R09: fffff520004b2219
R10: ffffc900025910cf R11: 31392d2068736168 R12: 00000000002000c0
R13: ffffc900025910c0 R14: 00000000fffef608 R15: ffff88811840e900
FS:  0000000000000000(0000) GS:ffff8881f7000000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffffc900025910c8 CR3: 0000000129d00000 CR4: 0000000000750ee0
PKRU: 55555554
Call Trace:
 <TASK>
 ? __die+0x23/0x70
 ? page_fault_oops+0x1ee/0x5c0
 ? __pfx_is_prefetch.constprop.0+0x10/0x10
 ? __pfx_page_fault_oops+0x10/0x10
 ? search_bpf_extables+0xfe/0x1c0
 ? fixup_exception+0x3b/0x470
 ? exc_page_fault+0xf6/0x110
 ? asm_exc_page_fault+0x26/0x30
 ? nexthop_select_path+0x197/0xbf0
 ? nexthop_select_path+0x197/0xbf0
 ? lock_is_held_type+0xe7/0x140
 vxlan_xmit+0x5b2/0x2340
 ? __lock_acquire+0x92b/0x3370
 ? __pfx_vxlan_xmit+0x10/0x10
 ? __pfx___lock_acquire+0x10/0x10
 ? __pfx_register_lock_class+0x10/0x10
 ? skb_network_protocol+0xce/0x2d0
 ? dev_hard_start_xmit+0xca/0x350
 ? __pfx_vxlan_xmit+0x10/0x10
 dev_hard_start_xmit+0xca/0x350
 __dev_queue_xmit+0x513/0x1e20
 ? __pfx___dev_queue_xmit+0x10/0x10
 ? __pfx_lock_release+0x10/0x10
 ? mark_held_locks+0x44/0x90
 ? skb_push+0x4c/0x80
 ? eth_header+0x81/0xe0
 ? __pfx_eth_header+0x10/0x10
 ? neigh_resolve_output+0x215/0x310
 ? ip6_finish_output2+0x2ba/0xc90
 ip6_finish_output2+0x2ba/0xc90
 ? lock_release+0x236/0x3e0
 ? ip6_mtu+0xbb/0x240
 ? __pfx_ip6_finish_output2+0x10/0x10
 ? find_held_lock+0x83/0xa0
 ? lock_is_held_type+0xe7/0x140
 ip6_finish_output+0x1ee/0x780
 ip6_output+0x138/0x460
 ? __pfx_ip6_output+0x10/0x10
 ? __pfx___lock_acquire+0x10/0x10
 ? __pfx_ip6_finish_output+0x10/0x10
 NF_HOOK.constprop.0+0xc0/0x420
 ? __pfx_NF_HOOK.constprop.0+0x10/0x10
 ? ndisc_send_skb+0x2c0/0x960
 ? __pfx_lock_release+0x10/0x10
 ? __local_bh_enable_ip+0x93/0x110
 ? lock_is_held_type+0xe7/0x140
 ndisc_send_skb+0x4be/0x960
 ? __pfx_ndisc_send_skb+0x10/0x10
 ? mark_held_locks+0x65/0x90
 ? find_held_lock+0x83/0xa0
 ndisc_send_ns+0xb0/0x110
 ? __pfx_ndisc_send_ns+0x10/0x10
 addrconf_dad_work+0x631/0x8e0
 ? lock_acquire+0x180/0x3f0
 ? __pfx_addrconf_dad_work+0x10/0x10
 ? mark_held_locks+0x24/0x90
 process_one_work+0x582/0x9c0
 ? __pfx_process_one_work+0x10/0x10
 ? __pfx_do_raw_spin_lock+0x10/0x10
 ? mark_held_locks+0x24/0x90
 worker_thread+0x93/0x630
 ? __kthread_parkme+0xdc/0x100
 ? __pfx_worker_thread+0x10/0x10
 kthread+0x1a5/0x1e0
 ? __pfx_kthread+0x10/0x10
 ret_from_fork+0x34/0x60
 ? __pfx_kthread+0x10/0x10
 ret_from_fork_asm+0x1b/0x30
RIP: 0000:0x0
Code: Unable to access opcode bytes at 0xffffffffffffffd6.
RSP: 0000:0000000000000000 EFLAGS: 00000000 ORIG_RAX: 0000000000000000
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
 </TASK>
Modules linked in:
CR2: ffffc900025910c8
---[ end trace 0000000000000000 ]---
RIP: 0010:nexthop_select_path+0x197/0xbf0
Code: c1 e4 05 be 08 00 00 00 4c 8b 35 a4 14 7e 01 4e 8d 6c 25 00 4a 8d 7c 25 08 48 01 dd e8 c2 25 15 ff 49 8d 7d 08 e8 39 13 15 ff <4d> 89 75 08 48 89 ef e8 7d 12 15 ff 48 8b 5d 00 e8 14 55 2f 00 85
RSP: 0018:ffff88810c36f260 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 00000000002000c0 RCX: ffffffffaf02dd77
RDX: dffffc0000000000 RSI: 0000000000000008 RDI: ffffc900025910c8
RBP: ffffc900025910c0 R08: 0000000000000001 R09: fffff520004b2219
R10: ffffc900025910cf R11: 31392d2068736168 R12: 00000000002000c0
R13: ffffc900025910c0 R14: 00000000fffef608 R15: ffff88811840e900
FS:  0000000000000000(0000) GS:ffff8881f7000000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffffffffffffffd6 CR3: 0000000129d00000 CR4: 0000000000750ee0
PKRU: 55555554
Kernel panic - not syncing: Fatal exception in interrupt
Kernel Offset: 0x2ca00000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
---[ end Kernel panic - not syncing: Fatal exception in interrupt ]---

Fix this problem by ensuring the MSB of hash is 0 using a right shift - the same approach used in fib_multipath_hash() and rt6_multipath_hash().

Fixes 1274e1cc4226 ("vxlan: ecmp support for mac fdb entries")
---
 include/net/vxlan.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
include/net/vxlan.h +2 -2
index cf1d870f7b9a8..e149a0b6f9a3c 100644
@@ -497,12 +497,12 @@ static inline void vxlan_flag_attr_error(int attrtype,
 }
 
 static inline bool vxlan_fdb_nh_path_select(struct nexthop *nh,
-					    int hash,
+					    u32 hash,
 					    struct vxlan_rdst *rdst)
 {
 	struct fib_nh_common *nhc;
 
-	nhc = nexthop_path_fdb_result(nh, hash);
+	nhc = nexthop_path_fdb_result(nh, hash >> 1);
 	if (unlikely(!nhc))
 		return false;
 
Signature · 1 line
2.40.1
author maintainer
#47 lore raw

PATCH 45/92 net/mlx5: fs_core: Make find_closest_ft more generic

In reply to Greg Kroah-Hartman

From: Jianbo Liu <[email protected]>

[ Upstream commit 618d28a535a0582617465d14e05f3881736a2962 ]

As find_closest_ft_recursive is called to find the closest FT, the first parameter of find_closest_ft can be changed from fs_prio to fs_node. Thus this function is extended to find the closest FT for the nodes of any type, not only prios, but also the sub namespaces.

Stable-dep-of: c635ca45a7a2 ("net/mlx5: fs_core: Skip the FTs in the same FS_TYPE_PRIO_CHAINS fs_prio")

---
 .../net/ethernet/mellanox/mlx5/core/fs_core.c | 29 +++++++++----------
 1 file changed, 14 insertions(+), 15 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +14 -15
index cb3f9de3d00ba..8b1a5f5637597 100644
@@ -818,18 +818,17 @@ static struct mlx5_flow_table *find_closest_ft_recursive(struct fs_node  *root,
 	return ft;
 }
 
-/* If reverse is false then return the first flow table in next priority of
- * prio in the tree, else return the last flow table in the previous priority
- * of prio in the tree.
+/* If reverse is false then return the first flow table next to the passed node
+ * in the tree, else return the last flow table before the node in the tree.
  */
-static struct mlx5_flow_table *find_closest_ft(struct fs_prio *prio, bool reverse)
+static struct mlx5_flow_table *find_closest_ft(struct fs_node *node, bool reverse)
 {
 	struct mlx5_flow_table *ft = NULL;
 	struct fs_node *curr_node;
 	struct fs_node *parent;
 
-	parent = prio->node.parent;
-	curr_node = &prio->node;
+	parent = node->parent;
+	curr_node = node;
 	while (!ft && parent) {
 		ft = find_closest_ft_recursive(parent, &curr_node->list, reverse);
 		curr_node = parent;
@@ -839,15 +838,15 @@ static struct mlx5_flow_table *find_closest_ft(struct fs_prio *prio, bool revers
 }
 
 /* Assuming all the tree is locked by mutex chain lock */
-static struct mlx5_flow_table *find_next_chained_ft(struct fs_prio *prio)
+static struct mlx5_flow_table *find_next_chained_ft(struct fs_node *node)
 {
-	return find_closest_ft(prio, false);
+	return find_closest_ft(node, false);
 }
 
 /* Assuming all the tree is locked by mutex chain lock */
-static struct mlx5_flow_table *find_prev_chained_ft(struct fs_prio *prio)
+static struct mlx5_flow_table *find_prev_chained_ft(struct fs_node *node)
 {
-	return find_closest_ft(prio, true);
+	return find_closest_ft(node, true);
 }
 
 static struct mlx5_flow_table *find_next_fwd_ft(struct mlx5_flow_table *ft,
@@ -859,7 +858,7 @@ static struct mlx5_flow_table *find_next_fwd_ft(struct mlx5_flow_table *ft,
 	next_ns = flow_act->action & MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_NS;
 	fs_get_obj(prio, next_ns ? ft->ns->node.parent : ft->node.parent);
 
-	return find_next_chained_ft(prio);
+	return find_next_chained_ft(&prio->node);
 }
 
 static int connect_fts_in_prio(struct mlx5_core_dev *dev,
@@ -890,7 +889,7 @@ static int connect_prev_fts(struct mlx5_core_dev *dev,
 {
 	struct mlx5_flow_table *prev_ft;
 
-	prev_ft = find_prev_chained_ft(prio);
+	prev_ft = find_prev_chained_ft(&prio->node);
 	if (prev_ft) {
 		struct fs_prio *prev_prio;
 
@@ -1036,7 +1035,7 @@ static int connect_flow_table(struct mlx5_core_dev *dev, struct mlx5_flow_table
 		if (err)
 			return err;
 
-		next_ft = first_ft ? first_ft : find_next_chained_ft(prio);
+		next_ft = first_ft ? first_ft : find_next_chained_ft(&prio->node);
 		err = connect_fwd_rules(dev, ft, next_ft);
 		if (err)
 			return err;
@@ -1111,7 +1110,7 @@ static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespa
 
 	tree_init_node(&ft->node, del_hw_flow_table, del_sw_flow_table);
 	next_ft = unmanaged ? ft_attr->next_ft :
-			      find_next_chained_ft(fs_prio);
+			      find_next_chained_ft(&fs_prio->node);
 	ft->def_miss_action = ns->def_miss_action;
 	ft->ns = ns;
 	err = root->cmds->create_flow_table(root, ft, ft_attr->max_fte, next_ft);
@@ -2086,7 +2085,7 @@ static struct mlx5_flow_table *find_next_ft(struct mlx5_flow_table *ft)
 
 	if (!list_is_last(&ft->node.list, &prio->node.children))
 		return list_next_entry(ft, node.list);
-	return find_next_chained_ft(prio);
+	return find_next_chained_ft(&prio->node);
 }
 
 static int update_root_ft_destroy(struct mlx5_flow_table *ft)
Signature · 1 line
2.40.1
author maintainer
#48 lore raw

PATCH 46/92 net/mlx5: fs_core: Skip the FTs in the same FS_TYPE_PRIO_CHAINS fs_prio

In reply to Greg Kroah-Hartman

From: Jianbo Liu <[email protected]>

[ Upstream commit c635ca45a7a2023904a1f851e99319af7b87017d ]

In the cited commit, new type of FS_TYPE_PRIO_CHAINS fs_prio was added to support multiple parallel namespaces for multi-chains. And we skip all the flow tables under the fs_node of this type unconditionally, when searching for the next or previous flow table to connect for a new table.

As this search function is also used for find new root table when the old one is being deleted, it will skip the entire FS_TYPE_PRIO_CHAINS fs_node next to the old root. However, new root table should be chosen from it if there is any table in it. Fix it by skipping only the flow tables in the same FS_TYPE_PRIO_CHAINS fs_node when finding the closest FT for a fs_node.

Besides, complete the connecting from FTs of previous priority of prio because there should be multiple prevs after this fs_prio type is introduced. And also the next FT should be chosen from the first flow table next to the prio in the same FS_TYPE_PRIO_CHAINS fs_prio, if this prio is the first child.

---
 .../net/ethernet/mellanox/mlx5/core/fs_core.c | 80 +++++++++++++++++--
 1 file changed, 72 insertions(+), 8 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +72 -8
index 8b1a5f5637597..161ad2ae40196 100644
@@ -802,7 +802,7 @@ static struct mlx5_flow_table *find_closest_ft_recursive(struct fs_node  *root,
 	struct fs_node *iter = list_entry(start, struct fs_node, list);
 	struct mlx5_flow_table *ft = NULL;
 
-	if (!root || root->type == FS_TYPE_PRIO_CHAINS)
+	if (!root)
 		return NULL;
 
 	list_for_each_advance_continue(iter, &root->children, reverse) {
@@ -818,19 +818,42 @@ static struct mlx5_flow_table *find_closest_ft_recursive(struct fs_node  *root,
 	return ft;
 }
 
+static struct fs_node *find_prio_chains_parent(struct fs_node *parent,
+					       struct fs_node **child)
+{
+	struct fs_node *node = NULL;
+
+	while (parent && parent->type != FS_TYPE_PRIO_CHAINS) {
+		node = parent;
+		parent = parent->parent;
+	}
+
+	if (child)
+		*child = node;
+
+	return parent;
+}
+
 /* If reverse is false then return the first flow table next to the passed node
  * in the tree, else return the last flow table before the node in the tree.
+ * If skip is true, skip the flow tables in the same prio_chains prio.
  */
-static struct mlx5_flow_table *find_closest_ft(struct fs_node *node, bool reverse)
+static struct mlx5_flow_table *find_closest_ft(struct fs_node *node, bool reverse,
+					       bool skip)
 {
+	struct fs_node *prio_chains_parent = NULL;
 	struct mlx5_flow_table *ft = NULL;
 	struct fs_node *curr_node;
 	struct fs_node *parent;
 
+	if (skip)
+		prio_chains_parent = find_prio_chains_parent(node, NULL);
 	parent = node->parent;
 	curr_node = node;
 	while (!ft && parent) {
-		ft = find_closest_ft_recursive(parent, &curr_node->list, reverse);
+		if (parent != prio_chains_parent)
+			ft = find_closest_ft_recursive(parent, &curr_node->list,
+						       reverse);
 		curr_node = parent;
 		parent = curr_node->parent;
 	}
@@ -840,13 +863,13 @@ static struct mlx5_flow_table *find_closest_ft(struct fs_node *node, bool revers
 /* Assuming all the tree is locked by mutex chain lock */
 static struct mlx5_flow_table *find_next_chained_ft(struct fs_node *node)
 {
-	return find_closest_ft(node, false);
+	return find_closest_ft(node, false, true);
 }
 
 /* Assuming all the tree is locked by mutex chain lock */
 static struct mlx5_flow_table *find_prev_chained_ft(struct fs_node *node)
 {
-	return find_closest_ft(node, true);
+	return find_closest_ft(node, true, true);
 }
 
 static struct mlx5_flow_table *find_next_fwd_ft(struct mlx5_flow_table *ft,
@@ -882,21 +905,55 @@ static int connect_fts_in_prio(struct mlx5_core_dev *dev,
 	return 0;
 }
 
+static struct mlx5_flow_table *find_closet_ft_prio_chains(struct fs_node *node,
+							  struct fs_node *parent,
+							  struct fs_node **child,
+							  bool reverse)
+{
+	struct mlx5_flow_table *ft;
+
+	ft = find_closest_ft(node, reverse, false);
+
+	if (ft && parent == find_prio_chains_parent(&ft->node, child))
+		return ft;
+
+	return NULL;
+}
+
 /* Connect flow tables from previous priority of prio to ft */
 static int connect_prev_fts(struct mlx5_core_dev *dev,
 			    struct mlx5_flow_table *ft,
 			    struct fs_prio *prio)
 {
+	struct fs_node *prio_parent, *parent = NULL, *child, *node;
 	struct mlx5_flow_table *prev_ft;
+	int err = 0;
+
+	prio_parent = find_prio_chains_parent(&prio->node, &child);
+
+	/* return directly if not under the first sub ns of prio_chains prio */
+	if (prio_parent && !list_is_first(&child->list, &prio_parent->children))
+		return 0;
 
 	prev_ft = find_prev_chained_ft(&prio->node);
-	if (prev_ft) {
+	while (prev_ft) {
 		struct fs_prio *prev_prio;
 
 		fs_get_obj(prev_prio, prev_ft->node.parent);
-		return connect_fts_in_prio(dev, prev_prio, ft);
+		err = connect_fts_in_prio(dev, prev_prio, ft);
+		if (err)
+			break;
+
+		if (!parent) {
+			parent = find_prio_chains_parent(&prev_prio->node, &child);
+			if (!parent)
+				break;
+		}
+
+		node = child;
+		prev_ft = find_closet_ft_prio_chains(node, parent, &child, true);
 	}
-	return 0;
+	return err;
 }
 
 static int update_root_ft_create(struct mlx5_flow_table *ft, struct fs_prio
@@ -2079,12 +2136,19 @@ EXPORT_SYMBOL(mlx5_del_flow_rules);
 /* Assuming prio->node.children(flow tables) is sorted by level */
 static struct mlx5_flow_table *find_next_ft(struct mlx5_flow_table *ft)
 {
+	struct fs_node *prio_parent, *child;
 	struct fs_prio *prio;
 
 	fs_get_obj(prio, ft->node.parent);
 
 	if (!list_is_last(&ft->node.list, &prio->node.children))
 		return list_next_entry(ft, node.list);
+
+	prio_parent = find_prio_chains_parent(&prio->node, &child);
+
+	if (prio_parent && list_is_first(&child->list, &prio_parent->children))
+		return find_closest_ft(&prio->node, false, false);
+
 	return find_next_chained_ft(&prio->node);
 }
 
Signature · 1 line
2.40.1
author maintainer
#49 lore raw

PATCH 47/92 prestera: fix fallback to previous version on same major version

In reply to Greg Kroah-Hartman

From: Jonas Gorski <[email protected]>

[ Upstream commit b755c25fbcd568821a3bb0e0d5c2daa5fcb00bba ]

When both supported and previous version have the same major version, and the firmwares are missing, the driver ends in a loop requesting the same (previous) version over and over again:

    [   76.327413] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.1.img firmware, fall-back to previous 4.0 version
    [   76.339802] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.0.img firmware, fall-back to previous 4.0 version
    [   76.352162] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.0.img firmware, fall-back to previous 4.0 version
    [   76.364502] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.0.img firmware, fall-back to previous 4.0 version
    [   76.376848] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.0.img firmware, fall-back to previous 4.0 version
    [   76.389183] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.0.img firmware, fall-back to previous 4.0 version
    [   76.401522] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.0.img firmware, fall-back to previous 4.0 version
    [   76.413860] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.0.img firmware, fall-back to previous 4.0 version
    [   76.426199] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.0.img firmware, fall-back to previous 4.0 version
    ...

Fix this by inverting the check to that we aren't yet at the previous version, and also check the minor version.

This also catches the case where both versions are the same, as it was after commit bb5dbf2cc64d ("net: marvell: prestera: add firmware v4.0 support").

With this fix applied:

    [   88.499622] Prestera DX 0000:01:00.0: missing latest mrvl/prestera/mvsw_prestera_fw-v4.1.img firmware, fall-back to previous 4.0 version
    [   88.511995] Prestera DX 0000:01:00.0: failed to request previous firmware: mrvl/prestera/mvsw_prestera_fw-v4.0.img
    [   88.522403] Prestera DX: probe of 0000:01:00.0 failed with error -2
---
 drivers/net/ethernet/marvell/prestera/prestera_pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/marvell/prestera/prestera_pci.c +2 -1
index a8d7b889ebeee..6bef633aa6330 100644
@@ -645,7 +645,8 @@ static int prestera_fw_get(struct prestera_fw *fw)
 
 	err = request_firmware_direct(&fw->bin, fw_path, fw->dev.dev);
 	if (err) {
-		if (ver_maj == PRESTERA_SUPP_FW_MAJ_VER) {
+		if (ver_maj != PRESTERA_PREV_FW_MAJ_VER ||
+		    ver_min != PRESTERA_PREV_FW_MIN_VER) {
 			ver_maj = PRESTERA_PREV_FW_MAJ_VER;
 			ver_min = PRESTERA_PREV_FW_MIN_VER;
 
Signature · 1 line
2.40.1
author maintainer
#50 lore raw

PATCH 48/92 tcp_metrics: fix addr_same() helper

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit e6638094d7af6c7b9dcca05ad009e79e31b4f670 ]

Because v4 and v6 families use separate inetpeer trees (respectively net->ipv4.peers and net->ipv6.peers), inetpeer_addr_cmp(a, b) assumes a & b share the same family.

tcp_metrics use a common hash table, where entries can have different families.

We must therefore make sure to not call inetpeer_addr_cmp() if the families do not match.

Fixes d39d14ffa24c ("net: Add helper function to compare inetpeer addresses")
---
 net/ipv4/tcp_metrics.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/tcp_metrics.c +1 -1
index d58e672be31c7..6a06d2e0e11db 100644
@@ -78,7 +78,7 @@ static void tcp_metric_set(struct tcp_metrics_block *tm,
 static bool addr_same(const struct inetpeer_addr *a,
 		      const struct inetpeer_addr *b)
 {
-	return inetpeer_addr_cmp(a, b) == 0;
+	return (a->family == b->family) && !inetpeer_addr_cmp(a, b);
 }
 
 struct tcpm_hash_bucket {
Signature · 1 line
2.40.1
author maintainer
#51 lore raw

PATCH 49/92 tcp_metrics: annotate data-races around tm->tcpm_stamp

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit 949ad62a5d5311d36fce2e14fe5fed3f936da51c ]

tm->tcpm_stamp can be read or written locklessly.

Add needed READ_ONCE()/WRITE_ONCE() to document this.

Also constify tcpm_check_stamp() dst argument.

---
 net/ipv4/tcp_metrics.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)
net/ipv4/tcp_metrics.c +13 -6
index 6a06d2e0e11db..cac6548340906 100644
@@ -97,7 +97,7 @@ static void tcpm_suck_dst(struct tcp_metrics_block *tm,
 	u32 msval;
 	u32 val;
 
-	tm->tcpm_stamp = jiffies;
+	WRITE_ONCE(tm->tcpm_stamp, jiffies);
 
 	val = 0;
 	if (dst_metric_locked(dst, RTAX_RTT))
@@ -131,9 +131,15 @@ static void tcpm_suck_dst(struct tcp_metrics_block *tm,
 
 #define TCP_METRICS_TIMEOUT		(60 * 60 * HZ)
 
-static void tcpm_check_stamp(struct tcp_metrics_block *tm, struct dst_entry *dst)
+static void tcpm_check_stamp(struct tcp_metrics_block *tm,
+			     const struct dst_entry *dst)
 {
-	if (tm && unlikely(time_after(jiffies, tm->tcpm_stamp + TCP_METRICS_TIMEOUT)))
+	unsigned long limit;
+
+	if (!tm)
+		return;
+	limit = READ_ONCE(tm->tcpm_stamp) + TCP_METRICS_TIMEOUT;
+	if (unlikely(time_after(jiffies, limit)))
 		tcpm_suck_dst(tm, dst, false);
 }
 
@@ -174,7 +180,8 @@ static struct tcp_metrics_block *tcpm_new(struct dst_entry *dst,
 		oldest = deref_locked(tcp_metrics_hash[hash].chain);
 		for (tm = deref_locked(oldest->tcpm_next); tm;
 		     tm = deref_locked(tm->tcpm_next)) {
-			if (time_before(tm->tcpm_stamp, oldest->tcpm_stamp))
+			if (time_before(READ_ONCE(tm->tcpm_stamp),
+					READ_ONCE(oldest->tcpm_stamp)))
 				oldest = tm;
 		}
 		tm = oldest;
@@ -434,7 +441,7 @@ void tcp_update_metrics(struct sock *sk)
 					       tp->reordering);
 		}
 	}
-	tm->tcpm_stamp = jiffies;
+	WRITE_ONCE(tm->tcpm_stamp, jiffies);
 out_unlock:
 	rcu_read_unlock();
 }
@@ -647,7 +654,7 @@ static int tcp_metrics_fill_info(struct sk_buff *msg,
 	}
 
 	if (nla_put_msecs(msg, TCP_METRICS_ATTR_AGE,
-			  jiffies - tm->tcpm_stamp,
+			  jiffies - READ_ONCE(tm->tcpm_stamp),
 			  TCP_METRICS_ATTR_PAD) < 0)
 		goto nla_put_failure;
 
Signature · 1 line
2.40.1
author maintainer
#52 lore raw

PATCH 50/92 tcp_metrics: annotate data-races around tm->tcpm_lock

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit 285ce119a3c6c4502585936650143e54c8692788 ]

tm->tcpm_lock can be read or written locklessly.

Add needed READ_ONCE()/WRITE_ONCE() to document this.

---
 net/ipv4/tcp_metrics.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/ipv4/tcp_metrics.c +4 -2
index cac6548340906..07bce467c71a7 100644
@@ -59,7 +59,8 @@ static inline struct net *tm_net(struct tcp_metrics_block *tm)
 static bool tcp_metric_locked(struct tcp_metrics_block *tm,
 			      enum tcp_metric_index idx)
 {
-	return tm->tcpm_lock & (1 << idx);
+	/* Paired with WRITE_ONCE() in tcpm_suck_dst() */
+	return READ_ONCE(tm->tcpm_lock) & (1 << idx);
 }
 
 static u32 tcp_metric_get(struct tcp_metrics_block *tm,
@@ -110,7 +111,8 @@ static void tcpm_suck_dst(struct tcp_metrics_block *tm,
 		val |= 1 << TCP_METRIC_CWND;
 	if (dst_metric_locked(dst, RTAX_REORDERING))
 		val |= 1 << TCP_METRIC_REORDERING;
-	tm->tcpm_lock = val;
+	/* Paired with READ_ONCE() in tcp_metric_locked() */
+	WRITE_ONCE(tm->tcpm_lock, val);
 
 	msval = dst_metric_raw(dst, RTAX_RTT);
 	tm->tcpm_vals[TCP_METRIC_RTT] = msval * USEC_PER_MSEC;
Signature · 1 line
2.40.1
author maintainer
#53 lore raw

PATCH 51/92 tcp_metrics: annotate data-races around tm->tcpm_vals[]

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit 8c4d04f6b443869d25e59822f7cec88d647028a9 ]

tm->tcpm_vals[] values can be read or written locklessly.

Add needed READ_ONCE()/WRITE_ONCE() to document this, and force use of tcp_metric_get() and tcp_metric_set()

Fixes 51c5d0c4b169 ("tcp: Maintain dynamic metrics in local cache.")
---
 net/ipv4/tcp_metrics.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)
net/ipv4/tcp_metrics.c +14 -9
index 07bce467c71a7..0c309fd4e04b2 100644
@@ -63,17 +63,19 @@ static bool tcp_metric_locked(struct tcp_metrics_block *tm,
 	return READ_ONCE(tm->tcpm_lock) & (1 << idx);
 }
 
-static u32 tcp_metric_get(struct tcp_metrics_block *tm,
+static u32 tcp_metric_get(const struct tcp_metrics_block *tm,
 			  enum tcp_metric_index idx)
 {
-	return tm->tcpm_vals[idx];
+	/* Paired with WRITE_ONCE() in tcp_metric_set() */
+	return READ_ONCE(tm->tcpm_vals[idx]);
 }
 
 static void tcp_metric_set(struct tcp_metrics_block *tm,
 			   enum tcp_metric_index idx,
 			   u32 val)
 {
-	tm->tcpm_vals[idx] = val;
+	/* Paired with READ_ONCE() in tcp_metric_get() */
+	WRITE_ONCE(tm->tcpm_vals[idx], val);
 }
 
 static bool addr_same(const struct inetpeer_addr *a,
@@ -115,13 +117,16 @@ static void tcpm_suck_dst(struct tcp_metrics_block *tm,
 	WRITE_ONCE(tm->tcpm_lock, val);
 
 	msval = dst_metric_raw(dst, RTAX_RTT);
-	tm->tcpm_vals[TCP_METRIC_RTT] = msval * USEC_PER_MSEC;
+	tcp_metric_set(tm, TCP_METRIC_RTT, msval * USEC_PER_MSEC);
 
 	msval = dst_metric_raw(dst, RTAX_RTTVAR);
-	tm->tcpm_vals[TCP_METRIC_RTTVAR] = msval * USEC_PER_MSEC;
-	tm->tcpm_vals[TCP_METRIC_SSTHRESH] = dst_metric_raw(dst, RTAX_SSTHRESH);
-	tm->tcpm_vals[TCP_METRIC_CWND] = dst_metric_raw(dst, RTAX_CWND);
-	tm->tcpm_vals[TCP_METRIC_REORDERING] = dst_metric_raw(dst, RTAX_REORDERING);
+	tcp_metric_set(tm, TCP_METRIC_RTTVAR, msval * USEC_PER_MSEC);
+	tcp_metric_set(tm, TCP_METRIC_SSTHRESH,
+		       dst_metric_raw(dst, RTAX_SSTHRESH));
+	tcp_metric_set(tm, TCP_METRIC_CWND,
+		       dst_metric_raw(dst, RTAX_CWND));
+	tcp_metric_set(tm, TCP_METRIC_REORDERING,
+		       dst_metric_raw(dst, RTAX_REORDERING));
 	if (fastopen_clear) {
 		tm->tcpm_fastopen.mss = 0;
 		tm->tcpm_fastopen.syn_loss = 0;
@@ -667,7 +672,7 @@ static int tcp_metrics_fill_info(struct sk_buff *msg,
 		if (!nest)
 			goto nla_put_failure;
 		for (i = 0; i < TCP_METRIC_MAX_KERNEL + 1; i++) {
-			u32 val = tm->tcpm_vals[i];
+			u32 val = tcp_metric_get(tm, i);
 
 			if (!val)
 				continue;
Signature · 1 line
2.40.1
author maintainer
#54 lore raw

PATCH 52/92 tcp_metrics: annotate data-races around tm->tcpm_net

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit d5d986ce42c71a7562d32c4e21e026b0f87befec ]

tm->tcpm_net can be read or written locklessly.

Instead of changing write_pnet() and read_pnet() and potentially hurt performance, add the needed READ_ONCE()/WRITE_ONCE() in tm_net() and tcpm_new().

Fixes 849e8a0ca8d5 ("tcp_metrics: Add a field tcpm_net and verify it matches on lookup")
---
 net/ipv4/tcp_metrics.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
net/ipv4/tcp_metrics.c +7 -4
index 0c309fd4e04b2..88b0d952a2888 100644
@@ -40,7 +40,7 @@ struct tcp_fastopen_metrics {
 
 struct tcp_metrics_block {
 	struct tcp_metrics_block __rcu	*tcpm_next;
-	possible_net_t			tcpm_net;
+	struct net			*tcpm_net;
 	struct inetpeer_addr		tcpm_saddr;
 	struct inetpeer_addr		tcpm_daddr;
 	unsigned long			tcpm_stamp;
@@ -51,9 +51,10 @@ struct tcp_metrics_block {
 	struct rcu_head			rcu_head;
 };
 
-static inline struct net *tm_net(struct tcp_metrics_block *tm)
+static inline struct net *tm_net(const struct tcp_metrics_block *tm)
 {
-	return read_pnet(&tm->tcpm_net);
+	/* Paired with the WRITE_ONCE() in tcpm_new() */
+	return READ_ONCE(tm->tcpm_net);
 }
 
 static bool tcp_metric_locked(struct tcp_metrics_block *tm,
@@ -197,7 +198,9 @@ static struct tcp_metrics_block *tcpm_new(struct dst_entry *dst,
 		if (!tm)
 			goto out_unlock;
 	}
-	write_pnet(&tm->tcpm_net, net);
+	/* Paired with the READ_ONCE() in tm_net() */
+	WRITE_ONCE(tm->tcpm_net, net);
+
 	tm->tcpm_saddr = *saddr;
 	tm->tcpm_daddr = *daddr;
 
Signature · 1 line
2.40.1
author maintainer
#55 lore raw

PATCH 53/92 tcp_metrics: fix data-race in tcpm_suck_dst() vs fastopen

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit ddf251fa2bc1d3699eec0bae6ed0bc373b8fda79 ]

Whenever tcpm_new() reclaims an old entry, tcpm_suck_dst() would overwrite data that could be read from tcp_fastopen_cache_get() or tcp_metrics_fill_info().

We need to acquire fastopen_seqlock to maintain consistency.

For newly allocated objects, tcpm_new() can switch to kzalloc() to avoid an extra fastopen_seqlock acquisition.

Fixes 1fe4c481ba63 ("net-tcp: Fast Open client - cookie cache")
Cc Yuchung Cheng <[email protected]>
---
 net/ipv4/tcp_metrics.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
net/ipv4/tcp_metrics.c +5 -4
index 88b0d952a2888..5df97aaac252e 100644
@@ -93,6 +93,7 @@ static struct tcpm_hash_bucket	*tcp_metrics_hash __read_mostly;
 static unsigned int		tcp_metrics_hash_log __read_mostly;
 
 static DEFINE_SPINLOCK(tcp_metrics_lock);
+static DEFINE_SEQLOCK(fastopen_seqlock);
 
 static void tcpm_suck_dst(struct tcp_metrics_block *tm,
 			  const struct dst_entry *dst,
@@ -129,11 +130,13 @@ static void tcpm_suck_dst(struct tcp_metrics_block *tm,
 	tcp_metric_set(tm, TCP_METRIC_REORDERING,
 		       dst_metric_raw(dst, RTAX_REORDERING));
 	if (fastopen_clear) {
+		write_seqlock(&fastopen_seqlock);
 		tm->tcpm_fastopen.mss = 0;
 		tm->tcpm_fastopen.syn_loss = 0;
 		tm->tcpm_fastopen.try_exp = 0;
 		tm->tcpm_fastopen.cookie.exp = false;
 		tm->tcpm_fastopen.cookie.len = 0;
+		write_sequnlock(&fastopen_seqlock);
 	}
 }
 
@@ -194,7 +197,7 @@ static struct tcp_metrics_block *tcpm_new(struct dst_entry *dst,
 		}
 		tm = oldest;
 	} else {
-		tm = kmalloc(sizeof(*tm), GFP_ATOMIC);
+		tm = kzalloc(sizeof(*tm), GFP_ATOMIC);
 		if (!tm)
 			goto out_unlock;
 	}
@@ -204,7 +207,7 @@ static struct tcp_metrics_block *tcpm_new(struct dst_entry *dst,
 	tm->tcpm_saddr = *saddr;
 	tm->tcpm_daddr = *daddr;
 
-	tcpm_suck_dst(tm, dst, true);
+	tcpm_suck_dst(tm, dst, reclaim);
 
 	if (likely(!reclaim)) {
 		tm->tcpm_next = tcp_metrics_hash[hash].chain;
@@ -556,8 +559,6 @@ bool tcp_peer_is_proven(struct request_sock *req, struct dst_entry *dst)
 	return ret;
 }
 
-static DEFINE_SEQLOCK(fastopen_seqlock);
-
 void tcp_fastopen_cache_get(struct sock *sk, u16 *mss,
 			    struct tcp_fastopen_cookie *cookie)
 {
Signature · 1 line
2.40.1
author maintainer
#56 lore raw

PATCH 54/92 scsi: zfcp: Defer fc_rport blocking until after ADISC response

In reply to Greg Kroah-Hartman

From: Steffen Maier <[email protected]>

commit e65851989001c0c9ba9177564b13b38201c0854c upstream.

Storage devices are free to send RSCNs, e.g. for internal state changes. If this happens on all connected paths, zfcp risks temporarily losing all paths at the same time. This has strong requirements on multipath configuration such as "no_path_retry queue".

Avoid such situations by deferring fc_rport blocking until after the ADISC response, when any actual state change of the remote port became clear. The already existing port recovery triggers explicitly block the fc_rport. The triggers are: on ADISC reject or timeout (typical cable pull case), and on ADISC indicating that the remote port has changed its WWPN or the port is meanwhile no longer open.

As a side effect, this also removes a confusing direct function call to another work item function zfcp_scsi_rport_work() instead of scheduling that other work item. It was probably done that way to have the rport block side effect immediate and synchronous to the caller.

---
 drivers/s390/scsi/zfcp_fc.c |    6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)
drivers/s390/scsi/zfcp_fc.c +1 -5
@@ -534,8 +534,7 @@ static void zfcp_fc_adisc_handler(void *
 
 	/* re-init to undo drop from zfcp_fc_adisc() */
 	port->d_id = ntoh24(adisc_resp->adisc_port_id);
-	/* port is good, unblock rport without going through erp */
-	zfcp_scsi_schedule_rport_register(port);
+	/* port is still good, nothing to do */
  out:
 	atomic_andnot(ZFCP_STATUS_PORT_LINK_TEST, &port->status);
 	put_device(&port->dev);
@@ -595,9 +594,6 @@ void zfcp_fc_link_test_work(struct work_
 	int retval;
 
 	set_worker_desc("zadisc%16llx", port->wwpn); /* < WORKER_DESC_LEN=24 */
-	get_device(&port->dev);
-	port->rport_task = RPORT_DEL;
-	zfcp_scsi_rport_work(&port->rport_work);
 
 	/* only issue one test command at one time per port */
 	if (atomic_read(&port->status) & ZFCP_STATUS_PORT_LINK_TEST)
author maintainer
#57 lore raw

PATCH 55/92 scsi: storvsc: Limit max_sectors for virtual Fibre Channel devices

In reply to Greg Kroah-Hartman

From: Michael Kelley <[email protected]>

commit 010c1e1c5741365dbbf44a5a5bb9f30192875c4c upstream.

The Hyper-V host is queried to get the max transfer size that it supports, and this value is used to set max_sectors for the synthetic SCSI controller. However, this max transfer size may be too large for virtual Fibre Channel devices, which are limited to 512 Kbytes. If a larger transfer size is used with a vFC device, Hyper-V always returns an error, and storvsc logs a message like this where the SRB status and SCSI status are both zero:

hv_storvsc <GUID>: tag#197 cmd 0x8a status: scsi 0x0 srb 0x0 hv 0xc0000001

Add logic to limit the max transfer size to 512 Kbytes for vFC devices.

Fixes 1d3e0980782f ("scsi: storvsc: Correct reporting of Hyper-V I/O size limits")
---
 drivers/scsi/storvsc_drv.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/scsi/storvsc_drv.c +4 -0
@@ -406,6 +406,7 @@ static void storvsc_on_channel_callback(
 #define STORVSC_FC_MAX_LUNS_PER_TARGET			255
 #define STORVSC_FC_MAX_TARGETS				128
 #define STORVSC_FC_MAX_CHANNELS				8
+#define STORVSC_FC_MAX_XFER_SIZE			((u32)(512 * 1024))
 
 #define STORVSC_IDE_MAX_LUNS_PER_TARGET			64
 #define STORVSC_IDE_MAX_TARGETS				1
@@ -2071,6 +2072,9 @@ static int storvsc_probe(struct hv_devic
 	 * protecting it from any weird value.
 	 */
 	max_xfer_bytes = round_down(stor_device->max_transfer_bytes, HV_HYP_PAGE_SIZE);
+	if (is_fc)
+		max_xfer_bytes = min(max_xfer_bytes, STORVSC_FC_MAX_XFER_SIZE);
+
 	/* max_hw_sectors_kb */
 	host->max_sectors = max_xfer_bytes >> 9;
 	/*
author maintainer
#58 lore raw

PATCH 56/92 libceph: fix potential hang in ceph_osdc_notify()

In reply to Greg Kroah-Hartman

From: Ilya Dryomov <[email protected]>

commit e6e2843230799230fc5deb8279728a7218b0d63c upstream.

If the cluster becomes unavailable, ceph_osdc_notify() may hang even with osd_request_timeout option set because linger_notify_finish_wait() waits for MWatchNotify NOTIFY_COMPLETE message with no associated OSD request in flight -- it's completely asynchronous.

Introduce an additional timeout, derived from the specified notify timeout. While at it, switch both waits to killable which is more correct.

---
 net/ceph/osd_client.c |   20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)
net/ceph/osd_client.c +14 -6
@@ -3330,17 +3330,24 @@ static int linger_reg_commit_wait(struct
 	int ret;
 
 	dout("%s lreq %p linger_id %llu\n", __func__, lreq, lreq->linger_id);
-	ret = wait_for_completion_interruptible(&lreq->reg_commit_wait);
+	ret = wait_for_completion_killable(&lreq->reg_commit_wait);
 	return ret ?: lreq->reg_commit_error;
 }
 
-static int linger_notify_finish_wait(struct ceph_osd_linger_request *lreq)
+static int linger_notify_finish_wait(struct ceph_osd_linger_request *lreq,
+				     unsigned long timeout)
 {
-	int ret;
+	long left;
 
 	dout("%s lreq %p linger_id %llu\n", __func__, lreq, lreq->linger_id);
-	ret = wait_for_completion_interruptible(&lreq->notify_finish_wait);
-	return ret ?: lreq->notify_finish_error;
+	left = wait_for_completion_killable_timeout(&lreq->notify_finish_wait,
+						ceph_timeout_jiffies(timeout));
+	if (left <= 0)
+		left = left ?: -ETIMEDOUT;
+	else
+		left = lreq->notify_finish_error; /* completed */
+
+	return left;
 }
 
 /*
@@ -4890,7 +4897,8 @@ int ceph_osdc_notify(struct ceph_osd_cli
 	linger_submit(lreq);
 	ret = linger_reg_commit_wait(lreq);
 	if (!ret)
-		ret = linger_notify_finish_wait(lreq);
+		ret = linger_notify_finish_wait(lreq,
+				 msecs_to_jiffies(2 * timeout * MSEC_PER_SEC));
 	else
 		dout("lreq %p failed to initiate notify %d\n", lreq, ret);
 
author maintainer
#59 lore raw

PATCH 57/92 USB: zaurus: Add ID for A-300/B-500/C-700

In reply to Greg Kroah-Hartman

From: Ross Maynard <[email protected]>

commit b99225b4fe297d07400f9e2332ecd7347b224f8d upstream.

The SL-A300, B500/5600, and C700 devices no longer auto-load because of "usbnet: Remove over-broad module alias from zaurus."
This patch adds IDs for those 3 devices.

---
 drivers/net/usb/cdc_ether.c |   21 +++++++++++++++++++++
 drivers/net/usb/zaurus.c    |   21 +++++++++++++++++++++
 2 files changed, 42 insertions(+)
drivers/net/usb/cdc_ether.c +21 -0
@@ -617,6 +617,13 @@ static const struct usb_device_id	produc
 	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
 			  | USB_DEVICE_ID_MATCH_DEVICE,
 	.idVendor		= 0x04DD,
+	.idProduct		= 0x8005,   /* A-300 */
+	ZAURUS_FAKE_INTERFACE,
+	.driver_info        = 0,
+}, {
+	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x04DD,
 	.idProduct		= 0x8006,	/* B-500/SL-5600 */
 	ZAURUS_MASTER_INTERFACE,
 	.driver_info		= 0,
@@ -624,11 +631,25 @@ static const struct usb_device_id	produc
 	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
 			  | USB_DEVICE_ID_MATCH_DEVICE,
 	.idVendor		= 0x04DD,
+	.idProduct		= 0x8006,   /* B-500/SL-5600 */
+	ZAURUS_FAKE_INTERFACE,
+	.driver_info        = 0,
+}, {
+	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x04DD,
 	.idProduct		= 0x8007,	/* C-700 */
 	ZAURUS_MASTER_INTERFACE,
 	.driver_info		= 0,
 }, {
 	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x04DD,
+	.idProduct		= 0x8007,   /* C-700 */
+	ZAURUS_FAKE_INTERFACE,
+	.driver_info        = 0,
+}, {
+	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
 		 | USB_DEVICE_ID_MATCH_DEVICE,
 	.idVendor               = 0x04DD,
 	.idProduct              = 0x9031,	/* C-750 C-760 */
drivers/net/usb/zaurus.c +21 -0
@@ -289,11 +289,25 @@ static const struct usb_device_id	produc
 	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
 			  | USB_DEVICE_ID_MATCH_DEVICE,
 	.idVendor		= 0x04DD,
+	.idProduct		= 0x8005,	/* A-300 */
+	ZAURUS_FAKE_INTERFACE,
+	.driver_info = (unsigned long)&bogus_mdlm_info,
+}, {
+	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x04DD,
 	.idProduct		= 0x8006,	/* B-500/SL-5600 */
 	ZAURUS_MASTER_INTERFACE,
 	.driver_info = ZAURUS_PXA_INFO,
 }, {
 	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x04DD,
+	.idProduct		= 0x8006,	/* B-500/SL-5600 */
+	ZAURUS_FAKE_INTERFACE,
+	.driver_info = (unsigned long)&bogus_mdlm_info,
+}, {
+	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
 	          | USB_DEVICE_ID_MATCH_DEVICE,
 	.idVendor		= 0x04DD,
 	.idProduct		= 0x8007,	/* C-700 */
@@ -301,6 +315,13 @@ static const struct usb_device_id	produc
 	.driver_info = ZAURUS_PXA_INFO,
 }, {
 	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x04DD,
+	.idProduct		= 0x8007,	/* C-700 */
+	ZAURUS_FAKE_INTERFACE,
+	.driver_info = (unsigned long)&bogus_mdlm_info,
+}, {
+	.match_flags    =   USB_DEVICE_ID_MATCH_INT_INFO
 		 | USB_DEVICE_ID_MATCH_DEVICE,
 	.idVendor               = 0x04DD,
 	.idProduct              = 0x9031,	/* C-750 C-760 */
author maintainer
#60 lore raw

PATCH 58/92 ceph: defer stopping mdsc delayed_work

In reply to Greg Kroah-Hartman

From: Xiubo Li <[email protected]>

commit e7e607bd00481745550389a29ecabe33e13d67cf upstream.

Flushing the dirty buffer may take a long time if the cluster is overloaded or if there is network issue. So we should ping the MDSs periodically to keep alive, else the MDS will blocklist the kclient.

---
 fs/ceph/mds_client.c |    4 ++--
 fs/ceph/mds_client.h |    5 +++++
 fs/ceph/super.c      |   10 ++++++++++
 3 files changed, 17 insertions(+), 2 deletions(-)
fs/ceph/mds_client.c +2 -2
@@ -4607,7 +4607,7 @@ static void delayed_work(struct work_str
 
 	dout("mdsc delayed_work\n");
 
-	if (mdsc->stopping)
+	if (mdsc->stopping >= CEPH_MDSC_STOPPING_FLUSHED)
 		return;
 
 	mutex_lock(&mdsc->mutex);
@@ -4786,7 +4786,7 @@ void send_flush_mdlog(struct ceph_mds_se
 void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc)
 {
 	dout("pre_umount\n");
-	mdsc->stopping = 1;
+	mdsc->stopping = CEPH_MDSC_STOPPING_BEGIN;
 
 	ceph_mdsc_iterate_sessions(mdsc, send_flush_mdlog, true);
 	ceph_mdsc_iterate_sessions(mdsc, lock_unlock_session, false);
fs/ceph/mds_client.h +5 -0
@@ -370,6 +370,11 @@ struct cap_wait {
 	int			want;
 };
 
+enum {
+       CEPH_MDSC_STOPPING_BEGIN = 1,
+       CEPH_MDSC_STOPPING_FLUSHED = 2,
+};
+
 /*
  * mds client state
  */
fs/ceph/super.c +10 -0
@@ -1227,6 +1227,16 @@ static void ceph_kill_sb(struct super_bl
 	ceph_mdsc_pre_umount(fsc->mdsc);
 	flush_fs_workqueues(fsc);
 
+	/*
+	 * Though the kill_anon_super() will finally trigger the
+	 * sync_filesystem() anyway, we still need to do it here
+	 * and then bump the stage of shutdown to stop the work
+	 * queue as earlier as possible.
+	 */
+	sync_filesystem(s);
+
+	fsc->mdsc->stopping = CEPH_MDSC_STOPPING_FLUSHED;
+
 	kill_anon_super(s);
 
 	fsc->client->extra_mon_dispatch = NULL;
author maintainer
#61 lore raw

PATCH 59/92 firmware: arm_scmi: Drop OF node reference in the transport channel setup

In reply to Greg Kroah-Hartman

From: Krzysztof Kozlowski <[email protected]>

commit da042eb4f061a0b54aedadcaa15391490c48e1ad upstream.

The OF node reference obtained from of_parse_phandle() should be dropped if node is not compatible with arm,scmi-shmem.

Fixes 507cd4d2c5eb ("firmware: arm_scmi: Add compatibility checks for shmem node")
---
 drivers/firmware/arm_scmi/mailbox.c |    4 +++-
 drivers/firmware/arm_scmi/smc.c     |    4 +++-
 2 files changed, 6 insertions(+), 2 deletions(-)
drivers/firmware/arm_scmi/mailbox.c +3 -1
@@ -106,8 +106,10 @@ static int mailbox_chan_setup(struct scm
 		return -ENOMEM;
 
 	shmem = of_parse_phandle(cdev->of_node, "shmem", idx);
-	if (!of_device_is_compatible(shmem, "arm,scmi-shmem"))
+	if (!of_device_is_compatible(shmem, "arm,scmi-shmem")) {
+		of_node_put(shmem);
 		return -ENXIO;
+	}
 
 	ret = of_address_to_resource(shmem, 0, &res);
 	of_node_put(shmem);
drivers/firmware/arm_scmi/smc.c +3 -1
@@ -78,8 +78,10 @@ static int smc_chan_setup(struct scmi_ch
 		return -ENOMEM;
 
 	np = of_parse_phandle(cdev->of_node, "shmem", 0);
-	if (!of_device_is_compatible(np, "arm,scmi-shmem"))
+	if (!of_device_is_compatible(np, "arm,scmi-shmem")) {
+		of_node_put(np);
 		return -ENXIO;
+	}
 
 	ret = of_address_to_resource(np, 0, &res);
 	of_node_put(np);
author maintainer
#62 lore raw

PATCH 60/92 exfat: use kvmalloc_array/kvfree instead of kmalloc_array/kfree

In reply to Greg Kroah-Hartman

From: gaoming <[email protected]>

commit daf60d6cca26e50d65dac374db92e58de745ad26 upstream.

The call stack shown below is a scenario in the Linux 4.19 kernel. Allocating memory failed where exfat fs use kmalloc_array due to system memory fragmentation, while the u-disk was inserted without recognition.
Devices such as u-disk using the exfat file system are pluggable and may be insert into the system at any time.
However, long-term running systems cannot guarantee the continuity of physical memory. Therefore, it's necessary to address this issue.

Binder:2632_6: page allocation failure: order:4,
mode:0x6040c0(GFP_KERNEL|__GFP_COMP), nodemask=(null)

Call trace:
[242178.097582]  dump_backtrace+0x0/0x4
[242178.097589]  dump_stack+0xf4/0x134
[242178.097598]  warn_alloc+0xd8/0x144
[242178.097603]  __alloc_pages_nodemask+0x1364/0x1384
[242178.097608]  kmalloc_order+0x2c/0x510
[242178.097612]  kmalloc_order_trace+0x40/0x16c
[242178.097618]  __kmalloc+0x360/0x408
[242178.097624]  load_alloc_bitmap+0x160/0x284
[242178.097628]  exfat_fill_super+0xa3c/0xe7c
[242178.097635]  mount_bdev+0x2e8/0x3a0
[242178.097638]  exfat_fs_mount+0x40/0x50
[242178.097643]  mount_fs+0x138/0x2e8
[242178.097649]  vfs_kern_mount+0x90/0x270
[242178.097655]  do_mount+0x798/0x173c
[242178.097659]  ksys_mount+0x114/0x1ac
[242178.097665]  __arm64_sys_mount+0x24/0x34
[242178.097671]  el0_svc_common+0xb8/0x1b8
[242178.097676]  el0_svc_handler+0x74/0x90
[242178.097681]  el0_svc+0x8/0x340

By analyzing the exfat code,we found that continuous physical memory is not required here,so kvmalloc_array is used can solve this problem.

---
 fs/exfat/balloc.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
fs/exfat/balloc.c +3 -3
@@ -69,7 +69,7 @@ static int exfat_allocate_bitmap(struct
 	}
 	sbi->map_sectors = ((need_map_size - 1) >>
 			(sb->s_blocksize_bits)) + 1;
-	sbi->vol_amap = kmalloc_array(sbi->map_sectors,
+	sbi->vol_amap = kvmalloc_array(sbi->map_sectors,
 				sizeof(struct buffer_head *), GFP_KERNEL);
 	if (!sbi->vol_amap)
 		return -ENOMEM;
@@ -84,7 +84,7 @@ static int exfat_allocate_bitmap(struct
 			while (j < i)
 				brelse(sbi->vol_amap[j++]);
 
-			kfree(sbi->vol_amap);
+			kvfree(sbi->vol_amap);
 			sbi->vol_amap = NULL;
 			return -EIO;
 		}
@@ -138,7 +138,7 @@ void exfat_free_bitmap(struct exfat_sb_i
 	for (i = 0; i < sbi->map_sectors; i++)
 		__brelse(sbi->vol_amap[i]);
 
-	kfree(sbi->vol_amap);
+	kvfree(sbi->vol_amap);
 }
 
 int exfat_set_bitmap(struct inode *inode, unsigned int clu, bool sync)
author maintainer
#63 lore raw

PATCH 61/92 exfat: release s_lock before calling dir_emit()

In reply to Greg Kroah-Hartman

From: Sungjong Seo <[email protected]>

commit ff84772fd45d486e4fc78c82e2f70ce5333543e6 upstream.

There is a potential deadlock reported by syzbot as below:

======================================================
WARNING: possible circular locking dependency detected
6.4.0-next-20230707-syzkaller #0 Not tainted
------------------------------------------------------
syz-executor330/5073 is trying to acquire lock:
ffff8880218527a0 (&mm->mmap_lock){++++}-{3:3}, at: mmap_read_lock_killable include/linux/mmap_lock.h:151 [inline] ffff8880218527a0 (&mm->mmap_lock){++++}-{3:3}, at: get_mmap_lock_carefully mm/memory.c:5293 [inline] ffff8880218527a0 (&mm->mmap_lock){++++}-{3:3}, at: lock_mm_and_find_vma+0x369/0x510 mm/memory.c:5344 but task is already holding lock:
ffff888019f760e0 (&sbi->s_lock){+.+.}-{3:3}, at: exfat_iterate+0x117/0xb50 fs/exfat/dir.c:232

which lock already depends on the new lock.

Chain exists of:
  &mm->mmap_lock --> mapping.invalidate_lock#3 --> &sbi->s_lock

 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  lock(&sbi->s_lock);
                               lock(mapping.invalidate_lock#3);
                               lock(&sbi->s_lock);

rlock(&mm->mmap_lock);

Let's try to avoid above potential deadlock condition by moving dir_emit*() out of sbi->s_lock coverage.

---
 fs/exfat/dir.c |   27 ++++++++++++---------------
 1 file changed, 12 insertions(+), 15 deletions(-)
fs/exfat/dir.c +12 -15
@@ -211,7 +211,10 @@ static void exfat_free_namebuf(struct ex
 	exfat_init_namebuf(nb);
 }
 
-/* skip iterating emit_dots when dir is empty */
+/*
+ * Before calling dir_emit*(), sbi->s_lock should be released
+ * because page fault can occur in dir_emit*().
+ */
 #define ITER_POS_FILLED_DOTS    (2)
 static int exfat_iterate(struct file *filp, struct dir_context *ctx)
 {
@@ -226,11 +229,10 @@ static int exfat_iterate(struct file *fi
 	int err = 0, fake_offset = 0;
 
 	exfat_init_namebuf(nb);
-	mutex_lock(&EXFAT_SB(sb)->s_lock);
 
 	cpos = ctx->pos;
 	if (!dir_emit_dots(filp, ctx))
-		goto unlock;
+		goto out;
 
 	if (ctx->pos == ITER_POS_FILLED_DOTS) {
 		cpos = 0;
@@ -242,16 +244,18 @@ static int exfat_iterate(struct file *fi
 	/* name buffer should be allocated before use */
 	err = exfat_alloc_namebuf(nb);
 	if (err)
-		goto unlock;
+		goto out;
 get_new:
+	mutex_lock(&EXFAT_SB(sb)->s_lock);
+
 	if (ei->flags == ALLOC_NO_FAT_CHAIN && cpos >= i_size_read(inode))
 		goto end_of_dir;
 
 	err = exfat_readdir(inode, &cpos, &de);
 	if (err) {
 		/*
-		 * At least we tried to read a sector.  Move cpos to next sector
-		 * position (should be aligned).
+		 * At least we tried to read a sector.
+		 * Move cpos to next sector position (should be aligned).
 		 */
 		if (err == -EIO) {
 			cpos += 1 << (sb->s_blocksize_bits);
@@ -274,16 +278,10 @@ get_new:
 		inum = iunique(sb, EXFAT_ROOT_INO);
 	}
 
-	/*
-	 * Before calling dir_emit(), sb_lock should be released.
-	 * Because page fault can occur in dir_emit() when the size
-	 * of buffer given from user is larger than one page size.
-	 */
 	mutex_unlock(&EXFAT_SB(sb)->s_lock);
 	if (!dir_emit(ctx, nb->lfn, strlen(nb->lfn), inum,
 			(de.attr & ATTR_SUBDIR) ? DT_DIR : DT_REG))
-		goto out_unlocked;
-	mutex_lock(&EXFAT_SB(sb)->s_lock);
+		goto out;
 	ctx->pos = cpos;
 	goto get_new;
 
@@ -291,9 +289,8 @@ end_of_dir:
 	if (!cpos && fake_offset)
 		cpos = ITER_POS_FILLED_DOTS;
 	ctx->pos = cpos;
-unlock:
 	mutex_unlock(&EXFAT_SB(sb)->s_lock);
-out_unlocked:
+out:
 	/*
 	 * To improve performance, free namebuf after unlock sb_lock.
 	 * If namebuf is not allocated, this function do nothing
author maintainer
#64 lore raw

PATCH 62/92 mtd: spinand: toshiba: Fix ecc_get_status

In reply to Greg Kroah-Hartman

From: Olivier Maignial <[email protected]>

commit 8544cda94dae6be3f1359539079c68bb731428b1 upstream.

Reading ECC status is failing.

tx58cxgxsxraix_ecc_get_status() is using on-stack buffer for SPINAND_GET_FEATURE_OP() output. It is not suitable for DMA needs of spi-mem.

Fix this by using the spi-mem operations dedicated buffer spinand->scratchbuf.

See
spinand->scratchbuf:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/include/linux/mtd/spinand.h?h=v6.3#n418
spi_mem_check_op():
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/spi/spi-mem.c?h=v6.3#n199

---
 drivers/mtd/nand/spi/toshiba.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/mtd/nand/spi/toshiba.c +2 -2
@@ -73,7 +73,7 @@ static int tx58cxgxsxraix_ecc_get_status
 {
 	struct nand_device *nand = spinand_to_nand(spinand);
 	u8 mbf = 0;
-	struct spi_mem_op op = SPINAND_GET_FEATURE_OP(0x30, &mbf);
+	struct spi_mem_op op = SPINAND_GET_FEATURE_OP(0x30, spinand->scratchbuf);
 
 	switch (status & STATUS_ECC_MASK) {
 	case STATUS_ECC_NO_BITFLIPS:
@@ -92,7 +92,7 @@ static int tx58cxgxsxraix_ecc_get_status
 		if (spi_mem_exec_op(spinand->spimem, &op))
 			return nanddev_get_ecc_conf(nand)->strength;
 
-		mbf >>= 4;
+		mbf = *(spinand->scratchbuf) >> 4;
 
 		if (WARN_ON(mbf > nanddev_get_ecc_conf(nand)->strength || !mbf))
 			return nanddev_get_ecc_conf(nand)->strength;
author maintainer
#65 lore raw

PATCH 63/92 mtd: rawnand: meson: fix OOB available bytes for ECC

In reply to Greg Kroah-Hartman

From: Arseniy Krasnov <[email protected]>

commit 7e6b04f9238eab0f684fafd158c1f32ea65b9eaa upstream.

It is incorrect to calculate number of OOB bytes for ECC engine using some "already known" ECC step size (1024 bytes here). Number of such bytes for ECC engine must be whole OOB except 2 bytes for bad block marker, while proper ECC step size and strength will be selected by ECC logic.

Fixes 8fae856c5350 ("mtd: rawnand: meson: add support for Amlogic NAND flash controller")
---
 drivers/mtd/nand/raw/meson_nand.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/mtd/nand/raw/meson_nand.c +1 -2
@@ -1180,7 +1180,6 @@ static int meson_nand_attach_chip(struct
 	struct meson_nfc *nfc = nand_get_controller_data(nand);
 	struct meson_nfc_nand_chip *meson_chip = to_meson_nand(nand);
 	struct mtd_info *mtd = nand_to_mtd(nand);
-	int nsectors = mtd->writesize / 1024;
 	int ret;
 
 	if (!mtd->name) {
@@ -1198,7 +1197,7 @@ static int meson_nand_attach_chip(struct
 	nand->options |= NAND_NO_SUBPAGE_WRITE;
 
 	ret = nand_ecc_choose_conf(nand, nfc->data->ecc_caps,
-				   mtd->oobsize - 2 * nsectors);
+				   mtd->oobsize - 2);
 	if (ret) {
 		dev_err(nfc->dev, "failed to ECC init\n");
 		return -EINVAL;
author maintainer
#66 lore raw

PATCH 64/92 arm64: dts: stratix10: fix incorrect I2C property for SCL signal

In reply to Greg Kroah-Hartman

From: Dinh Nguyen <[email protected]>

commit db66795f61354c373ecdadbdae1ed253a96c47cb upstream.

The correct dts property for the SCL falling time is "i2c-scl-falling-time-ns".

Fixes c8da1d15b8a4 ("arm64: dts: stratix10: i2c clock running out of spec")
---
 arch/arm64/boot/dts/altera/socfpga_stratix10_socdk.dts      |    2 +-
 arch/arm64/boot/dts/altera/socfpga_stratix10_socdk_nand.dts |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/altera/socfpga_stratix10_socdk.dts +1 -1
@@ -129,7 +129,7 @@
 	status = "okay";
 	clock-frequency = <100000>;
 	i2c-sda-falling-time-ns = <890>;  /* hcnt */
-	i2c-sdl-falling-time-ns = <890>;  /* lcnt */
+	i2c-scl-falling-time-ns = <890>;  /* lcnt */
 
 	adc@14 {
 		compatible = "lltc,ltc2497";
arch/arm64/boot/dts/altera/socfpga_stratix10_socdk_nand.dts +1 -1
@@ -162,7 +162,7 @@
 	status = "okay";
 	clock-frequency = <100000>;
 	i2c-sda-falling-time-ns = <890>;  /* hcnt */
-	i2c-sdl-falling-time-ns = <890>;  /* lcnt */
+	i2c-scl-falling-time-ns = <890>;  /* lcnt */
 
 	adc@14 {
 		compatible = "lltc,ltc2497";
author maintainer
#67 lore raw

PATCH 65/92 net: tun_chr_open(): set sk_uid from current_fsuid()

In reply to Greg Kroah-Hartman

From: Laszlo Ersek <[email protected]>

commit 9bc3047374d5bec163e83e743709e23753376f0c upstream.

Commit a096ccca6e50 initializes the "sk_uid" field in the protocol socket (struct sock) from the "/dev/net/tun" device node's owner UID. Per original commit 86741ec25462 ("net: core: Add a UID field to struct sock.", 2016-11-04), that's wrong: the idea is to cache the UID of the userspace process that creates the socket. Commit 86741ec25462 mentions socket() and accept(); with "tun", the action that creates the socket is open("/dev/net/tun").

Therefore the device node's owner UID is irrelevant. In most cases, "/dev/net/tun" will be owned by root, so in practice, commit a096ccca6e50 has no observable effect:

- before, "sk_uid" would be zero, due to undefined behavior (CVE-2023-1076),

- after, "sk_uid" would be zero, due to "/dev/net/tun" being owned by root.

What matters is the (fs)UID of the process performing the open(), so cache that in "sk_uid".

Cc Eric Dumazet <[email protected]>
Cc Lorenzo Colitti <[email protected]>
Cc Paolo Abeni <[email protected]>
Cc Pietro Borrello <[email protected]>
Fixes a096ccca6e50 ("tun: tun_chr_open(): correctly initialize socket uid")

Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2173435

---
 drivers/net/tun.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/tun.c +1 -1
@@ -3411,7 +3411,7 @@ static int tun_chr_open(struct inode *in
 	tfile->socket.file = file;
 	tfile->socket.ops = &tun_socket_ops;
 
-	sock_init_data_uid(&tfile->socket, &tfile->sk, inode->i_uid);
+	sock_init_data_uid(&tfile->socket, &tfile->sk, current_fsuid());
 
 	tfile->sk.sk_write_space = tun_sock_write_space;
 	tfile->sk.sk_sndbuf = INT_MAX;
author maintainer
#68 lore raw

PATCH 66/92 net: tap_open(): set sk_uid from current_fsuid()

In reply to Greg Kroah-Hartman

From: Laszlo Ersek <[email protected]>

commit 5c9241f3ceab3257abe2923a59950db0dc8bb737 upstream.

Commit 66b2c338adce initializes the "sk_uid" field in the protocol socket (struct sock) from the "/dev/tapX" device node's owner UID. Per original commit 86741ec25462 ("net: core: Add a UID field to struct sock.", 2016-11-04), that's wrong: the idea is to cache the UID of the userspace process that creates the socket. Commit 86741ec25462 mentions socket() and accept(); with "tap", the action that creates the socket is open("/dev/tapX").

Therefore the device node's owner UID is irrelevant. In most cases, "/dev/tapX" will be owned by root, so in practice, commit 66b2c338adce has no observable effect:

- before, "sk_uid" would be zero, due to undefined behavior (CVE-2023-1076),

- after, "sk_uid" would be zero, due to "/dev/tapX" being owned by root.

What matters is the (fs)UID of the process performing the open(), so cache that in "sk_uid".

Cc Eric Dumazet <[email protected]>
Cc Lorenzo Colitti <[email protected]>
Cc Paolo Abeni <[email protected]>
Cc Pietro Borrello <[email protected]>
Fixes 66b2c338adce ("tap: tap_open(): correctly initialize socket uid")

Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2173435

---
 drivers/net/tap.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/tap.c +1 -1
@@ -523,7 +523,7 @@ static int tap_open(struct inode *inode,
 	q->sock.state = SS_CONNECTED;
 	q->sock.file = file;
 	q->sock.ops = &tap_socket_ops;
-	sock_init_data_uid(&q->sock, &q->sk, inode->i_uid);
+	sock_init_data_uid(&q->sock, &q->sk, current_fsuid());
 	q->sk.sk_write_space = tap_sock_write_space;
 	q->sk.sk_destruct = tap_sock_destruct;
 	q->flags = IFF_VNET_HDR | IFF_NO_PI | IFF_TAP;
author maintainer
#69 lore raw

PATCH 67/92 wifi: mt76: mt7615: do not advertise 5 GHz on first phy of MT7615D (DBDC)

In reply to Greg Kroah-Hartman

From: Paul Fertser <[email protected]>

commit 421033deb91521aa6a9255e495cb106741a52275 upstream.

On DBDC devices the first (internal) phy is only capable of using 2.4 GHz band, and the 5 GHz band is exposed via a separate phy object, so avoid the false advertising.

---
 drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c +3 -3
@@ -123,12 +123,12 @@ mt7615_eeprom_parse_hw_band_cap(struct m
 	case MT_EE_5GHZ:
 		dev->mphy.cap.has_5ghz = true;
 		break;
-	case MT_EE_2GHZ:
-		dev->mphy.cap.has_2ghz = true;
-		break;
 	case MT_EE_DBDC:
 		dev->dbdc_support = true;
 		fallthrough;
+	case MT_EE_2GHZ:
+		dev->mphy.cap.has_2ghz = true;
+		break;
 	default:
 		dev->mphy.cap.has_2ghz = true;
 		dev->mphy.cap.has_5ghz = true;
author maintainer
#70 lore raw

PATCH 68/92 rbd: prevent busy loop when requesting exclusive lock

In reply to Greg Kroah-Hartman

From: Ilya Dryomov <[email protected]>

commit 9d01e07fd1bfb4daae156ab528aa196f5ac2b2bc upstream.

Due to rbd_try_acquire_lock() effectively swallowing all but EBLOCKLISTED error from rbd_try_lock() ("request lock anyway") and rbd_request_lock() returning ETIMEDOUT error not only for an actual notify timeout but also when the lock owner doesn't respond, a busy loop inside of rbd_acquire_lock() between rbd_try_acquire_lock() and rbd_request_lock() is possible.

Requesting the lock on EBUSY error (returned by get_lock_owner_info() if an incompatible lock or invalid lock owner is detected) makes very little sense. The same goes for ETIMEDOUT error (might pop up pretty much anywhere if osd_request_timeout option is set) and many others.

Just fail I/O requests on rbd_dev->acquiring_list immediately on any error from rbd_try_lock().

Cc [email protected] # 588159009d5b: rbd: retrieve and check lock owner twice before blocklisting
---
 drivers/block/rbd.c |   28 +++++++++++++++-------------
 1 file changed, 15 insertions(+), 13 deletions(-)
drivers/block/rbd.c +15 -13
@@ -3677,7 +3677,7 @@ static int rbd_lock(struct rbd_device *r
 	ret = ceph_cls_lock(osdc, &rbd_dev->header_oid, &rbd_dev->header_oloc,
 			    RBD_LOCK_NAME, CEPH_CLS_LOCK_EXCLUSIVE, cookie,
 			    RBD_LOCK_TAG, "", 0);
-	if (ret)
+	if (ret && ret != -EEXIST)
 		return ret;
 
 	__rbd_lock(rbd_dev, cookie);
@@ -3880,7 +3880,7 @@ static struct ceph_locker *get_lock_owne
 				 &rbd_dev->header_oloc, RBD_LOCK_NAME,
 				 &lock_type, &lock_tag, &lockers, &num_lockers);
 	if (ret) {
-		rbd_warn(rbd_dev, "failed to retrieve lockers: %d", ret);
+		rbd_warn(rbd_dev, "failed to get header lockers: %d", ret);
 		return ERR_PTR(ret);
 	}
 
@@ -3942,8 +3942,10 @@ static int find_watcher(struct rbd_devic
 	ret = ceph_osdc_list_watchers(osdc, &rbd_dev->header_oid,
 				      &rbd_dev->header_oloc, &watchers,
 				      &num_watchers);
-	if (ret)
+	if (ret) {
+		rbd_warn(rbd_dev, "failed to get watchers: %d", ret);
 		return ret;
+	}
 
 	sscanf(locker->id.cookie, RBD_LOCK_COOKIE_PREFIX " %llu", &cookie);
 	for (i = 0; i < num_watchers; i++) {
@@ -3987,8 +3989,12 @@ static int rbd_try_lock(struct rbd_devic
 		locker = refreshed_locker = NULL;
 
 		ret = rbd_lock(rbd_dev);
-		if (ret != -EBUSY)
+		if (!ret)
+			goto out;
+		if (ret != -EBUSY) {
+			rbd_warn(rbd_dev, "failed to lock header: %d", ret);
 			goto out;
+		}
 
 		/* determine if the current lock holder is still alive */
 		locker = get_lock_owner_info(rbd_dev);
@@ -4091,11 +4097,8 @@ static int rbd_try_acquire_lock(struct r
 
 	ret = rbd_try_lock(rbd_dev);
 	if (ret < 0) {
-		rbd_warn(rbd_dev, "failed to lock header: %d", ret);
-		if (ret == -EBLOCKLISTED)
-			goto out;
-
-		ret = 1; /* request lock anyway */
+		rbd_warn(rbd_dev, "failed to acquire lock: %d", ret);
+		goto out;
 	}
 	if (ret > 0) {
 		up_write(&rbd_dev->lock_rwsem);
@@ -6631,12 +6634,11 @@ static int rbd_add_acquire_lock(struct r
 		cancel_delayed_work_sync(&rbd_dev->lock_dwork);
 		if (!ret)
 			ret = -ETIMEDOUT;
-	}
 
-	if (ret) {
-		rbd_warn(rbd_dev, "failed to acquire exclusive lock: %ld", ret);
-		return ret;
+		rbd_warn(rbd_dev, "failed to acquire lock: %ld", ret);
 	}
+	if (ret)
+		return ret;
 
 	/*
 	 * The lock may have been released by now, unless automatic lock
author maintainer
#71 lore raw

PATCH 69/92 bpf: Disable preemption in bpf_event_output

In reply to Greg Kroah-Hartman

From: Jiri Olsa <[email protected]>

commit d62cc390c2e99ae267ffe4b8d7e2e08b6c758c32 upstream.

We received report [1] of kernel crash, which is caused by using nesting protection without disabled preemption.

The bpf_event_output can be called by programs executed by bpf_prog_run_array_cg function that disabled migration but keeps preemption enabled.

This can cause task to be preempted by another one inside the nesting protection and lead eventually to two tasks using same perf_sample_data buffer and cause crashes like:

  BUG: kernel NULL pointer dereference, address: 0000000000000001
  #PF: supervisor instruction fetch in kernel mode
  #PF: error_code(0x0010) - not-present page
  ...
  ? perf_output_sample+0x12a/0x9a0
  ? finish_task_switch.isra.0+0x81/0x280
  ? perf_event_output+0x66/0xa0
  ? bpf_event_output+0x13a/0x190
  ? bpf_event_output_data+0x22/0x40
  ? bpf_prog_dfc84bbde731b257_cil_sock4_connect+0x40a/0xacb
  ? xa_load+0x87/0xe0
  ? __cgroup_bpf_run_filter_sock_addr+0xc1/0x1a0
  ? release_sock+0x3e/0x90
  ? sk_setsockopt+0x1a1/0x12f0
  ? udp_pre_connect+0x36/0x50
  ? inet_dgram_connect+0x93/0xa0
  ? __sys_connect+0xb4/0xe0
  ? udp_setsockopt+0x27/0x40
  ? __pfx_udp_push_pending_frames+0x10/0x10
  ? __sys_setsockopt+0xdf/0x1a0
  ? __x64_sys_connect+0xf/0x20
  ? do_syscall_64+0x3a/0x90
  ? entry_SYSCALL_64_after_hwframe+0x72/0xdc

Fixing this by disabling preemption in bpf_event_output.

[1] https://github.com/cilium/cilium/issues/26756

---
 kernel/trace/bpf_trace.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
kernel/trace/bpf_trace.c +5 -1
@@ -662,7 +662,6 @@ static DEFINE_PER_CPU(struct bpf_trace_s
 u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size,
 		     void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy)
 {
-	int nest_level = this_cpu_inc_return(bpf_event_output_nest_level);
 	struct perf_raw_frag frag = {
 		.copy		= ctx_copy,
 		.size		= ctx_size,
@@ -679,8 +678,12 @@ u64 bpf_event_output(struct bpf_map *map
 	};
 	struct perf_sample_data *sd;
 	struct pt_regs *regs;
+	int nest_level;
 	u64 ret;
 
+	preempt_disable();
+	nest_level = this_cpu_inc_return(bpf_event_output_nest_level);
+
 	if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(bpf_misc_sds.sds))) {
 		ret = -EBUSY;
 		goto out;
@@ -695,6 +698,7 @@ u64 bpf_event_output(struct bpf_map *map
 	ret = __bpf_perf_event_output(regs, map, flags, sd);
 out:
 	this_cpu_dec(bpf_event_output_nest_level);
+	preempt_enable();
 	return ret;
 }
 
author maintainer
#72 lore raw

PATCH 70/92 open: make RESOLVE_CACHED correctly test for O_TMPFILE

In reply to Greg Kroah-Hartman

From: Aleksa Sarai <[email protected]>

commit a0fc452a5d7fed986205539259df1d60546f536c upstream.

O_TMPFILE is actually __O_TMPFILE|O_DIRECTORY. This means that the old fast-path check for RESOLVE_CACHED would reject all users passing O_DIRECTORY with -EAGAIN, when in fact the intended test was to check for __O_TMPFILE.

Cc [email protected] # v5.12+
Fixes 99668f618062 ("fs: expose LOOKUP_CACHED through openat2() RESOLVE_CACHED")

Message-Id: <[email protected]>

---
 fs/open.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/open.c +1 -1
@@ -1126,7 +1126,7 @@ inline int build_open_flags(const struct
 		lookup_flags |= LOOKUP_IN_ROOT;
 	if (how->resolve & RESOLVE_CACHED) {
 		/* Don't bother even trying for create/truncate/tmpfile open */
-		if (flags & (O_TRUNC | O_CREAT | O_TMPFILE))
+		if (flags & (O_TRUNC | O_CREAT | __O_TMPFILE))
 			return -EAGAIN;
 		lookup_flags |= LOOKUP_CACHED;
 	}
author maintainer
#73 lore raw

PATCH 71/92 drm/ttm: check null pointer before accessing when swapping

In reply to Greg Kroah-Hartman

From: Guchun Chen <[email protected]>

commit 2dedcf414bb01b8d966eb445db1d181d92304fb2 upstream.

Add a check to avoid null pointer dereference as below:

[   90.002283] general protection fault, probably for non-canonical
address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN NOPTI
[   90.002292] KASAN: null-ptr-deref in range
[0x0000000000000000-0x0000000000000007]
[   90.002346]  ? exc_general_protection+0x159/0x240
[   90.002352]  ? asm_exc_general_protection+0x26/0x30
[   90.002357]  ? ttm_bo_evict_swapout_allowable+0x322/0x5e0 [ttm]
[   90.002365]  ? ttm_bo_evict_swapout_allowable+0x42e/0x5e0 [ttm]
[   90.002373]  ttm_bo_swapout+0x134/0x7f0 [ttm]
[   90.002383]  ? __pfx_ttm_bo_swapout+0x10/0x10 [ttm]
[   90.002391]  ? lock_acquire+0x44d/0x4f0
[   90.002398]  ? ttm_device_swapout+0xa5/0x260 [ttm]
[   90.002412]  ? lock_acquired+0x355/0xa00
[   90.002416]  ? do_raw_spin_trylock+0xb6/0x190
[   90.002421]  ? __pfx_lock_acquired+0x10/0x10
[   90.002426]  ? ttm_global_swapout+0x25/0x210 [ttm]
[   90.002442]  ttm_device_swapout+0x198/0x260 [ttm]
[   90.002456]  ? __pfx_ttm_device_swapout+0x10/0x10 [ttm]
[   90.002472]  ttm_global_swapout+0x75/0x210 [ttm]
[   90.002486]  ttm_tt_populate+0x187/0x3f0 [ttm]
[   90.002501]  ttm_bo_handle_move_mem+0x437/0x590 [ttm]
[   90.002517]  ttm_bo_validate+0x275/0x430 [ttm]
[   90.002530]  ? __pfx_ttm_bo_validate+0x10/0x10 [ttm]
[   90.002544]  ? kasan_save_stack+0x33/0x60
[   90.002550]  ? kasan_set_track+0x25/0x30
[   90.002554]  ? __kasan_kmalloc+0x8f/0xa0
[   90.002558]  ? amdgpu_gtt_mgr_new+0x81/0x420 [amdgpu]
[   90.003023]  ? ttm_resource_alloc+0xf6/0x220 [ttm]
[   90.003038]  amdgpu_bo_pin_restricted+0x2dd/0x8b0 [amdgpu]
[   90.003210]  ? __x64_sys_ioctl+0x131/0x1a0
[   90.003210]  ? do_syscall_64+0x60/0x90
---
 drivers/gpu/drm/ttm/ttm_bo.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/ttm/ttm_bo.c +2 -1
@@ -606,7 +606,8 @@ static bool ttm_bo_evict_swapout_allowab
 
 	if (bo->pin_count) {
 		*locked = false;
-		*busy = false;
+		if (busy)
+			*busy = false;
 		return false;
 	}
 
author maintainer
#74 lore raw

PATCH 72/92 bpf, cpumap: Make sure kthread is running before map update returns

In reply to Greg Kroah-Hartman

From: Hou Tao <[email protected]>

commit 640a604585aa30f93e39b17d4d6ba69fcb1e66c9 upstream.

The following warning was reported when running stress-mode enabled xdp_redirect_cpu with some RT threads:

  ------------[ cut here ]------------
  WARNING: CPU: 4 PID: 65 at kernel/bpf/cpumap.c:135
  CPU: 4 PID: 65 Comm: kworker/4:1 Not tainted 6.5.0-rc2+ #1
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
  Workqueue: events cpu_map_kthread_stop
  RIP: 0010:put_cpu_map_entry+0xda/0x220
  ......
  Call Trace:
   <TASK>
   ? show_regs+0x65/0x70
   ? __warn+0xa5/0x240
   ......
   ? put_cpu_map_entry+0xda/0x220
   cpu_map_kthread_stop+0x41/0x60
   process_one_work+0x6b0/0xb80
   worker_thread+0x96/0x720
   kthread+0x1a5/0x1f0
   ret_from_fork+0x3a/0x70
   ret_from_fork_asm+0x1b/0x30
   </TASK>

The root cause is the same as commit 436901649731 ("bpf: cpumap: Fix memory leak in cpu_map_update_elem"). The kthread is stopped prematurely by kthread_stop() in cpu_map_kthread_stop(), and kthread() doesn't call cpu_map_kthread_run() at all but XDP program has already queued some frames or skbs into ptr_ring. So when __cpu_map_ring_cleanup() checks the ptr_ring, it will find it was not emptied and report a warning.

An alternative fix is to use __cpu_map_ring_cleanup() to drop these pending frames or skbs when kthread_stop() returns -EINTR, but it may confuse the user, because these frames or skbs have been handled correctly by XDP program. So instead of dropping these frames or skbs, just make sure the per-cpu kthread is running before __cpu_map_entry_alloc() returns.

After apply the fix, the error handle for kthread_stop() will be unnecessary because it will always return 0, so just remove it.

---
 kernel/bpf/cpumap.c |   21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)
kernel/bpf/cpumap.c +11 -10
@@ -26,6 +26,7 @@
 #include <linux/workqueue.h>
 #include <linux/kthread.h>
 #include <linux/capability.h>
+#include <linux/completion.h>
 #include <trace/events/xdp.h>
 
 #include <linux/netdevice.h>   /* netif_receive_skb_list */
@@ -70,6 +71,7 @@ struct bpf_cpu_map_entry {
 	struct rcu_head rcu;
 
 	struct work_struct kthread_stop_wq;
+	struct completion kthread_running;
 };
 
 struct bpf_cpu_map {
@@ -163,7 +165,6 @@ static void put_cpu_map_entry(struct bpf
 static void cpu_map_kthread_stop(struct work_struct *work)
 {
 	struct bpf_cpu_map_entry *rcpu;
-	int err;
 
 	rcpu = container_of(work, struct bpf_cpu_map_entry, kthread_stop_wq);
 
@@ -173,14 +174,7 @@ static void cpu_map_kthread_stop(struct
 	rcu_barrier();
 
 	/* kthread_stop will wake_up_process and wait for it to complete */
-	err = kthread_stop(rcpu->kthread);
-	if (err) {
-		/* kthread_stop may be called before cpu_map_kthread_run
-		 * is executed, so we need to release the memory related
-		 * to rcpu.
-		 */
-		put_cpu_map_entry(rcpu);
-	}
+	kthread_stop(rcpu->kthread);
 }
 
 static void cpu_map_bpf_prog_run_skb(struct bpf_cpu_map_entry *rcpu,
@@ -308,11 +302,11 @@ static int cpu_map_bpf_prog_run(struct b
 	return nframes;
 }
 
-
 static int cpu_map_kthread_run(void *data)
 {
 	struct bpf_cpu_map_entry *rcpu = data;
 
+	complete(&rcpu->kthread_running);
 	set_current_state(TASK_INTERRUPTIBLE);
 
 	/* When kthread gives stop order, then rcpu have been disconnected
@@ -475,6 +469,7 @@ __cpu_map_entry_alloc(struct bpf_map *ma
 		goto free_ptr_ring;
 
 	/* Setup kthread */
+	init_completion(&rcpu->kthread_running);
 	rcpu->kthread = kthread_create_on_node(cpu_map_kthread_run, rcpu, numa,
 					       "cpumap/%d/map:%d", cpu,
 					       map->id);
@@ -488,6 +483,12 @@ __cpu_map_entry_alloc(struct bpf_map *ma
 	kthread_bind(rcpu->kthread, cpu);
 	wake_up_process(rcpu->kthread);
 
+	/* Make sure kthread has been running, so kthread_stop() will not
+	 * stop the kthread prematurely and all pending frames or skbs
+	 * will be handled by the kthread before kthread_stop() returns.
+	 */
+	wait_for_completion(&rcpu->kthread_running);
+
 	return rcpu;
 
 free_prog:
author maintainer
#75 lore raw

PATCH 73/92 file: reinstate f_pos locking optimization for regular files

In reply to Greg Kroah-Hartman

From: Linus Torvalds <[email protected]>

commit 797964253d358cf8d705614dda394dbe30120223 upstream.

In commit 20ea1e7d13c1 ("file: always lock position for FMODE_ATOMIC_POS") we ended up always taking the file pos lock, because pidfd_getfd() could get a reference to the file even when it didn't have an elevated file count due to threading of other sharing cases.

But Mateusz Guzik reports that the extra locking is actually measurable, so let's re-introduce the optimization, and only force the locking for directory traversal.

Directories need the lock for correctness reasons, while regular files only need it for "POSIX semantics". Since pidfd_getfd() is about debuggers etc special things that are _way_ outside of POSIX, we can relax the rules for that case.

---
 fs/file.c |   18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)
fs/file.c +17 -1
@@ -1062,12 +1062,28 @@ unsigned long __fdget_raw(unsigned int f
 	return __fget_light(fd, 0);
 }
 
+/*
+ * Try to avoid f_pos locking. We only need it if the
+ * file is marked for FMODE_ATOMIC_POS, and it can be
+ * accessed multiple ways.
+ *
+ * Always do it for directories, because pidfd_getfd()
+ * can make a file accessible even if it otherwise would
+ * not be, and for directories this is a correctness
+ * issue, not a "POSIX requirement".
+ */
+static inline bool file_needs_f_pos_lock(struct file *file)
+{
+	return (file->f_mode & FMODE_ATOMIC_POS) &&
+		(file_count(file) > 1 || S_ISDIR(file_inode(file)->i_mode));
+}
+
 unsigned long __fdget_pos(unsigned int fd)
 {
 	unsigned long v = __fdget(fd);
 	struct file *file = (struct file *)(v & ~3);
 
-	if (file && (file->f_mode & FMODE_ATOMIC_POS)) {
+	if (file && file_needs_f_pos_lock(file)) {
 		v |= FDPUT_POS_UNLOCK;
 		mutex_lock(&file->f_pos_lock);
 	}
author maintainer
#76 lore raw

PATCH 74/92 fs/ntfs3: Use __GFP_NOWARN allocation at ntfs_load_attr_list()

In reply to Greg Kroah-Hartman

From: Tetsuo Handa <[email protected]>

commit ea303f72d70ce2f0b0aa94ab127085289768c5a6 upstream.

syzbot is reporting too large allocation at ntfs_load_attr_list(), for a crafted filesystem can have huge data_size.

---
 fs/ntfs3/attrlist.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/ntfs3/attrlist.c +2 -2
@@ -52,7 +52,7 @@ int ntfs_load_attr_list(struct ntfs_inod
 
 	if (!attr->non_res) {
 		lsize = le32_to_cpu(attr->res.data_size);
-		le = kmalloc(al_aligned(lsize), GFP_NOFS);
+		le = kmalloc(al_aligned(lsize), GFP_NOFS | __GFP_NOWARN);
 		if (!le) {
 			err = -ENOMEM;
 			goto out;
@@ -80,7 +80,7 @@ int ntfs_load_attr_list(struct ntfs_inod
 		if (err < 0)
 			goto out;
 
-		le = kmalloc(al_aligned(lsize), GFP_NOFS);
+		le = kmalloc(al_aligned(lsize), GFP_NOFS | __GFP_NOWARN);
 		if (!le) {
 			err = -ENOMEM;
 			goto out;
author maintainer
#77 lore raw

PATCH 75/92 fs/sysv: Null check to prevent null-ptr-deref bug

In reply to Greg Kroah-Hartman

From: Prince Kumar Maurya <[email protected]>

commit ea2b62f305893992156a798f665847e0663c9f41 upstream.

sb_getblk(inode->i_sb, parent) return a null ptr and taking lock on that leads to the null-ptr-deref bug.

Message-Id: <[email protected]>

---
 fs/sysv/itree.c |    4 ++++
 1 file changed, 4 insertions(+)
fs/sysv/itree.c +4 -0
@@ -145,6 +145,10 @@ static int alloc_branch(struct inode *in
 		 */
 		parent = block_to_cpu(SYSV_SB(inode->i_sb), branch[n-1].key);
 		bh = sb_getblk(inode->i_sb, parent);
+		if (!bh) {
+			sysv_free_block(inode->i_sb, branch[n].key);
+			break;
+		}
 		lock_buffer(bh);
 		memset(bh->b_data, 0, blocksize);
 		branch[n].bh = bh;
author maintainer
#78 lore raw

PATCH 76/92 Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb

In reply to Greg Kroah-Hartman

From: Sungwoo Kim <[email protected]>

commit 1728137b33c00d5a2b5110ed7aafb42e7c32e4a1 upstream.

l2cap_sock_release(sk) frees sk. However, sk's children are still alive and point to the already free'd sk's address.
To fix this, l2cap_sock_release(sk) also cleans sk's children.

==================================================================

BUG: KASAN: use-after-free in l2cap_sock_ready_cb+0xb7/0x100 net/bluetooth/l2cap_sock.c:1650
Read of size 8 at addr ffff888104617aa8 by task kworker/u3:0/276

CPU: 0 PID: 276 Comm: kworker/u3:0 Not tainted 6.2.0-00001-gef397bd4d5fb-dirty #59
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
Workqueue: hci2 hci_rx_work
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x72/0x95 lib/dump_stack.c:106
 print_address_description mm/kasan/report.c:306 [inline]
 print_report+0x175/0x478 mm/kasan/report.c:417
 kasan_report+0xb1/0x130 mm/kasan/report.c:517
 l2cap_sock_ready_cb+0xb7/0x100 net/bluetooth/l2cap_sock.c:1650
 l2cap_chan_ready+0x10e/0x1e0 net/bluetooth/l2cap_core.c:1386
 l2cap_config_req+0x753/0x9f0 net/bluetooth/l2cap_core.c:4480
 l2cap_bredr_sig_cmd net/bluetooth/l2cap_core.c:5739 [inline]
 l2cap_sig_channel net/bluetooth/l2cap_core.c:6509 [inline]
 l2cap_recv_frame+0xe2e/0x43c0 net/bluetooth/l2cap_core.c:7788
 l2cap_recv_acldata+0x6ed/0x7e0 net/bluetooth/l2cap_core.c:8506
 hci_acldata_packet net/bluetooth/hci_core.c:3813 [inline]
 hci_rx_work+0x66e/0xbc0 net/bluetooth/hci_core.c:4048
 process_one_work+0x4ea/0x8e0 kernel/workqueue.c:2289
 worker_thread+0x364/0x8e0 kernel/workqueue.c:2436
 kthread+0x1b9/0x200 kernel/kthread.c:376
 ret_from_fork+0x2c/0x50 arch/x86/entry/entry_64.S:308
 </TASK>

Allocated by task 288:
 kasan_save_stack+0x22/0x50 mm/kasan/common.c:45
 kasan_set_track+0x25/0x30 mm/kasan/common.c:52
 ____kasan_kmalloc mm/kasan/common.c:374 [inline]
 __kasan_kmalloc+0x82/0x90 mm/kasan/common.c:383
 kasan_kmalloc include/linux/kasan.h:211 [inline]
 __do_kmalloc_node mm/slab_common.c:968 [inline]
 __kmalloc+0x5a/0x140 mm/slab_common.c:981
 kmalloc include/linux/slab.h:584 [inline]
 sk_prot_alloc+0x113/0x1f0 net/core/sock.c:2040
 sk_alloc+0x36/0x3c0 net/core/sock.c:2093
 l2cap_sock_alloc.constprop.0+0x39/0x1c0 net/bluetooth/l2cap_sock.c:1852
 l2cap_sock_create+0x10d/0x220 net/bluetooth/l2cap_sock.c:1898
 bt_sock_create+0x183/0x290 net/bluetooth/af_bluetooth.c:132
 __sock_create+0x226/0x380 net/socket.c:1518
 sock_create net/socket.c:1569 [inline]
 __sys_socket_create net/socket.c:1606 [inline]
 __sys_socket_create net/socket.c:1591 [inline]
 __sys_socket+0x112/0x200 net/socket.c:1639
 __do_sys_socket net/socket.c:1652 [inline]
 __se_sys_socket net/socket.c:1650 [inline]
 __x64_sys_socket+0x40/0x50 net/socket.c:1650
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x3f/0x90 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x72/0xdc

Freed by task 288:
 kasan_save_stack+0x22/0x50 mm/kasan/common.c:45
 kasan_set_track+0x25/0x30 mm/kasan/common.c:52
 kasan_save_free_info+0x2e/0x50 mm/kasan/generic.c:523
 ____kasan_slab_free mm/kasan/common.c:236 [inline]
 ____kasan_slab_free mm/kasan/common.c:200 [inline]
 __kasan_slab_free+0x10a/0x190 mm/kasan/common.c:244
 kasan_slab_free include/linux/kasan.h:177 [inline]
 slab_free_hook mm/slub.c:1781 [inline]
 slab_free_freelist_hook mm/slub.c:1807 [inline]
 slab_free mm/slub.c:3787 [inline]
 __kmem_cache_free+0x88/0x1f0 mm/slub.c:3800
 sk_prot_free net/core/sock.c:2076 [inline]
 __sk_destruct+0x347/0x430 net/core/sock.c:2168
 sk_destruct+0x9c/0xb0 net/core/sock.c:2183
 __sk_free+0x82/0x220 net/core/sock.c:2194
 sk_free+0x7c/0xa0 net/core/sock.c:2205
 sock_put include/net/sock.h:1991 [inline]
 l2cap_sock_kill+0x256/0x2b0 net/bluetooth/l2cap_sock.c:1257
 l2cap_sock_release+0x1a7/0x220 net/bluetooth/l2cap_sock.c:1428
 __sock_release+0x80/0x150 net/socket.c:650
 sock_close+0x19/0x30 net/socket.c:1368
 __fput+0x17a/0x5c0 fs/file_table.c:320
 task_work_run+0x132/0x1c0 kernel/task_work.c:179
 resume_user_mode_work include/linux/resume_user_mode.h:49 [inline]
 exit_to_user_mode_loop kernel/entry/common.c:171 [inline]
 exit_to_user_mode_prepare+0x113/0x120 kernel/entry/common.c:203
 __syscall_exit_to_user_mode_work kernel/entry/common.c:285 [inline]
 syscall_exit_to_user_mode+0x21/0x50 kernel/entry/common.c:296
 do_syscall_64+0x4c/0x90 arch/x86/entry/common.c:86
 entry_SYSCALL_64_after_hwframe+0x72/0xdc

The buggy address belongs to the object at ffff888104617800
 which belongs to the cache kmalloc-1k of size 1024

The buggy address is located 680 bytes inside of
1024-byte region [ffff888104617800, ffff888104617c00)

The buggy address belongs to the physical page:
page:00000000dbca6a80 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff888104614000 pfn:0x104614
head:00000000dbca6a80 order:2 compound_mapcount:0 subpages_mapcount:0 compound_pincount:0
flags: 0x200000000010200(slab|head|node=0|zone=2)
raw: 0200000000010200 ffff888100041dc0 ffffea0004212c10 ffffea0004234b10
raw: ffff888104614000 0000000000080002 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff888104617980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff888104617a00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff888104617a80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                  ^
 ffff888104617b00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff888104617b80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================

Ack: This bug is found by FuzzBT with a modified Syzkaller. Other contributors are Ruoyu Wu and Hui Peng.

---
 net/bluetooth/l2cap_sock.c |    2 ++
 1 file changed, 2 insertions(+)
net/bluetooth/l2cap_sock.c +2 -0
@@ -45,6 +45,7 @@ static const struct proto_ops l2cap_sock
 static void l2cap_sock_init(struct sock *sk, struct sock *parent);
 static struct sock *l2cap_sock_alloc(struct net *net, struct socket *sock,
 				     int proto, gfp_t prio, int kern);
+static void l2cap_sock_cleanup_listen(struct sock *parent);
 
 bool l2cap_is_socket(struct socket *sock)
 {
@@ -1414,6 +1415,7 @@ static int l2cap_sock_release(struct soc
 	if (!sk)
 		return 0;
 
+	l2cap_sock_cleanup_listen(sk);
 	bt_sock_unlink(&l2cap_sk_list, sk);
 
 	err = l2cap_sock_shutdown(sock, SHUT_RDWR);
author maintainer
#79 lore raw

PATCH 77/92 net: usbnet: Fix WARNING in usbnet_start_xmit/usb_submit_urb

In reply to Greg Kroah-Hartman

From: Alan Stern <[email protected]>

commit 5e1627cb43ddf1b24b92eb26f8d958a3f5676ccb upstream.

The syzbot fuzzer identified a problem in the usbnet driver:

usb 1-1: BOGUS urb xfer, pipe 3 != type 1

WARNING: CPU: 0 PID: 754 at drivers/usb/core/urb.c:504 usb_submit_urb+0xed6/0x1880 drivers/usb/core/urb.c:504
Modules linked in:
CPU: 0 PID: 754 Comm: kworker/0:2 Not tainted 6.4.0-rc7-syzkaller-00014-g692b7dc87ca6 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/27/2023
Workqueue: mld mld_ifc_work
RIP: 0010:usb_submit_urb+0xed6/0x1880 drivers/usb/core/urb.c:504
Code: 7c 24 18 e8 2c b4 5b fb 48 8b 7c 24 18 e8 42 07 f0 fe 41 89 d8 44 89 e1 4c 89 ea 48 89 c6 48 c7 c7 a0 c9 fc 8a e8 5a 6f 23 fb <0f> 0b e9 58 f8 ff ff e8 fe b3 5b fb 48 81 c5 c0 05 00 00 e9 84 f7
RSP: 0018:ffffc9000463f568 EFLAGS: 00010086
RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000
RDX: ffff88801eb28000 RSI: ffffffff814c03b7 RDI: 0000000000000001
RBP: ffff8881443b7190 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000003
R13: ffff88802a77cb18 R14: 0000000000000003 R15: ffff888018262500
FS:  0000000000000000(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000556a99c15a18 CR3: 0000000028c71000 CR4: 0000000000350ef0
Call Trace:
 <TASK>
 usbnet_start_xmit+0xfe5/0x2190 drivers/net/usb/usbnet.c:1453
 __netdev_start_xmit include/linux/netdevice.h:4918 [inline]
 netdev_start_xmit include/linux/netdevice.h:4932 [inline]
 xmit_one net/core/dev.c:3578 [inline]
 dev_hard_start_xmit+0x187/0x700 net/core/dev.c:3594

...

This bug is caused by the fact that usbnet trusts the bulk endpoint addresses its probe routine receives in the driver_info structure, and it does not check to see that these endpoints actually exist and have the expected type and directions.

The fix is simply to add such a check.

---
 drivers/net/usb/usbnet.c |    6 ++++++
 1 file changed, 6 insertions(+)
drivers/net/usb/usbnet.c +6 -0
@@ -1771,6 +1771,10 @@ usbnet_probe (struct usb_interface *udev
 	} else if (!info->in || !info->out)
 		status = usbnet_get_endpoints (dev, udev);
 	else {
+		u8 ep_addrs[3] = {
+			info->in + USB_DIR_IN, info->out + USB_DIR_OUT, 0
+		};
+
 		dev->in = usb_rcvbulkpipe (xdev, info->in);
 		dev->out = usb_sndbulkpipe (xdev, info->out);
 		if (!(info->flags & FLAG_NO_SETINT))
@@ -1780,6 +1784,8 @@ usbnet_probe (struct usb_interface *udev
 		else
 			status = 0;
 
+		if (status == 0 && !usb_check_bulk_endpoints(udev, ep_addrs))
+			status = -EINVAL;
 	}
 	if (status >= 0 && dev->status)
 		status = init_status (dev, udev);
author maintainer
#80 lore raw

PATCH 78/92 fs: Protect reconfiguration of sb read-write from racing writes

In reply to Greg Kroah-Hartman

From: Jan Kara <[email protected]>

commit c541dce86c537714b6761a79a969c1623dfa222b upstream.

The reconfigure / remount code takes a lot of effort to protect filesystem's reconfiguration code from racing writes on remounting read-only. However during remounting read-only filesystem to read-write mode userspace writes can start immediately once we clear SB_RDONLY flag. This is inconvenient for example for ext4 because we need to do some writes to the filesystem (such as preparation of quota files) before we can take userspace writes so we are clearing SB_RDONLY flag before we are fully ready to accept userpace writes and syzbot has found a way to exploit this [1]. Also as far as I'm reading the code the filesystem remount code was protected from racing writes in the legacy mount path by the mount's MNT_READONLY flag so this is relatively new problem. It is actually fairly easy to protect remount read-write from racing writes using sb->s_readonly_remount flag so let's just do that instead of having to workaround these races in the filesystem code.

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

Message-Id: <[email protected]>

---
 fs/super.c |   11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
fs/super.c +10 -1
@@ -863,6 +863,7 @@ int reconfigure_super(struct fs_context
 	struct super_block *sb = fc->root->d_sb;
 	int retval;
 	bool remount_ro = false;
+	bool remount_rw = false;
 	bool force = fc->sb_flags & SB_FORCE;
 
 	if (fc->sb_flags_mask & ~MS_RMT_MASK)
@@ -880,7 +881,7 @@ int reconfigure_super(struct fs_context
 		    bdev_read_only(sb->s_bdev))
 			return -EACCES;
 #endif
-
+		remount_rw = !(fc->sb_flags & SB_RDONLY) && sb_rdonly(sb);
 		remount_ro = (fc->sb_flags & SB_RDONLY) && !sb_rdonly(sb);
 	}
 
@@ -910,6 +911,14 @@ int reconfigure_super(struct fs_context
 			if (retval)
 				return retval;
 		}
+	} else if (remount_rw) {
+		/*
+		 * We set s_readonly_remount here to protect filesystem's
+		 * reconfigure code from writes from userspace until
+		 * reconfigure finishes.
+		 */
+		sb->s_readonly_remount = 1;
+		smp_wmb();
 	}
 
 	if (fc->ops->reconfigure) {
author maintainer
#81 lore raw

PATCH 79/92 ext2: Drop fragment support

In reply to Greg Kroah-Hartman

From: Jan Kara <[email protected]>

commit 404615d7f1dcd4cca200e9a7a9df3a1dcae1dd62 upstream.

Ext2 has fields in superblock reserved for subblock allocation support. However that never landed. Drop the many years dead code.

---
 fs/ext2/ext2.h  |   12 ------------
 fs/ext2/super.c |   23 ++++-------------------
 2 files changed, 4 insertions(+), 31 deletions(-)
fs/ext2/ext2.h +0 -12
@@ -70,10 +70,7 @@ struct mb_cache;
  * second extended-fs super-block data in memory
  */
 struct ext2_sb_info {
-	unsigned long s_frag_size;	/* Size of a fragment in bytes */
-	unsigned long s_frags_per_block;/* Number of fragments per block */
 	unsigned long s_inodes_per_block;/* Number of inodes per block */
-	unsigned long s_frags_per_group;/* Number of fragments in a group */
 	unsigned long s_blocks_per_group;/* Number of blocks in a group */
 	unsigned long s_inodes_per_group;/* Number of inodes in a group */
 	unsigned long s_itb_per_group;	/* Number of inode table blocks per group */
@@ -188,15 +185,6 @@ static inline struct ext2_sb_info *EXT2_
 #define EXT2_FIRST_INO(s)		(EXT2_SB(s)->s_first_ino)
 
 /*
- * Macro-instructions used to manage fragments
- */
-#define EXT2_MIN_FRAG_SIZE		1024
-#define	EXT2_MAX_FRAG_SIZE		4096
-#define EXT2_MIN_FRAG_LOG_SIZE		  10
-#define EXT2_FRAG_SIZE(s)		(EXT2_SB(s)->s_frag_size)
-#define EXT2_FRAGS_PER_BLOCK(s)		(EXT2_SB(s)->s_frags_per_block)
-
-/*
  * Structure of a blocks group descriptor
  */
 struct ext2_group_desc
fs/ext2/super.c +4 -19
@@ -670,10 +670,9 @@ static int ext2_setup_super (struct supe
 		es->s_max_mnt_count = cpu_to_le16(EXT2_DFL_MAX_MNT_COUNT);
 	le16_add_cpu(&es->s_mnt_count, 1);
 	if (test_opt (sb, DEBUG))
-		ext2_msg(sb, KERN_INFO, "%s, %s, bs=%lu, fs=%lu, gc=%lu, "
+		ext2_msg(sb, KERN_INFO, "%s, %s, bs=%lu, gc=%lu, "
 			"bpg=%lu, ipg=%lu, mo=%04lx]",
 			EXT2FS_VERSION, EXT2FS_DATE, sb->s_blocksize,
-			sbi->s_frag_size,
 			sbi->s_groups_count,
 			EXT2_BLOCKS_PER_GROUP(sb),
 			EXT2_INODES_PER_GROUP(sb),
@@ -1012,14 +1011,7 @@ static int ext2_fill_super(struct super_
 		}
 	}
 
-	sbi->s_frag_size = EXT2_MIN_FRAG_SIZE <<
-				   le32_to_cpu(es->s_log_frag_size);
-	if (sbi->s_frag_size == 0)
-		goto cantfind_ext2;
-	sbi->s_frags_per_block = sb->s_blocksize / sbi->s_frag_size;
-
 	sbi->s_blocks_per_group = le32_to_cpu(es->s_blocks_per_group);
-	sbi->s_frags_per_group = le32_to_cpu(es->s_frags_per_group);
 	sbi->s_inodes_per_group = le32_to_cpu(es->s_inodes_per_group);
 
 	sbi->s_inodes_per_block = sb->s_blocksize / EXT2_INODE_SIZE(sb);
@@ -1045,11 +1037,10 @@ static int ext2_fill_super(struct super_
 		goto failed_mount;
 	}
 
-	if (sb->s_blocksize != sbi->s_frag_size) {
+	if (es->s_log_frag_size != es->s_log_block_size) {
 		ext2_msg(sb, KERN_ERR,
-			"error: fragsize %lu != blocksize %lu"
-			"(not supported yet)",
-			sbi->s_frag_size, sb->s_blocksize);
+			"error: fragsize log %u != blocksize log %u",
+			le32_to_cpu(es->s_log_frag_size), sb->s_blocksize_bits);
 		goto failed_mount;
 	}
 
@@ -1059,12 +1050,6 @@ static int ext2_fill_super(struct super_
 			sbi->s_blocks_per_group);
 		goto failed_mount;
 	}
-	if (sbi->s_frags_per_group > sb->s_blocksize * 8) {
-		ext2_msg(sb, KERN_ERR,
-			"error: #fragments per group too big: %lu",
-			sbi->s_frags_per_group);
-		goto failed_mount;
-	}
 	if (sbi->s_inodes_per_group < sbi->s_inodes_per_block ||
 	    sbi->s_inodes_per_group > sb->s_blocksize * 8) {
 		ext2_msg(sb, KERN_ERR,
author maintainer
#82 lore raw

PATCH 80/92 mtd: rawnand: omap_elm: Fix incorrect type in assignment

In reply to Greg Kroah-Hartman

From: Roger Quadros <[email protected]>

[ Upstream commit d8403b9eeee66d5dd81ecb9445800b108c267ce3 ]

Once the ECC word endianness is converted to BE32, we force cast it to u32 so we can use elm_write_reg() which in turn uses writel().

Fixes below sparse warnings:

   drivers/mtd/nand/raw/omap_elm.c:180:37: sparse:     expected unsigned int [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:180:37: sparse:     got restricted __be32 [usertype]
   drivers/mtd/nand/raw/omap_elm.c:185:37: sparse:     expected unsigned int [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:185:37: sparse:     got restricted __be32 [usertype]
   drivers/mtd/nand/raw/omap_elm.c:190:37: sparse:     expected unsigned int [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:190:37: sparse:     got restricted __be32 [usertype]
Quoted 1 line
Quoted 1 line

drivers/mtd/nand/raw/omap_elm.c:200:40: sparse: sparse: restricted __be32 degrades to integer

drivers/mtd/nand/raw/omap_elm.c:206:39: sparse: sparse: restricted __be32 degrades to integer

   drivers/mtd/nand/raw/omap_elm.c:210:37: sparse:     expected unsigned int [assigned] [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:210:37: sparse:     got restricted __be32 [usertype]
   drivers/mtd/nand/raw/omap_elm.c:213:37: sparse:     expected unsigned int [assigned] [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:213:37: sparse:     got restricted __be32 [usertype]
   drivers/mtd/nand/raw/omap_elm.c:216:37: sparse:     expected unsigned int [assigned] [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:216:37: sparse:     got restricted __be32 [usertype]
   drivers/mtd/nand/raw/omap_elm.c:219:37: sparse:     expected unsigned int [assigned] [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:219:37: sparse:     got restricted __be32 [usertype]
   drivers/mtd/nand/raw/omap_elm.c:222:37: sparse:     expected unsigned int [assigned] [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:222:37: sparse:     got restricted __be32 [usertype]
   drivers/mtd/nand/raw/omap_elm.c:225:37: sparse:     expected unsigned int [assigned] [usertype] val
   drivers/mtd/nand/raw/omap_elm.c:225:37: sparse:     got restricted __be32 [usertype]

drivers/mtd/nand/raw/omap_elm.c:228:39: sparse: sparse: restricted __be32 degrades to integer

---
 drivers/mtd/nand/raw/omap_elm.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)
drivers/mtd/nand/raw/omap_elm.c +12 -12
index 2b21ce04b3ec6..1a48347be3fe4 100644
@@ -177,17 +177,17 @@ static void elm_load_syndrome(struct elm_info *info,
 			switch (info->bch_type) {
 			case BCH8_ECC:
 				/* syndrome fragment 0 = ecc[9-12B] */
-				val = cpu_to_be32(*(u32 *) &ecc[9]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[9]);
 				elm_write_reg(info, offset, val);
 
 				/* syndrome fragment 1 = ecc[5-8B] */
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[5]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[5]);
 				elm_write_reg(info, offset, val);
 
 				/* syndrome fragment 2 = ecc[1-4B] */
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[1]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[1]);
 				elm_write_reg(info, offset, val);
 
 				/* syndrome fragment 3 = ecc[0B] */
@@ -197,35 +197,35 @@ static void elm_load_syndrome(struct elm_info *info,
 				break;
 			case BCH4_ECC:
 				/* syndrome fragment 0 = ecc[20-52b] bits */
-				val = (cpu_to_be32(*(u32 *) &ecc[3]) >> 4) |
+				val = ((__force u32)cpu_to_be32(*(u32 *)&ecc[3]) >> 4) |
 					((ecc[2] & 0xf) << 28);
 				elm_write_reg(info, offset, val);
 
 				/* syndrome fragment 1 = ecc[0-20b] bits */
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[0]) >> 12;
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[0]) >> 12;
 				elm_write_reg(info, offset, val);
 				break;
 			case BCH16_ECC:
-				val = cpu_to_be32(*(u32 *) &ecc[22]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[22]);
 				elm_write_reg(info, offset, val);
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[18]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[18]);
 				elm_write_reg(info, offset, val);
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[14]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[14]);
 				elm_write_reg(info, offset, val);
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[10]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[10]);
 				elm_write_reg(info, offset, val);
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[6]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[6]);
 				elm_write_reg(info, offset, val);
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[2]);
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[2]);
 				elm_write_reg(info, offset, val);
 				offset += 4;
-				val = cpu_to_be32(*(u32 *) &ecc[0]) >> 16;
+				val = (__force u32)cpu_to_be32(*(u32 *)&ecc[0]) >> 16;
 				elm_write_reg(info, offset, val);
 				break;
 			default:
Signature · 1 line
2.40.1
author maintainer
#83 lore raw

PATCH 81/92 mtd: rawnand: rockchip: fix oobfree offset and description

In reply to Greg Kroah-Hartman

From: Johan Jonker <[email protected]>

[ Upstream commit d0ca3b92b7a6f42841ea9da8492aaf649db79780 ]

Rockchip boot blocks are written per 4 x 512 byte sectors per page. Each page with boot blocks must have a page address (PA) pointer in OOB to the next page.

The currently advertised free OOB area starts at offset 6, like if 4 PA bytes were located right after the BBM. This is wrong as the PA bytes are located right before the ECC bytes.

Fix the layout by allowing access to all bytes between the BBM and the PA bytes instead of reserving 4 bytes right after the BBM.

This change breaks existing jffs2 users.

Fixes 058e0e847d54 ("mtd: rawnand: rockchip: NFC driver for RK3308, RK2928 and others")
---
 drivers/mtd/nand/raw/rockchip-nand-controller.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)
drivers/mtd/nand/raw/rockchip-nand-controller.c +4 -7
index b5405bc7ca3a3..45cdc6ca210ca 100644
@@ -562,9 +562,10 @@ static int rk_nfc_write_page_raw(struct nand_chip *chip, const u8 *buf,
 		 *    BBM  OOB1 OOB2 OOB3 |......|  PA0  PA1  PA2  PA3
 		 *
 		 * The rk_nfc_ooblayout_free() function already has reserved
-		 * these 4 bytes with:
+		 * these 4 bytes together with 2 bytes for BBM
+		 * by reducing it's length:
 		 *
-		 * oob_region->offset = NFC_SYS_DATA_SIZE + 2;
+		 * oob_region->length = rknand->metadata_size - NFC_SYS_DATA_SIZE - 2;
 		 */
 		if (!i)
 			memcpy(rk_nfc_oob_ptr(chip, i),
@@ -935,12 +936,8 @@ static int rk_nfc_ooblayout_free(struct mtd_info *mtd, int section,
 	if (section)
 		return -ERANGE;
 
-	/*
-	 * The beginning of the OOB area stores the reserved data for the NFC,
-	 * the size of the reserved data is NFC_SYS_DATA_SIZE bytes.
-	 */
 	oob_region->length = rknand->metadata_size - NFC_SYS_DATA_SIZE - 2;
-	oob_region->offset = NFC_SYS_DATA_SIZE + 2;
+	oob_region->offset = 2;
 
 	return 0;
 }
Signature · 1 line
2.40.1
author maintainer
#84 lore raw

PATCH 82/92 mtd: rawnand: rockchip: Align hwecc vs. raw page helper layouts

In reply to Greg Kroah-Hartman

From: Johan Jonker <[email protected]>

[ Upstream commit ea690ad78dd611e3906df5b948a516000b05c1cb ]

Currently, read/write_page_hwecc() and read/write_page_raw() are not
aligned: there is a mismatch in the OOB bytes which are not read/written at the same offset in both cases (raw vs. hwecc).

This is a real problem when relying on the presence of the Page Addresses (PA) when using the NAND chip as a boot device, as the BootROM expects additional data in the OOB area at specific locations.

Rockchip boot blocks are written per 4 x 512 byte sectors per page. Each page with boot blocks must have a page address (PA) pointer in OOB to the next page. Pages are written in a pattern depending on the NAND chip ID.

Generate boot block page address and pattern for hwecc in user space and copy PA data to/from the already reserved last 4 bytes before ECC in the chip->oob_poi data layout.

Align the different helpers. This change breaks existing jffs2 users.

Fixes 058e0e847d54 ("mtd: rawnand: rockchip: NFC driver for RK3308, RK2928 and others")
---
 .../mtd/nand/raw/rockchip-nand-controller.c   | 34 ++++++++++++-------
 1 file changed, 21 insertions(+), 13 deletions(-)
drivers/mtd/nand/raw/rockchip-nand-controller.c +21 -13
index 45cdc6ca210ca..99242bd684375 100644
@@ -598,7 +598,7 @@ static int rk_nfc_write_page_hwecc(struct nand_chip *chip, const u8 *buf,
 	int pages_per_blk = mtd->erasesize / mtd->writesize;
 	int ret = 0, i, boot_rom_mode = 0;
 	dma_addr_t dma_data, dma_oob;
-	u32 reg;
+	u32 tmp;
 	u8 *oob;
 
 	nand_prog_page_begin_op(chip, page, 0, NULL, 0);
@@ -625,6 +625,13 @@ static int rk_nfc_write_page_hwecc(struct nand_chip *chip, const u8 *buf,
 	 *
 	 *   0xFF 0xFF 0xFF 0xFF | BBM OOB1 OOB2 OOB3 | ...
 	 *
+	 * The code here just swaps the first 4 bytes with the last
+	 * 4 bytes without losing any data.
+	 *
+	 * The chip->oob_poi data layout:
+	 *
+	 *    BBM  OOB1 OOB2 OOB3 |......|  PA0  PA1  PA2  PA3
+	 *
 	 * Configure the ECC algorithm supported by the boot ROM.
 	 */
 	if ((page < (pages_per_blk * rknand->boot_blks)) &&
@@ -635,21 +642,17 @@ static int rk_nfc_write_page_hwecc(struct nand_chip *chip, const u8 *buf,
 	}
 
 	for (i = 0; i < ecc->steps; i++) {
-		if (!i) {
-			reg = 0xFFFFFFFF;
-		} else {
+		if (!i)
+			oob = chip->oob_poi + (ecc->steps - 1) * NFC_SYS_DATA_SIZE;
+		else
 			oob = chip->oob_poi + (i - 1) * NFC_SYS_DATA_SIZE;
-			reg = oob[0] | oob[1] << 8 | oob[2] << 16 |
-			      oob[3] << 24;
-		}
 
-		if (!i && boot_rom_mode)
-			reg = (page & (pages_per_blk - 1)) * 4;
+		tmp = oob[0] | oob[1] << 8 | oob[2] << 16 | oob[3] << 24;
 
 		if (nfc->cfg->type == NFC_V9)
-			nfc->oob_buf[i] = reg;
+			nfc->oob_buf[i] = tmp;
 		else
-			nfc->oob_buf[i * (oob_step / 4)] = reg;
+			nfc->oob_buf[i * (oob_step / 4)] = tmp;
 	}
 
 	dma_data = dma_map_single(nfc->dev, (void *)nfc->page_buf,
@@ -812,12 +815,17 @@ static int rk_nfc_read_page_hwecc(struct nand_chip *chip, u8 *buf, int oob_on,
 		goto timeout_err;
 	}
 
-	for (i = 1; i < ecc->steps; i++) {
-		oob = chip->oob_poi + (i - 1) * NFC_SYS_DATA_SIZE;
+	for (i = 0; i < ecc->steps; i++) {
+		if (!i)
+			oob = chip->oob_poi + (ecc->steps - 1) * NFC_SYS_DATA_SIZE;
+		else
+			oob = chip->oob_poi + (i - 1) * NFC_SYS_DATA_SIZE;
+
 		if (nfc->cfg->type == NFC_V9)
 			tmp = nfc->oob_buf[i];
 		else
 			tmp = nfc->oob_buf[i * (oob_step / 4)];
+
 		*oob++ = (u8)tmp;
 		*oob++ = (u8)(tmp >> 8);
 		*oob++ = (u8)(tmp >> 16);
Signature · 1 line
2.40.1
author maintainer
#85 lore raw

PATCH 83/92 mtd: rawnand: fsl_upm: Fix an off-by one test in fun_exec_op()

In reply to Greg Kroah-Hartman

From: Christophe JAILLET <[email protected]>

[ Upstream commit c6abce60338aa2080973cd95be0aedad528bb41f ]

'op-cs' is copied in 'fun->mchip_number' which is used to access the 'mchip_offsets' and the 'rnb_gpio' arrays.
These arrays have NAND_MAX_CHIPS elements, so the index must be below this limit.

Fix the sanity check in order to avoid the NAND_MAX_CHIPS value. This would lead to out-of-bound accesses.

---
 drivers/mtd/nand/raw/fsl_upm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mtd/nand/raw/fsl_upm.c +1 -1
index b3cc427100a22..636e65328bb32 100644
@@ -135,7 +135,7 @@ static int fun_exec_op(struct nand_chip *chip, const struct nand_operation *op,
 	unsigned int i;
 	int ret;
 
-	if (op->cs > NAND_MAX_CHIPS)
+	if (op->cs >= NAND_MAX_CHIPS)
 		return -EINVAL;
 
 	if (check_only)
Signature · 1 line
2.40.1
author maintainer
#86 lore raw

PATCH 84/92 powerpc/mm/altmap: Fix altmap boundary check

In reply to Greg Kroah-Hartman

From: Aneesh Kumar K.V <[email protected]>

[ Upstream commit 6722b25712054c0f903b839b8f5088438dd04df3 ]

altmap->free includes the entire free space from which altmap blocks can be allocated. So when checking whether the kernel is doing altmap block free, compute the boundary correctly, otherwise memory hotunplug can fail.

Fixes 9ef34630a461 ("powerpc/mm: Fallback to RAM if the altmap is unusable")
---
 arch/powerpc/mm/init_64.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
arch/powerpc/mm/init_64.c +1 -2
index b76cd49d521b9..db040f34c0046 100644
@@ -313,8 +313,7 @@ void __ref vmemmap_free(unsigned long start, unsigned long end,
 	start = ALIGN_DOWN(start, page_size);
 	if (altmap) {
 		alt_start = altmap->base_pfn;
-		alt_end = altmap->base_pfn + altmap->reserve +
-			  altmap->free + altmap->alloc + altmap->align;
+		alt_end = altmap->base_pfn + altmap->reserve + altmap->free;
 	}
 
 	pr_debug("vmemmap_free %lx...%lx\n", start, end);
Signature · 1 line
2.40.1
author maintainer
#87 lore raw

PATCH 85/92 drm/fsl-dcu: Use drm_plane_helper_destroy()

In reply to Greg Kroah-Hartman

From: Thomas Zimmermann <[email protected]>

[ Upstream commit a4d847df8b440cc667c18f99abe84ccaa83afac4 ]

Replace the driver's own function with drm_plane_helper_destroy(). No functional changes.

Stable-dep-of: ee31742bf176 ("drm/imx/ipuv3: Fix front porch adjustment upon hactive aligning")

---
 drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)
drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c +1 -7
index 8fe953d6e0a94..8a9e6db71e46b 100644
@@ -170,16 +170,10 @@ static const struct drm_plane_helper_funcs fsl_dcu_drm_plane_helper_funcs = {
 	.atomic_update = fsl_dcu_drm_plane_atomic_update,
 };
 
-static void fsl_dcu_drm_plane_destroy(struct drm_plane *plane)
-{
-	drm_plane_cleanup(plane);
-	kfree(plane);
-}
-
 static const struct drm_plane_funcs fsl_dcu_drm_plane_funcs = {
 	.atomic_duplicate_state = drm_atomic_helper_plane_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_plane_destroy_state,
-	.destroy = fsl_dcu_drm_plane_destroy,
+	.destroy = drm_plane_helper_destroy,
 	.disable_plane = drm_atomic_helper_disable_plane,
 	.reset = drm_atomic_helper_plane_reset,
 	.update_plane = drm_atomic_helper_update_plane,
Signature · 1 line
2.40.1
author maintainer
#88 lore raw

PATCH 86/92 drm/imx/ipuv3: Fix front porch adjustment upon hactive aligning

In reply to Greg Kroah-Hartman

From: Alexander Stein <[email protected]>

[ Upstream commit ee31742bf17636da1304af77b2cb1c29b5dda642 ]

When hactive is not aligned to 8 pixels, it is aligned accordingly and hfront porch needs to be reduced the same amount. Unfortunately the front porch is set to the difference rather than reducing it. There are some Samsung TVs which can't cope with a front porch of instead of 70.

---
 drivers/gpu/drm/imx/ipuv3-crtc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/imx/ipuv3-crtc.c +1 -1
index f7863d6dea804..ba5b16618c237 100644
@@ -311,7 +311,7 @@ static void ipu_crtc_mode_set_nofb(struct drm_crtc *crtc)
 		dev_warn(ipu_crtc->dev, "8-pixel align hactive %d -> %d\n",
 			 sig_cfg.mode.hactive, new_hactive);
 
-		sig_cfg.mode.hfront_porch = new_hactive - sig_cfg.mode.hactive;
+		sig_cfg.mode.hfront_porch -= new_hactive - sig_cfg.mode.hactive;
 		sig_cfg.mode.hactive = new_hactive;
 	}
 
Signature · 1 line
2.40.1
author maintainer
#89 lore raw

PATCH 87/92 selftests/rseq: check if libc rseq support is registered

In reply to Greg Kroah-Hartman

From: Michael Jeanson <[email protected]>

[ Upstream commit d1a997ba4c1bf65497d956aea90de42a6398f73a ]

When checking for libc rseq support in the library constructor, don't only depend on the symbols presence, check that the registration was completed.

This targets a scenario where the libc has rseq support but it is not wired for the current architecture in 'bits/rseq.h', we want to fallback to our internal registration mechanism.

Stable-dep-of: 3bcbc20942db ("selftests/rseq: Play nice with binaries statically linked against glibc 2.35+")

---
 tools/testing/selftests/rseq/rseq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
tools/testing/selftests/rseq/rseq.c +2 -1
index 986b9458efb26..4177f9507bbee 100644
@@ -111,7 +111,8 @@ void rseq_init(void)
 	libc_rseq_offset_p = dlsym(RTLD_NEXT, "__rseq_offset");
 	libc_rseq_size_p = dlsym(RTLD_NEXT, "__rseq_size");
 	libc_rseq_flags_p = dlsym(RTLD_NEXT, "__rseq_flags");
-	if (libc_rseq_size_p && libc_rseq_offset_p && libc_rseq_flags_p) {
+	if (libc_rseq_size_p && libc_rseq_offset_p && libc_rseq_flags_p &&
+			*libc_rseq_size_p != 0) {
 		/* rseq registration owned by glibc */
 		rseq_offset = *libc_rseq_offset_p;
 		rseq_size = *libc_rseq_size_p;
Signature · 1 line
2.40.1
author maintainer
#90 lore raw

PATCH 88/92 selftests/rseq: Play nice with binaries statically linked against glibc 2.35+

In reply to Greg Kroah-Hartman

From: Sean Christopherson <[email protected]>

[ Upstream commit 3bcbc20942db5d738221cca31a928efc09827069 ]

To allow running rseq and KVM's rseq selftests as statically linked binaries, initialize the various "trampoline" pointers to point directly at the expect glibc symbols, and skip the dlysm() lookups if the rseq size is non-zero, i.e. the binary is statically linked *and* the libc registered its own rseq.

Define weak versions of the symbols so as not to break linking against libc versions that don't support rseq in any capacity.

The KVM selftests in particular are often statically linked so that they can be run on targets with very limited runtime environments, i.e. test machines.

Fixes 233e667e1ae3 ("selftests/rseq: Uplift rseq selftests for compatibility with glibc-2.35")
Cc Aaron Lewis <[email protected]>

Message-Id: <[email protected]>

---
 tools/testing/selftests/rseq/rseq.c | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)
tools/testing/selftests/rseq/rseq.c +22 -6
index 4177f9507bbee..b736a5169aad0 100644
@@ -32,9 +32,17 @@
 #include "../kselftest.h"
 #include "rseq.h"
 
-static const ptrdiff_t *libc_rseq_offset_p;
-static const unsigned int *libc_rseq_size_p;
-static const unsigned int *libc_rseq_flags_p;
+/*
+ * Define weak versions to play nice with binaries that are statically linked
+ * against a libc that doesn't support registering its own rseq.
+ */
+__weak ptrdiff_t __rseq_offset;
+__weak unsigned int __rseq_size;
+__weak unsigned int __rseq_flags;
+
+static const ptrdiff_t *libc_rseq_offset_p = &__rseq_offset;
+static const unsigned int *libc_rseq_size_p = &__rseq_size;
+static const unsigned int *libc_rseq_flags_p = &__rseq_flags;
 
 /* Offset from the thread pointer to the rseq area.  */
 ptrdiff_t rseq_offset;
@@ -108,9 +116,17 @@ int rseq_unregister_current_thread(void)
 static __attribute__((constructor))
 void rseq_init(void)
 {
-	libc_rseq_offset_p = dlsym(RTLD_NEXT, "__rseq_offset");
-	libc_rseq_size_p = dlsym(RTLD_NEXT, "__rseq_size");
-	libc_rseq_flags_p = dlsym(RTLD_NEXT, "__rseq_flags");
+	/*
+	 * If the libc's registered rseq size isn't already valid, it may be
+	 * because the binary is dynamically linked and not necessarily due to
+	 * libc not having registered a restartable sequence.  Try to find the
+	 * symbols if that's the case.
+	 */
+	if (!*libc_rseq_size_p) {
+		libc_rseq_offset_p = dlsym(RTLD_NEXT, "__rseq_offset");
+		libc_rseq_size_p = dlsym(RTLD_NEXT, "__rseq_size");
+		libc_rseq_flags_p = dlsym(RTLD_NEXT, "__rseq_flags");
+	}
 	if (libc_rseq_size_p && libc_rseq_offset_p && libc_rseq_flags_p &&
 			*libc_rseq_size_p != 0) {
 		/* rseq registration owned by glibc */
Signature · 1 line
2.40.1
author maintainer
#91 lore raw

PATCH 89/92 soundwire: bus: pm_runtime_request_resume on peripheral attachment

In reply to Greg Kroah-Hartman

From: Pierre-Louis Bossart <[email protected]>

[ Upstream commit e557bca49b812908f380c56b5b4b2f273848b676 ]

In typical use cases, the peripheral becomes pm_runtime active as a result of the ALSA/ASoC framework starting up a DAI. The parent/child hierarchy guarantees that the manager device will be fully resumed beforehand.

There is however a corner case where the manager device may become pm_runtime active, but without ALSA/ASoC requesting any functionality from the peripherals. In this case, the hardware peripheral device will report as ATTACHED and its initialization routine will be executed. If this initialization routine initiates any sort of deferred processing, there is a possibility that the manager could suspend without the peripheral suspend sequence being invoked: from the pm_runtime framework perspective, the peripheral is *already* suspended.

To avoid such disconnects between hardware state and pm_runtime state, this patch adds an asynchronous pm_request_resume() upon successful attach/initialization which will result in the proper resume/suspend sequence to be followed on the peripheral side.

Stable-dep-of: c40d6b3249b1 ("soundwire: fix enumeration completion")

---
 drivers/soundwire/bus.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
drivers/soundwire/bus.c +12 -0
index b7cdfa65157c6..cc4cca0325b98 100644
@@ -1841,6 +1841,18 @@ int sdw_handle_slave_status(struct sdw_bus *bus,
 				__func__, slave->dev_num);
 
 			complete(&slave->initialization_complete);
+
+			/*
+			 * If the manager became pm_runtime active, the peripherals will be
+			 * restarted and attach, but their pm_runtime status may remain
+			 * suspended. If the 'update_slave_status' callback initiates
+			 * any sort of deferred processing, this processing would not be
+			 * cancelled on pm_runtime suspend.
+			 * To avoid such zombie states, we queue a request to resume.
+			 * This would be a no-op in case the peripheral was being resumed
+			 * by e.g. the ALSA/ASoC framework.
+			 */
+			pm_request_resume(&slave->dev);
 		}
 	}
 
Signature · 1 line
2.40.1
author maintainer
#92 lore raw

PATCH 90/92 soundwire: fix enumeration completion

In reply to Greg Kroah-Hartman

From: Johan Hovold <[email protected]>

[ Upstream commit c40d6b3249b11d60e09d81530588f56233d9aa44 ]

The soundwire subsystem uses two completion structures that allow drivers to wait for soundwire device to become enumerated on the bus and initialised by their drivers, respectively.

The code implementing the signalling is currently broken as it does not signal all current and future waiters and also uses the wrong reinitialisation function, which can potentially lead to memory corruption if there are still waiters on the queue.

Not signalling future waiters specifically breaks sound card probe deferrals as codec drivers can not tell that the soundwire device is already attached when being reprobed. Some codec runtime PM implementations suffer from similar problems as waiting for enumeration during resume can also timeout despite the device already having been enumerated.

Fixes fb9469e54fa7 ("soundwire: bus: fix race condition with enumeration_complete signaling")
Fixes a90def068127 ("soundwire: bus: fix race condition with initialization_complete signaling")
Cc Pierre-Louis Bossart <[email protected]>
Cc Rander Wang <[email protected]>
---
 drivers/soundwire/bus.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/soundwire/bus.c +4 -4
index cc4cca0325b98..230a3250f3154 100644
@@ -828,8 +828,8 @@ static void sdw_modify_slave_status(struct sdw_slave *slave,
 			"%s: initializing enumeration and init completion for Slave %d\n",
 			__func__, slave->dev_num);
 
-		init_completion(&slave->enumeration_complete);
-		init_completion(&slave->initialization_complete);
+		reinit_completion(&slave->enumeration_complete);
+		reinit_completion(&slave->initialization_complete);
 
 	} else if ((status == SDW_SLAVE_ATTACHED) &&
 		   (slave->status == SDW_SLAVE_UNATTACHED)) {
@@ -837,7 +837,7 @@ static void sdw_modify_slave_status(struct sdw_slave *slave,
 			"%s: signaling enumeration completion for Slave %d\n",
 			__func__, slave->dev_num);
 
-		complete(&slave->enumeration_complete);
+		complete_all(&slave->enumeration_complete);
 	}
 	slave->status = status;
 	mutex_unlock(&bus->bus_lock);
@@ -1840,7 +1840,7 @@ int sdw_handle_slave_status(struct sdw_bus *bus,
 				"%s: signaling initialization completion for Slave %d\n",
 				__func__, slave->dev_num);
 
-			complete(&slave->initialization_complete);
+			complete_all(&slave->initialization_complete);
 
 			/*
 			 * If the manager became pm_runtime active, the peripherals will be
Signature · 1 line
2.40.1
author maintainer
#93 lore raw

PATCH 91/92 PM / wakeirq: support enabling wake-up irq after runtime_suspend called

In reply to Greg Kroah-Hartman

From: Chunfeng Yun <[email protected]>

[ Upstream commit 259714100d98b50bf04d36a21bf50ca8b829fc11 ]

When the dedicated wake IRQ is level trigger, and it uses the device's low-power status as the wakeup source, that means if the device is not in low-power state, the wake IRQ will be triggered if enabled; For this case, need enable the wake IRQ after running the device's ->runtime_suspend() which make it enter low-power state.

e.g.
Assume the wake IRQ is a low level trigger type, and the wakeup signal comes from the low-power status of the device.
The wakeup signal is low level at running time (0), and becomes high level when the device enters low-power state (runtime_suspend
(1) is called), a wakeup event at (2) make the device exit low-power state, then the wakeup signal also becomes low level.

------------------
| ^ ^|
---------------- | | --------------
|<---(0)--->|<--(1)--| (3) (2) (4)

if enable the wake IRQ before running runtime_suspend during (0), a wake IRQ will arise, it causes resume immediately;
it works if enable wake IRQ ( e.g. at (3) or (4)) after running ->runtime_suspend().

This patch introduces a new status WAKE_IRQ_DEDICATED_REVERSE to optionally support enabling wake IRQ after running ->runtime_suspend().

Stable-dep-of: 8527beb12087 ("PM: sleep: wakeirq: fix wake irq arming")

---
 drivers/base/power/power.h   |   7 ++-
 drivers/base/power/runtime.c |   6 ++-
 drivers/base/power/wakeirq.c | 101 +++++++++++++++++++++++++++--------
 include/linux/pm_wakeirq.h   |   9 +++-
 4 files changed, 96 insertions(+), 27 deletions(-)
drivers/base/power/power.h +5 -2
index 54292cdd7808b..0eb7f02b3ad59 100644
@@ -25,8 +25,10 @@ extern u64 pm_runtime_active_time(struct device *dev);
 
 #define WAKE_IRQ_DEDICATED_ALLOCATED	BIT(0)
 #define WAKE_IRQ_DEDICATED_MANAGED	BIT(1)
+#define WAKE_IRQ_DEDICATED_REVERSE	BIT(2)
 #define WAKE_IRQ_DEDICATED_MASK		(WAKE_IRQ_DEDICATED_ALLOCATED | \
-					 WAKE_IRQ_DEDICATED_MANAGED)
+					 WAKE_IRQ_DEDICATED_MANAGED | \
+					 WAKE_IRQ_DEDICATED_REVERSE)
 
 struct wake_irq {
 	struct device *dev;
@@ -39,7 +41,8 @@ extern void dev_pm_arm_wake_irq(struct wake_irq *wirq);
 extern void dev_pm_disarm_wake_irq(struct wake_irq *wirq);
 extern void dev_pm_enable_wake_irq_check(struct device *dev,
 					 bool can_change_status);
-extern void dev_pm_disable_wake_irq_check(struct device *dev);
+extern void dev_pm_disable_wake_irq_check(struct device *dev, bool cond_disable);
+extern void dev_pm_enable_wake_irq_complete(struct device *dev);
 
 #ifdef CONFIG_PM_SLEEP
 
drivers/base/power/runtime.c +4 -2
index c1142a7a4fe65..5824d41a0b745 100644
@@ -673,6 +673,8 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 	if (retval)
 		goto fail;
 
+	dev_pm_enable_wake_irq_complete(dev);
+
  no_callback:
 	__update_runtime_status(dev, RPM_SUSPENDED);
 	pm_runtime_deactivate_timer(dev);
@@ -718,7 +720,7 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 	return retval;
 
  fail:
-	dev_pm_disable_wake_irq_check(dev);
+	dev_pm_disable_wake_irq_check(dev, true);
 	__update_runtime_status(dev, RPM_ACTIVE);
 	dev->power.deferred_resume = false;
 	wake_up_all(&dev->power.wait_queue);
@@ -901,7 +903,7 @@ static int rpm_resume(struct device *dev, int rpmflags)
 
 	callback = RPM_GET_CALLBACK(dev, runtime_resume);
 
-	dev_pm_disable_wake_irq_check(dev);
+	dev_pm_disable_wake_irq_check(dev, false);
 	retval = rpm_callback(callback, dev);
 	if (retval) {
 		__update_runtime_status(dev, RPM_SUSPENDED);
drivers/base/power/wakeirq.c +80 -21

146 lines not on this page · raw message

include/linux/pm_wakeirq.h +7 -2
index cd5b62db90845..e63a63aa47a37 100644
@@ -17,8 +17,8 @@
 #ifdef CONFIG_PM
 
 extern int dev_pm_set_wake_irq(struct device *dev, int irq);
-extern int dev_pm_set_dedicated_wake_irq(struct device *dev,
-					 int irq);
+extern int dev_pm_set_dedicated_wake_irq(struct device *dev, int irq);
+extern int dev_pm_set_dedicated_wake_irq_reverse(struct device *dev, int irq);
 extern void dev_pm_clear_wake_irq(struct device *dev);
 extern void dev_pm_enable_wake_irq(struct device *dev);
 extern void dev_pm_disable_wake_irq(struct device *dev);
@@ -35,6 +35,11 @@ static inline int dev_pm_set_dedicated_wake_irq(struct device *dev, int irq)
 	return 0;
 }
 
+static inline int dev_pm_set_dedicated_wake_irq_reverse(struct device *dev, int irq)
+{
+	return 0;
+}
+
 static inline void dev_pm_clear_wake_irq(struct device *dev)
 {
 }
Signature · 1 line
2.40.1
author maintainer
#94 lore raw

PATCH 92/92 PM: sleep: wakeirq: fix wake irq arming

In reply to Greg Kroah-Hartman

From: Johan Hovold <[email protected]>

[ Upstream commit 8527beb12087238d4387607597b4020bc393c4b4 ]

The decision whether to enable a wake irq during suspend can not be done based on the runtime PM state directly as a driver may use wake irqs without implementing runtime PM. Such drivers specifically leave the state set to the default 'suspended' and the wake irq is thus never enabled at suspend.

Add a new wake irq flag to track whether a dedicated wake irq has been enabled at runtime suspend and therefore must not be enabled at system suspend.

Note that pm_runtime_enabled() can not be used as runtime PM is always disabled during late suspend.

Fixes 69728051f5bf ("PM / wakeirq: Fix unbalanced IRQ enable for wakeirq")
Cc 4.16+ <[email protected]> # 4.16+
---
 drivers/base/power/power.h   |  1 +
 drivers/base/power/wakeirq.c | 12 ++++++++----
 2 files changed, 9 insertions(+), 4 deletions(-)
drivers/base/power/power.h +1 -0
index 0eb7f02b3ad59..922ed457db191 100644
@@ -29,6 +29,7 @@ extern u64 pm_runtime_active_time(struct device *dev);
 #define WAKE_IRQ_DEDICATED_MASK		(WAKE_IRQ_DEDICATED_ALLOCATED | \
 					 WAKE_IRQ_DEDICATED_MANAGED | \
 					 WAKE_IRQ_DEDICATED_REVERSE)
+#define WAKE_IRQ_DEDICATED_ENABLED	BIT(3)
 
 struct wake_irq {
 	struct device *dev;
drivers/base/power/wakeirq.c +8 -4
index 0004db4a9d3b9..6f2cdd8643afa 100644
@@ -314,8 +314,10 @@ void dev_pm_enable_wake_irq_check(struct device *dev,
 	return;
 
 enable:
-	if (!can_change_status || !(wirq->status & WAKE_IRQ_DEDICATED_REVERSE))
+	if (!can_change_status || !(wirq->status & WAKE_IRQ_DEDICATED_REVERSE)) {
 		enable_irq(wirq->irq);
+		wirq->status |= WAKE_IRQ_DEDICATED_ENABLED;
+	}
 }
 
 /**
@@ -336,8 +338,10 @@ void dev_pm_disable_wake_irq_check(struct device *dev, bool cond_disable)
 	if (cond_disable && (wirq->status & WAKE_IRQ_DEDICATED_REVERSE))
 		return;
 
-	if (wirq->status & WAKE_IRQ_DEDICATED_MANAGED)
+	if (wirq->status & WAKE_IRQ_DEDICATED_MANAGED) {
+		wirq->status &= ~WAKE_IRQ_DEDICATED_ENABLED;
 		disable_irq_nosync(wirq->irq);
+	}
 }
 
 /**
@@ -376,7 +380,7 @@ void dev_pm_arm_wake_irq(struct wake_irq *wirq)
 
 	if (device_may_wakeup(wirq->dev)) {
 		if (wirq->status & WAKE_IRQ_DEDICATED_ALLOCATED &&
-		    !pm_runtime_status_suspended(wirq->dev))
+		    !(wirq->status & WAKE_IRQ_DEDICATED_ENABLED))
 			enable_irq(wirq->irq);
 
 		enable_irq_wake(wirq->irq);
@@ -399,7 +403,7 @@ void dev_pm_disarm_wake_irq(struct wake_irq *wirq)
 		disable_irq_wake(wirq->irq);
 
 		if (wirq->status & WAKE_IRQ_DEDICATED_ALLOCATED &&
-		    !pm_runtime_status_suspended(wirq->dev))
+		    !(wirq->status & WAKE_IRQ_DEDICATED_ENABLED))
 			disable_irq_nosync(wirq->irq);
 	}
 }
Signature · 1 line
2.40.1
#95 lore raw
In reply to Greg Kroah-Hartman
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

I will do some more runs to see if TASKS03 hang is a new thing but it could be related to the existing issues.

thanks,

- Joel

Quoted 412 lines

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

    Linux 5.15.126-rc1

Johan Hovold <[email protected]>
PM: sleep: wakeirq: fix wake irq arming

Chunfeng Yun <[email protected]>
PM / wakeirq: support enabling wake-up irq after runtime_suspend called

Johan Hovold <[email protected]>
soundwire: fix enumeration completion

Pierre-Louis Bossart <[email protected]>
soundwire: bus: pm_runtime_request_resume on peripheral attachment

Sean Christopherson <[email protected]>
selftests/rseq: Play nice with binaries statically linked against glibc 2.35+

Michael Jeanson <[email protected]>
selftests/rseq: check if libc rseq support is registered

Alexander Stein <[email protected]>
drm/imx/ipuv3: Fix front porch adjustment upon hactive aligning

Thomas Zimmermann <[email protected]>
drm/fsl-dcu: Use drm_plane_helper_destroy()

Aneesh Kumar K.V <[email protected]>
powerpc/mm/altmap: Fix altmap boundary check

Christophe JAILLET <[email protected]>
mtd: rawnand: fsl_upm: Fix an off-by one test in fun_exec_op()

Johan Jonker <[email protected]>
mtd: rawnand: rockchip: Align hwecc vs. raw page helper layouts

Johan Jonker <[email protected]>
mtd: rawnand: rockchip: fix oobfree offset and description

Roger Quadros <[email protected]>
mtd: rawnand: omap_elm: Fix incorrect type in assignment

Jan Kara <[email protected]>
ext2: Drop fragment support

Jan Kara <[email protected]>
fs: Protect reconfiguration of sb read-write from racing writes

Alan Stern <[email protected]>
net: usbnet: Fix WARNING in usbnet_start_xmit/usb_submit_urb

Sungwoo Kim <[email protected]>
Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb

Prince Kumar Maurya <[email protected]>
fs/sysv: Null check to prevent null-ptr-deref bug

Tetsuo Handa <[email protected]>
fs/ntfs3: Use __GFP_NOWARN allocation at ntfs_load_attr_list()

Linus Torvalds <[email protected]>
file: reinstate f_pos locking optimization for regular files

Hou Tao <[email protected]>
bpf, cpumap: Make sure kthread is running before map update returns

Guchun Chen <[email protected]>
drm/ttm: check null pointer before accessing when swapping

Aleksa Sarai <[email protected]>
open: make RESOLVE_CACHED correctly test for O_TMPFILE

Jiri Olsa <[email protected]>
bpf: Disable preemption in bpf_event_output

Ilya Dryomov <[email protected]>
rbd: prevent busy loop when requesting exclusive lock

Paul Fertser <[email protected]>
wifi: mt76: mt7615: do not advertise 5 GHz on first phy of MT7615D (DBDC)

Laszlo Ersek <[email protected]>
net: tap_open(): set sk_uid from current_fsuid()

Laszlo Ersek <[email protected]>
net: tun_chr_open(): set sk_uid from current_fsuid()

Dinh Nguyen <[email protected]>
arm64: dts: stratix10: fix incorrect I2C property for SCL signal

Arseniy Krasnov <[email protected]>
mtd: rawnand: meson: fix OOB available bytes for ECC

Olivier Maignial <[email protected]>
mtd: spinand: toshiba: Fix ecc_get_status

Sungjong Seo <[email protected]>
exfat: release s_lock before calling dir_emit()

gaoming <[email protected]>
exfat: use kvmalloc_array/kvfree instead of kmalloc_array/kfree

Krzysztof Kozlowski <[email protected]>
firmware: arm_scmi: Drop OF node reference in the transport channel setup

Xiubo Li <[email protected]>
ceph: defer stopping mdsc delayed_work

Ross Maynard <[email protected]>

    USB: zaurus: Add ID for A-300/B-500/C-700

Ilya Dryomov <[email protected]>
libceph: fix potential hang in ceph_osdc_notify()

Michael Kelley <[email protected]>
scsi: storvsc: Limit max_sectors for virtual Fibre Channel devices

Steffen Maier <[email protected]>
scsi: zfcp: Defer fc_rport blocking until after ADISC response

Eric Dumazet <[email protected]>
tcp_metrics: fix data-race in tcpm_suck_dst() vs fastopen

Eric Dumazet <[email protected]>

    tcp_metrics: annotate data-races around tm->tcpm_net

Eric Dumazet <[email protected]>

    tcp_metrics: annotate data-races around tm->tcpm_vals[]

Eric Dumazet <[email protected]>

    tcp_metrics: annotate data-races around tm->tcpm_lock

Eric Dumazet <[email protected]>

    tcp_metrics: annotate data-races around tm->tcpm_stamp

Eric Dumazet <[email protected]>
tcp_metrics: fix addr_same() helper

Jonas Gorski <[email protected]>
prestera: fix fallback to previous version on same major version

Jianbo Liu <[email protected]>
net/mlx5: fs_core: Skip the FTs in the same FS_TYPE_PRIO_CHAINS fs_prio

Jianbo Liu <[email protected]>
net/mlx5: fs_core: Make find_closest_ft more generic

Benjamin Poirier <[email protected]>
vxlan: Fix nexthop hash size

Yue Haibing <[email protected]>
ip6mr: Fix skb_under_panic in ip6mr_cache_report()

Alexandra Winter <[email protected]>
s390/qeth: Don't call dev_close/dev_open (DOWN/UP)

Lin Ma <[email protected]>
net: dcb: choose correct policy to parse DCB_ATTR_BCN

Mark Brown <[email protected]>
net: netsec: Ignore 'phy-mode' on SynQuacer in DT mode

Yuanjun Gong <[email protected]>
net: korina: handle clk prepare error in korina_probe()

Dan Carpenter <[email protected]>
net: ll_temac: fix error checking of irq_of_parse_and_map()

Yang Yingliang <[email protected]>
net: ll_temac: Switch to use dev_err_probe() helper

Tomas Glozar <[email protected]>
bpf: sockmap: Remove preempt_disable in sock_map_sk_acquire

valis <[email protected]>
net/sched: cls_route: No longer copy tcf_result on update to avoid use-after-free

valis <[email protected]>
net/sched: cls_fw: No longer copy tcf_result on update to avoid use-after-free

valis <[email protected]>
net/sched: cls_u32: No longer copy tcf_result on update to avoid use-after-free

Hou Tao <[email protected]>
bpf, cpumap: Handle skb as well when clean up ptr_ring

Kuniyuki Iwashima <[email protected]>
net/sched: taprio: Limit TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME to INT_MAX.

Eric Dumazet <[email protected]>
net: add missing data-race annotation for sk_ll_usec

Eric Dumazet <[email protected]>

    net: add missing data-race annotations around sk->sk_peek_off

Eric Dumazet <[email protected]>

    net: add missing READ_ONCE(sk->sk_rcvbuf) annotation

Eric Dumazet <[email protected]>

    net: add missing READ_ONCE(sk->sk_sndbuf) annotation

Eric Dumazet <[email protected]>

    net: add missing READ_ONCE(sk->sk_rcvlowat) annotation

Eric Dumazet <[email protected]>

    net: annotate data-races around sk->sk_max_pacing_rate

Konstantin Khorenko <[email protected]>
qed: Fix scheduling in a tasklet while getting stats

Prabhakar Kushwaha <[email protected]>
qed: Fix kernel-doc warnings

Chengfeng Ye <[email protected]>

    mISDN: hfcpci: Fix potential deadlock on &hc->lock

Jamal Hadi Salim <[email protected]>
net: sched: cls_u32: Fix match key mis-addressing

Georg Müller <[email protected]>
perf test uprobe_from_different_cu: Skip if there is no gcc

Yuanjun Gong <[email protected]>
net: dsa: fix value check in bcm_sf2_sw_probe()

Lin Ma <[email protected]>
rtnetlink: let rtnl_bridge_setlink checks IFLA_BRIDGE_MODE length

Lin Ma <[email protected]>
bpf: Add length check for SK_DIAG_BPF_STORAGE_REQ_MAP_FD parsing

Yuanjun Gong <[email protected]>
net/mlx5e: fix return value check in mlx5e_ipsec_remove_trailer()

Zhengchao Shao <[email protected]>
net/mlx5: DR, fix memory leak in mlx5dr_cmd_create_reformat_ctx

Ilan Peer <[email protected]>
wifi: cfg80211: Fix return value in scan logic

Heiko Carstens <[email protected]>
KVM: s390: fix sthyi error handling

[email protected] <[email protected]>
word-at-a-time: use the same return type for has_zero regardless of endianness

Cristian Marussi <[email protected]>
firmware: arm_scmi: Fix chan_free cleanup on SMC

Hugo Villeneuve <[email protected]>
arm64: dts: imx8mn-var-som: add missing pull-up for onboard PHY reset pinmux

Robin Murphy <[email protected]>
iommu/arm-smmu-v3: Document nesting-related errata

Robin Murphy <[email protected]>
iommu/arm-smmu-v3: Add explicit feature for nesting

Robin Murphy <[email protected]>

    iommu/arm-smmu-v3: Document MMU-700 erratum 2812531

Robin Murphy <[email protected]>

    iommu/arm-smmu-v3: Work around MMU-600 erratum 1076982

Suzuki K Poulose <[email protected]>
arm64: errata: Add detection for TRBE write to out-of-range

Suzuki K Poulose <[email protected]>
arm64: errata: Add workaround for TSB flush failures

Shay Drory <[email protected]>
net/mlx5: Free irqs only on shutdown callback

Peter Zijlstra <[email protected]>
perf: Fix function pointer case

Jens Axboe <[email protected]>
io_uring: gate iowait schedule on having pending requests

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

Diffstat:

 Documentation/arm64/silicon-errata.rst             |  12 +
 Makefile                                           |   4 +-
 arch/arm64/Kconfig                                 |  74 ++
 .../boot/dts/altera/socfpga_stratix10_socdk.dts    |   2 +-
 .../dts/altera/socfpga_stratix10_socdk_nand.dts    |   2 +-
 arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi  |   2 +-
 arch/arm64/include/asm/barrier.h                   |  16 +-
 arch/arm64/kernel/cpu_errata.c                     |  39 +
 arch/arm64/tools/cpucaps                           |   2 +
 arch/powerpc/include/asm/word-at-a-time.h          |   2 +-
 arch/powerpc/mm/init_64.c                          |   3 +-
 arch/s390/kernel/sthyi.c                           |   6 +-
 arch/s390/kvm/intercept.c                          |   9 +-
 drivers/base/power/power.h                         |   8 +-
 drivers/base/power/runtime.c                       |   6 +-
 drivers/base/power/wakeirq.c                       | 111 ++-
 drivers/block/rbd.c                                |  28 +-
 drivers/firmware/arm_scmi/mailbox.c                |   4 +-
 drivers/firmware/arm_scmi/smc.c                    |  21 +-
 drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c        |   8 +-
 drivers/gpu/drm/imx/ipuv3-crtc.c                   |   2 +-
 drivers/gpu/drm/ttm/ttm_bo.c                       |   3 +-
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c        |  50 ++
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h        |   8 +
 drivers/isdn/hardware/mISDN/hfcpci.c               |  10 +-
 drivers/mtd/nand/raw/fsl_upm.c                     |   2 +-
 drivers/mtd/nand/raw/meson_nand.c                  |   3 +-
 drivers/mtd/nand/raw/omap_elm.c                    |  24 +-
 drivers/mtd/nand/raw/rockchip-nand-controller.c    |  45 +-
 drivers/mtd/nand/spi/toshiba.c                     |   4 +-
 drivers/net/dsa/bcm_sf2.c                          |   8 +-
 drivers/net/ethernet/korina.c                      |   3 +-
 .../net/ethernet/marvell/prestera/prestera_pci.c   |   3 +-
 .../mellanox/mlx5/core/en_accel/ipsec_rxtx.c       |   4 +-
 drivers/net/ethernet/mellanox/mlx5/core/eq.c       |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  | 105 ++-
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  |  29 +
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  |   5 +-
 drivers/net/ethernet/qlogic/qed/qed.h              |   9 +-
 drivers/net/ethernet/qlogic/qed/qed_cxt.h          | 138 +--
 drivers/net/ethernet/qlogic/qed/qed_dev_api.h      | 361 ++++----
 drivers/net/ethernet/qlogic/qed/qed_fcoe.c         |  19 +-
 drivers/net/ethernet/qlogic/qed/qed_fcoe.h         |  17 +-
 drivers/net/ethernet/qlogic/qed/qed_hsi.h          | 922 +++++++++++----------
 drivers/net/ethernet/qlogic/qed/qed_hw.c           |  26 +-
 drivers/net/ethernet/qlogic/qed/qed_hw.h           | 214 ++---
 drivers/net/ethernet/qlogic/qed/qed_init_ops.h     |  58 +-
 drivers/net/ethernet/qlogic/qed/qed_int.h          | 274 +++---
 drivers/net/ethernet/qlogic/qed/qed_iscsi.c        |  19 +-
 drivers/net/ethernet/qlogic/qed/qed_iscsi.h        |  17 +-
 drivers/net/ethernet/qlogic/qed/qed_l2.c           |  19 +-
 drivers/net/ethernet/qlogic/qed/qed_l2.h           | 158 ++--
 drivers/net/ethernet/qlogic/qed/qed_ll2.h          | 130 +--
 drivers/net/ethernet/qlogic/qed/qed_main.c         |   6 +-
 drivers/net/ethernet/qlogic/qed/qed_mcp.h          | 757 +++++++++--------
 drivers/net/ethernet/qlogic/qed/qed_selftest.h     |  30 +-
 drivers/net/ethernet/qlogic/qed/qed_sp.h           | 215 +++--
 drivers/net/ethernet/qlogic/qed/qed_sriov.h        |  99 ++-
 drivers/net/ethernet/qlogic/qed/qed_vf.h           | 301 ++++---
 drivers/net/ethernet/qlogic/qede/qede_main.c       |   5 +-
 drivers/net/ethernet/socionext/netsec.c            |  11 +
 drivers/net/ethernet/xilinx/ll_temac_main.c        |  16 +-
 drivers/net/tap.c                                  |   2 +-
 drivers/net/tun.c                                  |   2 +-
 drivers/net/usb/cdc_ether.c                        |  21 +
 drivers/net/usb/usbnet.c                           |   6 +
 drivers/net/usb/zaurus.c                           |  21 +
 drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c |   6 +-
 drivers/s390/net/qeth_core.h                       |   1 -
 drivers/s390/net/qeth_core_main.c                  |   2 -
 drivers/s390/net/qeth_l2_main.c                    |   9 +-
 drivers/s390/net/qeth_l3_main.c                    |   8 +-
 drivers/s390/scsi/zfcp_fc.c                        |   6 +-
 drivers/scsi/storvsc_drv.c                         |   4 +
 drivers/soundwire/bus.c                            |  20 +-
 fs/ceph/mds_client.c                               |   4 +-
 fs/ceph/mds_client.h                               |   5 +
 fs/ceph/super.c                                    |  10 +
 fs/exfat/balloc.c                                  |   6 +-
 fs/exfat/dir.c                                     |  27 +-
 fs/ext2/ext2.h                                     |  12 -
 fs/ext2/super.c                                    |  23 +-
 fs/file.c                                          |  18 +-
 fs/ntfs3/attrlist.c                                |   4 +-
 fs/open.c                                          |   2 +-
 fs/super.c                                         |  11 +-
 fs/sysv/itree.c                                    |   4 +
 include/asm-generic/word-at-a-time.h               |   2 +-
 include/linux/pm_wakeirq.h                         |   9 +-
 include/linux/qed/qed_chain.h                      |  97 ++-
 include/linux/qed/qed_if.h                         | 255 +++---
 include/linux/qed/qed_iscsi_if.h                   |   2 +-
 include/linux/qed/qed_ll2_if.h                     |  42 +-
 include/linux/qed/qed_nvmetcp_if.h                 |  17 +
 include/net/vxlan.h                                |   4 +-
 io_uring/io_uring.c                                |  23 +-
 kernel/bpf/cpumap.c                                |  35 +-
 kernel/events/core.c                               |   8 +-
 kernel/trace/bpf_trace.c                           |   6 +-
 net/bluetooth/l2cap_sock.c                         |   2 +
 net/ceph/osd_client.c                              |  20 +-
 net/core/bpf_sk_storage.c                          |   5 +-
 net/core/rtnetlink.c                               |   8 +-
 net/core/sock.c                                    |  21 +-
 net/core/sock_map.c                                |   2 -
 net/dcb/dcbnl.c                                    |   2 +-
 net/ipv4/tcp_metrics.c                             |  70 +-
 net/ipv6/ip6mr.c                                   |   2 +-
 net/sched/cls_fw.c                                 |   1 -
 net/sched/cls_route.c                              |   1 -
 net/sched/cls_u32.c                                |  57 +-
 net/sched/sch_taprio.c                             |  15 +-
 net/unix/af_unix.c                                 |   2 +-
 net/wireless/scan.c                                |   2 +-
 .../tests/shell/test_uprobe_from_different_cu.sh   |   8 +-
 tools/testing/selftests/rseq/rseq.c                |  31 +-
 117 files changed, 3227 insertions(+), 2247 deletions(-)
#96 lore raw
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

Thanks,
Guenter

Quoted 421 lines

I will do some more runs to see if TASKS03 hang is a new thing but it could be related to the existing issues.

thanks,

- Joel

Quoted 412 lines

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

     Linux 5.15.126-rc1

Johan Hovold <[email protected]>
PM: sleep: wakeirq: fix wake irq arming

Chunfeng Yun <[email protected]>
PM / wakeirq: support enabling wake-up irq after runtime_suspend called

Johan Hovold <[email protected]>
soundwire: fix enumeration completion

Pierre-Louis Bossart <[email protected]>
soundwire: bus: pm_runtime_request_resume on peripheral attachment

Sean Christopherson <[email protected]>
selftests/rseq: Play nice with binaries statically linked against glibc 2.35+

Michael Jeanson <[email protected]>
selftests/rseq: check if libc rseq support is registered

Alexander Stein <[email protected]>
drm/imx/ipuv3: Fix front porch adjustment upon hactive aligning

Thomas Zimmermann <[email protected]>
drm/fsl-dcu: Use drm_plane_helper_destroy()

Aneesh Kumar K.V <[email protected]>
powerpc/mm/altmap: Fix altmap boundary check

Christophe JAILLET <[email protected]>
mtd: rawnand: fsl_upm: Fix an off-by one test in fun_exec_op()

Johan Jonker <[email protected]>
mtd: rawnand: rockchip: Align hwecc vs. raw page helper layouts

Johan Jonker <[email protected]>
mtd: rawnand: rockchip: fix oobfree offset and description

Roger Quadros <[email protected]>
mtd: rawnand: omap_elm: Fix incorrect type in assignment

Jan Kara <[email protected]>
ext2: Drop fragment support

Jan Kara <[email protected]>
fs: Protect reconfiguration of sb read-write from racing writes

Alan Stern <[email protected]>
net: usbnet: Fix WARNING in usbnet_start_xmit/usb_submit_urb

Sungwoo Kim <[email protected]>
Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb

Prince Kumar Maurya <[email protected]>
fs/sysv: Null check to prevent null-ptr-deref bug

Tetsuo Handa <[email protected]>
fs/ntfs3: Use __GFP_NOWARN allocation at ntfs_load_attr_list()

Linus Torvalds <[email protected]>
file: reinstate f_pos locking optimization for regular files

Hou Tao <[email protected]>
bpf, cpumap: Make sure kthread is running before map update returns

Guchun Chen <[email protected]>
drm/ttm: check null pointer before accessing when swapping

Aleksa Sarai <[email protected]>
open: make RESOLVE_CACHED correctly test for O_TMPFILE

Jiri Olsa <[email protected]>
bpf: Disable preemption in bpf_event_output

Ilya Dryomov <[email protected]>
rbd: prevent busy loop when requesting exclusive lock

Paul Fertser <[email protected]>
wifi: mt76: mt7615: do not advertise 5 GHz on first phy of MT7615D (DBDC)

Laszlo Ersek <[email protected]>
net: tap_open(): set sk_uid from current_fsuid()

Laszlo Ersek <[email protected]>
net: tun_chr_open(): set sk_uid from current_fsuid()

Dinh Nguyen <[email protected]>
arm64: dts: stratix10: fix incorrect I2C property for SCL signal

Arseniy Krasnov <[email protected]>
mtd: rawnand: meson: fix OOB available bytes for ECC

Olivier Maignial <[email protected]>
mtd: spinand: toshiba: Fix ecc_get_status

Sungjong Seo <[email protected]>
exfat: release s_lock before calling dir_emit()

gaoming <[email protected]>
exfat: use kvmalloc_array/kvfree instead of kmalloc_array/kfree

Krzysztof Kozlowski <[email protected]>
firmware: arm_scmi: Drop OF node reference in the transport channel setup

Xiubo Li <[email protected]>
ceph: defer stopping mdsc delayed_work

Ross Maynard <[email protected]>

     USB: zaurus: Add ID for A-300/B-500/C-700

Ilya Dryomov <[email protected]>
libceph: fix potential hang in ceph_osdc_notify()

Michael Kelley <[email protected]>
scsi: storvsc: Limit max_sectors for virtual Fibre Channel devices

Steffen Maier <[email protected]>
scsi: zfcp: Defer fc_rport blocking until after ADISC response

Eric Dumazet <[email protected]>
tcp_metrics: fix data-race in tcpm_suck_dst() vs fastopen

Eric Dumazet <[email protected]>

     tcp_metrics: annotate data-races around tm->tcpm_net

Eric Dumazet <[email protected]>

     tcp_metrics: annotate data-races around tm->tcpm_vals[]

Eric Dumazet <[email protected]>

     tcp_metrics: annotate data-races around tm->tcpm_lock

Eric Dumazet <[email protected]>

     tcp_metrics: annotate data-races around tm->tcpm_stamp

Eric Dumazet <[email protected]>
tcp_metrics: fix addr_same() helper

Jonas Gorski <[email protected]>
prestera: fix fallback to previous version on same major version

Jianbo Liu <[email protected]>
net/mlx5: fs_core: Skip the FTs in the same FS_TYPE_PRIO_CHAINS fs_prio

Jianbo Liu <[email protected]>
net/mlx5: fs_core: Make find_closest_ft more generic

Benjamin Poirier <[email protected]>
vxlan: Fix nexthop hash size

Yue Haibing <[email protected]>
ip6mr: Fix skb_under_panic in ip6mr_cache_report()

Alexandra Winter <[email protected]>
s390/qeth: Don't call dev_close/dev_open (DOWN/UP)

Lin Ma <[email protected]>
net: dcb: choose correct policy to parse DCB_ATTR_BCN

Mark Brown <[email protected]>
net: netsec: Ignore 'phy-mode' on SynQuacer in DT mode

Yuanjun Gong <[email protected]>
net: korina: handle clk prepare error in korina_probe()

Dan Carpenter <[email protected]>
net: ll_temac: fix error checking of irq_of_parse_and_map()

Yang Yingliang <[email protected]>
net: ll_temac: Switch to use dev_err_probe() helper

Tomas Glozar <[email protected]>
bpf: sockmap: Remove preempt_disable in sock_map_sk_acquire

valis <[email protected]>
net/sched: cls_route: No longer copy tcf_result on update to avoid use-after-free

valis <[email protected]>
net/sched: cls_fw: No longer copy tcf_result on update to avoid use-after-free

valis <[email protected]>
net/sched: cls_u32: No longer copy tcf_result on update to avoid use-after-free

Hou Tao <[email protected]>
bpf, cpumap: Handle skb as well when clean up ptr_ring

Kuniyuki Iwashima <[email protected]>
net/sched: taprio: Limit TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME to INT_MAX.

Eric Dumazet <[email protected]>
net: add missing data-race annotation for sk_ll_usec

Eric Dumazet <[email protected]>

     net: add missing data-race annotations around sk->sk_peek_off

Eric Dumazet <[email protected]>

     net: add missing READ_ONCE(sk->sk_rcvbuf) annotation

Eric Dumazet <[email protected]>

     net: add missing READ_ONCE(sk->sk_sndbuf) annotation

Eric Dumazet <[email protected]>

     net: add missing READ_ONCE(sk->sk_rcvlowat) annotation

Eric Dumazet <[email protected]>

     net: annotate data-races around sk->sk_max_pacing_rate

Konstantin Khorenko <[email protected]>
qed: Fix scheduling in a tasklet while getting stats

Prabhakar Kushwaha <[email protected]>
qed: Fix kernel-doc warnings

Chengfeng Ye <[email protected]>

     mISDN: hfcpci: Fix potential deadlock on &hc->lock

Jamal Hadi Salim <[email protected]>
net: sched: cls_u32: Fix match key mis-addressing

Georg Müller <[email protected]>
perf test uprobe_from_different_cu: Skip if there is no gcc

Yuanjun Gong <[email protected]>
net: dsa: fix value check in bcm_sf2_sw_probe()

Lin Ma <[email protected]>
rtnetlink: let rtnl_bridge_setlink checks IFLA_BRIDGE_MODE length

Lin Ma <[email protected]>
bpf: Add length check for SK_DIAG_BPF_STORAGE_REQ_MAP_FD parsing

Yuanjun Gong <[email protected]>
net/mlx5e: fix return value check in mlx5e_ipsec_remove_trailer()

Zhengchao Shao <[email protected]>
net/mlx5: DR, fix memory leak in mlx5dr_cmd_create_reformat_ctx

Ilan Peer <[email protected]>
wifi: cfg80211: Fix return value in scan logic

Heiko Carstens <[email protected]>
KVM: s390: fix sthyi error handling

[email protected] <[email protected]>
word-at-a-time: use the same return type for has_zero regardless of endianness

Cristian Marussi <[email protected]>
firmware: arm_scmi: Fix chan_free cleanup on SMC

Hugo Villeneuve <[email protected]>
arm64: dts: imx8mn-var-som: add missing pull-up for onboard PHY reset pinmux

Robin Murphy <[email protected]>
iommu/arm-smmu-v3: Document nesting-related errata

Robin Murphy <[email protected]>
iommu/arm-smmu-v3: Add explicit feature for nesting

Robin Murphy <[email protected]>

     iommu/arm-smmu-v3: Document MMU-700 erratum 2812531

Robin Murphy <[email protected]>

     iommu/arm-smmu-v3: Work around MMU-600 erratum 1076982

Suzuki K Poulose <[email protected]>
arm64: errata: Add detection for TRBE write to out-of-range

Suzuki K Poulose <[email protected]>
arm64: errata: Add workaround for TSB flush failures

Shay Drory <[email protected]>
net/mlx5: Free irqs only on shutdown callback

Peter Zijlstra <[email protected]>
perf: Fix function pointer case

Jens Axboe <[email protected]>
io_uring: gate iowait schedule on having pending requests

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

Diffstat:

  Documentation/arm64/silicon-errata.rst             |  12 +
  Makefile                                           |   4 +-
  arch/arm64/Kconfig                                 |  74 ++
  .../boot/dts/altera/socfpga_stratix10_socdk.dts    |   2 +-
  .../dts/altera/socfpga_stratix10_socdk_nand.dts    |   2 +-
  arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi  |   2 +-
  arch/arm64/include/asm/barrier.h                   |  16 +-
  arch/arm64/kernel/cpu_errata.c                     |  39 +
  arch/arm64/tools/cpucaps                           |   2 +
  arch/powerpc/include/asm/word-at-a-time.h          |   2 +-
  arch/powerpc/mm/init_64.c                          |   3 +-
  arch/s390/kernel/sthyi.c                           |   6 +-
  arch/s390/kvm/intercept.c                          |   9 +-
  drivers/base/power/power.h                         |   8 +-
  drivers/base/power/runtime.c                       |   6 +-
  drivers/base/power/wakeirq.c                       | 111 ++-
  drivers/block/rbd.c                                |  28 +-
  drivers/firmware/arm_scmi/mailbox.c                |   4 +-
  drivers/firmware/arm_scmi/smc.c                    |  21 +-
  drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c        |   8 +-
  drivers/gpu/drm/imx/ipuv3-crtc.c                   |   2 +-
  drivers/gpu/drm/ttm/ttm_bo.c                       |   3 +-
  drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c        |  50 ++
  drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h        |   8 +
  drivers/isdn/hardware/mISDN/hfcpci.c               |  10 +-
  drivers/mtd/nand/raw/fsl_upm.c                     |   2 +-
  drivers/mtd/nand/raw/meson_nand.c                  |   3 +-
  drivers/mtd/nand/raw/omap_elm.c                    |  24 +-
  drivers/mtd/nand/raw/rockchip-nand-controller.c    |  45 +-
  drivers/mtd/nand/spi/toshiba.c                     |   4 +-
  drivers/net/dsa/bcm_sf2.c                          |   8 +-
  drivers/net/ethernet/korina.c                      |   3 +-
  .../net/ethernet/marvell/prestera/prestera_pci.c   |   3 +-
  .../mellanox/mlx5/core/en_accel/ipsec_rxtx.c       |   4 +-
  drivers/net/ethernet/mellanox/mlx5/core/eq.c       |   2 +-
  drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  | 105 ++-
  drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h |   1 +
  drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  |  29 +
  .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  |   5 +-
  drivers/net/ethernet/qlogic/qed/qed.h              |   9 +-
  drivers/net/ethernet/qlogic/qed/qed_cxt.h          | 138 +--
  drivers/net/ethernet/qlogic/qed/qed_dev_api.h      | 361 ++++----
  drivers/net/ethernet/qlogic/qed/qed_fcoe.c         |  19 +-
  drivers/net/ethernet/qlogic/qed/qed_fcoe.h         |  17 +-
  drivers/net/ethernet/qlogic/qed/qed_hsi.h          | 922 +++++++++++----------
  drivers/net/ethernet/qlogic/qed/qed_hw.c           |  26 +-
  drivers/net/ethernet/qlogic/qed/qed_hw.h           | 214 ++---
  drivers/net/ethernet/qlogic/qed/qed_init_ops.h     |  58 +-
  drivers/net/ethernet/qlogic/qed/qed_int.h          | 274 +++---
  drivers/net/ethernet/qlogic/qed/qed_iscsi.c        |  19 +-
  drivers/net/ethernet/qlogic/qed/qed_iscsi.h        |  17 +-
  drivers/net/ethernet/qlogic/qed/qed_l2.c           |  19 +-
  drivers/net/ethernet/qlogic/qed/qed_l2.h           | 158 ++--
  drivers/net/ethernet/qlogic/qed/qed_ll2.h          | 130 +--
  drivers/net/ethernet/qlogic/qed/qed_main.c         |   6 +-
  drivers/net/ethernet/qlogic/qed/qed_mcp.h          | 757 +++++++++--------
  drivers/net/ethernet/qlogic/qed/qed_selftest.h     |  30 +-
  drivers/net/ethernet/qlogic/qed/qed_sp.h           | 215 +++--
  drivers/net/ethernet/qlogic/qed/qed_sriov.h        |  99 ++-
  drivers/net/ethernet/qlogic/qed/qed_vf.h           | 301 ++++---
  drivers/net/ethernet/qlogic/qede/qede_main.c       |   5 +-
  drivers/net/ethernet/socionext/netsec.c            |  11 +
  drivers/net/ethernet/xilinx/ll_temac_main.c        |  16 +-
  drivers/net/tap.c                                  |   2 +-
  drivers/net/tun.c                                  |   2 +-
  drivers/net/usb/cdc_ether.c                        |  21 +
  drivers/net/usb/usbnet.c                           |   6 +
  drivers/net/usb/zaurus.c                           |  21 +
  drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c |   6 +-
  drivers/s390/net/qeth_core.h                       |   1 -
  drivers/s390/net/qeth_core_main.c                  |   2 -
  drivers/s390/net/qeth_l2_main.c                    |   9 +-
  drivers/s390/net/qeth_l3_main.c                    |   8 +-
  drivers/s390/scsi/zfcp_fc.c                        |   6 +-
  drivers/scsi/storvsc_drv.c                         |   4 +
  drivers/soundwire/bus.c                            |  20 +-
  fs/ceph/mds_client.c                               |   4 +-
  fs/ceph/mds_client.h                               |   5 +
  fs/ceph/super.c                                    |  10 +
  fs/exfat/balloc.c                                  |   6 +-
  fs/exfat/dir.c                                     |  27 +-
  fs/ext2/ext2.h                                     |  12 -
  fs/ext2/super.c                                    |  23 +-
  fs/file.c                                          |  18 +-
  fs/ntfs3/attrlist.c                                |   4 +-
  fs/open.c                                          |   2 +-
  fs/super.c                                         |  11 +-
  fs/sysv/itree.c                                    |   4 +
  include/asm-generic/word-at-a-time.h               |   2 +-
  include/linux/pm_wakeirq.h                         |   9 +-
  include/linux/qed/qed_chain.h                      |  97 ++-
  include/linux/qed/qed_if.h                         | 255 +++---
  include/linux/qed/qed_iscsi_if.h                   |   2 +-
  include/linux/qed/qed_ll2_if.h                     |  42 +-
  include/linux/qed/qed_nvmetcp_if.h                 |  17 +
  include/net/vxlan.h                                |   4 +-
  io_uring/io_uring.c                                |  23 +-
  kernel/bpf/cpumap.c                                |  35 +-
  kernel/events/core.c                               |   8 +-
  kernel/trace/bpf_trace.c                           |   6 +-
  net/bluetooth/l2cap_sock.c                         |   2 +
  net/ceph/osd_client.c                              |  20 +-
  net/core/bpf_sk_storage.c                          |   5 +-
  net/core/rtnetlink.c                               |   8 +-
  net/core/sock.c                                    |  21 +-
  net/core/sock_map.c                                |   2 -
  net/dcb/dcbnl.c                                    |   2 +-
  net/ipv4/tcp_metrics.c                             |  70 +-
  net/ipv6/ip6mr.c                                   |   2 +-
  net/sched/cls_fw.c                                 |   1 -
  net/sched/cls_route.c                              |   1 -
  net/sched/cls_u32.c                                |  57 +-
  net/sched/sch_taprio.c                             |  15 +-
  net/unix/af_unix.c                                 |   2 +-
  net/wireless/scan.c                                |   2 +-
  .../tests/shell/test_uprobe_from_different_cu.sh   |   8 +-
  tools/testing/selftests/rseq/rseq.c                |  31 +-
  117 files changed, 3227 insertions(+), 2247 deletions(-)
#97 lore raw
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

thanks,

- Joel

#98 lore raw
Quoted 32 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

Quoted 8 lines

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Adding to this, it appears one of the CPUs is constantly firing and reprogramming hrtimer events for some reason every few 100 microseconds (I see this in gdb). My debug angle right now is to figure out why it does that but collecting a trace is hard as it appears even trace collection may not be happening once hung and the only traces I am getting are the ones after the hang recovers, not during the hang. I am also trying to see if multi_cpu_stop() can panic the kernel if it sits there too long.

- Joel

#99 lore raw
In reply to Joel Fernandes
Quoted 31 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

Quoted 8 lines

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield(). Example:

<0>[63298.624328] watchdog: BUG: soft lockup - CPU#0 stuck for 11s! [migration/0:11]
<4>[63298.624331] Modules linked in: 8021q ccm snd_seq_dummy snd_seq snd_seq_device bridge stp llc tun nf_nat_tftp nf_conntrack_tftp nf_nat_ftp nf_conntrack_ftp esp6 ah6 ip6t_REJECT ip6t_ipv6header vhost_vsock vhost vmw_vsock_virtio_transport_common vsock veth rfcomm xt_cgroup cmac algif_hash algif_skcipher af_alg xt_MASQUERADE uinput iwlmvm snd_soc_skl_ssp_clk iwl7000_mac80211 btusb snd_soc_kbl_da7219_max98357a btrtl btintel snd_soc_hdac_hdmi btbcm bluetooth snd_soc_dmic snd_soc_skl ecdh_generic ecc snd_soc_sst_ipc snd_soc_sst_dsp snd_soc_hdac_hda uvcvideo snd_soc_acpi_intel_match snd_soc_acpi snd_hda_ext_core videobuf2_vmalloc videobuf2_v4l2 videobuf2_common snd_intel_dspcfg videobuf2_memops snd_hda_codec snd_hwdep snd_hda_core iwlwifi snd_soc_da7219 snd_soc_max98357a fuse ip6table_nat cfg80211 lzo_rle lzo_compress zram joydev
<4>[63298.624357] CPU: 0 PID: 11 Comm: migration/0 Tainted: G     U  W         5.4.180-17902-g44152654f29b #1
<4>[63298.624358] Hardware name: Google Nami/Nami, BIOS Google_Nami.10775.145.0 09/19/2019
<4>[63298.624363] RIP: 0010:stop_machine_yield+0xb/0xd
<4>[63298.624366] Code: ff 74 b6 f0 ff 0f 75 b1 48 83 c7 08 e8 1f cb f9 ff eb a6 e8 a0 20 e3 ff eb bc e8 50 4b f5 ff 0f 1f 44 00 00 55 48 89 e5 f3 90 <5d> c3 0f 1f 44 00 00 55 48 89 e5 41 57 41 56 41 55 41 54 53 48 81
<4>[63298.624368] RSP: 0000:ffffbaf90006fe38 EFLAGS: 00000293 ORIG_RAX: ffffffffffffff13
<4>[63298.624370] RAX: 0000000000000000 RBX: ffffbaf90300bca8 RCX: 0000000000000000
<4>[63298.624371] RDX: 0000000000000000 RSI: 0000000000000008 RDI: ffffffffb0d46920
<4>[63298.624373] RBP: ffffbaf90006fe38 R08: 0000000000000002 R09: 0000398ecf9a0ac5
<4>[63298.624374] R10: 0000000000000171 R11: ffffffffaf9cfb11 R12: 0000000000000001
<4>[63298.624376] R13: ffff9b09baa22201 R14: ffffffffb0d46920 R15: 0000000000000001
<4>[63298.624377] FS:  0000000000000000(0000) GS:ffff9b09baa00000(0000) knlGS:0000000000000000
<4>[63298.624379] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4>[63298.624380] CR2: 0000153c00724820 CR3: 0000000171ab8005 CR4: 00000000003606f0
<4>[63298.624382] Call Trace:
<4>[63298.624386]  multi_cpu_stop+0x89/0x119
<4>[63298.624389]  ? stop_two_cpus+0x24d/0x24d
<4>[63298.624391]  cpu_stopper_thread+0x8f/0x111
<4>[63298.624394]  smpboot_thread_fn+0x174/0x212
<4>[63298.624397]  kthread+0x147/0x156
<4>[63298.624399]  ? cpu_report_death+0x43/0x43
<4>[63298.624401]  ? kthread_blkcg+0x2e/0x2e
<4>[63298.624404]  ret_from_fork+0x35/0x40
<0>[63298.624407] Kernel panic - not syncing: softlockup: hung tasks

I guess that is something different ?

Guenter

#100 lore raw
Quoted 37 lines · Guenter Roeck
Quoted 36 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

Quoted 6 lines
Quoted 3 lines

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

But in dmesg, it prints nothing for about 20-30 mins before recovering, then I get RCU stalls. It looks like this:

[  682.721962] kvm-clock: cpu 7, msr 199981c1, secondary cpu clock [  682.736830] kvm-guest: stealtime: cpu 7, msr 1f5db140 [  684.445875] smpboot: Booting Node 0 Processor 5 APIC 0x5 [  684.467831] kvm-clock: cpu 5, msr 19998141, secondary cpu clock [  684.555766] kvm-guest: stealtime: cpu 5, msr 1f55b140 [  687.356637] smpboot: Booting Node 0 Processor 4 APIC 0x4 [  687.377214] kvm-clock: cpu 4, msr 19998101, secondary cpu clock

[ 2885.473742] kvm-guest: stealtime: cpu 4, msr 1f51b140
[ 2886.456408] rcu: INFO: rcu_sched self-detected stall on CPU
[ 2886.457590] rcu_torture_fwd_prog_nr: Duration 15423 cver 170 gps 337
[ 2886.464934] rcu: 0-...!: (2 ticks this GP) idle=7eb/0/0x1 softirq=118271/118271 fqs=0 last_accelerate: e3cd/71c0 dyntick_enabled: 1
[ 2886.490837] (t=2199034 jiffies g=185489 q=4)
[ 2886.497297] rcu: rcu_sched kthread timer wakeup didn't happen for 2199031 jiffies! g185489 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402
[ 2886.514201] rcu: Possible timer handling issue on cpu=0 timer-softirq=441616
[ 2886.524593] rcu: rcu_sched kthread starved for 2199034 jiffies! g185489 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402 ->cpu=0
[ 2886.540067] rcu: Unless rcu_sched kthread gets sufficient CPU time, OOM is now expected behavior.
[ 2886.551967] rcu: RCU grace-period kthread stack dump:
[ 2886.558644] task:rcu_sched       state:I stack:14896 pid:   15 ppid:     2 flags:0x00004000
[ 2886.569640] Call Trace:
[ 2886.572940]  <TASK>
[ 2886.575902]  __schedule+0x284/0x6e0
[ 2886.580969]  schedule+0x53/0xa0
[ 2886.585231]  schedule_timeout+0x8f/0x130

In that huge gap, I connect gdb and dumped those stacks in above link.

On 5.15 stable you could repro it in about an hour and a half most of the time by running something like:
tools/testing/selftests/rcutorture/bin/kvm.sh --cpus 48 --duration 60 --configs TREE04

Let me know if you saw anything like this. I am currently trying to panic the kernel when the hang happens so I can get better traces.

thanks,

- Joel

#101 lore raw
Quoted 42 lines · Joel Fernandes
Quoted 41 lines · Guenter Roeck
Quoted 39 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Quoted 7 lines
Quoted 2 lines

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

That link gives me "Entry not found".

Guenter

#102 lore raw
Quoted 47 lines · Guenter Roeck
Quoted 45 lines · Joel Fernandes
Quoted 43 lines · Guenter Roeck
Quoted 31 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

Quoted 8 lines

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Quoted 6 lines
Quoted 4 lines

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

That link gives me "Entry not found".

Yeah that was weird. Here it is again: https://pastebin.com/raw/L3nv1kH2

#103 lore raw
Quoted 48 lines · Joel Fernandes
Quoted 47 lines · Guenter Roeck
Quoted 45 lines · Joel Fernandes
Quoted 43 lines · Guenter Roeck
Quoted 31 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

Quoted 8 lines

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

That link gives me "Entry not found".

Yeah that was weird. Here it is again: https://pastebin.com/raw/L3nv1kH2

I found a couple of crash reports from chromeos-5.10, one of them complaining about RCU issues. I sent you links via IM. Nothing from 5.15 or later, though.

Guenter

#104 lore raw
Quoted 54 lines · Guenter Roeck
Quoted 53 lines · Joel Fernandes
Quoted 52 lines · Guenter Roeck
Quoted 49 lines · Joel Fernandes
Quoted 37 lines · Guenter Roeck
Quoted 36 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

Quoted 6 lines
Quoted 3 lines

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

Quoted 6 lines
Quoted 3 lines
Quoted 1 line

That link gives me "Entry not found".

Yeah that was weird. Here it is again: https://pastebin.com/raw/L3nv1kH2

I found a couple of crash reports from chromeos-5.10, one of them complaining about RCU issues. I sent you links via IM. Nothing from 5.15 or later, though.

Is the crash showing the eternally refiring timer fixed by this commit?

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry")

This commit fixed something similar for me in v5.16.

	https://paulmck.livejournal.com/62071.html

							Thanx, Paul
#105 lore raw
Quoted 57 lines · Paul E. McKenney
Quoted 56 lines · Guenter Roeck
Quoted 55 lines · Joel Fernandes
Quoted 53 lines · Guenter Roeck
Quoted 42 lines · Joel Fernandes
Quoted 41 lines · Guenter Roeck
Quoted 39 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Quoted 7 lines
Quoted 2 lines

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

That link gives me "Entry not found".

Quoted 8 lines
Quoted 4 lines
Quoted 1 line

Yeah that was weird. Here it is again: https://pastebin.com/raw/L3nv1kH2

I found a couple of crash reports from chromeos-5.10, one of them complaining about RCU issues. I sent you links via IM. Nothing from 5.15 or later, though.

Is the crash showing the eternally refiring timer fixed by this commit?

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry")

Ah I was just replying, I have been seeing really good results after applying the following 3 commits since yesterday:

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry") 5417ddc1cf1f ("timers/nohz: Switch to ONESHOT_STOPPED in the low-res handler when the tick is stopped") a1ff03cd6fb9 ("tick: Detect and fix jiffies update stall")

5417ddc1cf1f also mentioned a "tick storm" which is exactly what I was seeing.

I did a lengthy test and everything is looking good. I'll send these out to the stable list.

thanks,

- Joel

#106 lore raw
Quoted 74 lines · Joel Fernandes
Quoted 57 lines · Paul E. McKenney
Quoted 56 lines · Guenter Roeck
Quoted 55 lines · Joel Fernandes
Quoted 53 lines · Guenter Roeck
Quoted 42 lines · Joel Fernandes
Quoted 41 lines · Guenter Roeck
Quoted 39 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Quoted 7 lines
Quoted 2 lines

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

That link gives me "Entry not found".

Quoted 8 lines
Quoted 4 lines
Quoted 1 line

Yeah that was weird. Here it is again: https://pastebin.com/raw/L3nv1kH2

I found a couple of crash reports from chromeos-5.10, one of them complaining about RCU issues. I sent you links via IM. Nothing from 5.15 or later, though.

Is the crash showing the eternally refiring timer fixed by this commit?

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry")

Ah I was just replying, I have been seeing really good results after applying the following 3 commits since yesterday:

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry") 5417ddc1cf1f ("timers/nohz: Switch to ONESHOT_STOPPED in the low-res handler when the tick is stopped") a1ff03cd6fb9 ("tick: Detect and fix jiffies update stall")

Quoted 5 lines

5417ddc1cf1f also mentioned a "tick storm" which is exactly what I was seeing.

I did a lengthy test and everything is looking good. I'll send these out to the stable list.

I just read your post for the first time. And just to humor you about my debugging which was very similar to yours, I got as far as this statement in your post (before looking for fixes in timer code):
<quote>
Further checking showed that the stuck CPU was in fact suffering from an interrupt storm, namely an interrupt storm of scheduling-clock interrupts. This spurred another code-inspection session.
</quote>

My detection of this came from gdb, within that 2000 second stall, I broke into the VM with --gdb and kept dumping the stuck CPU's stack with "thread X" and "bt". I noticed that it was always in the timer interrupt. Here were the
stacks: https://pastebin.com/raw/L3nv1kH2

Then I narrowed my search down to timer events by enabling
boot options ftrace_dump_on_oops and panic-on-stall ones, and noticed a storm of hrtimer_start coming out of the long stall. I was all but certain it was a tick storm and noticed it kept programming hrtimer to the same event.

Ah, then I just did a "git diff" in kernel/time/ between v5.15 and v6.1 and noticed the missing patches. ;-)

Though in my experience, I wasn't seeing a KTIME_MAX-type of value like you mentioned in the post. What I noticed is that the tick was never stopped, it just kept firing a bit earlier than was requested and in the interrupt exit path (of the delivered-too-early timer interrupt), it kept re-requesting the tick.

thanks,

- Joel

#107 lore raw
Quoted 100 lines · Joel Fernandes
Quoted 74 lines · Joel Fernandes
Quoted 57 lines · Paul E. McKenney
Quoted 56 lines · Guenter Roeck
Quoted 55 lines · Joel Fernandes
Quoted 53 lines · Guenter Roeck
Quoted 42 lines · Joel Fernandes
Quoted 41 lines · Guenter Roeck
Quoted 39 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Quoted 7 lines
Quoted 2 lines

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

That link gives me "Entry not found".

Quoted 8 lines
Quoted 4 lines
Quoted 1 line

Yeah that was weird. Here it is again: https://pastebin.com/raw/L3nv1kH2

I found a couple of crash reports from chromeos-5.10, one of them complaining about RCU issues. I sent you links via IM. Nothing from 5.15 or later, though.

Is the crash showing the eternally refiring timer fixed by this commit?

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry")

Ah I was just replying, I have been seeing really good results after applying the following 3 commits since yesterday:

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry") 5417ddc1cf1f ("timers/nohz: Switch to ONESHOT_STOPPED in the low-res handler when the tick is stopped") a1ff03cd6fb9 ("tick: Detect and fix jiffies update stall")

Quoted 5 lines

5417ddc1cf1f also mentioned a "tick storm" which is exactly what I was seeing.

I did a lengthy test and everything is looking good. I'll send these out to the stable list.

I just read your post for the first time. And just to humor you about my debugging which was very similar to yours, I got as far as this statement in your post (before looking for fixes in timer code):
<quote>
Further checking showed that the stuck CPU was in fact suffering from an interrupt storm, namely an interrupt storm of scheduling-clock interrupts. This spurred another code-inspection session.
</quote>

My detection of this came from gdb, within that 2000 second stall, I broke into the VM with --gdb and kept dumping the stuck CPU's stack with "thread X" and "bt". I noticed that it was always in the timer interrupt. Here were the
stacks: https://pastebin.com/raw/L3nv1kH2

Then I narrowed my search down to timer events by enabling
boot options ftrace_dump_on_oops and panic-on-stall ones, and noticed a storm of hrtimer_start coming out of the long stall. I was all but certain it was a tick storm and noticed it kept programming hrtimer to the same event.

Quoted 8 lines

Ah, then I just did a "git diff" in kernel/time/ between v5.15 and v6.1 and noticed the missing patches. ;-)

Though in my experience, I wasn't seeing a KTIME_MAX-type of value like you mentioned in the post. What I noticed is that the tick was never stopped, it just kept firing a bit earlier than was requested and in the interrupt exit path (of the delivered-too-early timer interrupt), it kept re-requesting the tick.

That "git diff" wouldn't have shown me much at the time, but I am very glad that you found it!

Thanx, Paul

#108 lore raw
In reply to Joel Fernandes
Quoted 65 lines · Joel Fernandes
Quoted 64 lines · Paul E. McKenney
Quoted 54 lines · Guenter Roeck
Quoted 53 lines · Joel Fernandes
Quoted 52 lines · Guenter Roeck
Quoted 49 lines · Joel Fernandes
Quoted 37 lines · Guenter Roeck
Quoted 36 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

Quoted 6 lines
Quoted 3 lines

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

Quoted 6 lines
Quoted 3 lines
Quoted 1 line

That link gives me "Entry not found".

Yeah that was weird. Here it is again: https://pastebin.com/raw/L3nv1kH2

I found a couple of crash reports from chromeos-5.10, one of them complaining about RCU issues. I sent you links via IM. Nothing from 5.15 or later, though.

Is the crash showing the eternally refiring timer fixed by this commit?

Quoted 9 lines
Quoted 1 line

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry")

Ah I was just replying, I have been seeing really good results after applying the following 3 commits since yesterday:

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry") 5417ddc1cf1f ("timers/nohz: Switch to ONESHOT_STOPPED in the low-res handler when the tick is stopped") a1ff03cd6fb9 ("tick: Detect and fix jiffies update stall")

Would those also apply to v5.10.y, or just 5.15.y ?

Thanks,
Guenter

Quoted 11 lines

5417ddc1cf1f also mentioned a "tick storm" which is exactly what I was seeing.

I did a lengthy test and everything is looking good. I'll send these out to the stable list.

thanks,

- Joel

#109 lore raw
Quoted 70 lines · Guenter Roeck
Quoted 65 lines · Joel Fernandes
Quoted 65 lines · Paul E. McKenney
Quoted 54 lines · Guenter Roeck
Quoted 53 lines · Joel Fernandes
Quoted 52 lines · Guenter Roeck
Quoted 49 lines · Joel Fernandes
Quoted 37 lines · Guenter Roeck
Quoted 36 lines · Joel Fernandes
Quoted 26 lines · Guenter Roeck
Quoted 15 lines · Joel Fernandes
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 8 lines

Not necesscarily new with 5.15 stable but 3 of the 19 rcutorture scenarios hang with this -rc: TREE04, TREE07, TASKS03.

5.15 has a known stop machine issue where it hangs after 1.5 hours with cpu hotplug rcutorture testing. Me and tglx are continuing to debug this. The issue does not show up on anything but 5.15 stable kernels and neither on mainline.

Quoted 3 lines

Do you by any have a crash pattern that we could possibly use to find the crash in ChromeOS crash logs ? No idea if that would help, but it could provide some additional data points.

The pattern shows as a hard hang, the system is unresponsive and all CPUs are stuck in stop_machine. Sometimes it recovers on its own from the hang and then RCU immediately gives stall warnings. It takes 1.5 hour to reproduce and sometimes never happens for several hours.

Quoted 6 lines
Quoted 3 lines

It appears related to CPU hotplug since gdb showed me most of the CPUs are spinning in multi_cpu_stop() / stop machine after the hang.

Hmm, we do see lots of soft lockups with multi_cpu_stop() in the backtrace, but not with v5.15.y but with v5.4.y. The actual hang is in stop_machine_yield().

Interesting. It looks similar as far as the stack dump in gdb goes, here are the stacks I dumped with the hang I referred to:
https://paste.debian.net/1288308/

Quoted 6 lines
Quoted 3 lines
Quoted 1 line

That link gives me "Entry not found".

Yeah that was weird. Here it is again: https://pastebin.com/raw/L3nv1kH2

I found a couple of crash reports from chromeos-5.10, one of them complaining about RCU issues. I sent you links via IM. Nothing from 5.15 or later, though.

Is the crash showing the eternally refiring timer fixed by this commit?

Quoted 8 lines
Quoted 6 lines
Quoted 1 line

53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry")

Ah I was just replying, I have been seeing really good results after applying
the following 3 commits since yesterday:
53e87e3cdc15 ("timers/nohz: Last resort update jiffies on nohz_full IRQ entry") 5417ddc1cf1f ("timers/nohz: Switch to ONESHOT_STOPPED in the low-res handler when the tick is stopped") a1ff03cd6fb9 ("tick: Detect and fix jiffies update stall")

Would those also apply to v5.10.y, or just 5.15.y ?

All apply to 5.10 but one. I am currently testing with it more and will post to stable for 5.10 as well.

Thanks,

- Joel

Quoted 11 lines

Thanks,
Guenter

Quoted 6 lines

5417ddc1cf1f also mentioned a "tick storm" which is exactly what I was seeing.
I did a lengthy test and everything is looking good. I'll send these out to the stable list.
thanks,
- Joel

maintainer Tested-by on the v1 series
#110 lore raw
In reply to Greg Kroah-Hartman

Hello,

Quoted 5 lines · On 2023-08-09T12:40:36+02:00 Greg Kroah-Hartman <[email protected]>…

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 8 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

This rc kernel passes DAMON functionality test[1] on my test machine. Attaching the test results summary below. Please note that I retrieved the kernel from linux-stable-rc tree[2].

[1] https://github.com/awslabs/damon-tests/tree/next/corr
[2] ae7f23cbf199 ("Linux 5.15.126-rc1")

Thanks,
SJ

[...]

---

ok 1 selftests: damon: debugfs_attrs.sh
ok 1 selftests: damon-tests: kunit.sh
ok 2 selftests: damon-tests: huge_count_read_write.sh
ok 3 selftests: damon-tests: buffer_overflow.sh
ok 4 selftests: damon-tests: rm_contexts.sh
ok 5 selftests: damon-tests: record_null_deref.sh
ok 6 selftests: damon-tests: dbgfs_target_ids_read_before_terminate_race.sh ok 7 selftests: damon-tests: dbgfs_target_ids_pid_leak.sh
ok 8 selftests: damon-tests: damo_tests.sh
ok 9 selftests: damon-tests: masim-record.sh
ok 10 selftests: damon-tests: build_i386.sh
ok 11 selftests: damon-tests: build_m68k.sh
ok 12 selftests: damon-tests: build_arm64.sh
ok 13 selftests: damon-tests: build_i386_idle_flag.sh
ok 14 selftests: damon-tests: build_i386_highpte.sh
ok 15 selftests: damon-tests: build_nomemcg.sh

PASS

#111 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

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

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.15.126-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-5.15.y
and the diffstat can be found below.

thanks,

greg k-h

SCMI with the SMC transport fails to build because of "[PATCH 5.15 11/92] firmware: arm_scmi: Fix chan_free cleanup on SMC" where the specific details have been reported there. Here is the build failure FWIW:

drivers/firmware/arm_scmi/smc.c:39:6: error: duplicate member 'irq'

  int irq;
      ^~~
drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:20: error: 'irq' undeclared (first use in this function); did you mean 'rq'?
   scmi_info->irq = irq;
                    ^~~
                    rq

drivers/firmware/arm_scmi/smc.c:118:20: note: each undeclared identifier is reported only once for each function it appears in
CC drivers/mmc/core/slot-gpio.o
host-make[5]: *** [scripts/Makefile.build:289: drivers/firmware/arm_scmi/smc.o] Error 1
host-make[5]: *** Waiting for unfinished jobs....

Signature · 1 line
Florian
Tested-by on the v1 series
#112 lore raw
In reply to Greg Kroah-Hartman

Hi Greg,

Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

No problems seen on x86_64 and aarch64.

Thanks,
Harshit

Quoted 10 lines

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

	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.15.126-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-5.15.y
and the diffstat can be found below.

thanks,

greg k-h

#113 lore raw
Quoted 8 lines · Harshit Mogalapalli

Hi Greg,

Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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.

Quoted 5 lines
Quoted 3 lines

Responses should be made by Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

No problems seen on x86_64 and aarch64.

fwiw, aarch64:allmodconfig doesn't compile.

Guenter

Quoted 15 lines

Thanks,
Harshit

Quoted 10 lines

The whole patch series can be found in one patch at:
    https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.15.126-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-5.15.y and the diffstat can be found below.

thanks,

greg k-h

#114 lore raw
In reply to Greg Kroah-Hartman
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

Building arm:allmodconfig ... failed
--------------
Error log:
drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'

drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared

Building arm64:defconfig ... failed
--------------
Error log:

drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'

drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared

That is because commit d80e159dbdbb ("firmware: arm_scmi: Fix chan free cleanup on SMC") is applied without its dependent commit(s).

Guenter

#115 lore raw
Quoted 21 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

Building arm:allmodconfig ... failed
--------------
Error log:
drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'

drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared

Building arm64:defconfig ... failed
--------------
Error log:

Quoted 7 lines

drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'

drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared

That is because commit d80e159dbdbb ("firmware: arm_scmi: Fix chan free cleanup on SMC") is applied without its dependent commit(s).

Indeed, we discussed this here: https://lore.kernel.org/all/20230810084529.53thk6dmlejbma3t@bogus/

Signature · 1 line
Florian
author maintainer
#116 lore raw
Quoted 24 lines · Florian Fainelli
Quoted 23 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

Building arm:allmodconfig ... failed
--------------
Error log:
drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'

drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared

Building arm64:defconfig ... failed
--------------
Error log:

drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'

Quoted 8 lines
Quoted 5 lines

drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared

That is because commit d80e159dbdbb ("firmware: arm_scmi: Fix chan free cleanup on SMC") is applied without its dependent commit(s).

Indeed, we discussed this here:
https://lore.kernel.org/all/20230810084529.53thk6dmlejbma3t@bogus/

Offending commit should now be dropped, thanks.

greg k-h

#117 lore raw
In reply to Greg Kroah-Hartman
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

Build results:

	total: 160 pass: 157 fail: 3

Failed builds:
arm:allmodconfig
arm64:defconfig
arm64:allmodconfig
Qemu test results:

	total: 501 pass: 423 fail: 78
Failed tests:
	<most arm>
	<all arm64/arm64be>

As already reported, plus:

Error log:
drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c:176:20: error: 'drm_plane_helper_destroy' undeclared here

for arm:multi_v7_defconfig

Side note: I am surprised about successful arm64 tests/builds since arm64:defconfig fails to build with obvious code errors.

drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'

drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared

Guenter

author maintainer
#118 lore raw
Quoted 22 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

Build results:

	total: 160 pass: 157 fail: 3

Failed builds:
arm:allmodconfig
arm64:defconfig
arm64:allmodconfig
Qemu test results:

	total: 501 pass: 423 fail: 78
Failed tests:
	<most arm>
	<all arm64/arm64be>
Quoted 4 lines

As already reported, plus:

Error log:
drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c:176:20: error: 'drm_plane_helper_destroy' undeclared here

Offending commit now dropped, Sasha's dep-bot went a little crazy there, and this wasn't needed, sorry for not catching that sooner.

Quoted 9 lines

for arm:multi_v7_defconfig

Side note: I am surprised about successful arm64 tests/builds since arm64:defconfig fails to build with obvious code errors.

drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'

drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared

Should now be fixed, thanks.

greg k-h

Tested-by on the v1 series
#119 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +0000. Anything received after that time might be too late.

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

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.15.126-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-5.15.y
and the diffstat can be found below.

thanks,

greg k-h

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

#120 lore raw
In reply to Greg Kroah-Hartman

Hello!

Quoted 14 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.15.126 release. There are 92 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 Fri, 11 Aug 2023 10:36:10 +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-5.15.126-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-5.15.y
and the diffstat can be found below.

Quoted 5 lines

thanks,

greg k-h

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

We are also seeing build failures on Arm and Arm64, with Clang 17 and GCC 8:

* arm, build
- clang-17-defconfig
- clang-17-lkftconfig
- clang-17-lkftconfig-no-kselftest-frag
- clang-lkftconfig
- clang-nightly-lkftconfig-kselftest
- gcc-8-defconfig

* arm64, build
- clang-17-defconfig
- clang-17-defconfig-40bc7ee5
- clang-17-lkftconfig
- clang-17-lkftconfig-no-kselftest-frag
- clang-lkftconfig
- clang-nightly-lkftconfig-kselftest
- gcc-8-defconfig
- gcc-8-defconfig-40bc7ee5

Failure is:

-----8<----- /builds/linux/drivers/firmware/arm_scmi/smc.c:39:13: error: duplicate member 'irq'
39 | int irq;
| ^~~
/builds/linux/drivers/firmware/arm_scmi/smc.c: In function 'smc_chan_setup':
/builds/linux/drivers/firmware/arm_scmi/smc.c:118:34: error: 'irq' undeclared (first use in this function); did you mean 'rq'?

  118 |                 scmi_info->irq = irq;
      |                                  ^~~
      |                                  rq

----->8-----

(Funnily enough, this was reported by Naresh [1] before this RC round, but we chalked it up to GCC-13 on an older branch.)

Greetings!

Daniel Díaz
[email protected]

[1] https://lore.kernel.org/stable/CA+G9fYvTjm2oa6mXR=HUe6gYuVaS2nFb_otuvPfmPeKHDoC+Tw@mail.gmail.com/