Skip to contentSkip to conversation

5.7.17-rc1 review

PATCH In review Greg Kroah-Hartman 5 participants 210 messages Aug 20, 2020 – Aug 21, 2020 Page 1 of 2

Subsystems (4)
lore
Share & tools
lore

Greg Kroah-Hartman

author maintainer
#0 lore raw

cover 5.7.17-rc1 review

This is the start of the stable review cycle for the 5.7.17 release. There are 204 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Sat, 22 Aug 2020 09:15:09 +0000. Anything received after that time might be too late.

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

	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.7.17-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-5.7.y
and the diffstat can be found below.

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

    Linux 5.7.17-rc1

hersen wu <[email protected]>
drm/amd/display: dchubbub p-state warning during surface planes switch

Stylon Wang <[email protected]>
drm/amd/display: Fix dmesg warning from setting abm level

Sandeep Raghuraman <[email protected]>
drm/amdgpu: Fix bug where DPM is not enabled after hibernate and resume

Xin Xiong <[email protected]>
drm: fix drm_dp_mst_port refcount leaks in drm_dp_mst_allocate_vcpi

Marius Iacob <[email protected]>
drm: Added orientation quirk for ASUS tablet model T103HAF

Tomi Valkeinen <[email protected]>
drm/tidss: fix modeset init for DPI panels

Tomi Valkeinen <[email protected]>
drm/omap: force runtime PM suspend on system suspend

Alex Deucher <[email protected]>
drm/amdgpu: fix ordering of psp suspend

Imre Deak <[email protected]>
drm/dp_mst: Fix the DDC I2C device registration of an MST port

Denis Efremov <[email protected]>

    drm/panfrost: Use kvfree() to free bo->sgts

Chris Wilson <[email protected]>
drm/i915/gt: Force the GT reset on shutdown

Denis Efremov <[email protected]>
drm/radeon: fix fb_div check in ni_init_smc_spll_table()

Sowjanya Komatineni <[email protected]>
ASoC: tegra: Enable audio mclk during tegra_asoc_utils_init()

Sowjanya Komatineni <[email protected]>
ASoC: tegra: Add audio mclk parent configuration

Sowjanya Komatineni <[email protected]>
ASoC: tegra: Use device managed resource APIs to get the clock

Hugh Dickins <[email protected]>
khugepaged: retract_page_tables() remember to test exit

Geert Uytterhoeven <[email protected]>
sh: landisk: Add missing initialization of sh_io_port_base

Zhang Rui <[email protected]>
perf/x86/rapl: Fix missing psys sysfs attributes

Daniel Díaz <[email protected]>
tools build feature: Quote CC and CXX for their arguments

Vincent Whitchurch <[email protected]>
perf bench mem: Always memset source before memcpy

Dinghao Liu <[email protected]>
ALSA: echoaudio: Fix potential Oops in snd_echo_resume()

Ondrej Mosnacek <[email protected]>

    crypto: algif_aead - fix uninitialized ctx->init

Andy Shevchenko <[email protected]>
mfd: dln2: Run event handler loop under spinlock

Dhananjay Phadke <[email protected]>
i2c: iproc: fix race between client unreg and isr

Tiezhu Yang <[email protected]>
test_kmod: avoid potential double free in trigger_config_run_type()

Colin Ian King <[email protected]>
fs/ufs: avoid potential u32 multiplication overflow

Eric Biggers <[email protected]>
fs/minix: remove expected error message in block_to_path()

Eric Biggers <[email protected]>
fs/minix: fix block limit check for V1 filesystems

Eric Biggers <[email protected]>
fs/minix: set s_maxbytes correctly

Tiezhu Yang <[email protected]>
lib/test_lockup.c: fix return value of test_lockup_init()

Trond Myklebust <[email protected]>
NFS: Fix flexfiles read failover

Jeffrey Mitchell <[email protected]>
nfs: Fix getxattr kernel panic and memory overflow

Wang Hai <[email protected]>
net: qcom/emac: add missed clk_disable_unprepare in error path of emac_clks_phase1_init

Krzysztof Kozlowski <[email protected]>
s390/Kconfig: add missing ZCRYPT dependency to VFIO_AP

Wang Hai <[email protected]>
s390/test_unwind: fix possible memleak in test_unwind()

Dan Carpenter <[email protected]>
drm/vmwgfx: Fix two list_for_each loop exit tests

Dan Carpenter <[email protected]>
drm/vmwgfx: Use correct vmw_legacy_display_unit pointer

Dan Carpenter <[email protected]>
vdpa: Fix pointer math bug in vdpasim_get_config()

Christophe Leroy <[email protected]>
recordmcount: Fix build failure on non arm64

Michael S. Tsirkin <[email protected]>
vdpa_sim: init iommu lock

Colin Ian King <[email protected]>
Input: sentelic - fix error return when fsp_reg_write fails

Andrii Nakryiko <[email protected]>
selftests/bpf: Prevent runqslower from racing on building bpftool

Pawan Gupta <[email protected]>
x86/bugs/multihit: Fix mitigation reporting when VMX is not in use

Dilip Kota <[email protected]>
x86/tsr: Fix tsc frequency enumeration bug on Lightning Mountain SoC

Dan Carpenter <[email protected]>
md-cluster: Fix potential error pointer dereference in resize_bitmaps()

Tero Kristo <[email protected]>
watchdog: rti-wdt: balance pm runtime enable calls

Krzysztof Sobota <[email protected]>
watchdog: initialize device before misc_register

Scott Mayhew <[email protected]>
nfs: nfs_file_write() should check for writeback errors

Ewan D. Milne <[email protected]>
scsi: lpfc: nvmet: Avoid hang / use-after-free again when destroying targetport

Stafford Horne <[email protected]>
openrisc: Fix oops caused when dumping stack

Jane Chu <[email protected]>
libnvdimm/security: ensure sysfs poll thread woke up and fetch updated attr

Jane Chu <[email protected]>
libnvdimm/security: fix a typo

Nicolas Saenz Julienne <[email protected]>
clk: bcm2835: Do not use prediv with bcm2711's PLLs

Zhihao Cheng <[email protected]>
ubifs: Fix wrong orphan node deletion in ubifs_jnl_update|rename

Scott Mayhew <[email protected]>
nfs: ensure correct writeback errors are returned on close()

Wolfram Sang <[email protected]>
i2c: rcar: avoid race when unregistering slave

Thomas Hebb <[email protected]>
tools build feature: Use CC and CXX from parent

Rayagonda Kokatanur <[email protected]>
pwm: bcm-iproc: handle clk_get_rate() return

Qais Yousef <[email protected]>
sched/uclamp: Fix a deadlock when enabling uclamp static key

Sagi Grimberg <[email protected]>
nvme: fix deadlock in disconnect during scan_work and/or ana_work

Xu Wang <[email protected]>
clk: clk-atlas6: fix return value check in atlas6_clk_init()

Konrad Dybcio <[email protected]>
clk: qcom: gcc-sdm660: Fix up gcc_mss_mnoc_bimc_axi_clk

Chao Yu <[email protected]>
f2fs: compress: fix to update isize when overwriting compressed file

Wolfram Sang <[email protected]>
i2c: rcar: slave: only send STOP event when we have been addressed

Liu Yi L <[email protected]>
iommu/vt-d: Enforce PASID devTLB field mask

Jonathan Marek <[email protected]>
clk: qcom: clk-alpha-pll: remove unused/incorrect PLL_CAL_VAL

Jonathan Marek <[email protected]>
clk: qcom: gcc: fix sm8150 GPU and NPU clocks

Colin Ian King <[email protected]>
iommu/omap: Check for failure of a call to omap_iommu_dump_ctx

Aneesh Kumar K.V <[email protected]>
selftests/powerpc: ptrace-pkey: Don't update expected UAMOR value

Aneesh Kumar K.V <[email protected]>
selftests/powerpc: ptrace-pkey: Update the test to mark an invalid pkey correctly

Aneesh Kumar K.V <[email protected]>
selftests/powerpc: ptrace-pkey: Rename variables to make it easier to follow code

Cristian Ciocaltea <[email protected]>
clk: actions: Fix h_clk for Actions S500 SoC

Chao Yu <[email protected]>

    f2fs: compress: fix to avoid memory leak on cc->cpages

Tyler Hicks <[email protected]>
ima: Fail rule parsing when appraise_flag=blacklist is unsupportable

Ming Lei <[email protected]>
dm rq: don't call blk_mq_queue_stopped() in dm_stop_queue()

Steve Longerbeam <[email protected]>
gpu: ipu-v3: image-convert: Wait for all EOFs before completing a tile

Steve Longerbeam <[email protected]>
gpu: ipu-v3: image-convert: Combine rotate/no-rotate irq handlers

Herbert Xu <[email protected]>
crypto: caam - Remove broken arc4 support

Sudeep Holla <[email protected]>
rtc: pl031: fix set_alarm by adding back call to alarm_irq_enable

Yan-Hsuan Chuang <[email protected]>
rtw88: pci: disable aspm for platform inter-op with module parameter

Yoshihiro Shimoda <[email protected]>
mmc: renesas_sdhi_internal_dmac: clean up the code for dma complete

Mark Zhang <[email protected]>
RDMA/counter: Allow manually bind QPs with different pids to same counter

Mark Zhang <[email protected]>
RDMA/counter: Only bind user QPs in auto mode

Vladimir Oltean <[email protected]>
devres: keep both device name and resource name in pretty name

Herbert Xu <[email protected]>
crypto: af_alg - Fix regression on empty requests

Johan Hovold <[email protected]>
USB: serial: ftdi_sio: fix break and sysrq handling

Johan Hovold <[email protected]>
USB: serial: ftdi_sio: clean up receive processing

Johan Hovold <[email protected]>
USB: serial: ftdi_sio: make process-packet buffer unsigned

Jesper Dangaard Brouer <[email protected]>
selftests/bpf: test_progs avoid minus shell exit codes

Jesper Dangaard Brouer <[email protected]>
selftests/bpf: test_progs use another shell exit on non-actions

Martin KaFai Lau <[email protected]>
bpf: selftests: Restore netns after each test

Jesper Dangaard Brouer <[email protected]>
selftests/bpf: Test_progs indicate to shell on non-actions

Qais Yousef <[email protected]>
sched/uclamp: Protect uclamp fast path code with static key

Yishai Hadas <[email protected]>
IB/uverbs: Set IOVA on IB MR in uverbs layer

Paul Kocialkowski <[email protected]>
media: rockchip: rga: Only set output CSC mode for RGB input

Paul Kocialkowski <[email protected]>
media: rockchip: rga: Introduce color fmt macros and refactor CSC mode logic

Dafna Hirschfeld <[email protected]>
media: staging: rkisp1: remove macro RKISP1_DIR_SINK_SRC

Sebastian Reichel <[email protected]>
rtc: cpcap: fix range

Jason Gunthorpe <[email protected]>
RDMA/ipoib: Fix ABBA deadlock with ipoib_reap_ah()

Kamal Heib <[email protected]>
RDMA/ipoib: Return void from ipoib_ib_dev_stop()

Qiushi Wu <[email protected]>
platform/chrome: cros_ec_ishtp: Fix a double-unlock issue

Kamal Dasu <[email protected]>
mtd: rawnand: brcmnand: ECC error handling on EDU transfers

Boris Brezillon <[email protected]>
mtd: rawnand: fsl_upm: Remove unused mtd var

Eric Dumazet <[email protected]>

    octeontx2-af: change (struct qmem)->entry_sz from u8 to u16

Charles Keepax <[email protected]>
mfd: arizona: Ensure 32k clock is put on driver unbind and error

Herbert Xu <[email protected]>

    crypto: algif_aead - Only wake up when ctx->more is zero

Paul Cercueil <[email protected]>

    drm/ingenic: Fix incorrect assumption about plane->index

Liu Ying <[email protected]>

    drm/imx: imx-ldb: Disable both channels for split mode in enc->disable()

Sibi Sankar <[email protected]>
remoteproc: qcom_q6v5_mss: Validate modem blob firmware size before load

Sibi Sankar <[email protected]>
remoteproc: qcom_q6v5_mss: Validate MBA firmware size before load

Sibi Sankar <[email protected]>
remoteproc: qcom: q6v5: Update running state before requesting stop

Bob Peterson <[email protected]>
gfs2: Never call gfs2_block_zero_range with an open transaction

Adrian Hunter <[email protected]>
perf intel-pt: Fix duplicate branch after CBR

Adrian Hunter <[email protected]>
perf intel-pt: Fix FUP packet state

Masami Hiramatsu <[email protected]>
perf probe: Fix memory leakage when the probe point is not found

Masami Hiramatsu <[email protected]>
perf probe: Fix wrong variable warning when the probe point is not found

Masami Hiramatsu <[email protected]>
bootconfig: Fix to find the initargs correctly

Kees Cook <[email protected]>
module: Correctly truncate sysfs sections output

Johannes Thumshirn <[email protected]>
dm: don't call report zones for more than the user requested

Anton Blanchard <[email protected]>
pseries: Fix 64 bit logical memory block panic

Jeff Layton <[email protected]>
ceph: handle zero-length feature mask in session messages

Jeff Layton <[email protected]>
ceph: set sec_context xattr on symlink creation

Ahmad Fatoum <[email protected]>
watchdog: f71808e_wdt: clear watchdog timeout occurred flag

Ahmad Fatoum <[email protected]>
watchdog: f71808e_wdt: remove use of wrong watchdog_info option

Ahmad Fatoum <[email protected]>
watchdog: f71808e_wdt: indicate WDIOF_CARDRESET support in watchdog_info.options

Steven Rostedt (VMware) <[email protected]>
tracing: Use trace_sched_process_free() instead of exit() for pid tracing

Kevin Hao <[email protected]>
tracing/hwlat: Honor the tracing_cpumask

Muchun Song <[email protected]>
kprobes: Fix NULL pointer dereference at kprobe_ftrace_handler

Chengming Zhou <[email protected]>
ftrace: Setup correct FTRACE_FL_REGS flags for module

Jia He <[email protected]>
mm/memory_hotplug: fix unpaired mem_hotplug_begin/done

Mike Kravetz <[email protected]>
cma: don't quit at first error when activating reserved areas

Michal Koutný <[email protected]>
mm/page_counter.c: fix protection usage propagation

Junxiao Bi <[email protected]>
ocfs2: change slot number type s16 to u16

David Hildenbrand <[email protected]>
mm/shuffle: don't move pages between zones and don't read garbage memmaps

Mike Kravetz <[email protected]>
hugetlbfs: remove call to huge_pte_alloc without i_mmap_rwsem

Hugh Dickins <[email protected]>
khugepaged: collapse_pte_mapped_thp() protect the pmd lock

Peter Xu <[email protected]>
mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible

Hugh Dickins <[email protected]>
khugepaged: collapse_pte_mapped_thp() flush the right range

Mikulas Patocka <[email protected]>
ext2: fix missing percpu_counter_inc

Mike Rapoport <[email protected]>
MIPS: SGI-IP27: always enable NUMA in Kconfig

Paul Cercueil <[email protected]>
MIPS: qi_lb60: Fix routing to audio amplifier

Huacai Chen <[email protected]>
MIPS: CPU#0 is not hotpluggable

Lukas Wunner <[email protected]>
driver core: Avoid binding drivers to dead devices

Johannes Berg <[email protected]>
mac80211: fix misplaced while instead of if

Coly Li <[email protected]>
bcache: fix overflow in offset_to_stripe()

Coly Li <[email protected]>
bcache: allocate meta data pages as compound pages

ChangSyun Peng <[email protected]>
md/raid5: Fix Force reconstruct-write io stuck in degraded raid5

Kees Cook <[email protected]>
selftests/seccomp: Set NNP for TSYNC ESRCH flag test

Kees Cook <[email protected]>
net/compat: Add missing sock updates for SCM_RIGHTS

Kees Cook <[email protected]>
pidfd: Add missing sock updates for pidfd_getfd()

Zenghui Yu <[email protected]>
irqchip/gic-v4.1: Ensure accessing the correct RD when writing INVALLR

Huacai Chen <[email protected]>

    irqchip/loongson-liointc: Fix misuse of gc->mask_cache

Jonathan McDowell <[email protected]>
net: stmmac: dwmac1000: provide multicast filter fallback

Jonathan McDowell <[email protected]>
net: ethernet: stmmac: Disable hardware multicast filter

Eugeniu Rosca <[email protected]>
media: vsp1: dl: Fix NULL pointer dereference on unbind

Mansur Alisha Shaik <[email protected]>
media: venus: fix multiple encoder crash

Paul Cercueil <[email protected]>
pinctrl: ingenic: Properly detect GPIO direction when configured for IRQ

Paul Cercueil <[email protected]>
pinctrl: ingenic: Enhance support for IRQ_TYPE_EDGE_BOTH

Michael Ellerman <[email protected]>
powerpc: Fix circular dependency between percpu.h and mmu.h

Michael Ellerman <[email protected]>
powerpc: Allow 4224 bytes of stack expansion for the signal frame

Christophe Leroy <[email protected]>
powerpc/ptdump: Fix build failure in hashpagetable.c

Paul Aurich <[email protected]>
cifs: Fix leak when handling lease break for cached root fid

Max Filippov <[email protected]>
xtensa: fix xtensa_pmu_setup prototype

Max Filippov <[email protected]>
xtensa: add missing exclusive access state management

Lorenzo Bianconi <[email protected]>
iio: imu: st_lsm6dsx: reset hw ts after resume

Alexandru Ardelean <[email protected]>
iio: dac: ad5592r: fix unbalanced mutex unlocks in ad5592r_read_raw()

Christian Eggers <[email protected]>
dt-bindings: iio: io-channel-mux: Fix compatible string in example code

Shaokun Zhang <[email protected]>
arm64: perf: Correct the event index in sysfs

Pavel Machek <[email protected]>
btrfs: fix return value mixup in btrfs_get_extent

Josef Bacik <[email protected]>
btrfs: make sure SB_I_VERSION doesn't get unset by remount

Qu Wenruo <[email protected]>
btrfs: trim: fix underflow in trim length to prevent access beyond device boundary

Filipe Manana <[email protected]>
btrfs: fix memory leaks after failure to lookup checksums during inode logging

Qu Wenruo <[email protected]>
btrfs: inode: fix NULL pointer dereference if inode doesn't need compression

Josef Bacik <[email protected]>
btrfs: only search for left_info if there is no right_info in try_merge_free_space

David Sterba <[email protected]>
btrfs: fix messages after changing compression level by remount

Josef Bacik <[email protected]>
btrfs: don't show full path of bind mounts in subvol=

Filipe Manana <[email protected]>
btrfs: fix race between page release and a fast fsync

Josef Bacik <[email protected]>
btrfs: don't WARN if we abort a transaction with EROFS

Josef Bacik <[email protected]>
btrfs: sysfs: use NOFS for device creation

Josef Bacik <[email protected]>
btrfs: return EROFS for BTRFS_FS_STATE_ERROR cases

Qu Wenruo <[email protected]>
btrfs: avoid possible signal interruption of btrfs_drop_snapshot() on relocation tree

David Sterba <[email protected]>
btrfs: add missing check for nocow and compression inode flags

Qu Wenruo <[email protected]>
btrfs: relocation: review the call sites which can be interrupted by signal

Josef Bacik <[email protected]>
btrfs: move the chunk_mutex in btrfs_read_chunk_tree

Josef Bacik <[email protected]>
btrfs: open device without device_list_mutex

Johannes Thumshirn <[email protected]>
btrfs: pass checksum type via BTRFS_IOC_FS_INFO ioctl

Anand Jain <[email protected]>
btrfs: don't traverse into the seed devices in show_devname

Filipe Manana <[email protected]>
btrfs: remove no longer needed use of log_writers for the log root tree

Filipe Manana <[email protected]>
btrfs: only commit delayed items at fsync if we are logging a directory

Filipe Manana <[email protected]>
btrfs: stop incremening log_batch for the log root tree when syncing log

Filipe Manana <[email protected]>
btrfs: only commit the delayed inode when doing a full fsync

Tom Rix <[email protected]>
btrfs: ref-verify: fix memory leak in add_block_entry

Qu Wenruo <[email protected]>
btrfs: don't allocate anonymous block device for user invisible roots

Qu Wenruo <[email protected]>
btrfs: free anon block device right after subvolume deletion

David Sterba <[email protected]>
btrfs: allow use of global block reserve for balance item deletion

Ansuel Smith <[email protected]>
PCI: qcom: Add support for tx term offset for rev 2.1.0

Ansuel Smith <[email protected]>
PCI: qcom: Define some PARF params needed for ipq8064 SoC

Rajat Jain <[email protected]>
PCI: Add device even if driver attach failed

Kai-Heng Feng <[email protected]>

    PCI: Mark AMD Navi10 GPU rev 0x00 ATS as broken

Ashok Raj <[email protected]>
PCI/ATS: Add pci_pri_supported() to check device or associated PF

Rafael J. Wysocki <[email protected]>
PCI: hotplug: ACPI: Fix context refcounting in acpiphp_grab_context()

Guenter Roeck <[email protected]>
genirq/PM: Always unlock IRQ descriptor in rearm_wake_irq()

Thomas Gleixner <[email protected]>
genirq/affinity: Make affinity setting if activated opt-in

Steve French <[email protected]>
smb3: warn on confusing error scenario with sec=krb5

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

Diffstat:

 Documentation/admin-guide/hw-vuln/multihit.rst     |   4 +
 .../bindings/iio/multiplexer/io-channel-mux.txt    |   2 +-
 Makefile                                           |   4 +-
 arch/arm64/kernel/perf_event.c                     |  13 +-
 arch/mips/Kconfig                                  |   1 +
 arch/mips/boot/dts/ingenic/qi_lb60.dts             |   2 +-
 arch/mips/kernel/topology.c                        |   2 +-
 arch/openrisc/kernel/stacktrace.c                  |  18 ++-
 arch/powerpc/include/asm/percpu.h                  |   4 +-
 arch/powerpc/mm/fault.c                            |   7 +-
 arch/powerpc/mm/ptdump/hashpagetable.c             |   2 +-
 arch/powerpc/platforms/pseries/hotplug-memory.c    |   2 +-
 arch/s390/Kconfig                                  |   1 +
 arch/s390/lib/test_unwind.c                        |   1 +
 arch/sh/boards/mach-landisk/setup.c                |   3 +
 arch/x86/events/rapl.c                             |   2 +-
 arch/x86/kernel/apic/vector.c                      |   4 +
 arch/x86/kernel/cpu/bugs.c                         |   8 +-
 arch/x86/kernel/tsc_msr.c                          |   9 +-
 arch/xtensa/include/asm/thread_info.h              |   4 +
 arch/xtensa/kernel/asm-offsets.c                   |   3 +
 arch/xtensa/kernel/entry.S                         |  11 ++
 arch/xtensa/kernel/perf_event.c                    |   2 +-
 crypto/af_alg.c                                    |  11 +-
 crypto/algif_aead.c                                |  10 +-
 crypto/algif_skcipher.c                            |  11 +-
 drivers/base/dd.c                                  |   4 +-
 drivers/clk/actions/owl-s500.c                     |   2 +-
 drivers/clk/bcm/clk-bcm2835.c                      |  25 +++-
 drivers/clk/qcom/clk-alpha-pll.c                   |   2 -
 drivers/clk/qcom/gcc-sdm660.c                      |   3 +
 drivers/clk/qcom/gcc-sm8150.c                      |   8 +-
 drivers/clk/sirf/clk-atlas6.c                      |   2 +-
 drivers/crypto/caam/caamalg.c                      |  29 -----
 drivers/crypto/caam/compat.h                       |   1 -
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c            |   8 +-
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c  |  23 ++++
 .../drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c |  69 +++++++++-
 drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.c   |   5 +-
 drivers/gpu/drm/drm_dp_mst_topology.c              |  35 ++---
 drivers/gpu/drm/drm_panel_orientation_quirks.c     |   6 +
 drivers/gpu/drm/i915/gt/intel_gt.c                 |   5 +
 drivers/gpu/drm/imx/imx-ldb.c                      |   7 +-
 drivers/gpu/drm/ingenic/ingenic-drm.c              |   2 +-
 drivers/gpu/drm/omapdrm/dss/dispc.c                |   1 +
 drivers/gpu/drm/omapdrm/dss/dsi.c                  |   1 +
 drivers/gpu/drm/omapdrm/dss/dss.c                  |   1 +
 drivers/gpu/drm/omapdrm/dss/venc.c                 |   1 +
 drivers/gpu/drm/panfrost/panfrost_gem.c            |   2 +-
 drivers/gpu/drm/panfrost/panfrost_mmu.c            |   2 +-
 drivers/gpu/drm/radeon/ni_dpm.c                    |   2 +-
 drivers/gpu/drm/tidss/tidss_kms.c                  |   2 +-
 drivers/gpu/drm/vmwgfx/vmwgfx_kms.c                |   8 +-
 drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c                |   5 +-
 drivers/gpu/ipu-v3/ipu-image-convert.c             | 145 +++++++++++++--------
 drivers/i2c/busses/i2c-bcm-iproc.c                 |  13 +-
 drivers/i2c/busses/i2c-rcar.c                      |  15 ++-
 drivers/iio/dac/ad5592r-base.c                     |   4 +-
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx.h            |   3 +-
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c     |  23 ++--
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c       |   2 +-
 drivers/infiniband/core/counters.c                 |   4 +-
 drivers/infiniband/core/uverbs_cmd.c               |   4 +
 drivers/infiniband/hw/cxgb4/mem.c                  |   1 -
 drivers/infiniband/hw/mlx4/mr.c                    |   1 -
 drivers/infiniband/ulp/ipoib/ipoib.h               |   2 +-
 drivers/infiniband/ulp/ipoib/ipoib_ib.c            |  67 +++++-----
 drivers/infiniband/ulp/ipoib/ipoib_main.c          |   2 +
 drivers/input/mouse/sentelic.c                     |   2 +-
 drivers/iommu/intel-iommu.c                        |   2 +-
 drivers/iommu/omap-iommu-debug.c                   |   3 +
 drivers/irqchip/irq-gic-v3-its.c                   |  15 ++-
 drivers/irqchip/irq-loongson-liointc.c             |  10 +-
 drivers/md/bcache/bcache.h                         |   2 +-
 drivers/md/bcache/bset.c                           |   2 +-
 drivers/md/bcache/btree.c                          |   2 +-
 drivers/md/bcache/journal.c                        |   4 +-
 drivers/md/bcache/super.c                          |   2 +-
 drivers/md/bcache/writeback.c                      |  14 +-
 drivers/md/bcache/writeback.h                      |  19 ++-
 drivers/md/dm-rq.c                                 |   3 -
 drivers/md/dm.c                                    |   3 +-
 drivers/md/md-cluster.c                            |   1 +
 drivers/md/raid5.c                                 |   3 +-
 drivers/media/platform/qcom/venus/pm_helpers.c     |   4 +
 drivers/media/platform/rockchip/rga/rga-hw.c       |  29 +++--
 drivers/media/platform/rockchip/rga/rga-hw.h       |   5 +
 drivers/media/platform/vsp1/vsp1_dl.c              |   2 +
 drivers/mfd/arizona-core.c                         |  18 +++
 drivers/mfd/dln2.c                                 |   4 +
 drivers/mmc/host/renesas_sdhi_internal_dmac.c      |  18 ++-
 drivers/mtd/nand/raw/brcmnand/brcmnand.c           |  26 ++++
 drivers/mtd/nand/raw/fsl_upm.c                     |   1 -
 drivers/net/ethernet/marvell/octeontx2/af/common.h |   2 +-
 drivers/net/ethernet/qualcomm/emac/emac.c          |  17 ++-
 .../net/ethernet/stmicro/stmmac/dwmac-ipq806x.c    |   1 +
 .../net/ethernet/stmicro/stmmac/dwmac1000_core.c   |   3 +
 drivers/net/wireless/realtek/rtw88/pci.c           |   9 ++
 drivers/nvdimm/security.c                          |  13 +-
 drivers/nvme/host/core.c                           |  15 +++
 drivers/nvme/host/fabrics.c                        |   2 +-
 drivers/nvme/host/fabrics.h                        |   3 +-
 drivers/nvme/host/fc.c                             |   1 +
 drivers/nvme/host/multipath.c                      |  18 ++-
 drivers/nvme/host/nvme.h                           |   1 +
 drivers/nvme/host/rdma.c                           |  10 +-
 drivers/nvme/host/tcp.c                            |  15 ++-
 drivers/pci/ats.c                                  |  15 +++
 drivers/pci/bus.c                                  |   6 +-
 drivers/pci/controller/dwc/pcie-qcom.c             |  41 +++++-
 drivers/pci/hotplug/acpiphp_glue.c                 |  14 +-
 drivers/pci/quirks.c                               |   5 +-
 drivers/pinctrl/pinctrl-ingenic.c                  |   9 +-
 drivers/platform/chrome/cros_ec_ishtp.c            |   4 +-
 drivers/pwm/pwm-bcm-iproc.c                        |   9 +-
 drivers/remoteproc/qcom_q6v5.c                     |   2 +
 drivers/remoteproc/qcom_q6v5_mss.c                 |  11 +-
 drivers/rtc/rtc-cpcap.c                            |   2 +-
 drivers/rtc/rtc-pl031.c                            |   1 +
 drivers/scsi/lpfc/lpfc_nvmet.c                     |   2 +-
 drivers/staging/media/rkisp1/rkisp1-isp.c          |  25 ++--
 drivers/usb/serial/ftdi_sio.c                      |  57 ++++----
 drivers/vdpa/vdpa_sim/vdpa_sim.c                   |   3 +-
 drivers/watchdog/f71808e_wdt.c                     |  13 +-
 drivers/watchdog/rti_wdt.c                         |   2 +
 drivers/watchdog/watchdog_dev.c                    |  18 +--
 fs/btrfs/ctree.h                                   |   4 +-
 fs/btrfs/disk-io.c                                 |  13 +-
 fs/btrfs/extent-io-tree.h                          |   2 +
 fs/btrfs/extent-tree.c                             |  23 +++-
 fs/btrfs/extent_io.c                               |  18 ++-
 fs/btrfs/free-space-cache.c                        |   4 +-
 fs/btrfs/inode.c                                   |  20 ++-
 fs/btrfs/ioctl.c                                   |  46 +++++--
 fs/btrfs/ref-verify.c                              |   2 +
 fs/btrfs/relocation.c                              |  12 +-
 fs/btrfs/scrub.c                                   |   2 +-
 fs/btrfs/super.c                                   |  51 +++++---
 fs/btrfs/sysfs.c                                   |   3 +
 fs/btrfs/transaction.c                             |   5 +-
 fs/btrfs/tree-log.c                                |  43 ++----
 fs/btrfs/volumes.c                                 |  48 ++++++-
 fs/ceph/dir.c                                      |   4 +
 fs/ceph/mds_client.c                               |   6 +-
 fs/cifs/smb2misc.c                                 |  73 ++++++++---
 fs/cifs/smb2pdu.c                                  |   2 +
 fs/ext2/ialloc.c                                   |   3 +-
 fs/f2fs/compress.c                                 |   2 +
 fs/f2fs/data.c                                     |   4 +
 fs/gfs2/bmap.c                                     |  69 +++++-----
 fs/minix/inode.c                                   |  12 +-
 fs/minix/itree_v1.c                                |  12 +-
 fs/minix/itree_v2.c                                |  13 +-
 fs/minix/minix.h                                   |   1 -
 fs/nfs/file.c                                      |  17 ++-
 fs/nfs/flexfilelayout/flexfilelayout.c             |  50 +++++--
 fs/nfs/nfs4file.c                                  |   5 +-
 fs/nfs/nfs4proc.c                                  |   2 -
 fs/nfs/nfs4xdr.c                                   |   6 +-
 fs/nfs/pnfs.c                                      |   4 +-
 fs/nfs/pnfs.h                                      |   2 +-
 fs/ocfs2/ocfs2.h                                   |   4 +-
 fs/ocfs2/suballoc.c                                |   4 +-
 fs/ocfs2/super.c                                   |   4 +-
 fs/ubifs/journal.c                                 |  10 +-
 fs/ufs/super.c                                     |   2 +-
 include/crypto/if_alg.h                            |   4 +-
 include/linux/fs.h                                 |  10 ++
 include/linux/hugetlb.h                            |   8 +-
 include/linux/intel-iommu.h                        |   4 +-
 include/linux/irq.h                                |  13 ++
 include/linux/pci-ats.h                            |   4 +
 include/net/sock.h                                 |   4 +
 include/uapi/linux/btrfs.h                         |  14 +-
 init/main.c                                        |  14 +-
 kernel/irq/manage.c                                |   6 +-
 kernel/irq/pm.c                                    |   8 +-
 kernel/kprobes.c                                   |   7 +
 kernel/module.c                                    |  22 +++-
 kernel/pid.c                                       |   7 +-
 kernel/sched/core.c                                |  81 +++++++++++-
 kernel/sched/cpufreq_schedutil.c                   |   2 +-
 kernel/sched/sched.h                               |  47 ++++++-
 kernel/trace/ftrace.c                              |  15 ++-
 kernel/trace/trace_events.c                        |   4 +-
 kernel/trace/trace_hwlat.c                         |   5 +-
 lib/devres.c                                       |  11 +-
 lib/test_kmod.c                                    |   2 +-
 lib/test_lockup.c                                  |   4 +-
 mm/cma.c                                           |  23 ++--
 mm/hugetlb.c                                       |  39 +++---
 mm/khugepaged.c                                    |  70 +++++-----
 mm/memory_hotplug.c                                |   5 +-
 mm/page_counter.c                                  |   6 +-
 mm/rmap.c                                          |   2 +-
 mm/shuffle.c                                       |  18 +--
 net/compat.c                                       |   1 +
 net/core/sock.c                                    |  21 +++
 net/mac80211/sta_info.c                            |   2 +-
 scripts/recordmcount.c                             |   2 +
 security/integrity/ima/ima_policy.c                |  15 ++-
 sound/pci/echoaudio/echoaudio.c                    |   2 -
 sound/soc/tegra/tegra_alc5632.c                    |   7 +-
 sound/soc/tegra/tegra_asoc_utils.c                 | 113 ++++++++--------
 sound/soc/tegra/tegra_asoc_utils.h                 |   1 -
 sound/soc/tegra/tegra_max98090.c                   |  22 +---
 sound/soc/tegra/tegra_rt5640.c                     |  22 +---
 sound/soc/tegra/tegra_rt5677.c                     |   7 +-
 sound/soc/tegra/tegra_sgtl5000.c                   |   7 +-
 sound/soc/tegra/tegra_wm8753.c                     |  22 +---
 sound/soc/tegra/tegra_wm8903.c                     |  22 +---
 sound/soc/tegra/tegra_wm9712.c                     |   8 +-
 sound/soc/tegra/trimslice.c                        |  18 +--
 tools/build/Makefile.feature                       |   2 +-
 tools/build/feature/Makefile                       |   2 -
 tools/perf/bench/mem-functions.c                   |  21 +--
 .../perf/util/intel-pt-decoder/intel-pt-decoder.c  |  29 ++---
 tools/perf/util/probe-finder.c                     |   5 +-
 tools/testing/selftests/bpf/Makefile               |   5 +-
 tools/testing/selftests/bpf/test_progs.c           |  33 ++++-
 tools/testing/selftests/bpf/test_progs.h           |   2 +
 .../testing/selftests/powerpc/ptrace/ptrace-pkey.c |  55 ++++----
 tools/testing/selftests/seccomp/seccomp_bpf.c      |   5 +
 223 files changed, 1815 insertions(+), 933 deletions(-)
author maintainer
#1 lore raw

PATCH 1/204 smb3: warn on confusing error scenario with sec=krb5

From: Steve French <[email protected]>

commit 0a018944eee913962bce8ffebbb121960d5125d9 upstream.

When mounting with Kerberos, users have been confused about the default error returned in scenarios in which either keyutils is not installed or the user did not properly acquire a krb5 ticket. Log a warning message in the case that "ENOKEY" is returned from the get_spnego_key upcall so that users can better understand why mount failed in those two cases.

---
 fs/cifs/smb2pdu.c |    2 ++
 1 file changed, 2 insertions(+)
fs/cifs/smb2pdu.c +2 -0
@@ -1347,6 +1347,8 @@ SMB2_auth_kerberos(struct SMB2_sess_data
 	spnego_key = cifs_get_spnego_key(ses);
 	if (IS_ERR(spnego_key)) {
 		rc = PTR_ERR(spnego_key);
+		if (rc == -ENOKEY)
+			cifs_dbg(VFS, "Verify user has a krb5 ticket and keyutils is installed\n");
 		spnego_key = NULL;
 		goto out;
 	}
author maintainer
#2 lore raw

PATCH 2/204 genirq/affinity: Make affinity setting if activated opt-in

In reply to Greg Kroah-Hartman

From: Thomas Gleixner <[email protected]>

commit f0c7baca180046824e07fc5f1326e83a8fd150c7 upstream.

John reported that on a RK3288 system the perf per CPU interrupts are all affine to CPU0 and provided the analysis:

"It looks like what happens is that because the interrupts are not per-CPU
in the hardware, armpmu_request_irq() calls irq_force_affinity() while the interrupt is deactivated and then request_irq() with IRQF_PERCPU | IRQF_NOBALANCING.

Now when irq_startup() runs with IRQ_STARTUP_NORMAL, it calls irq_setup_affinity() which returns early because IRQF_PERCPU and IRQF_NOBALANCING are set, leaving the interrupt on its original CPU."

This was broken by the recent commit which blocked interrupt affinity setting in hardware before activation of the interrupt. While this works in general, it does not work for this particular case. As contrary to the initial analysis not all interrupt chip drivers implement an activate callback, the safe cure is to make the deferred interrupt affinity setting at activation time opt-in.

Implement the necessary core logic and make the two irqchip implementations for which this is required opt-in. In hindsight this would have been the right thing to do, but ...

Fixes baedb87d1b53 ("genirq/affinity: Handle affinity setting on inactive interrupts correctly")
---
 arch/x86/kernel/apic/vector.c    |    4 ++++
 drivers/irqchip/irq-gic-v3-its.c |    5 ++++-
 include/linux/irq.h              |   13 +++++++++++++
 kernel/irq/manage.c              |    6 +++++-
 4 files changed, 26 insertions(+), 2 deletions(-)
arch/x86/kernel/apic/vector.c +4 -0
@@ -560,6 +560,10 @@ static int x86_vector_alloc_irqs(struct
 		 * as that can corrupt the affinity move state.
 		 */
 		irqd_set_handle_enforce_irqctx(irqd);
+
+		/* Don't invoke affinity setter on deactivated interrupts */
+		irqd_set_affinity_on_activate(irqd);
+
 		/*
 		 * Legacy vectors are already assigned when the IOAPIC
 		 * takes them over. They stay on the same vector. This is
drivers/irqchip/irq-gic-v3-its.c +4 -1
@@ -3399,6 +3399,7 @@ static int its_irq_domain_alloc(struct i
 	msi_alloc_info_t *info = args;
 	struct its_device *its_dev = info->scratchpad[0].ptr;
 	struct its_node *its = its_dev->its;
+	struct irq_data *irqd;
 	irq_hw_number_t hwirq;
 	int err;
 	int i;
@@ -3418,7 +3419,9 @@ static int its_irq_domain_alloc(struct i
 
 		irq_domain_set_hwirq_and_chip(domain, virq + i,
 					      hwirq + i, &its_irq_chip, its_dev);
-		irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(virq + i)));
+		irqd = irq_get_irq_data(virq + i);
+		irqd_set_single_target(irqd);
+		irqd_set_affinity_on_activate(irqd);
 		pr_debug("ID:%d pID:%d vID:%d\n",
 			 (int)(hwirq + i - its_dev->event_map.lpi_base),
 			 (int)(hwirq + i), virq + i);
include/linux/irq.h +13 -0
@@ -213,6 +213,8 @@ struct irq_data {
  *				  required
  * IRQD_HANDLE_ENFORCE_IRQCTX	- Enforce that handle_irq_*() is only invoked
  *				  from actual interrupt context.
+ * IRQD_AFFINITY_ON_ACTIVATE	- Affinity is set on activation. Don't call
+ *				  irq_chip::irq_set_affinity() when deactivated.
  */
 enum {
 	IRQD_TRIGGER_MASK		= 0xf,
@@ -237,6 +239,7 @@ enum {
 	IRQD_CAN_RESERVE		= (1 << 26),
 	IRQD_MSI_NOMASK_QUIRK		= (1 << 27),
 	IRQD_HANDLE_ENFORCE_IRQCTX	= (1 << 28),
+	IRQD_AFFINITY_ON_ACTIVATE	= (1 << 29),
 };
 
 #define __irqd_to_state(d) ACCESS_PRIVATE((d)->common, state_use_accessors)
@@ -421,6 +424,16 @@ static inline bool irqd_msi_nomask_quirk
 	return __irqd_to_state(d) & IRQD_MSI_NOMASK_QUIRK;
 }
 
+static inline void irqd_set_affinity_on_activate(struct irq_data *d)
+{
+	__irqd_to_state(d) |= IRQD_AFFINITY_ON_ACTIVATE;
+}
+
+static inline bool irqd_affinity_on_activate(struct irq_data *d)
+{
+	return __irqd_to_state(d) & IRQD_AFFINITY_ON_ACTIVATE;
+}
+
 #undef __irqd_to_state
 
 static inline irq_hw_number_t irqd_to_hwirq(struct irq_data *d)
kernel/irq/manage.c +5 -1
@@ -320,12 +320,16 @@ static bool irq_set_affinity_deactivated
 	struct irq_desc *desc = irq_data_to_desc(data);
 
 	/*
+	 * Handle irq chips which can handle affinity only in activated
+	 * state correctly
+	 *
 	 * If the interrupt is not yet activated, just store the affinity
 	 * mask and do not call the chip driver at all. On activation the
 	 * driver has to make sure anyway that the interrupt is in a
 	 * useable state so startup works.
 	 */
-	if (!IS_ENABLED(CONFIG_IRQ_DOMAIN_HIERARCHY) || irqd_is_activated(data))
+	if (!IS_ENABLED(CONFIG_IRQ_DOMAIN_HIERARCHY) ||
+	    irqd_is_activated(data) || !irqd_affinity_on_activate(data))
 		return false;
 
 	cpumask_copy(desc->irq_common_data.affinity, mask);
author maintainer
#3 lore raw

PATCH 3/204 genirq/PM: Always unlock IRQ descriptor in rearm_wake_irq()

In reply to Greg Kroah-Hartman

From: Guenter Roeck <[email protected]>

commit e27b1636e9337d1a1d174b191e53d0f86421a822 upstream.

rearm_wake_irq() does not unlock the irq descriptor if the interrupt is not suspended or if wakeup is not enabled on it.

Restucture the exit conditions so the unlock is always ensured.

---
 kernel/irq/pm.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
kernel/irq/pm.c +6 -2
@@ -185,14 +185,18 @@ void rearm_wake_irq(unsigned int irq)
 	unsigned long flags;
 	struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, IRQ_GET_DESC_CHECK_GLOBAL);
 
-	if (!desc || !(desc->istate & IRQS_SUSPENDED) ||
-	    !irqd_is_wakeup_set(&desc->irq_data))
+	if (!desc)
 		return;
 
+	if (!(desc->istate & IRQS_SUSPENDED) ||
+	    !irqd_is_wakeup_set(&desc->irq_data))
+		goto unlock;
+
 	desc->istate &= ~IRQS_SUSPENDED;
 	irqd_set(&desc->irq_data, IRQD_WAKEUP_ARMED);
 	__enable_irq(desc);
 
+unlock:
 	irq_put_desc_busunlock(desc, flags);
 }
 
author maintainer
#4 lore raw

PATCH 4/204 PCI: hotplug: ACPI: Fix context refcounting in acpiphp_grab_context()

In reply to Greg Kroah-Hartman

From: Rafael J. Wysocki <[email protected]>

commit dae68d7fd4930315389117e9da35b763f12238f9 upstream.

If context is not NULL in acpiphp_grab_context(), but the is_going_away flag is set for the device's parent, the reference counter of the context needs to be decremented before returning NULL or the context will never be freed, so make that happen.

Fixes edf5bf34d408 ("ACPI / dock: Use callback pointers from devices' ACPI hotplug contexts")
Cc 3.15+ <[email protected]> # 3.15+
---
 drivers/pci/hotplug/acpiphp_glue.c |   14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)
drivers/pci/hotplug/acpiphp_glue.c +11 -3
@@ -122,13 +122,21 @@ static struct acpiphp_context *acpiphp_g
 	struct acpiphp_context *context;
 
 	acpi_lock_hp_context();
+
 	context = acpiphp_get_context(adev);
-	if (!context || context->func.parent->is_going_away) {
-		acpi_unlock_hp_context();
-		return NULL;
+	if (!context)
+		goto unlock;
+
+	if (context->func.parent->is_going_away) {
+		acpiphp_put_context(context);
+		context = NULL;
+		goto unlock;
 	}
+
 	get_bridge(context->func.parent);
 	acpiphp_put_context(context);
+
+unlock:
 	acpi_unlock_hp_context();
 	return context;
 }
author maintainer
#5 lore raw

PATCH 5/204 PCI/ATS: Add pci_pri_supported() to check device or associated PF

In reply to Greg Kroah-Hartman

From: Ashok Raj <[email protected]>

commit 3f9a7a13fe4cb6e119e4e4745fbf975d30bfac9b upstream.

For SR-IOV, the PF PRI is shared between the PF and any associated VFs, and the PRI Capability is allowed for PFs but not for VFs. Searching for the PRI Capability on a VF always fails, even if its associated PF supports PRI.

Add pci_pri_supported() to check whether device or its associated PF supports PRI.

[bhelgaas: commit log, avoid "!!"]
Fixes b16d0cb9e2fc ("iommu/vt-d: Always enable PASID/PRI PCI capabilities before ATS")
---
 drivers/iommu/intel-iommu.c |    2 +-
 drivers/pci/ats.c           |   15 +++++++++++++++
 include/linux/pci-ats.h     |    4 ++++
 3 files changed, 20 insertions(+), 1 deletion(-)
drivers/iommu/intel-iommu.c +1 -1
@@ -2645,7 +2645,7 @@ static struct dmar_domain *dmar_insert_o
 			}
 
 			if (info->ats_supported && ecap_prs(iommu->ecap) &&
-			    pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI))
+			    pci_pri_supported(pdev))
 				info->pri_supported = 1;
 		}
 	}
drivers/pci/ats.c +15 -0
@@ -309,6 +309,21 @@ int pci_prg_resp_pasid_required(struct p
 
 	return pdev->pasid_required;
 }
+
+/**
+ * pci_pri_supported - Check if PRI is supported.
+ * @pdev: PCI device structure
+ *
+ * Returns true if PRI capability is present, false otherwise.
+ */
+bool pci_pri_supported(struct pci_dev *pdev)
+{
+	/* VFs share the PF PRI */
+	if (pci_physfn(pdev)->pri_cap)
+		return true;
+	return false;
+}
+EXPORT_SYMBOL_GPL(pci_pri_supported);
 #endif /* CONFIG_PCI_PRI */
 
 #ifdef CONFIG_PCI_PASID
include/linux/pci-ats.h +4 -0
@@ -25,6 +25,10 @@ int pci_enable_pri(struct pci_dev *pdev,
 void pci_disable_pri(struct pci_dev *pdev);
 int pci_reset_pri(struct pci_dev *pdev);
 int pci_prg_resp_pasid_required(struct pci_dev *pdev);
+bool pci_pri_supported(struct pci_dev *pdev);
+#else
+static inline bool pci_pri_supported(struct pci_dev *pdev)
+{ return false; }
 #endif /* CONFIG_PCI_PRI */
 
 #ifdef CONFIG_PCI_PASID
author maintainer
#6 lore raw

PATCH 6/204 PCI: Mark AMD Navi10 GPU rev 0x00 ATS as broken

In reply to Greg Kroah-Hartman

From: Kai-Heng Feng <[email protected]>

commit 45beb31d3afb651bb5c41897e46bd4fa9980c51c upstream.

We are seeing AMD Radeon Pro W5700 doesn't work when IOMMU is enabled:

iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=63:00.0 address=0x42b5b01a0] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=63:00.0 address=0x42b5b01c0]

The error also makes graphics driver fail to probe the device.

It appears to be the same issue as commit 5e89cd303e3a ("PCI: Mark AMD Navi14 GPU rev 0xc5 ATS as broken") addresses, and indeed the same ATS quirk can workaround the issue.

See-also: 5e89cd303e3a ("PCI: Mark AMD Navi14 GPU rev 0xc5 ATS as broken")
See-also: d28ca864c493 ("PCI: Mark AMD Stoney Radeon R7 GPU ATS as broken")
See-also: 9b44b0b09dec ("PCI: Mark AMD Stoney GPU ATS as broken")
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=208725

---
 drivers/pci/quirks.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/pci/quirks.c +4 -1
@@ -5207,7 +5207,8 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SE
  */
 static void quirk_amd_harvest_no_ats(struct pci_dev *pdev)
 {
-	if (pdev->device == 0x7340 && pdev->revision != 0xc5)
+	if ((pdev->device == 0x7312 && pdev->revision != 0x00) ||
+	    (pdev->device == 0x7340 && pdev->revision != 0xc5))
 		return;
 
 	pci_info(pdev, "disabling ATS\n");
@@ -5218,6 +5219,8 @@ static void quirk_amd_harvest_no_ats(str
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x98e4, quirk_amd_harvest_no_ats);
 /* AMD Iceland dGPU */
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x6900, quirk_amd_harvest_no_ats);
+/* AMD Navi10 dGPU */
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x7312, quirk_amd_harvest_no_ats);
 /* AMD Navi14 dGPU */
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x7340, quirk_amd_harvest_no_ats);
 #endif /* CONFIG_PCI_ATS */
author maintainer
#7 lore raw

PATCH 7/204 PCI: Add device even if driver attach failed

In reply to Greg Kroah-Hartman

From: Rajat Jain <[email protected]>

commit 2194bc7c39610be7cabe7456c5f63a570604f015 upstream.

device_attach() returning failure indicates a driver error while trying to probe the device. In such a scenario, the PCI device should still be added in the system and be visible to the user.

When device_attach() fails, merely warn about it and keep the PCI device in the system.

This partially reverts ab1a187bba5c ("PCI: Check device_attach() return value always").

---
 drivers/pci/bus.c |    6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)
drivers/pci/bus.c +1 -5
@@ -322,12 +322,8 @@ void pci_bus_add_device(struct pci_dev *
 
 	dev->match_driver = true;
 	retval = device_attach(&dev->dev);
-	if (retval < 0 && retval != -EPROBE_DEFER) {
+	if (retval < 0 && retval != -EPROBE_DEFER)
 		pci_warn(dev, "device attach failed (%d)\n", retval);
-		pci_proc_detach_device(dev);
-		pci_remove_sysfs_dev_files(dev);
-		return;
-	}
 
 	pci_dev_assign_added(dev, true);
 }
author maintainer
#8 lore raw

PATCH 8/204 PCI: qcom: Define some PARF params needed for ipq8064 SoC

In reply to Greg Kroah-Hartman

From: Ansuel Smith <[email protected]>

commit 5149901e9e6deca487c01cc434a3ac4125c7b00b upstream.

Set some specific value for Tx De-Emphasis, Tx Swing and Rx equalization needed on some ipq8064 based device (Netgear R7800 for example). Without this the system locks on kernel load.

---
 drivers/pci/controller/dwc/pcie-qcom.c |   24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
drivers/pci/controller/dwc/pcie-qcom.c +24 -0
@@ -77,6 +77,18 @@
 #define DBI_RO_WR_EN				1
 
 #define PERST_DELAY_US				1000
+/* PARF registers */
+#define PCIE20_PARF_PCS_DEEMPH			0x34
+#define PCS_DEEMPH_TX_DEEMPH_GEN1(x)		((x) << 16)
+#define PCS_DEEMPH_TX_DEEMPH_GEN2_3_5DB(x)	((x) << 8)
+#define PCS_DEEMPH_TX_DEEMPH_GEN2_6DB(x)	((x) << 0)
+
+#define PCIE20_PARF_PCS_SWING			0x38
+#define PCS_SWING_TX_SWING_FULL(x)		((x) << 8)
+#define PCS_SWING_TX_SWING_LOW(x)		((x) << 0)
+
+#define PCIE20_PARF_CONFIG_BITS		0x50
+#define PHY_RX0_EQ(x)				((x) << 24)
 
 #define PCIE20_v3_PARF_SLV_ADDR_SPACE_SIZE	0x358
 #define SLV_ADDR_SPACE_SZ			0x10000000
@@ -286,6 +298,7 @@ static int qcom_pcie_init_2_1_0(struct q
 	struct qcom_pcie_resources_2_1_0 *res = &pcie->res.v2_1_0;
 	struct dw_pcie *pci = pcie->pci;
 	struct device *dev = pci->dev;
+	struct device_node *node = dev->of_node;
 	u32 val;
 	int ret;
 
@@ -330,6 +343,17 @@ static int qcom_pcie_init_2_1_0(struct q
 	val &= ~BIT(0);
 	writel(val, pcie->parf + PCIE20_PARF_PHY_CTRL);
 
+	if (of_device_is_compatible(node, "qcom,pcie-ipq8064")) {
+		writel(PCS_DEEMPH_TX_DEEMPH_GEN1(24) |
+			       PCS_DEEMPH_TX_DEEMPH_GEN2_3_5DB(24) |
+			       PCS_DEEMPH_TX_DEEMPH_GEN2_6DB(34),
+		       pcie->parf + PCIE20_PARF_PCS_DEEMPH);
+		writel(PCS_SWING_TX_SWING_FULL(120) |
+			       PCS_SWING_TX_SWING_LOW(120),
+		       pcie->parf + PCIE20_PARF_PCS_SWING);
+		writel(PHY_RX0_EQ(4), pcie->parf + PCIE20_PARF_CONFIG_BITS);
+	}
+
 	/* enable external reference clock */
 	val = readl(pcie->parf + PCIE20_PARF_PHY_REFCLK);
 	val |= BIT(16);
author maintainer
#9 lore raw

PATCH 9/204 PCI: qcom: Add support for tx term offset for rev 2.1.0

In reply to Greg Kroah-Hartman

From: Ansuel Smith <[email protected]>

commit de3c4bf648975ea0b1d344d811e9b0748907b47c upstream.

Add tx term offset support to pcie qcom driver need in some revision of the ipq806x SoC. Ipq8064 needs tx term offset set to 7.

---
 drivers/pci/controller/dwc/pcie-qcom.c |   17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)
drivers/pci/controller/dwc/pcie-qcom.c +16 -1
@@ -45,7 +45,13 @@
 #define PCIE_CAP_CPL_TIMEOUT_DISABLE		0x10
 
 #define PCIE20_PARF_PHY_CTRL			0x40
+#define PHY_CTRL_PHY_TX0_TERM_OFFSET_MASK	GENMASK(20, 16)
+#define PHY_CTRL_PHY_TX0_TERM_OFFSET(x)		((x) << 16)
+
 #define PCIE20_PARF_PHY_REFCLK			0x4C
+#define PHY_REFCLK_SSP_EN			BIT(16)
+#define PHY_REFCLK_USE_PAD			BIT(12)
+
 #define PCIE20_PARF_DBI_BASE_ADDR		0x168
 #define PCIE20_PARF_SLV_ADDR_SPACE_SIZE		0x16C
 #define PCIE20_PARF_MHI_CLOCK_RESET_CTRL	0x174
@@ -354,9 +360,18 @@ static int qcom_pcie_init_2_1_0(struct q
 		writel(PHY_RX0_EQ(4), pcie->parf + PCIE20_PARF_CONFIG_BITS);
 	}
 
+	if (of_device_is_compatible(node, "qcom,pcie-ipq8064")) {
+		/* set TX termination offset */
+		val = readl(pcie->parf + PCIE20_PARF_PHY_CTRL);
+		val &= ~PHY_CTRL_PHY_TX0_TERM_OFFSET_MASK;
+		val |= PHY_CTRL_PHY_TX0_TERM_OFFSET(7);
+		writel(val, pcie->parf + PCIE20_PARF_PHY_CTRL);
+	}
+
 	/* enable external reference clock */
 	val = readl(pcie->parf + PCIE20_PARF_PHY_REFCLK);
-	val |= BIT(16);
+	val &= ~PHY_REFCLK_USE_PAD;
+	val |= PHY_REFCLK_SSP_EN;
 	writel(val, pcie->parf + PCIE20_PARF_PHY_REFCLK);
 
 	ret = reset_control_deassert(res->phy_reset);
author maintainer
#10 lore raw

PATCH 10/204 btrfs: allow use of global block reserve for balance item deletion

In reply to Greg Kroah-Hartman

From: David Sterba <[email protected]>

commit 3502a8c0dc1bd4b4970b59b06e348f22a1c05581 upstream.

On a filesystem with exhausted metadata, but still enough to start balance, it's possible to hit this error:

[324402.053842] BTRFS info (device loop0): 1 enospc errors during balance
[324402.060769] BTRFS info (device loop0): balance: ended with status: -28
[324402.172295] BTRFS: error (device loop0) in reset_balance_state:3321: errno=-28 No space left

It fails inside reset_balance_state and turns the filesystem to read-only, which is unnecessary and should be fixed too, but the problem is caused by lack for space when the balance item is deleted. This is a one-time operation and from the same rank as unlink that is allowed to use the global block reserve. So do the same for the balance item.

Status of the filesystem (100GiB) just after the balance fails:

$ btrfs fi df mnt
Data, single: total=80.01GiB, used=38.58GiB
System, single: total=4.00MiB, used=16.00KiB
Metadata, single: total=19.99GiB, used=19.48GiB
GlobalReserve, single: total=512.00MiB, used=50.11MiB
---
 fs/btrfs/volumes.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/volumes.c +1 -1
@@ -3235,7 +3235,7 @@ static int del_balance_item(struct btrfs
 	if (!path)
 		return -ENOMEM;
 
-	trans = btrfs_start_transaction(root, 0);
+	trans = btrfs_start_transaction_fallback_global_rsv(root, 0);
 	if (IS_ERR(trans)) {
 		btrfs_free_path(path);
 		return PTR_ERR(trans);
author maintainer
#11 lore raw

PATCH 11/204 btrfs: free anon block device right after subvolume deletion

In reply to Greg Kroah-Hartman

From: Qu Wenruo <[email protected]>

commit 082b6c970f02fefd278c7833880cda29691a5f34 upstream.

[BUG]
When a lot of subvolumes are created, there is a user report about transaction aborted caused by slow anonymous block device reclaim:

BTRFS: Transaction aborted (error -24)

  WARNING: CPU: 17 PID: 17041 at fs/btrfs/transaction.c:1576 create_pending_snapshot+0xbc4/0xd10 [btrfs]
  RIP: 0010:create_pending_snapshot+0xbc4/0xd10 [btrfs]
  Call Trace:
   create_pending_snapshots+0x82/0xa0 [btrfs]
   btrfs_commit_transaction+0x275/0x8c0 [btrfs]
   btrfs_mksubvol+0x4b9/0x500 [btrfs]
   btrfs_ioctl_snap_create_transid+0x174/0x180 [btrfs]
   btrfs_ioctl_snap_create_v2+0x11c/0x180 [btrfs]
   btrfs_ioctl+0x11a4/0x2da0 [btrfs]
   do_vfs_ioctl+0xa9/0x640
   ksys_ioctl+0x67/0x90
   __x64_sys_ioctl+0x1a/0x20
   do_syscall_64+0x5a/0x110
   entry_SYSCALL_64_after_hwframe+0x44/0xa9
  ---[ end trace 33f2f83f3d5250e9 ]---
  BTRFS: error (device sda1) in create_pending_snapshot:1576: errno=-24 unknown
  BTRFS info (device sda1): forced readonly
  BTRFS warning (device sda1): Skipping commit of aborted transaction.
  BTRFS: error (device sda1) in cleanup_transaction:1831: errno=-24 unknown

[CAUSE]
The anonymous device pool is shared and its size is 1M. It's possible to hit that limit if the subvolume deletion is not fast enough and the subvolumes to be cleaned keep the ids allocated.

[WORKAROUND]
We can't avoid the anon device pool exhaustion but we can shorten the time the id is attached to the subvolume root once the subvolume becomes invisible to the user.

---
 fs/btrfs/inode.c |    2 ++
 1 file changed, 2 insertions(+)
fs/btrfs/inode.c +2 -0
@@ -4049,6 +4049,8 @@ int btrfs_delete_subvolume(struct inode
 		}
 	}
 
+	free_anon_bdev(dest->anon_dev);
+	dest->anon_dev = 0;
 out_end_trans:
 	trans->block_rsv = NULL;
 	trans->bytes_reserved = 0;
author maintainer
#12 lore raw

PATCH 12/204 btrfs: dont allocate anonymous block device for user invisible roots

In reply to Greg Kroah-Hartman

From: Qu Wenruo <[email protected]>

commit 851fd730a743e072badaf67caf39883e32439431 upstream.

[BUG]
When a lot of subvolumes are created, there is a user report about transaction aborted:

BTRFS: Transaction aborted (error -24)

  WARNING: CPU: 17 PID: 17041 at fs/btrfs/transaction.c:1576 create_pending_snapshot+0xbc4/0xd10 [btrfs]
  RIP: 0010:create_pending_snapshot+0xbc4/0xd10 [btrfs]
  Call Trace:
   create_pending_snapshots+0x82/0xa0 [btrfs]
   btrfs_commit_transaction+0x275/0x8c0 [btrfs]
   btrfs_mksubvol+0x4b9/0x500 [btrfs]
   btrfs_ioctl_snap_create_transid+0x174/0x180 [btrfs]
   btrfs_ioctl_snap_create_v2+0x11c/0x180 [btrfs]
   btrfs_ioctl+0x11a4/0x2da0 [btrfs]
   do_vfs_ioctl+0xa9/0x640
   ksys_ioctl+0x67/0x90
   __x64_sys_ioctl+0x1a/0x20
   do_syscall_64+0x5a/0x110
   entry_SYSCALL_64_after_hwframe+0x44/0xa9
  ---[ end trace 33f2f83f3d5250e9 ]---
  BTRFS: error (device sda1) in create_pending_snapshot:1576: errno=-24 unknown
  BTRFS info (device sda1): forced readonly
  BTRFS warning (device sda1): Skipping commit of aborted transaction.
  BTRFS: error (device sda1) in cleanup_transaction:1831: errno=-24 unknown

[CAUSE]
The error is EMFILE (Too many files open) and comes from the anonymous block device allocation. The ids are in a shared pool of size 1<<20.

The ids are assigned to live subvolumes, ie. the root structure exists in memory (eg. after creation or after the root appears in some path). The pool could be exhausted if the numbers are not reclaimed fast enough, after subvolume deletion or if other system component uses the anon block devices.

[WORKAROUND]
Since it's not possible to completely solve the problem, we can only minimize the time the id is allocated to a subvolume root.

Firstly, we can reduce the use of anon_dev by trees that are not subvolume roots, like data reloc tree.

This patch will do extra check on root objectid, to skip roots that don't need anon_dev. Currently it's only data reloc tree and orphan roots.

---
 fs/btrfs/disk-io.c |   13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)
fs/btrfs/disk-io.c +10 -3
@@ -1432,9 +1432,16 @@ static int btrfs_init_fs_root(struct btr
 	spin_lock_init(&root->ino_cache_lock);
 	init_waitqueue_head(&root->ino_cache_wait);
 
-	ret = get_anon_bdev(&root->anon_dev);
-	if (ret)
-		goto fail;
+	/*
+	 * Don't assign anonymous block device to roots that are not exposed to
+	 * userspace, the id pool is limited to 1M
+	 */
+	if (is_fstree(root->root_key.objectid) &&
+	    btrfs_root_refs(&root->root_item) > 0) {
+		ret = get_anon_bdev(&root->anon_dev);
+		if (ret)
+			goto fail;
+	}
 
 	mutex_lock(&root->objectid_mutex);
 	ret = btrfs_find_highest_objectid(root,
author maintainer
#13 lore raw

PATCH 13/204 btrfs: ref-verify: fix memory leak in add_block_entry

In reply to Greg Kroah-Hartman

From: Tom Rix <[email protected]>

commit d60ba8de1164e1b42e296ff270c622a070ef8fe7 upstream.

clang static analysis flags this error

fs/btrfs/ref-verify.c:290:3: warning: Potential leak of memory pointed to by 're' [unix.Malloc]

                kfree(be);
                ^~~~~

The problem is in this block of code:

	if (root_objectid) {
		struct root_entry *exist_re;

		exist_re = insert_root_entry(&exist->roots, re);
		if (exist_re)
			kfree(re);
	}

There is no 'else' block freeing when root_objectid is 0. Add the missing kfree to the else branch.

Fixes fd708b81d972 ("Btrfs: add a extent ref verify tool")
---
 fs/btrfs/ref-verify.c |    2 ++
 1 file changed, 2 insertions(+)
fs/btrfs/ref-verify.c +2 -0
@@ -286,6 +286,8 @@ static struct block_entry *add_block_ent
 			exist_re = insert_root_entry(&exist->roots, re);
 			if (exist_re)
 				kfree(re);
+		} else {
+			kfree(re);
 		}
 		kfree(be);
 		return exist;
author maintainer
#14 lore raw

PATCH 14/204 btrfs: only commit the delayed inode when doing a full fsync

In reply to Greg Kroah-Hartman

From: Filipe Manana <[email protected]>

commit 8c8648dd1f6d62aeb912deeb788b6ac33cb782e7 upstream.

Commit 2c2c452b0cafdc ("Btrfs: fix fsync when extend references are added to an inode") forced a commit of the delayed inode when logging an inode in order to ensure we would end up logging the inode item during a full fsync. By committing the delayed inode, we updated the inode item in the fs/subvolume tree and then later when copying items from leafs modified in the current transaction into the log tree (with copy_inode_items_to_log()) we ended up copying the inode item from the fs/subvolume tree into the log tree. Logging an up to date version of the inode item is required to make sure at log replay time we get the link count fixup triggered among other things (replay xattr deletes, etc). The test case generic/040 from fstests exercises the bug which that commit fixed.

However for a fast fsync we don't need to commit the delayed inode because we always log an up to date version of the inode item based on the struct btrfs_inode we have in-memory. We started doing this for fast fsyncs since commit e4545de5b035c7 ("Btrfs: fix fsync data loss after append write").

So just stop committing the delayed inode if we are doing a fast fsync, we are only wasting time and adding contention on fs/subvolume tree.

This patch is part of a series that has the following patches:

1/4 btrfs: only commit the delayed inode when doing a full fsync 2/4 btrfs: only commit delayed items at fsync if we are logging a directory 3/4 btrfs: stop incremening log_batch for the log root tree when syncing log 4/4 btrfs: remove no longer needed use of log_writers for the log root tree

After the entire patchset applied I saw about 12% decrease on max latency reported by dbench. The test was done on a qemu vm, with 8 cores, 16Gb of ram, using kvm and using a raw NVMe device directly (no intermediary fs on the host). The test was invoked like the following:

mkfs.btrfs -f /dev/sdk
mount -o ssd -o nospace_cache /dev/sdk /mnt/sdk
dbench -D /mnt/sdk -t 300 8
umount /mnt/dsk

---
 fs/btrfs/tree-log.c |   12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
fs/btrfs/tree-log.c +7 -5
@@ -5158,7 +5158,7 @@ static int btrfs_log_inode(struct btrfs_
 	struct btrfs_key max_key;
 	struct btrfs_root *log = root->log_root;
 	int err = 0;
-	int ret;
+	int ret = 0;
 	bool fast_search = false;
 	u64 ino = btrfs_ino(inode);
 	struct extent_map_tree *em_tree = &inode->extent_tree;
@@ -5195,14 +5195,16 @@ static int btrfs_log_inode(struct btrfs_
 
 	/*
 	 * Only run delayed items if we are a dir or a new file.
-	 * Otherwise commit the delayed inode only, which is needed in
-	 * order for the log replay code to mark inodes for link count
-	 * fixup (create temporary BTRFS_TREE_LOG_FIXUP_OBJECTID items).
+	 * Otherwise commit the delayed inode only if the full sync flag is set,
+	 * as we want to make sure an up to date version is in the subvolume
+	 * tree so copy_inode_items_to_log() / copy_items() can find it and copy
+	 * it to the log tree. For a non full sync, we always log the inode item
+	 * based on the in-memory struct btrfs_inode which is always up to date.
 	 */
 	if (S_ISDIR(inode->vfs_inode.i_mode) ||
 	    inode->generation > fs_info->last_trans_committed)
 		ret = btrfs_commit_inode_delayed_items(trans, inode);
-	else
+	else if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags))
 		ret = btrfs_commit_inode_delayed_inode(inode);
 
 	if (ret) {
author maintainer
#15 lore raw

PATCH 15/204 btrfs: stop incremening log_batch for the log root tree when syncing log

In reply to Greg Kroah-Hartman

From: Filipe Manana <[email protected]>

commit 28a9579561bcb9082715e720eac93012e708ab94 upstream.

We are incrementing the log_batch atomic counter of the root log tree but we never use that counter, it's used only for the log trees of subvolume roots. We started doing it when we moved the log_batch and log_write counters from the global, per fs, btrfs_fs_info structure, into the btrfs_root structure in commit 7237f1833601dc ("Btrfs: fix tree logs parallel sync").

So just stop doing it for the log root tree and add a comment over the field declaration so inform it's used only for log trees of subvolume roots.

This patch is part of a series that has the following patches:

1/4 btrfs: only commit the delayed inode when doing a full fsync 2/4 btrfs: only commit delayed items at fsync if we are logging a directory 3/4 btrfs: stop incremening log_batch for the log root tree when syncing log 4/4 btrfs: remove no longer needed use of log_writers for the log root tree

After the entire patchset applied I saw about 12% decrease on max latency reported by dbench. The test was done on a qemu vm, with 8 cores, 16Gb of ram, using kvm and using a raw NVMe device directly (no intermediary fs on the host). The test was invoked like the following:

mkfs.btrfs -f /dev/sdk
mount -o ssd -o nospace_cache /dev/sdk /mnt/sdk
dbench -D /mnt/sdk -t 300 8
umount /mnt/dsk

---
 fs/btrfs/ctree.h    |    1 +
 fs/btrfs/tree-log.c |    1 -
 2 files changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/ctree.h +1 -0
@@ -1040,6 +1040,7 @@ struct btrfs_root {
 	struct list_head log_ctxs[2];
 	atomic_t log_writers;
 	atomic_t log_commit[2];
+	/* Used only for log trees of subvolumes, not for the log root tree */
 	atomic_t log_batch;
 	int log_transid;
 	/* No matter the commit succeeds or not*/
fs/btrfs/tree-log.c +0 -1
@@ -3125,7 +3125,6 @@ int btrfs_sync_log(struct btrfs_trans_ha
 	btrfs_init_log_ctx(&root_log_ctx, NULL);
 
 	mutex_lock(&log_root_tree->log_mutex);
-	atomic_inc(&log_root_tree->log_batch);
 	atomic_inc(&log_root_tree->log_writers);
 
 	index2 = log_root_tree->log_transid % 2;
author maintainer
#16 lore raw

PATCH 16/204 btrfs: only commit delayed items at fsync if we are logging a directory

In reply to Greg Kroah-Hartman

From: Filipe Manana <[email protected]>

commit 5aa7d1a7f4a2f8ca6be1f32415e9365d026e8fa7 upstream.

When logging an inode we are committing its delayed items if either the inode is a directory or if it is a new inode, created in the current transaction.

We need to do it for directories, since new directory indexes are stored as delayed items of the inode and when logging a directory we need to be able to access all indexes from the fs/subvolume tree in order to figure out which index ranges need to be logged.

However for new inodes that are not directories, we do not need to do it because the only type of delayed item they can have is the inode item, and we are guaranteed to always log an up to date version of the inode item:

*) for a full fsync we do it by committing the delayed inode and then
copying the item from the fs/subvolume tree with copy_inode_items_to_log();

*) for a fast fsync we always log the inode item based on the contents of
the in-memory struct btrfs_inode. We guarantee this is always done since commit e4545de5b035c7 ("Btrfs: fix fsync data loss after append write").

So stop running delayed items for a new inodes that are not directories, since that forces committing the delayed inode into the fs/subvolume tree, wasting time and adding contention to the tree when a full fsync is not required. We will only do it in case a fast fsync is needed.

This patch is part of a series that has the following patches:

1/4 btrfs: only commit the delayed inode when doing a full fsync 2/4 btrfs: only commit delayed items at fsync if we are logging a directory 3/4 btrfs: stop incremening log_batch for the log root tree when syncing log 4/4 btrfs: remove no longer needed use of log_writers for the log root tree

After the entire patchset applied I saw about 12% decrease on max latency reported by dbench. The test was done on a qemu vm, with 8 cores, 16Gb of ram, using kvm and using a raw NVMe device directly (no intermediary fs on the host). The test was invoked like the following:

mkfs.btrfs -f /dev/sdk
mount -o ssd -o nospace_cache /dev/sdk /mnt/sdk
dbench -D /mnt/sdk -t 300 8
umount /mnt/dsk

---
 fs/btrfs/tree-log.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
fs/btrfs/tree-log.c +5 -4
@@ -5150,7 +5150,6 @@ static int btrfs_log_inode(struct btrfs_
 			   const loff_t end,
 			   struct btrfs_log_ctx *ctx)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_path *path;
 	struct btrfs_path *dst_path;
 	struct btrfs_key min_key;
@@ -5193,15 +5192,17 @@ static int btrfs_log_inode(struct btrfs_
 	max_key.offset = (u64)-1;
 
 	/*
-	 * Only run delayed items if we are a dir or a new file.
+	 * Only run delayed items if we are a directory. We want to make sure
+	 * all directory indexes hit the fs/subvolume tree so we can find them
+	 * and figure out which index ranges have to be logged.
+	 *
 	 * Otherwise commit the delayed inode only if the full sync flag is set,
 	 * as we want to make sure an up to date version is in the subvolume
 	 * tree so copy_inode_items_to_log() / copy_items() can find it and copy
 	 * it to the log tree. For a non full sync, we always log the inode item
 	 * based on the in-memory struct btrfs_inode which is always up to date.
 	 */
-	if (S_ISDIR(inode->vfs_inode.i_mode) ||
-	    inode->generation > fs_info->last_trans_committed)
+	if (S_ISDIR(inode->vfs_inode.i_mode))
 		ret = btrfs_commit_inode_delayed_items(trans, inode);
 	else if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags))
 		ret = btrfs_commit_inode_delayed_inode(inode);
author maintainer
#17 lore raw

PATCH 17/204 btrfs: remove no longer needed use of log_writers for the log root tree

In reply to Greg Kroah-Hartman

From: Filipe Manana <[email protected]>

commit a93e01682e283f6de09d6ce8f805dc52a2e942fb upstream.

When syncing the log, we used to update the log root tree without holding neither the log_mutex of the subvolume root nor the log_mutex of log root tree.

We used to have two critical sections delimited by the log_mutex of the log root tree, so in the first one we incremented the log_writers of the log root tree and on the second one we decremented it and waited for the log_writers counter to go down to zero. This was because the update of the log root tree happened between the two critical sections.

The use of two critical sections allowed a little bit more of parallelism and required the use of the log_writers counter, necessary to make sure we didn't miss any log root tree update when we have multiple tasks trying to sync the log in parallel.

However after commit 06989c799f0481 ("Btrfs: fix race updating log root item during fsync") the log root tree update was moved into a critical section delimited by the subvolume's log_mutex. Later another commit moved the log tree update from that critical section into the second critical section delimited by the log_mutex of the log root tree. Both commits addressed different bugs.

The end result is that the first critical section delimited by the log_mutex of the log root tree became pointless, since there's nothing done between it and the second critical section, we just have an unlock of the log_mutex followed by a lock operation. This means we can merge both critical sections, as the first one does almost nothing now, and we can stop using the log_writers counter of the log root tree, which was incremented in the first critical section and decremented in the second criticial section, used to make sure no one in the second critical section started writeback of the log root tree before some other task updated it.

So just remove the mutex_unlock() followed by mutex_lock() of the log root tree, as well as the use of the log_writers counter for the log root tree.

This patch is part of a series that has the following patches:

1/4 btrfs: only commit the delayed inode when doing a full fsync 2/4 btrfs: only commit delayed items at fsync if we are logging a directory 3/4 btrfs: stop incremening log_batch for the log root tree when syncing log 4/4 btrfs: remove no longer needed use of log_writers for the log root tree

After the entire patchset applied I saw about 12% decrease on max latency reported by dbench. The test was done on a qemu vm, with 8 cores, 16Gb of ram, using kvm and using a raw NVMe device directly (no intermediary fs on the host). The test was invoked like the following:

mkfs.btrfs -f /dev/sdk
mount -o ssd -o nospace_cache /dev/sdk /mnt/sdk
dbench -D /mnt/sdk -t 300 8
umount /mnt/dsk

---
 fs/btrfs/ctree.h    |    1 +
 fs/btrfs/tree-log.c |   13 -------------
 2 files changed, 1 insertion(+), 13 deletions(-)
fs/btrfs/ctree.h +1 -0
@@ -1038,6 +1038,7 @@ struct btrfs_root {
 	wait_queue_head_t log_writer_wait;
 	wait_queue_head_t log_commit_wait[2];
 	struct list_head log_ctxs[2];
+	/* Used only for log trees of subvolumes, not for the log root tree */
 	atomic_t log_writers;
 	atomic_t log_commit[2];
 	/* Used only for log trees of subvolumes, not for the log root tree */
fs/btrfs/tree-log.c +0 -13
@@ -3125,28 +3125,17 @@ int btrfs_sync_log(struct btrfs_trans_ha
 	btrfs_init_log_ctx(&root_log_ctx, NULL);
 
 	mutex_lock(&log_root_tree->log_mutex);
-	atomic_inc(&log_root_tree->log_writers);
 
 	index2 = log_root_tree->log_transid % 2;
 	list_add_tail(&root_log_ctx.list, &log_root_tree->log_ctxs[index2]);
 	root_log_ctx.log_transid = log_root_tree->log_transid;
 
-	mutex_unlock(&log_root_tree->log_mutex);
-
-	mutex_lock(&log_root_tree->log_mutex);
-
 	/*
 	 * Now we are safe to update the log_root_tree because we're under the
 	 * log_mutex, and we're a current writer so we're holding the commit
 	 * open until we drop the log_mutex.
 	 */
 	ret = update_log_root(trans, log, &new_root_item);
-
-	if (atomic_dec_and_test(&log_root_tree->log_writers)) {
-		/* atomic_dec_and_test implies a barrier */
-		cond_wake_up_nomb(&log_root_tree->log_writer_wait);
-	}
-
 	if (ret) {
 		if (!list_empty(&root_log_ctx.list))
 			list_del_init(&root_log_ctx.list);
@@ -3192,8 +3181,6 @@ int btrfs_sync_log(struct btrfs_trans_ha
 				root_log_ctx.log_transid - 1);
 	}
 
-	wait_for_writer(log_root_tree);
-
 	/*
 	 * now that we've moved on to the tree of log tree roots,
 	 * check the full commit flag again
author maintainer
#18 lore raw

PATCH 18/204 btrfs: dont traverse into the seed devices in show_devname

In reply to Greg Kroah-Hartman

From: Anand Jain <[email protected]>

commit 4faf55b03823e96c44dc4e364520000ed3b12fdb upstream.

->show_devname currently shows the lowest devid in the list. As the seed devices have the lowest devid in the sprouted filesystem, the userland tool such as findmnt end up seeing seed device instead of the device from the read-writable sprouted filesystem. As shown below.

mount /dev/sda /btrfs
mount: /btrfs: WARNING: device write-protected, mounted read-only.

findmnt --output SOURCE,TARGET,UUID /btrfs
SOURCE TARGET UUID
/dev/sda /btrfs 899f7027-3e46-4626-93e7-7d4c9ad19111

btrfs dev add -f /dev/sdb /btrfs

umount /btrfs
mount /dev/sdb /btrfs

findmnt --output SOURCE,TARGET,UUID /btrfs
SOURCE TARGET UUID
/dev/sda /btrfs 899f7027-3e46-4626-93e7-7d4c9ad19111

All sprouts from a single seed will show the same seed device and the same fsid. That's confusing.
This is causing problems in our prototype as there isn't any reference to the sprout file-system(s) which is being used for actual read and write.

This was added in the patch which implemented the show_devname in btrfs commit 9c5085c14798 ("Btrfs: implement ->show_devname").
I tried to look for any particular reason that we need to show the seed device, there isn't any.

So instead, do not traverse through the seed devices, just show the lowest devid in the sprouted fsid.

After the patch:

mount /dev/sda /btrfs
mount: /btrfs: WARNING: device write-protected, mounted read-only.

findmnt --output SOURCE,TARGET,UUID /btrfs
SOURCE TARGET UUID
/dev/sda /btrfs 899f7027-3e46-4626-93e7-7d4c9ad19111

btrfs dev add -f /dev/sdb /btrfs
mount -o rw,remount /dev/sdb /btrfs

findmnt --output SOURCE,TARGET,UUID /btrfs
SOURCE TARGET UUID
/dev/sdb /btrfs 595ca0e6-b82e-46b5-b9e2-c72a6928be48

mount /dev/sda /btrfs1
mount: /btrfs1: WARNING: device write-protected, mounted read-only.

btrfs dev add -f /dev/sdc /btrfs1

findmnt --output SOURCE,TARGET,UUID /btrfs1
SOURCE TARGET UUID
/dev/sdc /btrfs1 ca1dbb7a-8446-4f95-853c-a20f3f82bdbb

cat /proc/self/mounts | grep btrfs
/dev/sdb /btrfs btrfs rw,relatime,noacl,space_cache,subvolid=5,subvol=/ 0 0 /dev/sdc /btrfs1 btrfs ro,relatime,noacl,space_cache,subvolid=5,subvol=/ 0 0

---
 fs/btrfs/super.c |   21 +++++++--------------
 1 file changed, 7 insertions(+), 14 deletions(-)
fs/btrfs/super.c +7 -14
@@ -2294,9 +2294,7 @@ static int btrfs_unfreeze(struct super_b
 static int btrfs_show_devname(struct seq_file *m, struct dentry *root)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(root->d_sb);
-	struct btrfs_fs_devices *cur_devices;
 	struct btrfs_device *dev, *first_dev = NULL;
-	struct list_head *head;
 
 	/*
 	 * Lightweight locking of the devices. We should not need
@@ -2306,18 +2304,13 @@ static int btrfs_show_devname(struct seq
 	 * least until the rcu_read_unlock.
 	 */
 	rcu_read_lock();
-	cur_devices = fs_info->fs_devices;
-	while (cur_devices) {
-		head = &cur_devices->devices;
-		list_for_each_entry_rcu(dev, head, dev_list) {
-			if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state))
-				continue;
-			if (!dev->name)
-				continue;
-			if (!first_dev || dev->devid < first_dev->devid)
-				first_dev = dev;
-		}
-		cur_devices = cur_devices->seed;
+	list_for_each_entry_rcu(dev, &fs_info->fs_devices->devices, dev_list) {
+		if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state))
+			continue;
+		if (!dev->name)
+			continue;
+		if (!first_dev || dev->devid < first_dev->devid)
+			first_dev = dev;
 	}
 
 	if (first_dev)
author maintainer
#19 lore raw

PATCH 19/204 btrfs: pass checksum type via BTRFS_IOC_FS_INFO ioctl

In reply to Greg Kroah-Hartman

From: Johannes Thumshirn <[email protected]>

commit 137c541821a83debb63b3fa8abdd1cbc41bdf3a1 upstream.

With the recent addition of filesystem checksum types other than CRC32c, it is not anymore hard-coded which checksum type a btrfs filesystem uses.

Up to now there is no good way to read the filesystem checksum, apart from reading the filesystem UUID and then query sysfs for the checksum type.

Add a new csum_type and csum_size fields to the BTRFS_IOC_FS_INFO ioctl command which usually is used to query filesystem features. Also add a flags member indicating that the kernel responded with a set csum_type and csum_size field.

For compatibility reasons, only return the csum_type and csum_size if the BTRFS_FS_INFO_FLAG_CSUM_INFO flag was passed to the kernel. Also clear any unknown flags so we don't pass false positives to user-space newer than the kernel.

To simplify further additions to the ioctl, also switch the padding to a u8 array. Pahole was used to verify the result of this switch:

The csum members are added before flags, which might look odd, but this is to keep the alignment requirements and not to introduce holes in the structure.

  $ pahole -C btrfs_ioctl_fs_info_args fs/btrfs/btrfs.ko
  struct btrfs_ioctl_fs_info_args {
	  __u64                      max_id;               /*     0     8 */
	  __u64                      num_devices;          /*     8     8 */
	  __u8                       fsid[16];             /*    16    16 */
	  __u32                      nodesize;             /*    32     4 */
	  __u32                      sectorsize;           /*    36     4 */
	  __u32                      clone_alignment;      /*    40     4 */
	  __u16                      csum_type;            /*    44     2 */
	  __u16                      csum_size;            /*    46     2 */
	  __u64                      flags;                /*    48     8 */
	  __u8                       reserved[968];        /*    56   968 */
	  /* size: 1024, cachelines: 16, members: 10 */

};

Fixes 3951e7f050ac ("btrfs: add xxhash64 to checksumming algorithms")
Fixes 3831bf0094ab ("btrfs: add sha256 to checksumming algorithm")
---
 fs/btrfs/ioctl.c           |   16 +++++++++++++---
 include/uapi/linux/btrfs.h |   14 ++++++++++++--
 2 files changed, 25 insertions(+), 5 deletions(-)
fs/btrfs/ioctl.c +13 -3
@@ -3197,11 +3197,15 @@ static long btrfs_ioctl_fs_info(struct b
 	struct btrfs_ioctl_fs_info_args *fi_args;
 	struct btrfs_device *device;
 	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+	u64 flags_in;
 	int ret = 0;
 
-	fi_args = kzalloc(sizeof(*fi_args), GFP_KERNEL);
-	if (!fi_args)
-		return -ENOMEM;
+	fi_args = memdup_user(arg, sizeof(*fi_args));
+	if (IS_ERR(fi_args))
+		return PTR_ERR(fi_args);
+
+	flags_in = fi_args->flags;
+	memset(fi_args, 0, sizeof(*fi_args));
 
 	rcu_read_lock();
 	fi_args->num_devices = fs_devices->num_devices;
@@ -3217,6 +3221,12 @@ static long btrfs_ioctl_fs_info(struct b
 	fi_args->sectorsize = fs_info->sectorsize;
 	fi_args->clone_alignment = fs_info->sectorsize;
 
+	if (flags_in & BTRFS_FS_INFO_FLAG_CSUM_INFO) {
+		fi_args->csum_type = btrfs_super_csum_type(fs_info->super_copy);
+		fi_args->csum_size = btrfs_super_csum_size(fs_info->super_copy);
+		fi_args->flags |= BTRFS_FS_INFO_FLAG_CSUM_INFO;
+	}
+
 	if (copy_to_user(arg, fi_args, sizeof(*fi_args)))
 		ret = -EFAULT;
 
include/uapi/linux/btrfs.h +12 -2
@@ -243,6 +243,13 @@ struct btrfs_ioctl_dev_info_args {
 	__u8 path[BTRFS_DEVICE_PATH_NAME_MAX];	/* out */
 };
 
+/*
+ * Retrieve information about the filesystem
+ */
+
+/* Request information about checksum type and size */
+#define BTRFS_FS_INFO_FLAG_CSUM_INFO			(1 << 0)
+
 struct btrfs_ioctl_fs_info_args {
 	__u64 max_id;				/* out */
 	__u64 num_devices;			/* out */
@@ -250,8 +257,11 @@ struct btrfs_ioctl_fs_info_args {
 	__u32 nodesize;				/* out */
 	__u32 sectorsize;			/* out */
 	__u32 clone_alignment;			/* out */
-	__u32 reserved32;
-	__u64 reserved[122];			/* pad to 1k */
+	/* See BTRFS_FS_INFO_FLAG_* */
+	__u16 csum_type;			/* out */
+	__u16 csum_size;			/* out */
+	__u64 flags;				/* in/out */
+	__u8 reserved[968];			/* pad to 1k */
 };
 
 /*
author maintainer
#20 lore raw

PATCH 20/204 btrfs: open device without device_list_mutex

In reply to Greg Kroah-Hartman

From: Josef Bacik <[email protected]>

commit 18c850fdc5a801bad4977b0f1723761d42267e45 upstream.

There's long existed a lockdep splat because we open our bdev's under the ->device_list_mutex at mount time, which acquires the bd_mutex. Usually this goes unnoticed, but if you do loopback devices at all suddenly the bd_mutex comes with a whole host of other dependencies, which results in the splat when you mount a btrfs file system.

======================================================
WARNING: possible circular locking dependency detected 5.8.0-0.rc3.1.fc33.x86_64+debug #1 Not tainted
------------------------------------------------------
systemd-journal/509 is trying to acquire lock:
ffff970831f84db0 (&fs_info->reloc_mutex){+.+.}-{3:3}, at: btrfs_record_root_in_trans+0x44/0x70 [btrfs]

but task is already holding lock:
ffff97083144d598 (sb_pagefaults){.+.+}-{0:0}, at: btrfs_page_mkwrite+0x59/0x560 [btrfs]

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #6 (sb_pagefaults){.+.+}-{0:0}:

       __sb_start_write+0x13e/0x220
       btrfs_page_mkwrite+0x59/0x560 [btrfs]
       do_page_mkwrite+0x4f/0x130
       do_wp_page+0x3b0/0x4f0
       handle_mm_fault+0xf47/0x1850
       do_user_addr_fault+0x1fc/0x4b0
       exc_page_fault+0x88/0x300
       asm_exc_page_fault+0x1e/0x30

-> #5 (&mm->mmap_lock#2){++++}-{3:3}:

       __might_fault+0x60/0x80
       _copy_from_user+0x20/0xb0
       get_sg_io_hdr+0x9a/0xb0
       scsi_cmd_ioctl+0x1ea/0x2f0
       cdrom_ioctl+0x3c/0x12b4
       sr_block_ioctl+0xa4/0xd0
       block_ioctl+0x3f/0x50
       ksys_ioctl+0x82/0xc0
       __x64_sys_ioctl+0x16/0x20
       do_syscall_64+0x52/0xb0
       entry_SYSCALL_64_after_hwframe+0x44/0xa9

-> #4 (&cd->lock){+.+.}-{3:3}:

       __mutex_lock+0x7b/0x820
       sr_block_open+0xa2/0x180
       __blkdev_get+0xdd/0x550
       blkdev_get+0x38/0x150
       do_dentry_open+0x16b/0x3e0
       path_openat+0x3c9/0xa00
       do_filp_open+0x75/0x100
       do_sys_openat2+0x8a/0x140
       __x64_sys_openat+0x46/0x70
       do_syscall_64+0x52/0xb0
       entry_SYSCALL_64_after_hwframe+0x44/0xa9

-> #3 (&bdev->bd_mutex){+.+.}-{3:3}:

       __mutex_lock+0x7b/0x820
       __blkdev_get+0x6a/0x550
       blkdev_get+0x85/0x150
       blkdev_get_by_path+0x2c/0x70
       btrfs_get_bdev_and_sb+0x1b/0xb0 [btrfs]
       open_fs_devices+0x88/0x240 [btrfs]
       btrfs_open_devices+0x92/0xa0 [btrfs]
       btrfs_mount_root+0x250/0x490 [btrfs]
       legacy_get_tree+0x30/0x50
       vfs_get_tree+0x28/0xc0
       vfs_kern_mount.part.0+0x71/0xb0
       btrfs_mount+0x119/0x380 [btrfs]
       legacy_get_tree+0x30/0x50
       vfs_get_tree+0x28/0xc0
       do_mount+0x8c6/0xca0
       __x64_sys_mount+0x8e/0xd0
       do_syscall_64+0x52/0xb0
       entry_SYSCALL_64_after_hwframe+0x44/0xa9

-> #2 (&fs_devs->device_list_mutex){+.+.}-{3:3}:

       __mutex_lock+0x7b/0x820
       btrfs_run_dev_stats+0x36/0x420 [btrfs]
       commit_cowonly_roots+0x91/0x2d0 [btrfs]
       btrfs_commit_transaction+0x4e6/0x9f0 [btrfs]
       btrfs_sync_file+0x38a/0x480 [btrfs]
       __x64_sys_fdatasync+0x47/0x80
       do_syscall_64+0x52/0xb0
       entry_SYSCALL_64_after_hwframe+0x44/0xa9

-> #1 (&fs_info->tree_log_mutex){+.+.}-{3:3}:

       __mutex_lock+0x7b/0x820
       btrfs_commit_transaction+0x48e/0x9f0 [btrfs]
       btrfs_sync_file+0x38a/0x480 [btrfs]
       __x64_sys_fdatasync+0x47/0x80
       do_syscall_64+0x52/0xb0
       entry_SYSCALL_64_after_hwframe+0x44/0xa9

-> #0 (&fs_info->reloc_mutex){+.+.}-{3:3}:

       __lock_acquire+0x1241/0x20c0
       lock_acquire+0xb0/0x400
       __mutex_lock+0x7b/0x820
       btrfs_record_root_in_trans+0x44/0x70 [btrfs]
       start_transaction+0xd2/0x500 [btrfs]
       btrfs_dirty_inode+0x44/0xd0 [btrfs]
       file_update_time+0xc6/0x120
       btrfs_page_mkwrite+0xda/0x560 [btrfs]
       do_page_mkwrite+0x4f/0x130
       do_wp_page+0x3b0/0x4f0
       handle_mm_fault+0xf47/0x1850
       do_user_addr_fault+0x1fc/0x4b0
       exc_page_fault+0x88/0x300
       asm_exc_page_fault+0x1e/0x30

other info that might help us debug this:

Chain exists of:
  &fs_info->reloc_mutex --> &mm->mmap_lock#2 --> sb_pagefaults

Possible unsafe locking scenario:
     CPU0                    CPU1
     ----                    ----
 lock(sb_pagefaults);
                             lock(&mm->mmap_lock#2);
                             lock(sb_pagefaults);

lock(&fs_info->reloc_mutex);

*** DEADLOCK ***

3 locks held by systemd-journal/509:
#0: ffff97083bdec8b8 (&mm->mmap_lock#2){++++}-{3:3}, at: do_user_addr_fault+0x12e/0x4b0 #1: ffff97083144d598 (sb_pagefaults){.+.+}-{0:0}, at: btrfs_page_mkwrite+0x59/0x560 [btrfs] #2: ffff97083144d6a8 (sb_internal){.+.+}-{0:0}, at: start_transaction+0x3f8/0x500 [btrfs]

stack backtrace:
CPU: 0 PID: 509 Comm: systemd-journal Not tainted 5.8.0-0.rc3.1.fc33.x86_64+debug #1
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
Call Trace:
 dump_stack+0x92/0xc8
 check_noncircular+0x134/0x150
 __lock_acquire+0x1241/0x20c0
 lock_acquire+0xb0/0x400
 ? btrfs_record_root_in_trans+0x44/0x70 [btrfs]
 ? lock_acquire+0xb0/0x400
 ? btrfs_record_root_in_trans+0x44/0x70 [btrfs]
 __mutex_lock+0x7b/0x820
 ? btrfs_record_root_in_trans+0x44/0x70 [btrfs]
 ? kvm_sched_clock_read+0x14/0x30
 ? sched_clock+0x5/0x10
 ? sched_clock_cpu+0xc/0xb0
 btrfs_record_root_in_trans+0x44/0x70 [btrfs]
 start_transaction+0xd2/0x500 [btrfs]
 btrfs_dirty_inode+0x44/0xd0 [btrfs]
 file_update_time+0xc6/0x120
 btrfs_page_mkwrite+0xda/0x560 [btrfs]
 ? sched_clock+0x5/0x10
 do_page_mkwrite+0x4f/0x130
 do_wp_page+0x3b0/0x4f0
 handle_mm_fault+0xf47/0x1850
 do_user_addr_fault+0x1fc/0x4b0
 exc_page_fault+0x88/0x300
 ? asm_exc_page_fault+0x8/0x30
 asm_exc_page_fault+0x1e/0x30
RIP: 0033:0x7fa3972fdbfe
Code: Bad RIP value.

Fix this by not holding the ->device_list_mutex at this point. The device_list_mutex exists to protect us from modifying the device list while the file system is running.

However it can also be modified by doing a scan on a device. But this action is specifically protected by the uuid_mutex, which we are holding here. We cannot race with opening at this point because we have the ->s_mount lock held during the mount. Not having the ->device_list_mutex here is perfectly safe as we're not going to change the devices at this point.

---
 fs/btrfs/volumes.c |   21 ++++++++++++++++++---
 1 file changed, 18 insertions(+), 3 deletions(-)
fs/btrfs/volumes.c +18 -3
@@ -245,7 +245,9 @@ static int __btrfs_map_block(struct btrf
  *
  * global::fs_devs - add, remove, updates to the global list
  *
- * does not protect: manipulation of the fs_devices::devices list!
+ * does not protect: manipulation of the fs_devices::devices list in general
+ * but in mount context it could be used to exclude list modifications by eg.
+ * scan ioctl
  *
  * btrfs_device::name - renames (write side), read is RCU
  *
@@ -258,6 +260,9 @@ static int __btrfs_map_block(struct btrf
  * may be used to exclude some operations from running concurrently without any
  * modifications to the list (see write_all_supers)
  *
+ * Is not required at mount and close times, because our device list is
+ * protected by the uuid_mutex at that point.
+ *
  * balance_mutex
  * -------------
  * protects balance structures (status, state) and context accessed from
@@ -603,6 +608,11 @@ static int btrfs_free_stale_devices(cons
 	return ret;
 }
 
+/*
+ * This is only used on mount, and we are protected from competing things
+ * messing with our fs_devices by the uuid_mutex, thus we do not need the
+ * fs_devices->device_list_mutex here.
+ */
 static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices,
 			struct btrfs_device *device, fmode_t flags,
 			void *holder)
@@ -1232,8 +1242,14 @@ int btrfs_open_devices(struct btrfs_fs_d
 	int ret;
 
 	lockdep_assert_held(&uuid_mutex);
+	/*
+	 * The device_list_mutex cannot be taken here in case opening the
+	 * underlying device takes further locks like bd_mutex.
+	 *
+	 * We also don't need the lock here as this is called during mount and
+	 * exclusion is provided by uuid_mutex
+	 */
 
-	mutex_lock(&fs_devices->device_list_mutex);
 	if (fs_devices->opened) {
 		fs_devices->opened++;
 		ret = 0;
@@ -1241,7 +1257,6 @@ int btrfs_open_devices(struct btrfs_fs_d
 		list_sort(NULL, &fs_devices->devices, devid_cmp);
 		ret = open_fs_devices(fs_devices, flags, holder);
 	}
-	mutex_unlock(&fs_devices->device_list_mutex);
 
 	return ret;
 }
author maintainer
#21 lore raw

PATCH 21/204 btrfs: move the chunk_mutex in btrfs_read_chunk_tree

In reply to Greg Kroah-Hartman

From: Josef Bacik <[email protected]>

commit 01d01caf19ff7c537527d352d169c4368375c0a1 upstream.

We are currently getting this lockdep splat in btrfs/161:

======================================================
WARNING: possible circular locking dependency detected
5.8.0-rc5+ #20 Tainted: G E
------------------------------------------------------
mount/678048 is trying to acquire lock:
ffff9b769f15b6e0 (&fs_devs->device_list_mutex){+.+.}-{3:3}, at: clone_fs_devices+0x4d/0x170 [btrfs]

but task is already holding lock:
ffff9b76abdb08d0 (&fs_info->chunk_mutex){+.+.}-{3:3}, at: btrfs_read_chunk_tree+0x6a/0x800 [btrfs]

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (&fs_info->chunk_mutex){+.+.}-{3:3}:

	 __mutex_lock+0x8b/0x8f0
	 btrfs_init_new_device+0x2d2/0x1240 [btrfs]
	 btrfs_ioctl+0x1de/0x2d20 [btrfs]
	 ksys_ioctl+0x87/0xc0
	 __x64_sys_ioctl+0x16/0x20
	 do_syscall_64+0x52/0xb0
	 entry_SYSCALL_64_after_hwframe+0x44/0xa9

-> #0 (&fs_devs->device_list_mutex){+.+.}-{3:3}:

	 __lock_acquire+0x1240/0x2460
	 lock_acquire+0xab/0x360
	 __mutex_lock+0x8b/0x8f0
	 clone_fs_devices+0x4d/0x170 [btrfs]
	 btrfs_read_chunk_tree+0x330/0x800 [btrfs]
	 open_ctree+0xb7c/0x18ce [btrfs]
	 btrfs_mount_root.cold+0x13/0xfa [btrfs]
	 legacy_get_tree+0x30/0x50
	 vfs_get_tree+0x28/0xc0
	 fc_mount+0xe/0x40
	 vfs_kern_mount.part.0+0x71/0x90
	 btrfs_mount+0x13b/0x3e0 [btrfs]
	 legacy_get_tree+0x30/0x50
	 vfs_get_tree+0x28/0xc0
	 do_mount+0x7de/0xb30
	 __x64_sys_mount+0x8e/0xd0
	 do_syscall_64+0x52/0xb0
	 entry_SYSCALL_64_after_hwframe+0x44/0xa9

other info that might help us debug this:

   Possible unsafe locking scenario:
	 CPU0                    CPU1
	 ----                    ----
    lock(&fs_info->chunk_mutex);
				 lock(&fs_devs->device_list_mutex);
				 lock(&fs_info->chunk_mutex);
    lock(&fs_devs->device_list_mutex);

*** DEADLOCK ***

3 locks held by mount/678048:
#0: ffff9b75ff5fb0e0 (&type->s_umount_key#63/1){+.+.}-{3:3}, at: alloc_super+0xb5/0x380 #1: ffffffffc0c2fbc8 (uuid_mutex){+.+.}-{3:3}, at: btrfs_read_chunk_tree+0x54/0x800 [btrfs] #2: ffff9b76abdb08d0 (&fs_info->chunk_mutex){+.+.}-{3:3}, at: btrfs_read_chunk_tree+0x6a/0x800 [btrfs]

  stack backtrace:
  CPU: 2 PID: 678048 Comm: mount Tainted: G            E     5.8.0-rc5+ #20
  Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./890FX Deluxe5, BIOS P1.40 05/03/2011
  Call Trace:
   dump_stack+0x96/0xd0
   check_noncircular+0x162/0x180
   __lock_acquire+0x1240/0x2460
   ? asm_sysvec_apic_timer_interrupt+0x12/0x20
   lock_acquire+0xab/0x360
   ? clone_fs_devices+0x4d/0x170 [btrfs]
   __mutex_lock+0x8b/0x8f0
   ? clone_fs_devices+0x4d/0x170 [btrfs]
   ? rcu_read_lock_sched_held+0x52/0x60
   ? cpumask_next+0x16/0x20
   ? module_assert_mutex_or_preempt+0x14/0x40
   ? __module_address+0x28/0xf0
   ? clone_fs_devices+0x4d/0x170 [btrfs]
   ? static_obj+0x4f/0x60
   ? lockdep_init_map_waits+0x43/0x200
   ? clone_fs_devices+0x4d/0x170 [btrfs]
   clone_fs_devices+0x4d/0x170 [btrfs]
   btrfs_read_chunk_tree+0x330/0x800 [btrfs]
   open_ctree+0xb7c/0x18ce [btrfs]
   ? super_setup_bdi_name+0x79/0xd0
   btrfs_mount_root.cold+0x13/0xfa [btrfs]
   ? vfs_parse_fs_string+0x84/0xb0
   ? rcu_read_lock_sched_held+0x52/0x60
   ? kfree+0x2b5/0x310
   legacy_get_tree+0x30/0x50
   vfs_get_tree+0x28/0xc0
   fc_mount+0xe/0x40
   vfs_kern_mount.part.0+0x71/0x90
   btrfs_mount+0x13b/0x3e0 [btrfs]
   ? cred_has_capability+0x7c/0x120
   ? rcu_read_lock_sched_held+0x52/0x60
   ? legacy_get_tree+0x30/0x50
   legacy_get_tree+0x30/0x50
   vfs_get_tree+0x28/0xc0
   do_mount+0x7de/0xb30
   ? memdup_user+0x4e/0x90
   __x64_sys_mount+0x8e/0xd0
   do_syscall_64+0x52/0xb0
   entry_SYSCALL_64_after_hwframe+0x44/0xa9

This is because btrfs_read_chunk_tree() can come upon DEV_EXTENT's and then read the device, which takes the device_list_mutex. The device_list_mutex needs to be taken before the chunk_mutex, so this is a problem. We only really need the chunk mutex around adding the chunk, so move the mutex around read_one_chunk.

An argument could be made that we don't even need the chunk_mutex here as it's during mount, and we are protected by various other locks. However we already have special rules for ->device_list_mutex, and I'd rather not have another special case for ->chunk_mutex.

---
 fs/btrfs/volumes.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/btrfs/volumes.c +2 -2
@@ -7068,7 +7068,6 @@ int btrfs_read_chunk_tree(struct btrfs_f
 	 * otherwise we don't need it.
 	 */
 	mutex_lock(&uuid_mutex);
-	mutex_lock(&fs_info->chunk_mutex);
 
 	/*
 	 * It is possible for mount and umount to race in such a way that
@@ -7113,7 +7112,9 @@ int btrfs_read_chunk_tree(struct btrfs_f
 		} else if (found_key.type == BTRFS_CHUNK_ITEM_KEY) {
 			struct btrfs_chunk *chunk;
 			chunk = btrfs_item_ptr(leaf, slot, struct btrfs_chunk);
+			mutex_lock(&fs_info->chunk_mutex);
 			ret = read_one_chunk(&found_key, leaf, chunk);
+			mutex_unlock(&fs_info->chunk_mutex);
 			if (ret)
 				goto error;
 		}
@@ -7143,7 +7144,6 @@ int btrfs_read_chunk_tree(struct btrfs_f
 	}
 	ret = 0;
 error:
-	mutex_unlock(&fs_info->chunk_mutex);
 	mutex_unlock(&uuid_mutex);
 
 	btrfs_free_path(path);
author maintainer
#22 lore raw

PATCH 22/204 btrfs: relocation: review the call sites which can be interrupted by signal

In reply to Greg Kroah-Hartman

From: Qu Wenruo <[email protected]>

commit 44d354abf33e92a5e73b965c84caf5a5d5e58a0b upstream.

Since most metadata reservation calls can return -EINTR when get interrupted by fatal signal, we need to review the all the metadata reservation call sites.

In relocation code, the metadata reservation happens in the following sites:

- btrfs_block_rsv_refill() in merge_reloc_root() merge_reloc_root() is a pretty critical section, we don't want to be interrupted by signal, so change the flush status to BTRFS_RESERVE_FLUSH_LIMIT, so it won't get interrupted by signal. Since such change can be ENPSPC-prone, also shrink the amount of metadata to reserve least amount avoid deadly ENOSPC there.

- btrfs_block_rsv_refill() in reserve_metadata_space()
It calls with BTRFS_RESERVE_FLUSH_LIMIT, which won't get interrupted by signal.

- btrfs_block_rsv_refill() in prepare_to_relocate()

- btrfs_block_rsv_add() in prepare_to_relocate()

- btrfs_block_rsv_refill() in relocate_block_group()

- btrfs_delalloc_reserve_metadata() in relocate_file_extent_cluster()

- btrfs_start_transaction() in relocate_block_group()

- btrfs_start_transaction() in create_reloc_inode()
Can be interrupted by fatal signal and we can handle it easily. For these call sites, just catch the -EINTR value in btrfs_balance() and count them as canceled.

---
 fs/btrfs/relocation.c |   12 ++++++++++--
 fs/btrfs/volumes.c    |   17 ++++++++++++++++-
 2 files changed, 26 insertions(+), 3 deletions(-)
fs/btrfs/relocation.c +10 -2
@@ -2402,12 +2402,20 @@ static noinline_for_stack int merge_relo
 		btrfs_unlock_up_safe(path, 0);
 	}
 
-	min_reserved = fs_info->nodesize * (BTRFS_MAX_LEVEL - 1) * 2;
+	/*
+	 * In merge_reloc_root(), we modify the upper level pointer to swap the
+	 * tree blocks between reloc tree and subvolume tree.  Thus for tree
+	 * block COW, we COW at most from level 1 to root level for each tree.
+	 *
+	 * Thus the needed metadata size is at most root_level * nodesize,
+	 * and * 2 since we have two trees to COW.
+	 */
+	min_reserved = fs_info->nodesize * btrfs_root_level(root_item) * 2;
 	memset(&next_key, 0, sizeof(next_key));
 
 	while (1) {
 		ret = btrfs_block_rsv_refill(root, rc->block_rsv, min_reserved,
-					     BTRFS_RESERVE_FLUSH_ALL);
+					     BTRFS_RESERVE_FLUSH_LIMIT);
 		if (ret) {
 			err = ret;
 			goto out;
fs/btrfs/volumes.c +16 -1
@@ -4154,7 +4154,22 @@ int btrfs_balance(struct btrfs_fs_info *
 	mutex_lock(&fs_info->balance_mutex);
 	if (ret == -ECANCELED && atomic_read(&fs_info->balance_pause_req))
 		btrfs_info(fs_info, "balance: paused");
-	else if (ret == -ECANCELED && atomic_read(&fs_info->balance_cancel_req))
+	/*
+	 * Balance can be canceled by:
+	 *
+	 * - Regular cancel request
+	 *   Then ret == -ECANCELED and balance_cancel_req > 0
+	 *
+	 * - Fatal signal to "btrfs" process
+	 *   Either the signal caught by wait_reserve_ticket() and callers
+	 *   got -EINTR, or caught by btrfs_should_cancel_balance() and
+	 *   got -ECANCELED.
+	 *   Either way, in this case balance_cancel_req = 0, and
+	 *   ret == -EINTR or ret == -ECANCELED.
+	 *
+	 * So here we only check the return value to catch canceled balance.
+	 */
+	else if (ret == -ECANCELED || ret == -EINTR)
 		btrfs_info(fs_info, "balance: canceled");
 	else
 		btrfs_info(fs_info, "balance: ended with status: %d", ret);
author maintainer
#23 lore raw

PATCH 23/204 btrfs: add missing check for nocow and compression inode flags

In reply to Greg Kroah-Hartman

From: David Sterba <[email protected]>

commit f37c563bab4297024c300b05c8f48430e323809d upstream.

User Forza reported on IRC that some invalid combinations of file attributes are accepted by chattr.

The NODATACOW and compression file flags/attributes are mutually exclusive, but they could be set by 'chattr +c +C' on an empty file. The nodatacow will be in effect because it's checked first in btrfs_run_delalloc_range.

Extend the flag validation to catch the following cases:

- input flags are conflicting
- old and new flags are conflicting
- initialize the local variable with inode flags after inode ls locked

Inode attributes take precedence over mount options and are an independent setting.

Nocompress would be a no-op with nodatacow, but we don't want to mix any compression-related options with nodatacow.

---
 fs/btrfs/ioctl.c |   30 ++++++++++++++++++++++--------
 1 file changed, 22 insertions(+), 8 deletions(-)
fs/btrfs/ioctl.c +22 -8
@@ -164,8 +164,11 @@ static int btrfs_ioctl_getflags(struct f
 	return 0;
 }
 
-/* Check if @flags are a supported and valid set of FS_*_FL flags */
-static int check_fsflags(unsigned int flags)
+/*
+ * Check if @flags are a supported and valid set of FS_*_FL flags and that
+ * the old and new flags are not conflicting
+ */
+static int check_fsflags(unsigned int old_flags, unsigned int flags)
 {
 	if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \
 		      FS_NOATIME_FL | FS_NODUMP_FL | \
@@ -174,9 +177,19 @@ static int check_fsflags(unsigned int fl
 		      FS_NOCOW_FL))
 		return -EOPNOTSUPP;
 
+	/* COMPR and NOCOMP on new/old are valid */
 	if ((flags & FS_NOCOMP_FL) && (flags & FS_COMPR_FL))
 		return -EINVAL;
 
+	if ((flags & FS_COMPR_FL) && (flags & FS_NOCOW_FL))
+		return -EINVAL;
+
+	/* NOCOW and compression options are mutually exclusive */
+	if ((old_flags & FS_NOCOW_FL) && (flags & (FS_COMPR_FL | FS_NOCOMP_FL)))
+		return -EINVAL;
+	if ((flags & FS_NOCOW_FL) && (old_flags & (FS_COMPR_FL | FS_NOCOMP_FL)))
+		return -EINVAL;
+
 	return 0;
 }
 
@@ -190,7 +203,7 @@ static int btrfs_ioctl_setflags(struct f
 	unsigned int fsflags, old_fsflags;
 	int ret;
 	const char *comp = NULL;
-	u32 binode_flags = binode->flags;
+	u32 binode_flags;
 
 	if (!inode_owner_or_capable(inode))
 		return -EPERM;
@@ -201,22 +214,23 @@ static int btrfs_ioctl_setflags(struct f
 	if (copy_from_user(&fsflags, arg, sizeof(fsflags)))
 		return -EFAULT;
 
-	ret = check_fsflags(fsflags);
-	if (ret)
-		return ret;
-
 	ret = mnt_want_write_file(file);
 	if (ret)
 		return ret;
 
 	inode_lock(inode);
-
 	fsflags = btrfs_mask_fsflags_for_type(inode, fsflags);
 	old_fsflags = btrfs_inode_flags_to_fsflags(binode->flags);
+
 	ret = vfs_ioc_setflags_prepare(inode, old_fsflags, fsflags);
 	if (ret)
 		goto out_unlock;
 
+	ret = check_fsflags(old_fsflags, fsflags);
+	if (ret)
+		goto out_unlock;
+
+	binode_flags = binode->flags;
 	if (fsflags & FS_SYNC_FL)
 		binode_flags |= BTRFS_INODE_SYNC;
 	else
author maintainer
#24 lore raw

PATCH 24/204 btrfs: avoid possible signal interruption of btrfs_drop_snapshot() on relocation tree

In reply to Greg Kroah-Hartman

From: Qu Wenruo <[email protected]>

commit f3e3d9cc35252a70a2fd698762c9687718268ec6 upstream.

[BUG]
There is a bug report about bad signal timing could lead to read-only fs during balance:

BTRFS info (device xvdb): balance: start -d -m -s
BTRFS info (device xvdb): relocating block group 73001861120 flags metadata BTRFS info (device xvdb): found 12236 extents, stage: move data extents BTRFS info (device xvdb): relocating block group 71928119296 flags data BTRFS info (device xvdb): found 3 extents, stage: move data extents BTRFS info (device xvdb): found 3 extents, stage: update data pointers BTRFS info (device xvdb): relocating block group 60922265600 flags metadata
BTRFS: error (device xvdb) in btrfs_drop_snapshot:5505: errno=-4 unknown BTRFS info (device xvdb): forced readonly
BTRFS info (device xvdb): balance: ended with status: -4

[CAUSE]
The direct cause is the -EINTR from the following call chain when a fatal signal is pending:

relocate_block_group()
|- clean_dirty_subvols()

    |- btrfs_drop_snapshot()
       |- btrfs_start_transaction()
          |- btrfs_delayed_refs_rsv_refill()
             |- btrfs_reserve_metadata_bytes()
                |- __reserve_metadata_bytes()
                   |- wait_reserve_ticket()
                      |- prepare_to_wait_event();
                      |- ticket->error = -EINTR;

Normally this behavior is fine for most btrfs_start_transaction() callers, as they need to catch any other error, same for the signal, and exit ASAP.

However for balance, especially for the clean_dirty_subvols() case, we're already doing cleanup works, getting -EINTR from btrfs_drop_snapshot() could cause a lot of unexpected problems.

Quoted 1 line

From the mentioned forced read-only report, to later balance error due

to half dropped reloc trees.

[FIX]
Fix this problem by using btrfs_join_transaction() if btrfs_drop_snapshot() is called from relocation context.

Since btrfs_join_transaction() won't get interrupted by signal, we can continue the cleanup.

---
 fs/btrfs/extent-tree.c |    9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
fs/btrfs/extent-tree.c +8 -1
@@ -5313,7 +5313,14 @@ int btrfs_drop_snapshot(struct btrfs_roo
 		goto out;
 	}
 
-	trans = btrfs_start_transaction(tree_root, 0);
+	/*
+	 * Use join to avoid potential EINTR from transaction start. See
+	 * wait_reserve_ticket and the whole reservation callchain.
+	 */
+	if (for_reloc)
+		trans = btrfs_join_transaction(tree_root);
+	else
+		trans = btrfs_start_transaction(tree_root, 0);
 	if (IS_ERR(trans)) {
 		err = PTR_ERR(trans);
 		goto out_free;
author maintainer
#25 lore raw

PATCH 25/204 btrfs: return EROFS for BTRFS_FS_STATE_ERROR cases

In reply to Greg Kroah-Hartman

From: Josef Bacik <[email protected]>

commit fbabd4a36faaf74c83142d0b3d950c11ec14fda1 upstream.

Eric reported seeing this message while running generic/475

BTRFS: error (device dm-3) in btrfs_sync_log:3084: errno=-117 Filesystem corrupted

Full stack trace:

BTRFS: error (device dm-0) in btrfs_commit_transaction:2323: errno=-5 IO failure (Error while writing out transaction) BTRFS info (device dm-0): forced readonly
BTRFS warning (device dm-0): Skipping commit of aborted transaction.

  ------------[ cut here ]------------
  BTRFS: error (device dm-0) in cleanup_transaction:1894: errno=-5 IO failure
  BTRFS: Transaction aborted (error -117)
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c6480 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c6488 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c6490 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c6498 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c64a0 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c64a8 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c64b0 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c64b8 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3555 rw 0,0 sector 0x1c64c0 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3572 rw 0,0 sector 0x1b85e8 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3572 rw 0,0 sector 0x1b85f0 len 4096 err no 10
  WARNING: CPU: 3 PID: 23985 at fs/btrfs/tree-log.c:3084 btrfs_sync_log+0xbc8/0xd60 [btrfs]
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d4288 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d4290 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d4298 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d42a0 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d42a8 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d42b0 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d42b8 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d42c0 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d42c8 len 4096 err no 10
  BTRFS warning (device dm-0): direct IO failed ino 3548 rw 0,0 sector 0x1d42d0 len 4096 err no 10
  CPU: 3 PID: 23985 Comm: fsstress Tainted: G        W    L    5.8.0-rc4-default+ #1181
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba527-rebuilt.opensuse.org 04/01/2014
  RIP: 0010:btrfs_sync_log+0xbc8/0xd60 [btrfs]
  RSP: 0018:ffff909a44d17bd0 EFLAGS: 00010286
  RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000001
  RDX: ffff8f3be41cb940 RSI: ffffffffb0108d2b RDI: ffffffffb0108ff7
  RBP: ffff909a44d17e70 R08: 0000000000000000 R09: 0000000000000000
  R10: 0000000000000000 R11: 0000000000037988 R12: ffff8f3bd20e4000
  R13: ffff8f3bd20e4428 R14: 00000000ffffff8b R15: ffff909a44d17c70
  FS:  00007f6a6ed3fb80(0000) GS:ffff8f3c3dc00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 00007f6a6ed3e000 CR3: 00000000525c0003 CR4: 0000000000160ee0
  Call Trace:
   ? finish_wait+0x90/0x90
   ? __mutex_unlock_slowpath+0x45/0x2a0
   ? lock_acquire+0xa3/0x440
   ? lockref_put_or_lock+0x9/0x30
   ? dput+0x20/0x4a0
   ? dput+0x20/0x4a0
   ? do_raw_spin_unlock+0x4b/0xc0
   ? _raw_spin_unlock+0x1f/0x30
   btrfs_sync_file+0x335/0x490 [btrfs]
   do_fsync+0x38/0x70
   __x64_sys_fsync+0x10/0x20
   do_syscall_64+0x50/0xe0
   entry_SYSCALL_64_after_hwframe+0x44/0xa9
  RIP: 0033:0x7f6a6ef1b6e3
  Code: Bad RIP value.
  RSP: 002b:00007ffd01e20038 EFLAGS: 00000246 ORIG_RAX: 000000000000004a
  RAX: ffffffffffffffda RBX: 000000000007a120 RCX: 00007f6a6ef1b6e3
  RDX: 00007ffd01e1ffa0 RSI: 00007ffd01e1ffa0 RDI: 0000000000000003
  RBP: 0000000000000003 R08: 0000000000000001 R09: 00007ffd01e2004c
  R10: 0000000000000000 R11: 0000000000000246 R12: 000000000000009f
  R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
  irq event stamp: 0
  hardirqs last  enabled at (0): [<0000000000000000>] 0x0
  hardirqs last disabled at (0): [<ffffffffb007fe0b>] copy_process+0x67b/0x1b00
  softirqs last  enabled at (0): [<ffffffffb007fe0b>] copy_process+0x67b/0x1b00
  softirqs last disabled at (0): [<0000000000000000>] 0x0
  ---[ end trace af146e0e38433456 ]---
  BTRFS: error (device dm-0) in btrfs_sync_log:3084: errno=-117 Filesystem corrupted

This ret came from btrfs_write_marked_extents(). If we get an aborted transaction via EIO before, we'll see it in btree_write_cache_pages() and return EUCLEAN, which gets printed as "Filesystem corrupted".

Except we shouldn't be returning EUCLEAN here, we need to be returning EROFS because EUCLEAN is reserved for actual corruption, not IO errors.

We are inconsistent about our handling of BTRFS_FS_STATE_ERROR elsewhere, but we want to use EROFS for this particular case. The original transaction abort has the real error code for why we ended up with an aborted transaction, all subsequent actions just need to return EROFS because they may not have a trans handle and have no idea about the original cause of the abort.

After patch "btrfs: don't WARN if we abort a transaction with EROFS" the stacktrace will not be dumped either.

---
 fs/btrfs/extent_io.c   |    2 +-
 fs/btrfs/scrub.c       |    2 +-
 fs/btrfs/transaction.c |    5 ++++-
 3 files changed, 6 insertions(+), 3 deletions(-)
fs/btrfs/extent_io.c +1 -1
@@ -4110,7 +4110,7 @@ retry:
 	if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
 		ret = flush_write_bio(&epd);
 	} else {
-		ret = -EUCLEAN;
+		ret = -EROFS;
 		end_write_bio(&epd, ret);
 	}
 	return ret;
fs/btrfs/scrub.c +1 -1
@@ -3761,7 +3761,7 @@ static noinline_for_stack int scrub_supe
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
 
 	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
-		return -EIO;
+		return -EROFS;
 
 	/* Seed devices of a new filesystem has their own generation. */
 	if (scrub_dev->fs_devices != fs_info->fs_devices)
fs/btrfs/transaction.c +4 -1
@@ -937,7 +937,10 @@ static int __btrfs_end_transaction(struc
 	if (TRANS_ABORTED(trans) ||
 	    test_bit(BTRFS_FS_STATE_ERROR, &info->fs_state)) {
 		wake_up_process(info->transaction_kthread);
-		err = -EIO;
+		if (TRANS_ABORTED(trans))
+			err = trans->aborted;
+		else
+			err = -EROFS;
 	}
 
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
author maintainer
#26 lore raw

PATCH 26/204 btrfs: sysfs: use NOFS for device creation

In reply to Greg Kroah-Hartman

From: Josef Bacik <[email protected]>

commit a47bd78d0c44621efb98b525d04d60dc4d1a79b0 upstream.

Dave hit this splat during testing btrfs/078:

======================================================
WARNING: possible circular locking dependency detected
5.8.0-rc6-default+ #1191 Not tainted
------------------------------------------------------
kswapd0/75 is trying to acquire lock:
ffffa040e9d04ff8 (&delayed_node->mutex){+.+.}-{3:3}, at: __btrfs_release_delayed_node.part.0+0x3f/0x310 [btrfs]

but task is already holding lock:
ffffffff8b0c8040 (fs_reclaim){+.+.}-{0:0}, at: __fs_reclaim_acquire+0x5/0x30

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #2 (fs_reclaim){+.+.}-{0:0}:

	 __lock_acquire+0x56f/0xaa0
	 lock_acquire+0xa3/0x440
	 fs_reclaim_acquire.part.0+0x25/0x30
	 __kmalloc_track_caller+0x49/0x330
	 kstrdup+0x2e/0x60
	 __kernfs_new_node.constprop.0+0x44/0x250
	 kernfs_new_node+0x25/0x50
	 kernfs_create_link+0x34/0xa0
	 sysfs_do_create_link_sd+0x5e/0xd0
	 btrfs_sysfs_add_devices_dir+0x65/0x100 [btrfs]
	 btrfs_init_new_device+0x44c/0x12b0 [btrfs]
	 btrfs_ioctl+0xc3c/0x25c0 [btrfs]
	 ksys_ioctl+0x68/0xa0
	 __x64_sys_ioctl+0x16/0x20
	 do_syscall_64+0x50/0xe0
	 entry_SYSCALL_64_after_hwframe+0x44/0xa9

-> #1 (&fs_info->chunk_mutex){+.+.}-{3:3}:

	 __lock_acquire+0x56f/0xaa0
	 lock_acquire+0xa3/0x440
	 __mutex_lock+0xa0/0xaf0
	 btrfs_chunk_alloc+0x137/0x3e0 [btrfs]
	 find_free_extent+0xb44/0xfb0 [btrfs]
	 btrfs_reserve_extent+0x9b/0x180 [btrfs]
	 btrfs_alloc_tree_block+0xc1/0x350 [btrfs]
	 alloc_tree_block_no_bg_flush+0x4a/0x60 [btrfs]
	 __btrfs_cow_block+0x143/0x7a0 [btrfs]
	 btrfs_cow_block+0x15f/0x310 [btrfs]
	 push_leaf_right+0x150/0x240 [btrfs]
	 split_leaf+0x3cd/0x6d0 [btrfs]
	 btrfs_search_slot+0xd14/0xf70 [btrfs]
	 btrfs_insert_empty_items+0x64/0xc0 [btrfs]
	 __btrfs_commit_inode_delayed_items+0xb2/0x840 [btrfs]
	 btrfs_async_run_delayed_root+0x10e/0x1d0 [btrfs]
	 btrfs_work_helper+0x2f9/0x650 [btrfs]
	 process_one_work+0x22c/0x600
	 worker_thread+0x50/0x3b0
	 kthread+0x137/0x150
	 ret_from_fork+0x1f/0x30

-> #0 (&delayed_node->mutex){+.+.}-{3:3}:

	 check_prev_add+0x98/0xa20
	 validate_chain+0xa8c/0x2a00
	 __lock_acquire+0x56f/0xaa0
	 lock_acquire+0xa3/0x440
	 __mutex_lock+0xa0/0xaf0
	 __btrfs_release_delayed_node.part.0+0x3f/0x310 [btrfs]
	 btrfs_evict_inode+0x3bf/0x560 [btrfs]
	 evict+0xd6/0x1c0
	 dispose_list+0x48/0x70
	 prune_icache_sb+0x54/0x80
	 super_cache_scan+0x121/0x1a0
	 do_shrink_slab+0x175/0x420
	 shrink_slab+0xb1/0x2e0
	 shrink_node+0x192/0x600
	 balance_pgdat+0x31f/0x750
	 kswapd+0x206/0x510
	 kthread+0x137/0x150
	 ret_from_fork+0x1f/0x30

other info that might help us debug this:

  Chain exists of:
    &delayed_node->mutex --> &fs_info->chunk_mutex --> fs_reclaim

   Possible unsafe locking scenario:
	 CPU0                    CPU1
	 ----                    ----
    lock(fs_reclaim);
				 lock(&fs_info->chunk_mutex);
				 lock(fs_reclaim);
    lock(&delayed_node->mutex);

*** DEADLOCK ***

3 locks held by kswapd0/75:
#0: ffffffff8b0c8040 (fs_reclaim){+.+.}-{0:0}, at: __fs_reclaim_acquire+0x5/0x30 #1: ffffffff8b0b50b8 (shrinker_rwsem){++++}-{3:3}, at: shrink_slab+0x54/0x2e0 #2: ffffa040e057c0e8 (&type->s_umount_key#26){++++}-{3:3}, at: trylock_super+0x16/0x50

  stack backtrace:
  CPU: 2 PID: 75 Comm: kswapd0 Not tainted 5.8.0-rc6-default+ #1191
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba527-rebuilt.opensuse.org 04/01/2014
  Call Trace:
   dump_stack+0x78/0xa0
   check_noncircular+0x16f/0x190
   check_prev_add+0x98/0xa20
   validate_chain+0xa8c/0x2a00
   __lock_acquire+0x56f/0xaa0
   lock_acquire+0xa3/0x440
   ? __btrfs_release_delayed_node.part.0+0x3f/0x310 [btrfs]
   __mutex_lock+0xa0/0xaf0
   ? __btrfs_release_delayed_node.part.0+0x3f/0x310 [btrfs]
   ? __lock_acquire+0x56f/0xaa0
   ? __btrfs_release_delayed_node.part.0+0x3f/0x310 [btrfs]
   ? lock_acquire+0xa3/0x440
   ? btrfs_evict_inode+0x138/0x560 [btrfs]
   ? btrfs_evict_inode+0x2fe/0x560 [btrfs]
   ? __btrfs_release_delayed_node.part.0+0x3f/0x310 [btrfs]
   __btrfs_release_delayed_node.part.0+0x3f/0x310 [btrfs]
   btrfs_evict_inode+0x3bf/0x560 [btrfs]
   evict+0xd6/0x1c0
   dispose_list+0x48/0x70
   prune_icache_sb+0x54/0x80
   super_cache_scan+0x121/0x1a0
   do_shrink_slab+0x175/0x420
   shrink_slab+0xb1/0x2e0
   shrink_node+0x192/0x600
   balance_pgdat+0x31f/0x750
   kswapd+0x206/0x510
   ? _raw_spin_unlock_irqrestore+0x3e/0x50
   ? finish_wait+0x90/0x90
   ? balance_pgdat+0x750/0x750
   kthread+0x137/0x150
   ? kthread_stop+0x2a0/0x2a0
   ret_from_fork+0x1f/0x30

This is because we're holding the chunk_mutex while adding this device and adding its sysfs entries. We actually hold different locks in different places when calling this function, the dev_replace semaphore for instance in dev replace, so instead of moving this call around simply wrap it's operations in NOFS.

---
 fs/btrfs/sysfs.c |    3 +++
 1 file changed, 3 insertions(+)
fs/btrfs/sysfs.c +3 -0
@@ -1273,7 +1273,9 @@ int btrfs_sysfs_add_devices_dir(struct b
 {
 	int error = 0;
 	struct btrfs_device *dev;
+	unsigned int nofs_flag;
 
+	nofs_flag = memalloc_nofs_save();
 	list_for_each_entry(dev, &fs_devices->devices, dev_list) {
 
 		if (one_device && one_device != dev)
@@ -1301,6 +1303,7 @@ int btrfs_sysfs_add_devices_dir(struct b
 			break;
 		}
 	}
+	memalloc_nofs_restore(nofs_flag);
 
 	return error;
 }
author maintainer
#27 lore raw

PATCH 27/204 btrfs: dont WARN if we abort a transaction with EROFS

In reply to Greg Kroah-Hartman

From: Josef Bacik <[email protected]>

commit f95ebdbed46a4d8b9fdb7bff109fdbb6fc9a6dc8 upstream.

If we got some sort of corruption via a read and call btrfs_handle_fs_error() we'll set BTRFS_FS_STATE_ERROR on the fs and complain. If a subsequent trans handle trips over this it'll get EROFS and then abort. However at that point we're not aborting for the original reason, we're aborting because we've been flipped read only. We do not need to WARN_ON() here.

---
 fs/btrfs/ctree.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/ctree.h +1 -1
@@ -3198,7 +3198,7 @@ do {								\
 	/* Report first abort since mount */			\
 	if (!test_and_set_bit(BTRFS_FS_STATE_TRANS_ABORTED,	\
 			&((trans)->fs_info->fs_state))) {	\
-		if ((errno) != -EIO) {				\
+		if ((errno) != -EIO && (errno) != -EROFS) {		\
 			WARN(1, KERN_DEBUG				\
 			"BTRFS: Transaction aborted (error %d)\n",	\
 			(errno));					\
author maintainer
#28 lore raw

PATCH 28/204 btrfs: fix race between page release and a fast fsync

In reply to Greg Kroah-Hartman

From: Filipe Manana <[email protected]>

commit 3d6448e631591756da36efb3ea6355ff6f383c3a upstream.

When releasing an extent map, done through the page release callback, we can race with an ongoing fast fsync and cause the fsync to miss a new extent and not log it. The steps for this to happen are the following:

1) A page is dirtied for some inode I;

2) Writeback for that page is triggered by a path other than fsync, for example by the system due to memory pressure;

3) When the ordered extent for the extent (a single 4K page) finishes, we unpin the corresponding extent map and set its generation to N, the current transaction's generation;

4) The btrfs_releasepage() callback is invoked by the system due to memory pressure for that no longer dirty page of inode I;

5) At the same time, some task calls fsync on inode I, joins transaction N, and at btrfs_log_inode() it sees that the inode does not have the full sync flag set, so we proceed with a fast fsync. But before we get into btrfs_log_changed_extents() and lock the inode's extent map tree:

6) Through btrfs_releasepage() we end up at try_release_extent_mapping() and we remove the extent map for the new 4Kb extent, because it is neither pinned anymore nor locked. By calling remove_extent_mapping(), we remove the extent map from the list of modified extents, since the extent map does not have the logging flag set. We unlock the inode's extent map tree;

7) The task doing the fast fsync now enters btrfs_log_changed_extents(), locks the inode's extent map tree and iterates its list of modified extents, which no longer has the 4Kb extent in it, so it does not log the extent;

8) The fsync finishes;

9) Before transaction N is committed, a power failure happens. After replaying the log, the 4K extent of inode I will be missing, since it was not logged due to the race with try_release_extent_mapping().

So fix this by teaching try_release_extent_mapping() to not remove an extent map if it's still in the list of modified extents.

Fixes ff44c6e36dc9dc ("Btrfs: do not hold the write_lock on the extent tree while logging")
---
 fs/btrfs/extent_io.c |   16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
fs/btrfs/extent_io.c +13 -3
@@ -4504,15 +4504,25 @@ int try_release_extent_mapping(struct pa
 				free_extent_map(em);
 				break;
 			}
-			if (!test_range_bit(tree, em->start,
-					    extent_map_end(em) - 1,
-					    EXTENT_LOCKED, 0, NULL)) {
+			if (test_range_bit(tree, em->start,
+					   extent_map_end(em) - 1,
+					   EXTENT_LOCKED, 0, NULL))
+				goto next;
+			/*
+			 * If it's not in the list of modified extents, used
+			 * by a fast fsync, we can remove it. If it's being
+			 * logged we can safely remove it since fsync took an
+			 * extra reference on the em.
+			 */
+			if (list_empty(&em->list) ||
+			    test_bit(EXTENT_FLAG_LOGGING, &em->flags)) {
 				set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
 					&btrfs_inode->runtime_flags);
 				remove_extent_mapping(map, em);
 				/* once for the rb tree */
 				free_extent_map(em);
 			}
+next:
 			start = extent_map_end(em);
 			write_unlock(&map->lock);
 
author maintainer
#29 lore raw

PATCH 29/204 btrfs: dont show full path of bind mounts in subvol=

In reply to Greg Kroah-Hartman

From: Josef Bacik <[email protected]>

commit 3ef3959b29c4a5bd65526ab310a1a18ae533172a upstream.

Chris Murphy reported a problem where rpm ostree will bind mount a bunch of things for whatever voodoo it's doing. But when it does this /proc/mounts shows something like

/dev/sda /mnt/test btrfs rw,relatime,subvolid=256,subvol=/foo 0 0 /dev/sda /mnt/test/baz btrfs rw,relatime,subvolid=256,subvol=/foo/bar 0 0

Despite subvolid=256 being subvol=/foo. This is because we're just spitting out the dentry of the mount point, which in the case of bind mounts is the source path for the mountpoint. Instead we should spit out the path to the actual subvol. Fix this by looking up the name for the subvolid we have mounted. With this fix the same test looks like this

/dev/sda /mnt/test btrfs rw,relatime,subvolid=256,subvol=/foo 0 0 /dev/sda /mnt/test/baz btrfs rw,relatime,subvolid=256,subvol=/foo 0 0

---
 fs/btrfs/super.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
fs/btrfs/super.c +8 -2
@@ -1310,6 +1310,7 @@ static int btrfs_show_options(struct seq
 {
 	struct btrfs_fs_info *info = btrfs_sb(dentry->d_sb);
 	const char *compress_type;
+	const char *subvol_name;
 
 	if (btrfs_test_opt(info, DEGRADED))
 		seq_puts(seq, ",degraded");
@@ -1396,8 +1397,13 @@ static int btrfs_show_options(struct seq
 		seq_puts(seq, ",ref_verify");
 	seq_printf(seq, ",subvolid=%llu",
 		  BTRFS_I(d_inode(dentry))->root->root_key.objectid);
-	seq_puts(seq, ",subvol=");
-	seq_dentry(seq, dentry, " \t\n\\");
+	subvol_name = btrfs_get_subvol_name_from_objectid(info,
+			BTRFS_I(d_inode(dentry))->root->root_key.objectid);
+	if (!IS_ERR(subvol_name)) {
+		seq_puts(seq, ",subvol=");
+		seq_escape(seq, subvol_name, " \t\n\\");
+		kfree(subvol_name);
+	}
 	return 0;
 }
 
author maintainer
#30 lore raw

PATCH 30/204 btrfs: fix messages after changing compression level by remount

In reply to Greg Kroah-Hartman

From: David Sterba <[email protected]>

commit 27942c9971cc405c60432eca9395e514a2ae9f5e upstream.

Reported by Forza on IRC that remounting with compression options does not reflect the change in level, or at least it does not appear to do so according to the messages:

mount -o compress=zstd:1 /dev/sda /mnt
mount -o remount,compress=zstd:15 /mnt

does not print the change to the level to syslog:

  [   41.366060] BTRFS info (device vda): use zstd compression, level 1
  [   41.368254] BTRFS info (device vda): disk space caching is enabled
  [   41.390429] BTRFS info (device vda): disk space caching is enabled

What really happens is that the message is lost but the level is actualy changed.

There's another weird output, if compression is reset to 'no':

  [   45.413776] BTRFS info (device vda): use no compression, level 4

To fix that, save the previous compression level and print the message in that case too and use separate message for 'no' compression.

---
 fs/btrfs/super.c |   14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)
fs/btrfs/super.c +9 -5
@@ -440,6 +440,7 @@ int btrfs_parse_options(struct btrfs_fs_
 	char *compress_type;
 	bool compress_force = false;
 	enum btrfs_compression_type saved_compress_type;
+	int saved_compress_level;
 	bool saved_compress_force;
 	int no_compress = 0;
 
@@ -522,6 +523,7 @@ int btrfs_parse_options(struct btrfs_fs_
 				info->compress_type : BTRFS_COMPRESS_NONE;
 			saved_compress_force =
 				btrfs_test_opt(info, FORCE_COMPRESS);
+			saved_compress_level = info->compress_level;
 			if (token == Opt_compress ||
 			    token == Opt_compress_force ||
 			    strncmp(args[0].from, "zlib", 4) == 0) {
@@ -566,6 +568,8 @@ int btrfs_parse_options(struct btrfs_fs_
 				no_compress = 0;
 			} else if (strncmp(args[0].from, "no", 2) == 0) {
 				compress_type = "no";
+				info->compress_level = 0;
+				info->compress_type = 0;
 				btrfs_clear_opt(info->mount_opt, COMPRESS);
 				btrfs_clear_opt(info->mount_opt, FORCE_COMPRESS);
 				compress_force = false;
@@ -586,11 +590,11 @@ int btrfs_parse_options(struct btrfs_fs_
 				 */
 				btrfs_clear_opt(info->mount_opt, FORCE_COMPRESS);
 			}
-			if ((btrfs_test_opt(info, COMPRESS) &&
-			     (info->compress_type != saved_compress_type ||
-			      compress_force != saved_compress_force)) ||
-			    (!btrfs_test_opt(info, COMPRESS) &&
-			     no_compress == 1)) {
+			if (no_compress == 1) {
+				btrfs_info(info, "use no compression");
+			} else if ((info->compress_type != saved_compress_type) ||
+				   (compress_force != saved_compress_force) ||
+				   (info->compress_level != saved_compress_level)) {
 				btrfs_info(info, "%s %s compression, level %d",
 					   (compress_force) ? "force" : "use",
 					   compress_type, info->compress_level);
author maintainer
#31 lore raw

PATCH 31/204 btrfs: only search for left_info if there is no right_info in try_merge_free_space

In reply to Greg Kroah-Hartman

From: Josef Bacik <[email protected]>

commit bf53d4687b8f3f6b752f091eb85f62369a515dfd upstream.

In try_to_merge_free_space we attempt to find entries to the left and right of the entry we are adding to see if they can be merged. We search for an entry past our current info (saved into right_info), and then if right_info exists and it has a rb_prev() we save the rb_prev() into left_info.

However there's a slight problem in the case that we have a right_info, but no entry previous to that entry. At that point we will search for an entry just before the info we're attempting to insert. This will simply find right_info again, and assign it to left_info, making them both the same pointer.

Now if right_info _can_ be merged with the range we're inserting, we'll add it to the info and free right_info. However further down we'll access left_info, which was right_info, and thus get a use-after-free.

Fix this by only searching for the left entry if we don't find a right entry at all.

The CVE referenced had a specially crafted file system that could trigger this use-after-free. However with the tree checker improvements we no longer trigger the conditions for the UAF. But the original conditions still apply, hence this fix.

Reference: CVE-2019-19448

Fixes 963030817060 ("Btrfs: use hybrid extents+bitmap rb tree for free space")
---
 fs/btrfs/free-space-cache.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/btrfs/free-space-cache.c +2 -2
@@ -2286,7 +2286,7 @@ out:
 static bool try_merge_free_space(struct btrfs_free_space_ctl *ctl,
 			  struct btrfs_free_space *info, bool update_stat)
 {
-	struct btrfs_free_space *left_info;
+	struct btrfs_free_space *left_info = NULL;
 	struct btrfs_free_space *right_info;
 	bool merged = false;
 	u64 offset = info->offset;
@@ -2302,7 +2302,7 @@ static bool try_merge_free_space(struct
 	if (right_info && rb_prev(&right_info->offset_index))
 		left_info = rb_entry(rb_prev(&right_info->offset_index),
 				     struct btrfs_free_space, offset_index);
-	else
+	else if (!right_info)
 		left_info = tree_search_offset(ctl, offset - 1, 0, 0);
 
 	/* See try_merge_free_space() comment. */
author maintainer
#32 lore raw

PATCH 32/204 btrfs: inode: fix NULL pointer dereference if inode doesnt need compression

In reply to Greg Kroah-Hartman

From: Qu Wenruo <[email protected]>

commit 1e6e238c3002ea3611465ce5f32777ddd6a40126 upstream.

[BUG]
There is a bug report of NULL pointer dereference caused in compress_file_extent():

  Oops: Kernel access of bad area, sig: 11 [#1]
  LE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA pSeries
  Workqueue: btrfs-delalloc btrfs_delalloc_helper [btrfs]
  NIP [c008000006dd4d34] compress_file_range.constprop.41+0x75c/0x8a0 [btrfs]
  LR [c008000006dd4d1c] compress_file_range.constprop.41+0x744/0x8a0 [btrfs]
  Call Trace:
  [c000000c69093b00] [c008000006dd4d1c] compress_file_range.constprop.41+0x744/0x8a0 [btrfs] (unreliable)
  [c000000c69093bd0] [c008000006dd4ebc] async_cow_start+0x44/0xa0 [btrfs]
  [c000000c69093c10] [c008000006e14824] normal_work_helper+0xdc/0x598 [btrfs]
  [c000000c69093c80] [c0000000001608c0] process_one_work+0x2c0/0x5b0
  [c000000c69093d10] [c000000000160c38] worker_thread+0x88/0x660
  [c000000c69093db0] [c00000000016b55c] kthread+0x1ac/0x1c0
  [c000000c69093e20] [c00000000000b660] ret_from_kernel_thread+0x5c/0x7c
  ---[ end trace f16954aa20d822f6 ]---

[CAUSE]
For the following execution route of compress_file_range(), it's possible to hit NULL pointer dereference:

compress_file_extent()

 |- pages = NULL;
 |- start = async_chunk->start = 0;
 |- end = async_chunk = 4095;
 |- nr_pages = 1;
 |- inode_need_compress() == false; <<< Possible, see later explanation
 |  Now, we have nr_pages = 1, pages = NULL
 |- cont:
 |- 		ret = cow_file_range_inline();
 |- 		if (ret <= 0) {
 |-		for (i = 0; i < nr_pages; i++) {
 |-			WARN_ON(pages[i]->mapping);	<<< Crash

To enter above call execution branch, we need the following race:

Thread 1 (chattr) | Thread 2 (writeback)
--------------------------+------------------------------

                          | btrfs_run_delalloc_range
                          | |- inode_need_compress = true
                          | |- cow_file_range_async()
btrfs_ioctl_set_flag()    |
|- binode_flags |=        |
   BTRFS_INODE_NOCOMPRESS |
                          | compress_file_range()
                          | |- inode_need_compress = false
                          | |- nr_page = 1 while pages = NULL
                          | |  Then hit the crash

[FIX]
This patch will fix it by checking @pages before doing accessing it. This patch is only designed as a hot fix and easy to backport.

More elegant fix may make btrfs only check inode_need_compress() once to avoid such race, but that would be another story.

Fixes 4d3a800ebb12 ("btrfs: merge nr_pages input and output parameter in compress_pages")
CC [email protected] # 4.14.x: cecc8d9038d16: btrfs: Move free_pages_out label in inline extent handling branch in compress_file_range
---
 fs/btrfs/inode.c |   16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)
fs/btrfs/inode.c +11 -5
@@ -650,12 +650,18 @@ cont:
 						     page_error_op |
 						     PAGE_END_WRITEBACK);
 
-			for (i = 0; i < nr_pages; i++) {
-				WARN_ON(pages[i]->mapping);
-				put_page(pages[i]);
+			/*
+			 * Ensure we only free the compressed pages if we have
+			 * them allocated, as we can still reach here with
+			 * inode_need_compress() == false.
+			 */
+			if (pages) {
+				for (i = 0; i < nr_pages; i++) {
+					WARN_ON(pages[i]->mapping);
+					put_page(pages[i]);
+				}
+				kfree(pages);
 			}
-			kfree(pages);
-
 			return 0;
 		}
 	}
author maintainer
#33 lore raw

PATCH 33/204 btrfs: fix memory leaks after failure to lookup checksums during inode logging

In reply to Greg Kroah-Hartman

From: Filipe Manana <[email protected]>

commit 4f26433e9b3eb7a55ed70d8f882ae9cd48ba448b upstream.

While logging an inode, at copy_items(), if we fail to lookup the checksums for an extent we release the destination path, free the ins_data array and then return immediately. However a previous iteration of the for loop may have added checksums to the ordered_sums list, in which case we leak the memory used by them.

So fix this by making sure we iterate the ordered_sums list and free all its checksums before returning.

Fixes 3650860b90cc2a ("Btrfs: remove almost all of the BUG()'s from tree-log.c")
---
 fs/btrfs/tree-log.c |    8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
fs/btrfs/tree-log.c +2 -6
@@ -4040,11 +4040,8 @@ static noinline int copy_items(struct bt
 						fs_info->csum_root,
 						ds + cs, ds + cs + cl - 1,
 						&ordered_sums, 0);
-				if (ret) {
-					btrfs_release_path(dst_path);
-					kfree(ins_data);
-					return ret;
-				}
+				if (ret)
+					break;
 			}
 		}
 	}
@@ -4057,7 +4054,6 @@ static noinline int copy_items(struct bt
 	 * we have to do this after the loop above to avoid changing the
 	 * log tree while trying to change the log tree.
 	 */
-	ret = 0;
 	while (!list_empty(&ordered_sums)) {
 		struct btrfs_ordered_sum *sums = list_entry(ordered_sums.next,
 						   struct btrfs_ordered_sum,
author maintainer
#34 lore raw

PATCH 34/204 btrfs: trim: fix underflow in trim length to prevent access beyond device boundary

In reply to Greg Kroah-Hartman

From: Qu Wenruo <[email protected]>

commit c57dd1f2f6a7cd1bb61802344f59ccdc5278c983 upstream.

[BUG]
The following script can lead to tons of beyond device boundary access:

mkfs.btrfs -f $dev -b 10G
mount $dev $mnt
trimfs $mnt
btrfs filesystem resize 1:-1G $mnt
trimfs $mnt

[CAUSE]
Since commit 929be17a9b49 ("btrfs: Switch btrfs_trim_free_extents to find_first_clear_extent_bit"), we try to avoid trimming ranges that's already trimmed.

So we check device->alloc_state by finding the first range which doesn't have CHUNK_TRIMMED and CHUNK_ALLOCATED not set.

But if we shrunk the device, that bits are not cleared, thus we could easily got a range starts beyond the shrunk device size.

This results the returned @start and @end are all beyond device size, then we call "end = min(end, device->total_bytes -1);" making @end smaller than device size.

Then finally we goes "len = end - start + 1", totally underflow the result, and lead to the beyond-device-boundary access.

[FIX]
This patch will fix the problem in two ways:

- Clear CHUNK_TRIMMED | CHUNK_ALLOCATED bits when shrinking device This is the root fix

- Add extra safety check when trimming free device extents We check and warn if the returned range is already beyond current device.

Fixes 929be17a9b49 ("btrfs: Switch btrfs_trim_free_extents to find_first_clear_extent_bit")
---
 fs/btrfs/extent-io-tree.h |    2 ++
 fs/btrfs/extent-tree.c    |   14 ++++++++++++++
 fs/btrfs/volumes.c        |    4 ++++
 3 files changed, 20 insertions(+)
fs/btrfs/extent-io-tree.h +2 -0
@@ -34,6 +34,8 @@ struct io_failure_record;
  */
 #define CHUNK_ALLOCATED				EXTENT_DIRTY
 #define CHUNK_TRIMMED				EXTENT_DEFRAG
+#define CHUNK_STATE_MASK			(CHUNK_ALLOCATED |		\
+						 CHUNK_TRIMMED)
 
 enum {
 	IO_TREE_FS_PINNED_EXTENTS,
fs/btrfs/extent-tree.c +14 -0
@@ -33,6 +33,7 @@
 #include "delalloc-space.h"
 #include "block-group.h"
 #include "discard.h"
+#include "rcu-string.h"
 
 #undef SCRAMBLE_DELAYED_REFS
 
@@ -5685,6 +5686,19 @@ static int btrfs_trim_free_extents(struc
 					    &start, &end,
 					    CHUNK_TRIMMED | CHUNK_ALLOCATED);
 
+		/* Check if there are any CHUNK_* bits left */
+		if (start > device->total_bytes) {
+			WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG));
+			btrfs_warn_in_rcu(fs_info,
+"ignoring attempt to trim beyond device size: offset %llu length %llu device %s device size %llu",
+					  start, end - start + 1,
+					  rcu_str_deref(device->name),
+					  device->total_bytes);
+			mutex_unlock(&fs_info->chunk_mutex);
+			ret = 0;
+			break;
+		}
+
 		/* Ensure we skip the reserved area in the first 1M */
 		start = max_t(u64, start, SZ_1M);
 
fs/btrfs/volumes.c +4 -0
@@ -4724,6 +4724,10 @@ again:
 	}
 
 	mutex_lock(&fs_info->chunk_mutex);
+	/* Clear all state bits beyond the shrunk device size */
+	clear_extent_bits(&device->alloc_state, new_size, (u64)-1,
+			  CHUNK_STATE_MASK);
+
 	btrfs_device_set_disk_total_bytes(device, new_size);
 	if (list_empty(&device->post_commit_list))
 		list_add_tail(&device->post_commit_list,
author maintainer
#35 lore raw

PATCH 35/204 btrfs: make sure SB_I_VERSION doesnt get unset by remount

In reply to Greg Kroah-Hartman

From: Josef Bacik <[email protected]>

commit faa008899a4db21a2df99833cb4ff6fa67009a20 upstream.

There's some inconsistency around SB_I_VERSION handling with mount and remount. Since we don't really want it to be off ever just work around this by making sure we don't get the flag cleared on remount.

There's a tiny cpu cost of setting the bit, otherwise all changes to i_version also change some of the times (ctime/mtime) so the inode needs to be synced. We wouldn't save anything by disabling it.

---
 fs/btrfs/super.c |    6 ++++++
 1 file changed, 6 insertions(+)
fs/btrfs/super.c +6 -0
@@ -1895,6 +1895,12 @@ static int btrfs_remount(struct super_bl
 		set_bit(BTRFS_FS_OPEN, &fs_info->flags);
 	}
 out:
+	/*
+	 * We need to set SB_I_VERSION here otherwise it'll get cleared by VFS,
+	 * since the absence of the flag means it can be toggled off by remount.
+	 */
+	*flags |= SB_I_VERSION;
+
 	wake_up_process(fs_info->transaction_kthread);
 	btrfs_remount_cleanup(fs_info, old_opts);
 	return 0;
author maintainer
#36 lore raw

PATCH 36/204 btrfs: fix return value mixup in btrfs_get_extent

In reply to Greg Kroah-Hartman

From: Pavel Machek <[email protected]>

commit 881a3a11c2b858fe9b69ef79ac5ee9978a266dc9 upstream.

btrfs_get_extent() sets variable ret, but out: error path expect error to be in variable err so the error code is lost.

Fixes 6bf9e4bd6a27 ("btrfs: inode: Verify inode mode to avoid NULL pointer dereference")
---
 fs/btrfs/inode.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/inode.c +1 -1
@@ -6640,7 +6640,7 @@ struct extent_map *btrfs_get_extent(stru
 	    extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
 		/* Only regular file could have regular/prealloc extent */
 		if (!S_ISREG(inode->vfs_inode.i_mode)) {
-			ret = -EUCLEAN;
+			err = -EUCLEAN;
 			btrfs_crit(fs_info,
 		"regular/prealloc extent found for non-regular inode %llu",
 				   btrfs_ino(inode));
author maintainer
#37 lore raw

PATCH 37/204 arm64: perf: Correct the event index in sysfs

In reply to Greg Kroah-Hartman

From: Shaokun Zhang <[email protected]>

commit 539707caa1a89ee4efc57b4e4231c20c46575ccc upstream.

When PMU event ID is equal or greater than 0x4000, it will be reduced by 0x4000 and it is not the raw number in the sysfs. Let's correct it and obtain the raw event ID.

Before this patch:
cat /sys/bus/event_source/devices/armv8_pmuv3_0/events/sample_feed event=0x001
After this patch:
cat /sys/bus/event_source/devices/armv8_pmuv3_0/events/sample_feed event=0x4001

---
 arch/arm64/kernel/perf_event.c |   13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)
arch/arm64/kernel/perf_event.c +8 -5
@@ -155,7 +155,7 @@ armv8pmu_events_sysfs_show(struct device
 
 	pmu_attr = container_of(attr, struct perf_pmu_events_attr, attr);
 
-	return sprintf(page, "event=0x%03llx\n", pmu_attr->id);
+	return sprintf(page, "event=0x%04llx\n", pmu_attr->id);
 }
 
 #define ARMV8_EVENT_ATTR(name, config)						\
@@ -244,10 +244,13 @@ armv8pmu_event_attr_is_visible(struct ko
 	    test_bit(pmu_attr->id, cpu_pmu->pmceid_bitmap))
 		return attr->mode;
 
-	pmu_attr->id -= ARMV8_PMUV3_EXT_COMMON_EVENT_BASE;
-	if (pmu_attr->id < ARMV8_PMUV3_MAX_COMMON_EVENTS &&
-	    test_bit(pmu_attr->id, cpu_pmu->pmceid_ext_bitmap))
-		return attr->mode;
+	if (pmu_attr->id >= ARMV8_PMUV3_EXT_COMMON_EVENT_BASE) {
+		u64 id = pmu_attr->id - ARMV8_PMUV3_EXT_COMMON_EVENT_BASE;
+
+		if (id < ARMV8_PMUV3_MAX_COMMON_EVENTS &&
+		    test_bit(id, cpu_pmu->pmceid_ext_bitmap))
+			return attr->mode;
+	}
 
 	return 0;
 }
author maintainer
#38 lore raw

PATCH 38/204 dt-bindings: iio: io-channel-mux: Fix compatible string in example code

In reply to Greg Kroah-Hartman

From: Christian Eggers <[email protected]>

commit add48ba425192c6e04ce70549129cacd01e2a09e upstream.

The correct compatible string is "gpio-mux" (see bindings/mux/gpio-mux.txt).

---
 Documentation/devicetree/bindings/iio/multiplexer/io-channel-mux.txt |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/devicetree/bindings/iio/multiplexer/io-channel-mux.txt +1 -1
@@ -21,7 +21,7 @@ controller state. The mux controller sta
 
 Example:
 	mux: mux-controller {
-		compatible = "mux-gpio";
+		compatible = "gpio-mux";
 		#mux-control-cells = <0>;
 
 		mux-gpios = <&pioA 0 GPIO_ACTIVE_HIGH>,
author maintainer
#39 lore raw

PATCH 39/204 iio: dac: ad5592r: fix unbalanced mutex unlocks in ad5592r_read_raw()

In reply to Greg Kroah-Hartman

From: Alexandru Ardelean <[email protected]>

commit 65afb0932a81c1de719ceee0db0b276094b10ac8 upstream.

There are 2 exit paths where the lock isn't held, but try to unlock the mutex when exiting. In these places we should just return from the function.

A neater approach would be to cleanup the ad5592r_read_raw(), but that would make this patch more difficult to backport to stable versions.

Fixes 56ca9db862bf3: ("iio: dac: Add support for the AD5592R/AD5593R ADCs/DACs")

---
 drivers/iio/dac/ad5592r-base.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/iio/dac/ad5592r-base.c +2 -2
@@ -415,7 +415,7 @@ static int ad5592r_read_raw(struct iio_d
 			s64 tmp = *val * (3767897513LL / 25LL);
 			*val = div_s64_rem(tmp, 1000000000LL, val2);
 
-			ret = IIO_VAL_INT_PLUS_MICRO;
+			return IIO_VAL_INT_PLUS_MICRO;
 		} else {
 			int mult;
 
@@ -446,7 +446,7 @@ static int ad5592r_read_raw(struct iio_d
 		ret =  IIO_VAL_INT;
 		break;
 	default:
-		ret = -EINVAL;
+		return -EINVAL;
 	}
 
 unlock:
author maintainer
#40 lore raw

PATCH 40/204 iio: imu: st_lsm6dsx: reset hw ts after resume

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

commit a1bab9396c2d98c601ce81c27567159dfbc10c19 upstream.

Reset hw time samples generator after system resume in order to avoid disalignment between system and device time reference since FIFO batching and time samples generator are disabled during suspend.

Fixes 213451076bd3 ("iio: imu: st_lsm6dsx: add hw timestamp support")
---
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx.h        |    3 +--
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c |   23 +++++++++++++++--------
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c   |    2 +-
 3 files changed, 17 insertions(+), 11 deletions(-)
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx.h +1 -2
@@ -436,8 +436,7 @@ int st_lsm6dsx_update_watermark(struct s
 				u16 watermark);
 int st_lsm6dsx_update_fifo(struct st_lsm6dsx_sensor *sensor, bool enable);
 int st_lsm6dsx_flush_fifo(struct st_lsm6dsx_hw *hw);
-int st_lsm6dsx_set_fifo_mode(struct st_lsm6dsx_hw *hw,
-			     enum st_lsm6dsx_fifo_mode fifo_mode);
+int st_lsm6dsx_resume_fifo(struct st_lsm6dsx_hw *hw);
 int st_lsm6dsx_read_fifo(struct st_lsm6dsx_hw *hw);
 int st_lsm6dsx_read_tagged_fifo(struct st_lsm6dsx_hw *hw);
 int st_lsm6dsx_check_odr(struct st_lsm6dsx_sensor *sensor, u32 odr, u8 *val);
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c +15 -8
@@ -184,8 +184,8 @@ static int st_lsm6dsx_update_decimators(
 	return err;
 }
 
-int st_lsm6dsx_set_fifo_mode(struct st_lsm6dsx_hw *hw,
-			     enum st_lsm6dsx_fifo_mode fifo_mode)
+static int st_lsm6dsx_set_fifo_mode(struct st_lsm6dsx_hw *hw,
+				    enum st_lsm6dsx_fifo_mode fifo_mode)
 {
 	unsigned int data;
 
@@ -302,6 +302,18 @@ static int st_lsm6dsx_reset_hw_ts(struct
 	return 0;
 }
 
+int st_lsm6dsx_resume_fifo(struct st_lsm6dsx_hw *hw)
+{
+	int err;
+
+	/* reset hw ts counter */
+	err = st_lsm6dsx_reset_hw_ts(hw);
+	if (err < 0)
+		return err;
+
+	return st_lsm6dsx_set_fifo_mode(hw, ST_LSM6DSX_FIFO_CONT);
+}
+
 /*
  * Set max bulk read to ST_LSM6DSX_MAX_WORD_LEN/ST_LSM6DSX_MAX_TAGGED_WORD_LEN
  * in order to avoid a kmalloc for each bus access
@@ -675,12 +687,7 @@ int st_lsm6dsx_update_fifo(struct st_lsm
 		goto out;
 
 	if (fifo_mask) {
-		/* reset hw ts counter */
-		err = st_lsm6dsx_reset_hw_ts(hw);
-		if (err < 0)
-			goto out;
-
-		err = st_lsm6dsx_set_fifo_mode(hw, ST_LSM6DSX_FIFO_CONT);
+		err = st_lsm6dsx_resume_fifo(hw);
 		if (err < 0)
 			goto out;
 	}
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -1
@@ -2451,7 +2451,7 @@ static int __maybe_unused st_lsm6dsx_res
 	}
 
 	if (hw->fifo_mask)
-		err = st_lsm6dsx_set_fifo_mode(hw, ST_LSM6DSX_FIFO_CONT);
+		err = st_lsm6dsx_resume_fifo(hw);
 
 	return err;
 }
author maintainer
#41 lore raw

PATCH 41/204 xtensa: add missing exclusive access state management

In reply to Greg Kroah-Hartman

From: Max Filippov <[email protected]>

commit a0fc1436f1f4f84e93144480bf30e0c958d135b6 upstream.

The result of the s32ex opcode is recorded in the ATOMCTL special register and must be retrieved with the getex opcode. Context switch between s32ex and getex may trash the ATOMCTL register and result in duplicate update or missing update of the atomic variable. Add atomctl8 field to the struct thread_info and use getex to swap ATOMCTL bit 8 as a part of context switch.
Clear exclusive access monitor on kernel entry.

Fixes f7c34874f04a ("xtensa: add exclusive atomics support")
---
 arch/xtensa/include/asm/thread_info.h |    4 ++++
 arch/xtensa/kernel/asm-offsets.c      |    3 +++
 arch/xtensa/kernel/entry.S            |   11 +++++++++++
 3 files changed, 18 insertions(+)
arch/xtensa/include/asm/thread_info.h +4 -0
@@ -55,6 +55,10 @@ struct thread_info {
 	mm_segment_t		addr_limit;	/* thread address space */
 
 	unsigned long		cpenable;
+#if XCHAL_HAVE_EXCLUSIVE
+	/* result of the most recent exclusive store */
+	unsigned long		atomctl8;
+#endif
 
 	/* Allocate storage for extra user states and coprocessor states. */
 #if XTENSA_HAVE_COPROCESSORS
arch/xtensa/kernel/asm-offsets.c +3 -0
@@ -93,6 +93,9 @@ int main(void)
 	DEFINE(THREAD_RA, offsetof (struct task_struct, thread.ra));
 	DEFINE(THREAD_SP, offsetof (struct task_struct, thread.sp));
 	DEFINE(THREAD_CPENABLE, offsetof (struct thread_info, cpenable));
+#if XCHAL_HAVE_EXCLUSIVE
+	DEFINE(THREAD_ATOMCTL8, offsetof (struct thread_info, atomctl8));
+#endif
 #if XTENSA_HAVE_COPROCESSORS
 	DEFINE(THREAD_XTREGS_CP0, offsetof(struct thread_info, xtregs_cp.cp0));
 	DEFINE(THREAD_XTREGS_CP1, offsetof(struct thread_info, xtregs_cp.cp1));
arch/xtensa/kernel/entry.S +11 -0
@@ -374,6 +374,11 @@ common_exception:
 	s32i	a2, a1, PT_LCOUNT
 #endif
 
+#if XCHAL_HAVE_EXCLUSIVE
+	/* Clear exclusive access monitor set by interrupted code */
+	clrex
+#endif
+
 	/* It is now save to restore the EXC_TABLE_FIXUP variable. */
 
 	rsr	a2, exccause
@@ -2020,6 +2025,12 @@ ENTRY(_switch_to)
 	s32i	a3, a4, THREAD_CPENABLE
 #endif
 
+#if XCHAL_HAVE_EXCLUSIVE
+	l32i	a3, a5, THREAD_ATOMCTL8
+	getex	a3
+	s32i	a3, a4, THREAD_ATOMCTL8
+#endif
+
 	/* Flush register file. */
 
 	spill_registers_kernel
author maintainer
#42 lore raw

PATCH 42/204 xtensa: fix xtensa_pmu_setup prototype

In reply to Greg Kroah-Hartman

From: Max Filippov <[email protected]>

commit 6d65d3769d1910379e1cfa61ebf387efc6bfb22c upstream.

Fix the following build error in configurations with CONFIG_XTENSA_VARIANT_HAVE_PERF_EVENTS=y:

arch/xtensa/kernel/perf_event.c:420:29: error: passing argument 3 of ‘cpuhp_setup_state’ from incompatible pointer type

Fixes 25a77b55e74c ("xtensa/perf: Convert the hotplug notifier to state machine callbacks")
---
 arch/xtensa/kernel/perf_event.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/xtensa/kernel/perf_event.c +1 -1
@@ -401,7 +401,7 @@ static struct pmu xtensa_pmu = {
 	.read = xtensa_pmu_read,
 };
 
-static int xtensa_pmu_setup(int cpu)
+static int xtensa_pmu_setup(unsigned int cpu)
 {
 	unsigned i;
 
author maintainer
#43 lore raw

PATCH 43/204 cifs: Fix leak when handling lease break for cached root fid

In reply to Greg Kroah-Hartman

From: Paul Aurich <[email protected]>

commit baf57b56d3604880ccb3956ec6c62ea894f5de99 upstream.

Handling a lease break for the cached root didn't free the smb2_lease_break_work allocation, resulting in a leak:

    unreferenced object 0xffff98383a5af480 (size 128):
      comm "cifsd", pid 684, jiffies 4294936606 (age 534.868s)
      hex dump (first 32 bytes):
        c0 ff ff ff 1f 00 00 00 88 f4 5a 3a 38 98 ff ff  ..........Z:8...
        88 f4 5a 3a 38 98 ff ff 80 88 d6 8a ff ff ff ff  ..Z:8...........
      backtrace:
        [<0000000068957336>] smb2_is_valid_oplock_break+0x1fa/0x8c0
        [<0000000073b70b9e>] cifs_demultiplex_thread+0x73d/0xcc0
        [<00000000905fa372>] kthread+0x11c/0x150
        [<0000000079378e4e>] ret_from_fork+0x22/0x30

Avoid this leak by only allocating when necessary.

Fixes a93864d93977 ("cifs: add lease tracking to the cached root fid")
CC Stable <[email protected]> # v4.18+
---
 fs/cifs/smb2misc.c |   73 +++++++++++++++++++++++++++++++++++++----------------
 1 file changed, 52 insertions(+), 21 deletions(-)
fs/cifs/smb2misc.c +52 -21
@@ -508,15 +508,31 @@ cifs_ses_oplock_break(struct work_struct
 	kfree(lw);
 }
 
+static void
+smb2_queue_pending_open_break(struct tcon_link *tlink, __u8 *lease_key,
+			      __le32 new_lease_state)
+{
+	struct smb2_lease_break_work *lw;
+
+	lw = kmalloc(sizeof(struct smb2_lease_break_work), GFP_KERNEL);
+	if (!lw) {
+		cifs_put_tlink(tlink);
+		return;
+	}
+
+	INIT_WORK(&lw->lease_break, cifs_ses_oplock_break);
+	lw->tlink = tlink;
+	lw->lease_state = new_lease_state;
+	memcpy(lw->lease_key, lease_key, SMB2_LEASE_KEY_SIZE);
+	queue_work(cifsiod_wq, &lw->lease_break);
+}
+
 static bool
-smb2_tcon_has_lease(struct cifs_tcon *tcon, struct smb2_lease_break *rsp,
-		    struct smb2_lease_break_work *lw)
+smb2_tcon_has_lease(struct cifs_tcon *tcon, struct smb2_lease_break *rsp)
 {
-	bool found;
 	__u8 lease_state;
 	struct list_head *tmp;
 	struct cifsFileInfo *cfile;
-	struct cifs_pending_open *open;
 	struct cifsInodeInfo *cinode;
 	int ack_req = le32_to_cpu(rsp->Flags &
 				  SMB2_NOTIFY_BREAK_LEASE_FLAG_ACK_REQUIRED);
@@ -546,22 +562,29 @@ smb2_tcon_has_lease(struct cifs_tcon *tc
 		cfile->oplock_level = lease_state;
 
 		cifs_queue_oplock_break(cfile);
-		kfree(lw);
 		return true;
 	}
 
-	found = false;
+	return false;
+}
+
+static struct cifs_pending_open *
+smb2_tcon_find_pending_open_lease(struct cifs_tcon *tcon,
+				  struct smb2_lease_break *rsp)
+{
+	__u8 lease_state = le32_to_cpu(rsp->NewLeaseState);
+	int ack_req = le32_to_cpu(rsp->Flags &
+				  SMB2_NOTIFY_BREAK_LEASE_FLAG_ACK_REQUIRED);
+	struct cifs_pending_open *open;
+	struct cifs_pending_open *found = NULL;
+
 	list_for_each_entry(open, &tcon->pending_opens, olist) {
 		if (memcmp(open->lease_key, rsp->LeaseKey,
 			   SMB2_LEASE_KEY_SIZE))
 			continue;
 
 		if (!found && ack_req) {
-			found = true;
-			memcpy(lw->lease_key, open->lease_key,
-			       SMB2_LEASE_KEY_SIZE);
-			lw->tlink = cifs_get_tlink(open->tlink);
-			queue_work(cifsiod_wq, &lw->lease_break);
+			found = open;
 		}
 
 		cifs_dbg(FYI, "found in the pending open list\n");
@@ -582,14 +605,7 @@ smb2_is_valid_lease_break(char *buffer)
 	struct TCP_Server_Info *server;
 	struct cifs_ses *ses;
 	struct cifs_tcon *tcon;
-	struct smb2_lease_break_work *lw;
-
-	lw = kmalloc(sizeof(struct smb2_lease_break_work), GFP_KERNEL);
-	if (!lw)
-		return false;
-
-	INIT_WORK(&lw->lease_break, cifs_ses_oplock_break);
-	lw->lease_state = rsp->NewLeaseState;
+	struct cifs_pending_open *open;
 
 	cifs_dbg(FYI, "Checking for lease break\n");
 
@@ -607,11 +623,27 @@ smb2_is_valid_lease_break(char *buffer)
 				spin_lock(&tcon->open_file_lock);
 				cifs_stats_inc(
 				    &tcon->stats.cifs_stats.num_oplock_brks);
-				if (smb2_tcon_has_lease(tcon, rsp, lw)) {
+				if (smb2_tcon_has_lease(tcon, rsp)) {
 					spin_unlock(&tcon->open_file_lock);
 					spin_unlock(&cifs_tcp_ses_lock);
 					return true;
 				}
+				open = smb2_tcon_find_pending_open_lease(tcon,
+									 rsp);
+				if (open) {
+					__u8 lease_key[SMB2_LEASE_KEY_SIZE];
+					struct tcon_link *tlink;
+
+					tlink = cifs_get_tlink(open->tlink);
+					memcpy(lease_key, open->lease_key,
+					       SMB2_LEASE_KEY_SIZE);
+					spin_unlock(&tcon->open_file_lock);
+					spin_unlock(&cifs_tcp_ses_lock);
+					smb2_queue_pending_open_break(tlink,
+								      lease_key,
+								      rsp->NewLeaseState);
+					return true;
+				}
 				spin_unlock(&tcon->open_file_lock);
 
 				if (tcon->crfid.is_valid &&
@@ -629,7 +661,6 @@ smb2_is_valid_lease_break(char *buffer)
 		}
 	}
 	spin_unlock(&cifs_tcp_ses_lock);
-	kfree(lw);
 	cifs_dbg(FYI, "Can not process lease break - no lease matched\n");
 	return false;
 }
author maintainer
#44 lore raw

PATCH 44/204 powerpc/ptdump: Fix build failure in hashpagetable.c

In reply to Greg Kroah-Hartman

From: Christophe Leroy <[email protected]>

commit 7c466b0807960edc13e4b855be85ea765df9a6cd upstream.

H_SUCCESS is only defined when CONFIG_PPC_PSERIES is defined.

!= H_SUCCESS means != 0. Modify the test accordingly.

---
 arch/powerpc/mm/ptdump/hashpagetable.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/mm/ptdump/hashpagetable.c +1 -1
@@ -259,7 +259,7 @@ static int pseries_find(unsigned long ea
 	for (i = 0; i < HPTES_PER_GROUP; i += 4, hpte_group += 4) {
 		lpar_rc = plpar_pte_read_4(0, hpte_group, (void *)ptes);
 
-		if (lpar_rc != H_SUCCESS)
+		if (lpar_rc)
 			continue;
 		for (j = 0; j < 4; j++) {
 			if (HPTE_V_COMPARE(ptes[j].v, want_v) &&
author maintainer
#45 lore raw

PATCH 45/204 powerpc: Allow 4224 bytes of stack expansion for the signal frame

In reply to Greg Kroah-Hartman

From: Michael Ellerman <[email protected]>

commit 63dee5df43a31f3844efabc58972f0a206ca4534 upstream.

We have powerpc specific logic in our page fault handling to decide if an access to an unmapped address below the stack pointer should expand the stack VMA.

The code was originally added in 2004 "ported from 2.4". The rough logic is that the stack is allowed to grow to 1MB with no extra checking. Over 1MB the access must be within 2048 bytes of the stack pointer, or be from a user instruction that updates the stack pointer.

The 2048 byte allowance below the stack pointer is there to cover the 288 byte "red zone" as well as the "about 1.5kB" needed by the signal delivery code.

Unfortunately since then the signal frame has expanded, and is now 4224 bytes on 64-bit kernels with transactional memory enabled. This means if a process has consumed more than 1MB of stack, and its stack pointer lies less than 4224 bytes from the next page boundary, signal delivery will fault when trying to expand the stack and the process will see a SEGV.

The total size of the signal frame is the size of struct rt_sigframe (which includes the red zone) plus __SIGNAL_FRAMESIZE (128 bytes on 64-bit).

The 2048 byte allowance was correct until 2008 as the signal frame was:

struct rt_sigframe {
struct ucontext uc; /* 0 1440 */
/* --- cacheline 11 boundary (1408 bytes) was 32 bytes ago --- */
long unsigned int _unused[2]; /* 1440 16 */
unsigned int tramp[6]; /* 1456 24 */
struct siginfo * pinfo; /* 1480 8 */
void * puc; /* 1488 8 */
struct siginfo info; /* 1496 128 */
/* --- cacheline 12 boundary (1536 bytes) was 88 bytes ago --- */
char abigap[288]; /* 1624 288 */

/* size: 1920, cachelines: 15, members: 7 */ /* padding: 8 */
};

1920 + 128 = 2048

Then in commit ce48b2100785 ("powerpc: Add VSX context save/restore, ptrace and signal support") (Jul 2008) the signal frame expanded to 2304 bytes:

struct rt_sigframe {
struct ucontext uc; /* 0 1696 */ <--
/* --- cacheline 13 boundary (1664 bytes) was 32 bytes ago --- */
long unsigned int _unused[2]; /* 1696 16 */
unsigned int tramp[6]; /* 1712 24 */
struct siginfo * pinfo; /* 1736 8 */
void * puc; /* 1744 8 */
struct siginfo info; /* 1752 128 */
/* --- cacheline 14 boundary (1792 bytes) was 88 bytes ago --- */
char abigap[288]; /* 1880 288 */

/* size: 2176, cachelines: 17, members: 7 */ /* padding: 8 */
};

2176 + 128 = 2304

At this point we should have been exposed to the bug, though as far as I know it was never reported. I no longer have a system old enough to easily test on.

Then in 2010 commit 320b2b8de126 ("mm: keep a guard page below a grow-down stack segment") caused our stack expansion code to never trigger, as there was always a VMA found for a write up to PAGE_SIZE below r1.

That meant the bug was hidden as we continued to expand the signal frame in commit 2b0a576d15e0 ("powerpc: Add new transactional memory state to the signal context") (Feb 2013):

struct rt_sigframe {
        struct ucontext    uc;                           /*     0  1696 */
        /* --- cacheline 13 boundary (1664 bytes) was 32 bytes ago --- */
        struct ucontext    uc_transact;                  /*  1696  1696 */	<--
        /* --- cacheline 26 boundary (3328 bytes) was 64 bytes ago --- */
        long unsigned int          _unused[2];           /*  3392    16 */
        unsigned int               tramp[6];             /*  3408    24 */
        struct siginfo *           pinfo;                /*  3432     8 */
        void *                     puc;                  /*  3440     8 */
        struct siginfo     info;                         /*  3448   128 */
        /* --- cacheline 27 boundary (3456 bytes) was 120 bytes ago --- */
        char                       abigap[288];          /*  3576   288 */

        /* size: 3872, cachelines: 31, members: 8 */
        /* padding: 8 */
        /* last cacheline: 32 bytes */
};

3872 + 128 = 4000

And commit 573ebfa6601f ("powerpc: Increase stack redzone for 64-bit userspace to 512 bytes") (Feb 2014):

struct rt_sigframe {
struct ucontext uc; /* 0 1696 */
/* --- cacheline 13 boundary (1664 bytes) was 32 bytes ago --- */
struct ucontext uc_transact; /* 1696 1696 */
/* --- cacheline 26 boundary (3328 bytes) was 64 bytes ago --- */
long unsigned int _unused[2]; /* 3392 16 */
unsigned int tramp[6]; /* 3408 24 */
struct siginfo * pinfo; /* 3432 8 */
void * puc; /* 3440 8 */
struct siginfo info; /* 3448 128 */
/* --- cacheline 27 boundary (3456 bytes) was 120 bytes ago --- */
char abigap[512]; /* 3576 512 */ <--

/* size: 4096, cachelines: 32, members: 8 */ /* padding: 8 */
};

4096 + 128 = 4224

Then finally in 2017, commit 1be7107fbe18 ("mm: larger stack guard gap, between vmas") exposed us to the existing bug, because it changed the stack VMA to be the correct/real size, meaning our stack expansion code is now triggered.

Fix it by increasing the allowance to 4224 bytes.

Hard-coding 4224 is obviously unsafe against future expansions of the signal frame in the same way as the existing code. We can't easily use sizeof() because the signal frame structure is not in a header. We will either fix that, or rip out all the custom stack expansion checking logic entirely.

Fixes ce48b2100785 ("powerpc: Add VSX context save/restore, ptrace and signal support")
Cc [email protected] # v2.6.27+
---
 arch/powerpc/mm/fault.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
arch/powerpc/mm/fault.c +5 -2
@@ -241,6 +241,9 @@ static bool bad_kernel_fault(struct pt_r
 	return false;
 }
 
+// This comes from 64-bit struct rt_sigframe + __SIGNAL_FRAMESIZE
+#define SIGFRAME_MAX_SIZE	(4096 + 128)
+
 static bool bad_stack_expansion(struct pt_regs *regs, unsigned long address,
 				struct vm_area_struct *vma, unsigned int flags,
 				bool *must_retry)
@@ -248,7 +251,7 @@ static bool bad_stack_expansion(struct p
 	/*
 	 * N.B. The POWER/Open ABI allows programs to access up to
 	 * 288 bytes below the stack pointer.
-	 * The kernel signal delivery code writes up to about 1.5kB
+	 * The kernel signal delivery code writes a bit over 4KB
 	 * below the stack pointer (r1) before decrementing it.
 	 * The exec code can write slightly over 640kB to the stack
 	 * before setting the user r1.  Thus we allow the stack to
@@ -273,7 +276,7 @@ static bool bad_stack_expansion(struct p
 		 * between the last mapped region and the stack will
 		 * expand the stack rather than segfaulting.
 		 */
-		if (address + 2048 >= uregs->gpr[1])
+		if (address + SIGFRAME_MAX_SIZE >= uregs->gpr[1])
 			return false;
 
 		if ((flags & FAULT_FLAG_WRITE) && (flags & FAULT_FLAG_USER) &&
author maintainer
#46 lore raw

PATCH 46/204 powerpc: Fix circular dependency between percpu.h and mmu.h

In reply to Greg Kroah-Hartman

From: Michael Ellerman <[email protected]>

commit 0c83b277ada72b585e6a3e52b067669df15bcedb upstream.

Recently random.h started including percpu.h (see commit f227e3ec3b5c ("random32: update the net random state on interrupt and activity")), which broke corenet64_smp_defconfig:

In file included from /linux/arch/powerpc/include/asm/paca.h:18,

                   from /linux/arch/powerpc/include/asm/percpu.h:13,
                   from /linux/include/linux/random.h:14,
                   from /linux/lib/uuid.c:14:
  /linux/arch/powerpc/include/asm/mmu.h:139:22: error: unknown type name 'next_tlbcam_idx'
    139 | DECLARE_PER_CPU(int, next_tlbcam_idx);

This is due to a circular header dependency:
asm/mmu.h includes asm/percpu.h, which includes asm/paca.h, which includes asm/mmu.h

Which means DECLARE_PER_CPU() isn't defined when mmu.h needs it.

We can fix it by moving the include of paca.h below the include of asm-generic/percpu.h.

This moves the include of paca.h out of the #ifdef __powerpc64__, but that is OK because paca.h is almost entirely inside #ifdef CONFIG_PPC64 anyway.

It also moves the include of paca.h out of the #ifdef CONFIG_SMP, which could possibly break something, but seems to have no ill effects.

Fixes f227e3ec3b5c ("random32: update the net random state on interrupt and activity")
---
 arch/powerpc/include/asm/percpu.h |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/include/asm/percpu.h +2 -2
@@ -10,8 +10,6 @@
 
 #ifdef CONFIG_SMP
 
-#include <asm/paca.h>
-
 #define __my_cpu_offset local_paca->data_offset
 
 #endif /* CONFIG_SMP */
@@ -19,4 +17,6 @@
 
 #include <asm-generic/percpu.h>
 
+#include <asm/paca.h>
+
 #endif /* _ASM_POWERPC_PERCPU_H_ */
author maintainer
#47 lore raw

PATCH 47/204 pinctrl: ingenic: Enhance support for IRQ_TYPE_EDGE_BOTH

In reply to Greg Kroah-Hartman

From: Paul Cercueil <[email protected]>

commit 1c95348ba327fe8621d3680890c2341523d3524a upstream.

Ingenic SoCs don't natively support registering an interrupt for both rising and falling edges. This has to be emulated in software.

Until now, this was emulated by switching back and forth between IRQ_TYPE_EDGE_RISING and IRQ_TYPE_EDGE_FALLING according to the level of the GPIO. While this worked most of the time, when used with GPIOs that need debouncing, some events would be lost. For instance, between the time a falling-edge interrupt happens and the interrupt handler configures the hardware for rising-edge, the level of the pin may have already risen, and the rising-edge event is lost.

To address that issue, instead of switching back and forth between IRQ_TYPE_EDGE_RISING and IRQ_TYPE_EDGE_FALLING, we now switch back and forth between IRQ_TYPE_LEVEL_LOW and IRQ_TYPE_LEVEL_HIGH. Since we always switch in the interrupt handler, they actually permit to detect level changes. In the example above, if the pin level rises before switching the IRQ type from IRQ_TYPE_LEVEL_LOW to IRQ_TYPE_LEVEL_HIGH, a new interrupt will raise as soon as the handler exits, and the rising-edge event will be properly detected.

---
 drivers/pinctrl/pinctrl-ingenic.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/pinctrl/pinctrl-ingenic.c +3 -3
@@ -1810,9 +1810,9 @@ static void ingenic_gpio_irq_ack(struct
 		 */
 		high = ingenic_gpio_get_value(jzgc, irq);
 		if (high)
-			irq_set_type(jzgc, irq, IRQ_TYPE_EDGE_FALLING);
+			irq_set_type(jzgc, irq, IRQ_TYPE_LEVEL_LOW);
 		else
-			irq_set_type(jzgc, irq, IRQ_TYPE_EDGE_RISING);
+			irq_set_type(jzgc, irq, IRQ_TYPE_LEVEL_HIGH);
 	}
 
 	if (jzgc->jzpc->info->version >= ID_JZ4760)
@@ -1848,7 +1848,7 @@ static int ingenic_gpio_irq_set_type(str
 		 */
 		bool high = ingenic_gpio_get_value(jzgc, irqd->hwirq);
 
-		type = high ? IRQ_TYPE_EDGE_FALLING : IRQ_TYPE_EDGE_RISING;
+		type = high ? IRQ_TYPE_LEVEL_LOW : IRQ_TYPE_LEVEL_HIGH;
 	}
 
 	irq_set_type(jzgc, irqd->hwirq, type);
author maintainer
#48 lore raw

PATCH 48/204 pinctrl: ingenic: Properly detect GPIO direction when configured for IRQ

In reply to Greg Kroah-Hartman

From: Paul Cercueil <[email protected]>

commit 84e7a946da71f678affacea301f6d5cb4d9784e8 upstream.

The PAT1 register contains information about the IRQ type (edge/level) for input GPIOs with IRQ enabled, and the direction for non-IRQ GPIOs. So it makes sense to read it only if the GPIO has no interrupt configured, otherwise input GPIOs configured for level IRQs are misdetected as output GPIOs.

Fixes ebd6651418b6 ("pinctrl: ingenic: Implement .get_direction for GPIO chips")
---
 drivers/pinctrl/pinctrl-ingenic.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/pinctrl/pinctrl-ingenic.c +2 -1
@@ -1955,7 +1955,8 @@ static int ingenic_gpio_get_direction(st
 	unsigned int pin = gc->base + offset;
 
 	if (jzpc->info->version >= ID_JZ4760) {
-		if (ingenic_get_pin_config(jzpc, pin, JZ4760_GPIO_PAT1))
+		if (ingenic_get_pin_config(jzpc, pin, JZ4760_GPIO_INT) ||
+		    ingenic_get_pin_config(jzpc, pin, JZ4760_GPIO_PAT1))
 			return GPIO_LINE_DIRECTION_IN;
 		return GPIO_LINE_DIRECTION_OUT;
 	}
author maintainer
#49 lore raw

PATCH 49/204 media: venus: fix multiple encoder crash

In reply to Greg Kroah-Hartman

From: Mansur Alisha Shaik <[email protected]>

commit e0eb34810113dbbf1ace57440cf48d514312a373 upstream.

Currently we are considering the instances which are available in core->inst list for load calculation in min_loaded_core() function, but this is incorrect because by the time we call decide_core() for second instance, the third instance not filled yet codec_freq_data pointer.

Solve this by considering the instances whose session has started.

---
 drivers/media/platform/qcom/venus/pm_helpers.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/media/platform/qcom/venus/pm_helpers.c +4 -0
@@ -496,6 +496,10 @@ min_loaded_core(struct venus_inst *inst,
 	list_for_each_entry(inst_pos, &core->instances, list) {
 		if (inst_pos == inst)
 			continue;
+
+		if (inst_pos->state != INST_START)
+			continue;
+
 		vpp_freq = inst_pos->clk_data.codec_freq_data->vpp_freq;
 		coreid = inst_pos->clk_data.core_id;
 
author maintainer
#50 lore raw

PATCH 50/204 media: vsp1: dl: Fix NULL pointer dereference on unbind

In reply to Greg Kroah-Hartman

From: Eugeniu Rosca <[email protected]>

commit c92d30e4b78dc331909f8c6056c2792aa14e2166 upstream.

In commit f3b98e3c4d2e16 ("media: vsp1: Provide support for extended command pools"), the vsp pointer used for referencing the VSP1 device structure from a command pool during vsp1_dl_ext_cmd_pool_destroy() was not populated.

Correctly assign the pointer to prevent the following null-pointer-dereference when removing the device:

[*] h3ulcb-kf #>
echo fea28000.vsp > /sys/bus/platform/devices/fea28000.vsp/driver/unbind

 Unable to handle kernel NULL pointer dereference at virtual address 0000000000000028
 Mem abort info:
   ESR = 0x96000006
   EC = 0x25: DABT (current EL), IL = 32 bits
   SET = 0, FnV = 0
   EA = 0, S1PTW = 0
 Data abort info:
   ISV = 0, ISS = 0x00000006
   CM = 0, WnR = 0
 user pgtable: 4k pages, 48-bit VAs, pgdp=00000007318be000
 [0000000000000028] pgd=00000007333a1003, pud=00000007333a6003, pmd=0000000000000000
 Internal error: Oops: 96000006 [#1] PREEMPT SMP
 Modules linked in:
 CPU: 1 PID: 486 Comm: sh Not tainted 5.7.0-rc6-arm64-renesas-00118-ge644645abf47 #185
 Hardware name: Renesas H3ULCB Kingfisher board based on r8a77951 (DT)
 pstate: 40000005 (nZcv daif -PAN -UAO)
 pc : vsp1_dlm_destroy+0xe4/0x11c
 lr : vsp1_dlm_destroy+0xc8/0x11c
 sp : ffff800012963b60
 x29: ffff800012963b60 x28: ffff0006f83fc440
 x27: 0000000000000000 x26: ffff0006f5e13e80
 x25: ffff0006f5e13ed0 x24: ffff0006f5e13ed0
 x23: ffff0006f5e13ed0 x22: dead000000000122
 x21: ffff0006f5e3a080 x20: ffff0006f5df2938
 x19: ffff0006f5df2980 x18: 0000000000000003
 x17: 0000000000000000 x16: 0000000000000016
 x15: 0000000000000003 x14: 00000000000393c0
 x13: ffff800011a5ec18 x12: ffff800011d8d000
 x11: ffff0006f83fcc68 x10: ffff800011a53d70
 x9 : ffff8000111f3000 x8 : 0000000000000000
 x7 : 0000000000210d00 x6 : 0000000000000000
 x5 : ffff800010872e60 x4 : 0000000000000004
 x3 : 0000000078068000 x2 : ffff800012781000
 x1 : 0000000000002c00 x0 : 0000000000000000
 Call trace:
  vsp1_dlm_destroy+0xe4/0x11c
  vsp1_wpf_destroy+0x10/0x20
  vsp1_entity_destroy+0x24/0x4c
  vsp1_destroy_entities+0x54/0x130
  vsp1_remove+0x1c/0x40
  platform_drv_remove+0x28/0x50
  __device_release_driver+0x178/0x220
  device_driver_detach+0x44/0xc0
  unbind_store+0xe0/0x104
  drv_attr_store+0x20/0x30
  sysfs_kf_write+0x48/0x70
  kernfs_fop_write+0x148/0x230
  __vfs_write+0x18/0x40
  vfs_write+0xdc/0x1c4
  ksys_write+0x68/0xf0
  __arm64_sys_write+0x18/0x20
  el0_svc_common.constprop.0+0x70/0x170
  do_el0_svc+0x20/0x80
  el0_sync_handler+0x134/0x1b0
  el0_sync+0x140/0x180
 Code: b40000c2 f9403a60 d2800084 a9400663 (f9401400)
 ---[ end trace 3875369841fb288a ]---
Fixes f3b98e3c4d2e16 ("media: vsp1: Provide support for extended command pools")
Cc [email protected] # v4.19+
---
 drivers/media/platform/vsp1/vsp1_dl.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/media/platform/vsp1/vsp1_dl.c +2 -0
@@ -431,6 +431,8 @@ vsp1_dl_cmd_pool_create(struct vsp1_devi
 	if (!pool)
 		return NULL;
 
+	pool->vsp1 = vsp1;
+
 	spin_lock_init(&pool->lock);
 	INIT_LIST_HEAD(&pool->free);
 
author maintainer
#51 lore raw

PATCH 51/204 net: ethernet: stmmac: Disable hardware multicast filter

In reply to Greg Kroah-Hartman

From: Jonathan McDowell <[email protected]>

commit df43dd526e6609769ae513a81443c7aa727c8ca3 upstream.

The IPQ806x does not appear to have a functional multicast ethernet address filter. This was observed as a failure to correctly receive IPv6 packets on a LAN to the all stations address. Checking the vendor driver shows that it does not attempt to enable the multicast filter and instead falls back to receiving all multicast packets, internally setting ALLMULTI.

Use the new fallback support in the dwmac1000 driver to correctly achieve the same with the mainline IPQ806x driver. Confirmed to fix IPv6 functionality on an RB3011 router.

---
 drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c +1 -0
@@ -351,6 +351,7 @@ static int ipq806x_gmac_probe(struct pla
 	plat_dat->has_gmac = true;
 	plat_dat->bsp_priv = gmac;
 	plat_dat->fix_mac_speed = ipq806x_gmac_fix_mac_speed;
+	plat_dat->multicast_filter_bins = 0;
 
 	err = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res);
 	if (err)
author maintainer
#52 lore raw

PATCH 52/204 net: stmmac: dwmac1000: provide multicast filter fallback

In reply to Greg Kroah-Hartman

From: Jonathan McDowell <[email protected]>

commit 592d751c1e174df5ff219946908b005eb48934b3 upstream.

If we don't have a hardware multicast filter available then instead of silently failing to listen for the requested ethernet broadcast addresses fall back to receiving all multicast packets, in a similar fashion to other drivers with no multicast filter.

---
 drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +3 -0
@@ -164,6 +164,9 @@ static void dwmac1000_set_filter(struct
 		value = GMAC_FRAME_FILTER_PR | GMAC_FRAME_FILTER_PCF;
 	} else if (dev->flags & IFF_ALLMULTI) {
 		value = GMAC_FRAME_FILTER_PM;	/* pass all multi */
+	} else if (!netdev_mc_empty(dev) && (mcbitslog2 == 0)) {
+		/* Fall back to all multicast if we've no filter */
+		value = GMAC_FRAME_FILTER_PM;
 	} else if (!netdev_mc_empty(dev)) {
 		struct netdev_hw_addr *ha;
 
author maintainer
#53 lore raw

PATCH 53/204 irqchip/loongson-liointc: Fix misuse of gc->mask_cache

In reply to Greg Kroah-Hartman

From: Huacai Chen <[email protected]>

commit c9c73a05413ea4a465cae1cb3593b01b190a233f upstream.

In gc->mask_cache bits, 1 means enabled and 0 means disabled, but in the loongson-liointc driver mask_cache is misused by reverting its meaning. This patch fix the bug and update the comments as well.

Fixes dbb152267908c4b2c3639492a ("irqchip: Add driver for Loongson I/O Local Interrupt Controller")
---
 drivers/irqchip/irq-loongson-liointc.c |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/irqchip/irq-loongson-liointc.c +5 -5
@@ -60,7 +60,7 @@ static void liointc_chained_handle_irq(s
 	if (!pending) {
 		/* Always blame LPC IRQ if we have that bug */
 		if (handler->priv->has_lpc_irq_errata &&
-			(handler->parent_int_map & ~gc->mask_cache &
+			(handler->parent_int_map & gc->mask_cache &
 			BIT(LIOINTC_ERRATA_IRQ)))
 			pending = BIT(LIOINTC_ERRATA_IRQ);
 		else
@@ -132,11 +132,11 @@ static void liointc_resume(struct irq_ch
 	irq_gc_lock_irqsave(gc, flags);
 	/* Disable all at first */
 	writel(0xffffffff, gc->reg_base + LIOINTC_REG_INTC_DISABLE);
-	/* Revert map cache */
+	/* Restore map cache */
 	for (i = 0; i < LIOINTC_CHIP_IRQ; i++)
 		writeb(priv->map_cache[i], gc->reg_base + i);
-	/* Revert mask cache */
-	writel(~gc->mask_cache, gc->reg_base + LIOINTC_REG_INTC_ENABLE);
+	/* Restore mask cache */
+	writel(gc->mask_cache, gc->reg_base + LIOINTC_REG_INTC_ENABLE);
 	irq_gc_unlock_irqrestore(gc, flags);
 }
 
@@ -244,7 +244,7 @@ int __init liointc_of_init(struct device
 	ct->chip.irq_mask_ack = irq_gc_mask_disable_reg;
 	ct->chip.irq_set_type = liointc_set_type;
 
-	gc->mask_cache = 0xffffffff;
+	gc->mask_cache = 0;
 	priv->gc = gc;
 
 	for (i = 0; i < LIOINTC_NUM_PARENT; i++) {
author maintainer
#54 lore raw

PATCH 54/204 irqchip/gic-v4.1: Ensure accessing the correct RD when writing INVALLR

In reply to Greg Kroah-Hartman

From: Zenghui Yu <[email protected]>

commit 3af9571cd585efafc2facbd8dbd407317ff898cf upstream.

The GICv4.1 spec tells us that it's CONSTRAINED UNPREDICTABLE to issue a register-based invalidation operation for a vPEID not mapped to that RD, or another RD within the same CommonLPIAff group.

To follow this rule, commit f3a059219bc7 ("irqchip/gic-v4.1: Ensure mutual exclusion between vPE affinity change and RD access") tried to address the race between the RD accesses and the vPE affinity change, but somehow forgot to take GICR_INVALLR into account. Let's take the vpe_lock before evaluating vpe->col_idx to fix it.

Fixes f3a059219bc7 ("irqchip/gic-v4.1: Ensure mutual exclusion between vPE affinity change and RD access")
---
 drivers/irqchip/irq-gic-v3-its.c |   10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/irqchip/irq-gic-v3-its.c +7 -3
@@ -3974,18 +3974,22 @@ static void its_vpe_4_1_deschedule(struc
 static void its_vpe_4_1_invall(struct its_vpe *vpe)
 {
 	void __iomem *rdbase;
+	unsigned long flags;
 	u64 val;
+	int cpu;
 
 	val  = GICR_INVALLR_V;
 	val |= FIELD_PREP(GICR_INVALLR_VPEID, vpe->vpe_id);
 
 	/* Target the redistributor this vPE is currently known on */
-	raw_spin_lock(&gic_data_rdist_cpu(vpe->col_idx)->rd_lock);
-	rdbase = per_cpu_ptr(gic_rdists->rdist, vpe->col_idx)->rd_base;
+	cpu = vpe_to_cpuid_lock(vpe, &flags);
+	raw_spin_lock(&gic_data_rdist_cpu(cpu)->rd_lock);
+	rdbase = per_cpu_ptr(gic_rdists->rdist, cpu)->rd_base;
 	gic_write_lpir(val, rdbase + GICR_INVALLR);
 
 	wait_for_syncr(rdbase);
-	raw_spin_unlock(&gic_data_rdist_cpu(vpe->col_idx)->rd_lock);
+	raw_spin_unlock(&gic_data_rdist_cpu(cpu)->rd_lock);
+	vpe_to_cpuid_unlock(vpe, flags);
 }
 
 static int its_vpe_4_1_set_vcpu_affinity(struct irq_data *d, void *vcpu_info)
author maintainer
#55 lore raw

PATCH 55/204 pidfd: Add missing sock updates for pidfd_getfd()

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

commit 4969f8a073977123504609d7310b42a588297aa4 upstream.

The sock counting (sock_update_netprioidx() and sock_update_classid()) was missing from pidfd's implementation of received fd installation. Add a call to the new __receive_sock() helper.

Cc Christian Brauner <[email protected]>
Cc Christoph Hellwig <[email protected]>
Cc Sargun Dhillon <[email protected]>
Cc Jakub Kicinski <[email protected]>
Fixes 8649c322f75c ("pid: Implement pidfd_getfd syscall")
---
 kernel/pid.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
kernel/pid.c +5 -2
@@ -42,6 +42,7 @@
 #include <linux/sched/signal.h>
 #include <linux/sched/task.h>
 #include <linux/idr.h>
+#include <net/sock.h>
 
 struct pid init_struct_pid = {
 	.count		= REFCOUNT_INIT(1),
@@ -624,10 +625,12 @@ static int pidfd_getfd(struct pid *pid,
 	}
 
 	ret = get_unused_fd_flags(O_CLOEXEC);
-	if (ret < 0)
+	if (ret < 0) {
 		fput(file);
-	else
+	} else {
+		__receive_sock(file);
 		fd_install(ret, file);
+	}
 
 	return ret;
 }
author maintainer
#56 lore raw

PATCH 56/204 net/compat: Add missing sock updates for SCM_RIGHTS

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

commit d9539752d23283db4692384a634034f451261e29 upstream.

Add missed sock updates to compat path via a new helper, which will be used more in coming patches. (The net/core/scm.c code is left as-is here to assist with -stable backports for the compat path.)

Cc Christoph Hellwig <[email protected]>
Cc Sargun Dhillon <[email protected]>
Cc Jakub Kicinski <[email protected]>
Fixes 48a87cc26c13 ("net: netprio: fd passed in SCM_RIGHTS datagram not set correctly")
Fixes d84295067fc7 ("net: net_cls: fd passed in SCM_RIGHTS datagram not set correctly")
---
 include/net/sock.h |    4 ++++
 net/compat.c       |    1 +
 net/core/sock.c    |   21 +++++++++++++++++++++
 3 files changed, 26 insertions(+)
include/net/sock.h +4 -0
@@ -890,6 +890,8 @@ static inline int sk_memalloc_socks(void
 {
 	return static_branch_unlikely(&memalloc_socks_key);
 }
+
+void __receive_sock(struct file *file);
 #else
 
 static inline int sk_memalloc_socks(void)
@@ -897,6 +899,8 @@ static inline int sk_memalloc_socks(void
 	return 0;
 }
 
+static inline void __receive_sock(struct file *file)
+{ }
 #endif
 
 static inline gfp_t sk_gfp_mask(const struct sock *sk, gfp_t gfp_mask)
net/compat.c +1 -0
@@ -307,6 +307,7 @@ void scm_detach_fds_compat(struct msghdr
 			break;
 		}
 		/* Bump the usage count and install the file. */
+		__receive_sock(fp[i]);
 		fd_install(new_fd, get_file(fp[i]));
 	}
 
net/core/sock.c +21 -0
@@ -2753,6 +2753,27 @@ int sock_no_mmap(struct file *file, stru
 }
 EXPORT_SYMBOL(sock_no_mmap);
 
+/*
+ * When a file is received (via SCM_RIGHTS, etc), we must bump the
+ * various sock-based usage counts.
+ */
+void __receive_sock(struct file *file)
+{
+	struct socket *sock;
+	int error;
+
+	/*
+	 * The resulting value of "error" is ignored here since we only
+	 * need to take action when the file is a socket and testing
+	 * "sock" for NULL is sufficient.
+	 */
+	sock = sock_from_file(file, &error);
+	if (sock) {
+		sock_update_netprioidx(&sock->sk->sk_cgrp_data);
+		sock_update_classid(&sock->sk->sk_cgrp_data);
+	}
+}
+
 ssize_t sock_no_sendpage(struct socket *sock, struct page *page, int offset, size_t size, int flags)
 {
 	ssize_t res;
author maintainer
#57 lore raw

PATCH 57/204 selftests/seccomp: Set NNP for TSYNC ESRCH flag test

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

commit e4d05028a07f505a08802a6d1b11674c149df2b3 upstream.

The TSYNC ESRCH flag test will fail for regular users because NNP was not set yet. Add NNP setting.

Fixes 51891498f2da ("seccomp: allow TSYNC and USER_NOTIF together")
---
 tools/testing/selftests/seccomp/seccomp_bpf.c |    5 +++++
 1 file changed, 5 insertions(+)
tools/testing/selftests/seccomp/seccomp_bpf.c +5 -0
@@ -3257,6 +3257,11 @@ TEST(user_notification_with_tsync)
 	int ret;
 	unsigned int flags;
 
+	ret = prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
+	ASSERT_EQ(0, ret) {
+		TH_LOG("Kernel does not support PR_SET_NO_NEW_PRIVS!");
+	}
+
 	/* these were exclusive */
 	flags = SECCOMP_FILTER_FLAG_NEW_LISTENER |
 		SECCOMP_FILTER_FLAG_TSYNC;
author maintainer
#58 lore raw

PATCH 58/204 md/raid5: Fix Force reconstruct-write io stuck in degraded raid5

In reply to Greg Kroah-Hartman

From: ChangSyun Peng <[email protected]>

commit a1c6ae3d9f3dd6aa5981a332a6f700cf1c25edef upstream.

In degraded raid5, we need to read parity to do reconstruct-write when data disks fail. However, we can not read parity from handle_stripe_dirtying() in force reconstruct-write mode.

Reproducible Steps:

1. Create degraded raid5
mdadm -C /dev/md2 --assume-clean -l5 -n3 /dev/sda2 /dev/sdb2 missing
2. Set rmw_level to 0
echo 0 > /sys/block/md2/md/rmw_level
3. IO to raid5

Now some io may be stuck in raid5. We can use handle_stripe_fill() to read the parity in this situation.

---
 drivers/md/raid5.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/md/raid5.c +2 -1
@@ -3604,6 +3604,7 @@ static int need_this_block(struct stripe
 	 * is missing/faulty, then we need to read everything we can.
 	 */
 	if (sh->raid_conf->level != 6 &&
+	    sh->raid_conf->rmw_level != PARITY_DISABLE_RMW &&
 	    sh->sector < sh->raid_conf->mddev->recovery_cp)
 		/* reconstruct-write isn't being forced */
 		return 0;
@@ -4839,7 +4840,7 @@ static void handle_stripe(struct stripe_
 	 * or to load a block that is being partially written.
 	 */
 	if (s.to_read || s.non_overwrite
-	    || (conf->level == 6 && s.to_write && s.failed)
+	    || (s.to_write && s.failed)
 	    || (s.syncing && (s.uptodate + s.compute < disks))
 	    || s.replacing
 	    || s.expanding)
author maintainer
#59 lore raw

PATCH 59/204 bcache: allocate meta data pages as compound pages

In reply to Greg Kroah-Hartman

From: Coly Li <[email protected]>

commit 5fe48867856367142d91a82f2cbf7a57a24cbb70 upstream.

There are some meta data of bcache are allocated by multiple pages, and they are used as bio bv_page for I/Os to the cache device. for example cache_set->uuids, cache->disk_buckets, journal_write->data, bset_tree->data.

For such meta data memory, all the allocated pages should be treated as a single memory block. Then the memory management and underlying I/O code can treat them more clearly.

This patch adds __GFP_COMP flag to all the location allocating >0 order pages for the above mentioned meta data. Then their pages are treated as compound pages now.

---
 drivers/md/bcache/bset.c    |    2 +-
 drivers/md/bcache/btree.c   |    2 +-
 drivers/md/bcache/journal.c |    4 ++--
 drivers/md/bcache/super.c   |    2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)
drivers/md/bcache/bset.c +1 -1
@@ -322,7 +322,7 @@ int bch_btree_keys_alloc(struct btree_ke
 
 	b->page_order = page_order;
 
-	t->data = (void *) __get_free_pages(gfp, b->page_order);
+	t->data = (void *) __get_free_pages(__GFP_COMP|gfp, b->page_order);
 	if (!t->data)
 		goto err;
 
drivers/md/bcache/btree.c +1 -1
@@ -785,7 +785,7 @@ int bch_btree_cache_alloc(struct cache_s
 	mutex_init(&c->verify_lock);
 
 	c->verify_ondisk = (void *)
-		__get_free_pages(GFP_KERNEL, ilog2(bucket_pages(c)));
+		__get_free_pages(GFP_KERNEL|__GFP_COMP, ilog2(bucket_pages(c)));
 
 	c->verify_data = mca_bucket_alloc(c, &ZERO_KEY, GFP_KERNEL);
 
drivers/md/bcache/journal.c +2 -2
@@ -999,8 +999,8 @@ int bch_journal_alloc(struct cache_set *
 	j->w[1].c = c;
 
 	if (!(init_fifo(&j->pin, JOURNAL_PIN, GFP_KERNEL)) ||
-	    !(j->w[0].data = (void *) __get_free_pages(GFP_KERNEL, JSET_BITS)) ||
-	    !(j->w[1].data = (void *) __get_free_pages(GFP_KERNEL, JSET_BITS)))
+	    !(j->w[0].data = (void *) __get_free_pages(GFP_KERNEL|__GFP_COMP, JSET_BITS)) ||
+	    !(j->w[1].data = (void *) __get_free_pages(GFP_KERNEL|__GFP_COMP, JSET_BITS)))
 		return -ENOMEM;
 
 	return 0;
drivers/md/bcache/super.c +1 -1
@@ -1775,7 +1775,7 @@ void bch_cache_set_unregister(struct cac
 }
 
 #define alloc_bucket_pages(gfp, c)			\
-	((void *) __get_free_pages(__GFP_ZERO|gfp, ilog2(bucket_pages(c))))
+	((void *) __get_free_pages(__GFP_ZERO|__GFP_COMP|gfp, ilog2(bucket_pages(c))))
 
 struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
 {
author maintainer
#60 lore raw

PATCH 60/204 bcache: fix overflow in offset_to_stripe()

In reply to Greg Kroah-Hartman

From: Coly Li <[email protected]>

commit 7a1481267999c02abf4a624515c1b5c7c1fccbd6 upstream.

offset_to_stripe() returns the stripe number (in type unsigned int) from an offset (in type uint64_t) by the following calculation,

	do_div(offset, d->stripe_size);

For large capacity backing device (e.g. 18TB) with small stripe size (e.g. 4KB), the result is 4831838208 and exceeds UINT_MAX. The actual returned value which caller receives is 536870912, due to the overflow.

Indeed in bcache_device_init(), bcache_device->nr_stripes is limited in range [1, INT_MAX]. Therefore all valid stripe numbers in bcache are in range [0, bcache_dev->nr_stripes - 1].

This patch adds a upper limition check in offset_to_stripe(): the max valid stripe number should be less than bcache_device->nr_stripes. If the calculated stripe number from do_div() is equal to or larger than bcache_device->nr_stripe, -EINVAL will be returned. (Normally nr_stripes is less than INT_MAX, exceeding upper limitation doesn't mean overflow, therefore -EOVERFLOW is not used as error code.)

This patch also changes nr_stripes' type of struct bcache_device from 'unsigned int' to 'int', and return value type of offset_to_stripe() from 'unsigned int' to 'int', to match their exact data ranges.

All locations where bcache_device->nr_stripes and offset_to_stripe() are referenced also get updated for the above type change.

---
 drivers/md/bcache/bcache.h    |    2 +-
 drivers/md/bcache/writeback.c |   14 +++++++++-----
 drivers/md/bcache/writeback.h |   19 +++++++++++++++++--
 3 files changed, 27 insertions(+), 8 deletions(-)
drivers/md/bcache/bcache.h +1 -1
@@ -264,7 +264,7 @@ struct bcache_device {
 #define BCACHE_DEV_UNLINK_DONE		2
 #define BCACHE_DEV_WB_RUNNING		3
 #define BCACHE_DEV_RATE_DW_RUNNING	4
-	unsigned int		nr_stripes;
+	int			nr_stripes;
 	unsigned int		stripe_size;
 	atomic_t		*stripe_sectors_dirty;
 	unsigned long		*full_dirty_stripes;
drivers/md/bcache/writeback.c +9 -5
@@ -523,15 +523,19 @@ void bcache_dev_sectors_dirty_add(struct
 				  uint64_t offset, int nr_sectors)
 {
 	struct bcache_device *d = c->devices[inode];
-	unsigned int stripe_offset, stripe, sectors_dirty;
+	unsigned int stripe_offset, sectors_dirty;
+	int stripe;
 
 	if (!d)
 		return;
 
+	stripe = offset_to_stripe(d, offset);
+	if (stripe < 0)
+		return;
+
 	if (UUID_FLASH_ONLY(&c->uuids[inode]))
 		atomic_long_add(nr_sectors, &c->flash_dev_dirty_sectors);
 
-	stripe = offset_to_stripe(d, offset);
 	stripe_offset = offset & (d->stripe_size - 1);
 
 	while (nr_sectors) {
@@ -571,12 +575,12 @@ static bool dirty_pred(struct keybuf *bu
 static void refill_full_stripes(struct cached_dev *dc)
 {
 	struct keybuf *buf = &dc->writeback_keys;
-	unsigned int start_stripe, stripe, next_stripe;
+	unsigned int start_stripe, next_stripe;
+	int stripe;
 	bool wrapped = false;
 
 	stripe = offset_to_stripe(&dc->disk, KEY_OFFSET(&buf->last_scanned));
-
-	if (stripe >= dc->disk.nr_stripes)
+	if (stripe < 0)
 		stripe = 0;
 
 	start_stripe = stripe;
drivers/md/bcache/writeback.h +17 -2
@@ -52,10 +52,22 @@ static inline uint64_t bcache_dev_sector
 	return ret;
 }
 
-static inline unsigned int offset_to_stripe(struct bcache_device *d,
+static inline int offset_to_stripe(struct bcache_device *d,
 					uint64_t offset)
 {
 	do_div(offset, d->stripe_size);
+
+	/* d->nr_stripes is in range [1, INT_MAX] */
+	if (unlikely(offset >= d->nr_stripes)) {
+		pr_err("Invalid stripe %llu (>= nr_stripes %d).\n",
+			offset, d->nr_stripes);
+		return -EINVAL;
+	}
+
+	/*
+	 * Here offset is definitly smaller than INT_MAX,
+	 * return it as int will never overflow.
+	 */
 	return offset;
 }
 
@@ -63,7 +75,10 @@ static inline bool bcache_dev_stripe_dir
 					   uint64_t offset,
 					   unsigned int nr_sectors)
 {
-	unsigned int stripe = offset_to_stripe(&dc->disk, offset);
+	int stripe = offset_to_stripe(&dc->disk, offset);
+
+	if (stripe < 0)
+		return false;
 
 	while (1) {
 		if (atomic_read(dc->disk.stripe_sectors_dirty + stripe))
author maintainer
#61 lore raw

PATCH 61/204 mac80211: fix misplaced while instead of if

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

commit 5981fe5b0529ba25d95f37d7faa434183ad618c5 upstream.

This never was intended to be a 'while' loop, it should've just been an 'if' instead of 'while'. Fix this.

I noticed this while applying another patch from Ben that intended to fix a busy loop at this spot.

---
 net/mac80211/sta_info.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/mac80211/sta_info.c +1 -1
@@ -1050,7 +1050,7 @@ static void __sta_info_destroy_part2(str
 	might_sleep();
 	lockdep_assert_held(&local->sta_mtx);
 
-	while (sta->sta_state == IEEE80211_STA_AUTHORIZED) {
+	if (sta->sta_state == IEEE80211_STA_AUTHORIZED) {
 		ret = sta_info_move_state(sta, IEEE80211_STA_ASSOC);
 		WARN_ON_ONCE(ret);
 	}
author maintainer
#62 lore raw

PATCH 62/204 driver core: Avoid binding drivers to dead devices

In reply to Greg Kroah-Hartman

From: Lukas Wunner <[email protected]>

commit 654888327e9f655a9d55ad477a9583e90e8c9b5c upstream.

Commit 3451a495ef24 ("driver core: Establish order of operations for device_add and device_del via bitflag") sought to prevent asynchronous driver binding to a device which is being removed. It added a per-device "dead" flag which is checked in the following code paths:

* asynchronous binding in __driver_attach_async_helper()
* synchronous binding in device_driver_attach()
* asynchronous binding in __device_attach_async_helper()

It did *not* check the flag upon:

* synchronous binding in __device_attach()

However __device_attach() may also be called asynchronously from:

deferred_probe_work_func()
bus_probe_device()

    device_initial_probe()
      __device_attach()

So if the commit's intention was to check the "dead" flag in all asynchronous code paths, then a check is also necessary in __device_attach(). Add the missing check.

Fixes 3451a495ef24 ("driver core: Establish order of operations for device_add and device_del via bitflag")
Cc Alexander Duyck <[email protected]>
---
 drivers/base/dd.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/base/dd.c +3 -1
@@ -846,7 +846,9 @@ static int __device_attach(struct device
 	int ret = 0;
 
 	device_lock(dev);
-	if (dev->driver) {
+	if (dev->p->dead) {
+		goto out_unlock;
+	} else if (dev->driver) {
 		if (device_is_bound(dev)) {
 			ret = 1;
 			goto out_unlock;
author maintainer
#63 lore raw

PATCH 63/204 MIPS: CPU#0 is not hotpluggable

In reply to Greg Kroah-Hartman

From: Huacai Chen <[email protected]>

commit 9cce844abf07b683cff5f0273977d5f8d0af94c7 upstream.

Now CPU#0 is not hotpluggable on MIPS, so prevent to create /sys/devices /system/cpu/cpu0/online which confuses some user-space tools.

---
 arch/mips/kernel/topology.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/mips/kernel/topology.c +1 -1
@@ -20,7 +20,7 @@ static int __init topology_init(void)
 	for_each_present_cpu(i) {
 		struct cpu *c = &per_cpu(cpu_devices, i);
 
-		c->hotpluggable = 1;
+		c->hotpluggable = !!i;
 		ret = register_cpu(c, i);
 		if (ret)
 			printk(KERN_WARNING "topology_init: register_cpu %d "
author maintainer
#64 lore raw

PATCH 64/204 MIPS: qi_lb60: Fix routing to audio amplifier

In reply to Greg Kroah-Hartman

From: Paul Cercueil <[email protected]>

commit 0889a67a9e7a56ba39af223d536630b20b877fda upstream.

The ROUT (right channel output of audio codec) was connected to INL (left channel of audio amplifier) instead of INR (right channel of audio amplifier).

Fixes 8ddebad15e9b ("MIPS: qi_lb60: Migrate to devicetree")
---
 arch/mips/boot/dts/ingenic/qi_lb60.dts |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/mips/boot/dts/ingenic/qi_lb60.dts +1 -1
@@ -69,7 +69,7 @@
 			"Speaker", "OUTL",
 			"Speaker", "OUTR",
 			"INL", "LOUT",
-			"INL", "ROUT";
+			"INR", "ROUT";
 
 		simple-audio-card,aux-devs = <&amp>;
 
author maintainer
#65 lore raw

PATCH 65/204 MIPS: SGI-IP27: always enable NUMA in Kconfig

In reply to Greg Kroah-Hartman

From: Mike Rapoport <[email protected]>

commit 6c86a3029ce3b44597526909f2e39a77a497f640 upstream.

When a configuration has NUMA disabled and SGI_IP27 enabled, the build fails:

CC kernel/bounds.s
CC arch/mips/kernel/asm-offsets.s
In file included from arch/mips/include/asm/topology.h:11,
from include/linux/topology.h:36,
from include/linux/gfp.h:9,
from include/linux/slab.h:15,
from include/linux/crypto.h:19,
from include/crypto/hash.h:11,
from include/linux/uio.h:10,
from include/linux/socket.h:8,
from include/linux/compat.h:15,
from arch/mips/kernel/asm-offsets.c:12:
include/linux/topology.h: In function 'numa_node_id':
arch/mips/include/asm/mach-ip27/topology.h:16:27: error: implicit declaration of function 'cputonasid'; did you mean 'cpu_vpe_id'? [-Werror=implicit-function-declaration]

 #define cpu_to_node(cpu) (cputonasid(cpu))
                           ^~~~~~~~~~
include/linux/topology.h:119:9: note: in expansion of macro 'cpu_to_node'
  return cpu_to_node(raw_smp_processor_id());
         ^~~~~~~~~~~
include/linux/topology.h: In function 'cpu_cpu_mask':
arch/mips/include/asm/mach-ip27/topology.h:19:7: error: implicit declaration of function 'hub_data' [-Werror=implicit-function-declaration]
      &hub_data(node)->h_cpus)
       ^~~~~~~~
include/linux/topology.h:210:9: note: in expansion of macro 'cpumask_of_node'
  return cpumask_of_node(cpu_to_node(cpu));
         ^~~~~~~~~~~~~~~
arch/mips/include/asm/mach-ip27/topology.h:19:21: error: invalid type argument of '->' (have 'int')
      &hub_data(node)->h_cpus)
                     ^~
include/linux/topology.h:210:9: note: in expansion of macro 'cpumask_of_node'
  return cpumask_of_node(cpu_to_node(cpu));
         ^~~~~~~~~~~~~~~

Before switch from discontigmem to sparsemem, there always was CONFIG_NEED_MULTIPLE_NODES=y because it was selected by DISCONTIGMEM. Without DISCONTIGMEM it is possible to have SPARSEMEM without NUMA for SGI_IP27 and as many things there rely on custom node definition, the build breaks.

As Thomas noted "... there are right now too many places in IP27 code, which assumes NUMA enabled", the simplest solution would be to always enable NUMA for SGI-IP27 builds.

Fixes 397dc00e249e ("mips: sgi-ip27: switch from DISCONTIGMEM to SPARSEMEM")
---
 arch/mips/Kconfig |    1 +
 1 file changed, 1 insertion(+)
arch/mips/Kconfig +1 -0
@@ -722,6 +722,7 @@ config SGI_IP27
 	select SYS_SUPPORTS_NUMA
 	select SYS_SUPPORTS_SMP
 	select MIPS_L1_CACHE_SHIFT_7
+	select NUMA
 	help
 	  This are the SGI Origin 200, Origin 2000 and Onyx 2 Graphics
 	  workstations.  To compile a Linux kernel that runs on these, say Y
author maintainer
#66 lore raw

PATCH 66/204 ext2: fix missing percpu_counter_inc

In reply to Greg Kroah-Hartman

From: Mikulas Patocka <[email protected]>

commit bc2fbaa4d3808aef82dd1064a8e61c16549fe956 upstream.

sbi->s_freeinodes_counter is only decreased by the ext2 code, it is never increased. This patch fixes it.

Note that sbi->s_freeinodes_counter is only used in the algorithm that tries to find the group for new allocations, so this bug is not easily visible (the only visibility is that the group finding algorithm selects inoptinal result).

---
 fs/ext2/ialloc.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/ext2/ialloc.c +2 -1
@@ -80,6 +80,7 @@ static void ext2_release_inode(struct su
 	if (dir)
 		le16_add_cpu(&desc->bg_used_dirs_count, -1);
 	spin_unlock(sb_bgl_lock(EXT2_SB(sb), group));
+	percpu_counter_inc(&EXT2_SB(sb)->s_freeinodes_counter);
 	if (dir)
 		percpu_counter_dec(&EXT2_SB(sb)->s_dirs_counter);
 	mark_buffer_dirty(bh);
@@ -528,7 +529,7 @@ got:
 		goto fail;
 	}
 
-	percpu_counter_add(&sbi->s_freeinodes_counter, -1);
+	percpu_counter_dec(&sbi->s_freeinodes_counter);
 	if (S_ISDIR(mode))
 		percpu_counter_inc(&sbi->s_dirs_counter);
 
author maintainer
#67 lore raw

PATCH 67/204 khugepaged: collapse_pte_mapped_thp() flush the right range

In reply to Greg Kroah-Hartman

From: Hugh Dickins <[email protected]>

commit 723a80dafed5c95889d48baab9aa433a6ffa0b4e upstream.

pmdp_collapse_flush() should be given the start address at which the huge page is mapped, haddr: it was given addr, which at that point has been used as a local variable, incremented to the end address of the extent.

Found by source inspection while chasing a hugepage locking bug, which I then could not explain by this. At first I thought this was very bad; then saw that all of the page translations that were not flushed would actually still point to the right pages afterwards, so harmless; then realized that I know nothing of how different architectures and models cache intermediate paging structures, so maybe it matters after all - particularly since the page table concerned is immediately freed.

Much easier to fix than to think about.

Fixes 27e1f8273113 ("khugepaged: enable collapse pmd for pte-mapped THP")
Cc Andrea Arcangeli <[email protected]>
Cc Mike Kravetz <[email protected]>
Cc Song Liu <[email protected]>
Cc <[email protected]> [5.4+]
---
 mm/khugepaged.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
mm/khugepaged.c +1 -1
@@ -1403,7 +1403,7 @@ void collapse_pte_mapped_thp(struct mm_s
 
 	/* step 4: collapse pmd */
 	ptl = pmd_lock(vma->vm_mm, pmd);
-	_pmd = pmdp_collapse_flush(vma, addr, pmd);
+	_pmd = pmdp_collapse_flush(vma, haddr, pmd);
 	spin_unlock(ptl);
 	mm_dec_nr_ptes(mm);
 	pte_free(mm, pmd_pgtable(_pmd));
author maintainer
#68 lore raw

PATCH 68/204 mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible

In reply to Greg Kroah-Hartman

From: Peter Xu <[email protected]>

commit 75802ca66354a39ab8e35822747cd08b3384a99a upstream.

This is found by code observation only.

Firstly, the worst case scenario should assume the whole range was covered by pmd sharing. The old algorithm might not work as expected for ranges like (1g-2m, 1g+2m), where the adjusted range should be (0, 1g+2m) but the expected range should be (0, 2g).

Since at it, remove the loop since it should not be required. With that, the new code should be faster too when the invalidating range is huge.

Mike said:

: With range (1g-2m, 1g+2m) within a vma (0, 2g) the existing code will only : adjust to (0, 1g+2m) which is incorrect.
:
: We should cc stable. The original reason for adjusting the range was to : prevent data corruption (getting wrong page). Since the range is not : always adjusted correctly, the potential for corruption still exists. :
: However, I am fairly confident that adjust_range_if_pmd_sharing_possible : is only gong to be called in two cases:
:
: 1) for a single page
: 2) for range == entire vma
:
: In those cases, the current code should produce the correct results. :
: To be safe, let's just cc stable.

Fixes 017b1660df89 ("mm: migration: fix migration of huge PMD shared pages")
Cc Andrea Arcangeli <[email protected]>
Cc Matthew Wilcox <[email protected]>
---
 mm/hugetlb.c |   24 ++++++++++--------------
 1 file changed, 10 insertions(+), 14 deletions(-)
mm/hugetlb.c +10 -14
@@ -5201,25 +5201,21 @@ static bool vma_shareable(struct vm_area
 void adjust_range_if_pmd_sharing_possible(struct vm_area_struct *vma,
 				unsigned long *start, unsigned long *end)
 {
-	unsigned long check_addr;
+	unsigned long a_start, a_end;
 
 	if (!(vma->vm_flags & VM_MAYSHARE))
 		return;
 
-	for (check_addr = *start; check_addr < *end; check_addr += PUD_SIZE) {
-		unsigned long a_start = check_addr & PUD_MASK;
-		unsigned long a_end = a_start + PUD_SIZE;
+	/* Extend the range to be PUD aligned for a worst case scenario */
+	a_start = ALIGN_DOWN(*start, PUD_SIZE);
+	a_end = ALIGN(*end, PUD_SIZE);
 
-		/*
-		 * If sharing is possible, adjust start/end if necessary.
-		 */
-		if (range_in_vma(vma, a_start, a_end)) {
-			if (a_start < *start)
-				*start = a_start;
-			if (a_end > *end)
-				*end = a_end;
-		}
-	}
+	/*
+	 * Intersect the range with the vma range, since pmd sharing won't be
+	 * across vma after all
+	 */
+	*start = max(vma->vm_start, a_start);
+	*end = min(vma->vm_end, a_end);
 }
 
 /*
author maintainer
#69 lore raw

PATCH 69/204 khugepaged: collapse_pte_mapped_thp() protect the pmd lock

In reply to Greg Kroah-Hartman

From: Hugh Dickins <[email protected]>

commit 119a5fc16105b2b9383a6e2a7800b2ef861b2975 upstream.

When retract_page_tables() removes a page table to make way for a huge pmd, it holds huge page lock, i_mmap_lock_write, mmap_write_trylock and pmd lock; but when collapse_pte_mapped_thp() does the same (to handle the case when the original mmap_write_trylock had failed), only mmap_write_trylock and pmd lock are held.

That's not enough. One machine has twice crashed under load, with "BUG:
spinlock bad magic" and GPF on 6b6b6b6b6b6b6b6b. Examining the second crash, page_vma_mapped_walk_done()'s spin_unlock of pvmw->ptl (serving page_referenced() on a file THP, that had found a page table at *pmd) discovers that the page table page and its lock have already been freed by the time it comes to unlock.

Follow the example of retract_page_tables(), but we only need one of huge page lock or i_mmap_lock_write to secure against this: because it's the narrower lock, and because it simplifies collapse_pte_mapped_thp() to know the hpage earlier, choose to rely on huge page lock here.

Fixes 27e1f8273113 ("khugepaged: enable collapse pmd for pte-mapped THP")
Cc Andrea Arcangeli <[email protected]>
Cc Mike Kravetz <[email protected]>
Cc Song Liu <[email protected]>
Cc <[email protected]> [5.4+]
---
 mm/khugepaged.c |   44 +++++++++++++++++++-------------------------
 1 file changed, 19 insertions(+), 25 deletions(-)
mm/khugepaged.c +19 -25
@@ -1313,7 +1313,7 @@ void collapse_pte_mapped_thp(struct mm_s
 {
 	unsigned long haddr = addr & HPAGE_PMD_MASK;
 	struct vm_area_struct *vma = find_vma(mm, haddr);
-	struct page *hpage = NULL;
+	struct page *hpage;
 	pte_t *start_pte, *pte;
 	pmd_t *pmd, _pmd;
 	spinlock_t *ptl;
@@ -1333,9 +1333,17 @@ void collapse_pte_mapped_thp(struct mm_s
 	if (!hugepage_vma_check(vma, vma->vm_flags | VM_HUGEPAGE))
 		return;
 
+	hpage = find_lock_page(vma->vm_file->f_mapping,
+			       linear_page_index(vma, haddr));
+	if (!hpage)
+		return;
+
+	if (!PageHead(hpage))
+		goto drop_hpage;
+
 	pmd = mm_find_pmd(mm, haddr);
 	if (!pmd)
-		return;
+		goto drop_hpage;
 
 	start_pte = pte_offset_map_lock(mm, pmd, haddr, &ptl);
 
@@ -1354,30 +1362,11 @@ void collapse_pte_mapped_thp(struct mm_s
 
 		page = vm_normal_page(vma, addr, *pte);
 
-		if (!page || !PageCompound(page))
-			goto abort;
-
-		if (!hpage) {
-			hpage = compound_head(page);
-			/*
-			 * The mapping of the THP should not change.
-			 *
-			 * Note that uprobe, debugger, or MAP_PRIVATE may
-			 * change the page table, but the new page will
-			 * not pass PageCompound() check.
-			 */
-			if (WARN_ON(hpage->mapping != vma->vm_file->f_mapping))
-				goto abort;
-		}
-
 		/*
-		 * Confirm the page maps to the correct subpage.
-		 *
-		 * Note that uprobe, debugger, or MAP_PRIVATE may change
-		 * the page table, but the new page will not pass
-		 * PageCompound() check.
+		 * Note that uprobe, debugger, or MAP_PRIVATE may change the
+		 * page table, but the new page will not be a subpage of hpage.
 		 */
-		if (WARN_ON(hpage + i != page))
+		if (hpage + i != page)
 			goto abort;
 		count++;
 	}
@@ -1396,7 +1385,7 @@ void collapse_pte_mapped_thp(struct mm_s
 	pte_unmap_unlock(start_pte, ptl);
 
 	/* step 3: set proper refcount and mm_counters. */
-	if (hpage) {
+	if (count) {
 		page_ref_sub(hpage, count);
 		add_mm_counter(vma->vm_mm, mm_counter_file(hpage), -count);
 	}
@@ -1407,10 +1396,15 @@ void collapse_pte_mapped_thp(struct mm_s
 	spin_unlock(ptl);
 	mm_dec_nr_ptes(mm);
 	pte_free(mm, pmd_pgtable(_pmd));
+
+drop_hpage:
+	unlock_page(hpage);
+	put_page(hpage);
 	return;
 
 abort:
 	pte_unmap_unlock(start_pte, ptl);
+	goto drop_hpage;
 }
 
 static int khugepaged_collapse_pte_mapped_thps(struct mm_slot *mm_slot)
author maintainer
#70 lore raw

PATCH 70/204 hugetlbfs: remove call to huge_pte_alloc without i_mmap_rwsem

In reply to Greg Kroah-Hartman

From: Mike Kravetz <[email protected]>

commit 34ae204f18519f0920bd50a644abd6fefc8dbfcf upstream.

Commit c0d0381ade79 ("hugetlbfs: use i_mmap_rwsem for more pmd sharing synchronization") requires callers of huge_pte_alloc to hold i_mmap_rwsem in at least read mode. This is because the explicit locking in huge_pmd_share (called by huge_pte_alloc) was removed. When restructuring the code, the call to huge_pte_alloc in the else block at the beginning of hugetlb_fault was missed.

Unfortunately, that else clause is exercised when there is no page table entry. This will likely lead to a call to huge_pmd_share. If huge_pmd_share thinks pmd sharing is possible, it will traverse the mapping tree (i_mmap) without holding i_mmap_rwsem. If someone else is modifying the tree, bad things such as addressing exceptions or worse could happen.

Simply remove the else clause. It should have been removed previously. The code following the else will call huge_pte_alloc with the appropriate locking.

To prevent this type of issue in the future, add routines to assert that i_mmap_rwsem is held, and call these routines in huge pmd sharing routines.

Fixes c0d0381ade79 ("hugetlbfs: use i_mmap_rwsem for more pmd sharing synchronization")
Cc Michal Hocko <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Naoya Horiguchi <[email protected]>
Cc "Aneesh Kumar K.V" <[email protected]>
Cc Andrea Arcangeli <[email protected]>
Cc "Kirill A.Shutemov" <[email protected]>
Cc Davidlohr Bueso <[email protected]>
Cc Prakash Sangappa <[email protected]>
---
 include/linux/fs.h      |   10 ++++++++++
 include/linux/hugetlb.h |    8 +++++---
 mm/hugetlb.c            |   15 +++++++--------
 mm/rmap.c               |    2 +-
 4 files changed, 23 insertions(+), 12 deletions(-)
include/linux/fs.h +10 -0
@@ -546,6 +546,16 @@ static inline void i_mmap_unlock_read(st
 	up_read(&mapping->i_mmap_rwsem);
 }
 
+static inline void i_mmap_assert_locked(struct address_space *mapping)
+{
+	lockdep_assert_held(&mapping->i_mmap_rwsem);
+}
+
+static inline void i_mmap_assert_write_locked(struct address_space *mapping)
+{
+	lockdep_assert_held_write(&mapping->i_mmap_rwsem);
+}
+
 /*
  * Might pages of this file be mapped into userspace?
  */
include/linux/hugetlb.h +5 -3
@@ -165,7 +165,8 @@ pte_t *huge_pte_alloc(struct mm_struct *
 			unsigned long addr, unsigned long sz);
 pte_t *huge_pte_offset(struct mm_struct *mm,
 		       unsigned long addr, unsigned long sz);
-int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep);
+int huge_pmd_unshare(struct mm_struct *mm, struct vm_area_struct *vma,
+				unsigned long *addr, pte_t *ptep);
 void adjust_range_if_pmd_sharing_possible(struct vm_area_struct *vma,
 				unsigned long *start, unsigned long *end);
 struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address,
@@ -204,8 +205,9 @@ static inline struct address_space *huge
 	return NULL;
 }
 
-static inline int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr,
-					pte_t *ptep)
+static inline int huge_pmd_unshare(struct mm_struct *mm,
+					struct vm_area_struct *vma,
+					unsigned long *addr, pte_t *ptep)
 {
 	return 0;
 }
mm/hugetlb.c +7 -8
@@ -3840,7 +3840,7 @@ void __unmap_hugepage_range(struct mmu_g
 			continue;
 
 		ptl = huge_pte_lock(h, mm, ptep);
-		if (huge_pmd_unshare(mm, &address, ptep)) {
+		if (huge_pmd_unshare(mm, vma, &address, ptep)) {
 			spin_unlock(ptl);
 			/*
 			 * We just unmapped a page of PMDs by clearing a PUD.
@@ -4427,10 +4427,6 @@ vm_fault_t hugetlb_fault(struct mm_struc
 		} else if (unlikely(is_hugetlb_entry_hwpoisoned(entry)))
 			return VM_FAULT_HWPOISON_LARGE |
 				VM_FAULT_SET_HINDEX(hstate_index(h));
-	} else {
-		ptep = huge_pte_alloc(mm, haddr, huge_page_size(h));
-		if (!ptep)
-			return VM_FAULT_OOM;
 	}
 
 	/*
@@ -4907,7 +4903,7 @@ unsigned long hugetlb_change_protection(
 		if (!ptep)
 			continue;
 		ptl = huge_pte_lock(h, mm, ptep);
-		if (huge_pmd_unshare(mm, &address, ptep)) {
+		if (huge_pmd_unshare(mm, vma, &address, ptep)) {
 			pages++;
 			spin_unlock(ptl);
 			shared_pmd = true;
@@ -5288,12 +5284,14 @@ out:
  * returns: 1 successfully unmapped a shared pte page
  *	    0 the underlying pte page is not shared, or it is the last user
  */
-int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep)
+int huge_pmd_unshare(struct mm_struct *mm, struct vm_area_struct *vma,
+					unsigned long *addr, pte_t *ptep)
 {
 	pgd_t *pgd = pgd_offset(mm, *addr);
 	p4d_t *p4d = p4d_offset(pgd, *addr);
 	pud_t *pud = pud_offset(p4d, *addr);
 
+	i_mmap_assert_write_locked(vma->vm_file->f_mapping);
 	BUG_ON(page_count(virt_to_page(ptep)) == 0);
 	if (page_count(virt_to_page(ptep)) == 1)
 		return 0;
@@ -5311,7 +5309,8 @@ pte_t *huge_pmd_share(struct mm_struct *
 	return NULL;
 }
 
-int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep)
+int huge_pmd_unshare(struct mm_struct *mm, struct vm_area_struct *vma,
+				unsigned long *addr, pte_t *ptep)
 {
 	return 0;
 }
mm/rmap.c +1 -1
@@ -1458,7 +1458,7 @@ static bool try_to_unmap_one(struct page
 			 * do this outside rmap routines.
 			 */
 			VM_BUG_ON(!(flags & TTU_RMAP_LOCKED));
-			if (huge_pmd_unshare(mm, &address, pvmw.pte)) {
+			if (huge_pmd_unshare(mm, vma, &address, pvmw.pte)) {
 				/*
 				 * huge_pmd_unshare unmapped an entire PMD
 				 * page.  There is no way of knowing exactly
author maintainer
#71 lore raw

PATCH 71/204 mm/shuffle: dont move pages between zones and dont read garbage memmaps

In reply to Greg Kroah-Hartman

From: David Hildenbrand <[email protected]>

commit 4a93025cbe4a0b19d1a25a2d763a3d2018bad0d9 upstream.

Especially with memory hotplug, we can have offline sections (with a garbage memmap) and overlapping zones. We have to make sure to only touch initialized memmaps (online sections managed by the buddy) and that the zone matches, to not move pages between zones.

To test if this can actually happen, I added a simple

	BUG_ON(page_zone(page_i) != page_zone(page_j));

right before the swap. When hotplugging a 256M DIMM to a 4G x86-64 VM and onlining the first memory block "online_movable" and the second memory block "online_kernel", it will trigger the BUG, as both zones (NORMAL and MOVABLE) overlap.

This might result in all kinds of weird situations (e.g., double allocations, list corruptions, unmovable allocations ending up in the movable zone).

Fixes e900a918b098 ("mm: shuffle initial free memory to improve memory-side-cache utilization")
Cc Andrew Morton <[email protected]>
Cc Johannes Weiner <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Minchan Kim <[email protected]>
Cc Huang Ying <[email protected]>
Cc Wei Yang <[email protected]>
Cc Mel Gorman <[email protected]>
Cc <[email protected]> [5.2+]
---
 mm/shuffle.c |   18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)
mm/shuffle.c +9 -9
@@ -58,25 +58,25 @@ module_param_call(shuffle, shuffle_store
  * For two pages to be swapped in the shuffle, they must be free (on a
  * 'free_area' lru), have the same order, and have the same migratetype.
  */
-static struct page * __meminit shuffle_valid_page(unsigned long pfn, int order)
+static struct page * __meminit shuffle_valid_page(struct zone *zone,
+						  unsigned long pfn, int order)
 {
-	struct page *page;
+	struct page *page = pfn_to_online_page(pfn);
 
 	/*
 	 * Given we're dealing with randomly selected pfns in a zone we
 	 * need to ask questions like...
 	 */
 
-	/* ...is the pfn even in the memmap? */
-	if (!pfn_valid_within(pfn))
+	/* ... is the page managed by the buddy? */
+	if (!page)
 		return NULL;
 
-	/* ...is the pfn in a present section or a hole? */
-	if (!pfn_in_present_section(pfn))
+	/* ... is the page assigned to the same zone? */
+	if (page_zone(page) != zone)
 		return NULL;
 
 	/* ...is the page free and currently on a free_area list? */
-	page = pfn_to_page(pfn);
 	if (!PageBuddy(page))
 		return NULL;
 
@@ -123,7 +123,7 @@ void __meminit __shuffle_zone(struct zon
 		 * page_j randomly selected in the span @zone_start_pfn to
 		 * @spanned_pages.
 		 */
-		page_i = shuffle_valid_page(i, order);
+		page_i = shuffle_valid_page(z, i, order);
 		if (!page_i)
 			continue;
 
@@ -137,7 +137,7 @@ void __meminit __shuffle_zone(struct zon
 			j = z->zone_start_pfn +
 				ALIGN_DOWN(get_random_long() % z->spanned_pages,
 						order_pages);
-			page_j = shuffle_valid_page(j, order);
+			page_j = shuffle_valid_page(z, j, order);
 			if (page_j && page_j != page_i)
 				break;
 		}
author maintainer
#72 lore raw

PATCH 72/204 ocfs2: change slot number type s16 to u16

In reply to Greg Kroah-Hartman

From: Junxiao Bi <[email protected]>

commit 38d51b2dd171ad973afc1f5faab825ed05a2d5e9 upstream.

Dan Carpenter reported the following static checker warning.

	fs/ocfs2/super.c:1269 ocfs2_parse_options() warn: '(-1)' 65535 can't fit into 32767 'mopt->slot'
	fs/ocfs2/suballoc.c:859 ocfs2_init_inode_steal_slot() warn: '(-1)' 65535 can't fit into 32767 'osb->s_inode_steal_slot'
	fs/ocfs2/suballoc.c:867 ocfs2_init_meta_steal_slot() warn: '(-1)' 65535 can't fit into 32767 'osb->s_meta_steal_slot'

That's because OCFS2_INVALID_SLOT is (u16)-1. Slot number in ocfs2 can be never negative, so change s16 to u16.

---
 fs/ocfs2/ocfs2.h    |    4 ++--
 fs/ocfs2/suballoc.c |    4 ++--
 fs/ocfs2/super.c    |    4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)
fs/ocfs2/ocfs2.h +2 -2
@@ -326,8 +326,8 @@ struct ocfs2_super
 	spinlock_t osb_lock;
 	u32 s_next_generation;
 	unsigned long osb_flags;
-	s16 s_inode_steal_slot;
-	s16 s_meta_steal_slot;
+	u16 s_inode_steal_slot;
+	u16 s_meta_steal_slot;
 	atomic_t s_num_inodes_stolen;
 	atomic_t s_num_meta_stolen;
 
fs/ocfs2/suballoc.c +2 -2
@@ -879,9 +879,9 @@ static void __ocfs2_set_steal_slot(struc
 {
 	spin_lock(&osb->osb_lock);
 	if (type == INODE_ALLOC_SYSTEM_INODE)
-		osb->s_inode_steal_slot = slot;
+		osb->s_inode_steal_slot = (u16)slot;
 	else if (type == EXTENT_ALLOC_SYSTEM_INODE)
-		osb->s_meta_steal_slot = slot;
+		osb->s_meta_steal_slot = (u16)slot;
 	spin_unlock(&osb->osb_lock);
 }
 
fs/ocfs2/super.c +2 -2
@@ -78,7 +78,7 @@ struct mount_options
 	unsigned long	commit_interval;
 	unsigned long	mount_opt;
 	unsigned int	atime_quantum;
-	signed short	slot;
+	unsigned short	slot;
 	int		localalloc_opt;
 	unsigned int	resv_level;
 	int		dir_resv_level;
@@ -1334,7 +1334,7 @@ static int ocfs2_parse_options(struct su
 				goto bail;
 			}
 			if (option)
-				mopt->slot = (s16)option;
+				mopt->slot = (u16)option;
 			break;
 		case Opt_commit:
 			if (match_int(&args[0], &option)) {
author maintainer
#73 lore raw

PATCH 73/204 mm/page_counter.c: fix protection usage propagation

In reply to Greg Kroah-Hartman

From: Michal Koutný <[email protected]>

commit a6f23d14ec7d7d02220ad8bb2774be3322b9aeec upstream.

When workload runs in cgroups that aren't directly below root cgroup and their parent specifies reclaim protection, it may end up ineffective.

The reason is that propagate_protected_usage() is not called in all hierarchy up. All the protected usage is incorrectly accumulated in the workload's parent. This means that siblings_low_usage is overestimated and effective protection underestimated. Even though it is transitional phenomenon (uncharge path does correct propagation and fixes the wrong children_low_usage), it can undermine the intended protection unexpectedly.

We have noticed this problem while seeing a swap out in a descendant of a protected memcg (intermediate node) while the parent was conveniently under its protection limit and the memory pressure was external to that hierarchy. Michal has pinpointed this down to the wrong siblings_low_usage which led to the unwanted reclaim.

The fix is simply updating children_low_usage in respective ancestors also in the charging path.

---
 mm/page_counter.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
mm/page_counter.c +3 -3
@@ -72,7 +72,7 @@ void page_counter_charge(struct page_cou
 		long new;
 
 		new = atomic_long_add_return(nr_pages, &c->usage);
-		propagate_protected_usage(counter, new);
+		propagate_protected_usage(c, new);
 		/*
 		 * This is indeed racy, but we can live with some
 		 * inaccuracy in the watermark.
@@ -116,7 +116,7 @@ bool page_counter_try_charge(struct page
 		new = atomic_long_add_return(nr_pages, &c->usage);
 		if (new > c->max) {
 			atomic_long_sub(nr_pages, &c->usage);
-			propagate_protected_usage(counter, new);
+			propagate_protected_usage(c, new);
 			/*
 			 * This is racy, but we can live with some
 			 * inaccuracy in the failcnt.
@@ -125,7 +125,7 @@ bool page_counter_try_charge(struct page
 			*fail = c;
 			goto failed;
 		}
-		propagate_protected_usage(counter, new);
+		propagate_protected_usage(c, new);
 		/*
 		 * Just like with failcnt, we can live with some
 		 * inaccuracy in the watermark.
author maintainer
#74 lore raw

PATCH 74/204 cma: dont quit at first error when activating reserved areas

In reply to Greg Kroah-Hartman

From: Mike Kravetz <[email protected]>

commit 3a5139f1c5bb76d69756fb8f13fffa173e261153 upstream.

The routine cma_init_reserved_areas is designed to activate all reserved cma areas. It quits when it first encounters an error. This can leave some areas in a state where they are reserved but not activated. There is no feedback to code which performed the reservation. Attempting to allocate memory from areas in such a state will result in a BUG.

Modify cma_init_reserved_areas to always attempt to activate all areas. The called routine, cma_activate_area is responsible for leaving the area in a valid state. No one is making active use of returned error codes, so change the routine to void.

How to reproduce: This example uses kernelcore, hugetlb and cma as an easy way to reproduce. However, this is a more general cma issue.

Two node x86 VM 16GB total, 8GB per node
Kernel command line parameters, kernelcore=4G hugetlb_cma=8G Related boot time messages,
hugetlb_cma: reserve 8192 MiB, up to 4096 MiB per node
cma: Reserved 4096 MiB at 0x0000000100000000
hugetlb_cma: reserved 4096 MiB on node 0
cma: Reserved 4096 MiB at 0x0000000300000000
hugetlb_cma: reserved 4096 MiB on node 1
cma: CMA area hugetlb could not be activated

 # echo 8 > /sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages
  BUG: kernel NULL pointer dereference, address: 0000000000000000
  #PF: supervisor read access in kernel mode
  #PF: error_code(0x0000) - not-present page
  PGD 0 P4D 0
  Oops: 0000 [#1] SMP PTI
  ...
  Call Trace:
    bitmap_find_next_zero_area_off+0x51/0x90
    cma_alloc+0x1a5/0x310
    alloc_fresh_huge_page+0x78/0x1a0
    alloc_pool_huge_page+0x6f/0xf0
    set_max_huge_pages+0x10c/0x250
    nr_hugepages_store_common+0x92/0x120
    ? __kmalloc+0x171/0x270
    kernfs_fop_write+0xc1/0x1a0
    vfs_write+0xc7/0x1f0
    ksys_write+0x5f/0xe0
    do_syscall_64+0x4d/0x90
    entry_SYSCALL_64_after_hwframe+0x44/0xa9
---
 mm/cma.c |   23 +++++++++--------------
 1 file changed, 9 insertions(+), 14 deletions(-)
mm/cma.c +9 -14
@@ -93,17 +93,15 @@ static void cma_clear_bitmap(struct cma
 	mutex_unlock(&cma->lock);
 }
 
-static int __init cma_activate_area(struct cma *cma)
+static void __init cma_activate_area(struct cma *cma)
 {
 	unsigned long base_pfn = cma->base_pfn, pfn = base_pfn;
 	unsigned i = cma->count >> pageblock_order;
 	struct zone *zone;
 
 	cma->bitmap = bitmap_zalloc(cma_bitmap_maxno(cma), GFP_KERNEL);
-	if (!cma->bitmap) {
-		cma->count = 0;
-		return -ENOMEM;
-	}
+	if (!cma->bitmap)
+		goto out_error;
 
 	WARN_ON_ONCE(!pfn_valid(pfn));
 	zone = page_zone(pfn_to_page(pfn));
@@ -133,25 +131,22 @@ static int __init cma_activate_area(stru
 	spin_lock_init(&cma->mem_head_lock);
 #endif
 
-	return 0;
+	return;
 
 not_in_zone:
-	pr_err("CMA area %s could not be activated\n", cma->name);
 	bitmap_free(cma->bitmap);
+out_error:
 	cma->count = 0;
-	return -EINVAL;
+	pr_err("CMA area %s could not be activated\n", cma->name);
+	return;
 }
 
 static int __init cma_init_reserved_areas(void)
 {
 	int i;
 
-	for (i = 0; i < cma_area_count; i++) {
-		int ret = cma_activate_area(&cma_areas[i]);
-
-		if (ret)
-			return ret;
-	}
+	for (i = 0; i < cma_area_count; i++)
+		cma_activate_area(&cma_areas[i]);
 
 	return 0;
 }
author maintainer
#75 lore raw

PATCH 75/204 mm/memory_hotplug: fix unpaired mem_hotplug_begin/done

In reply to Greg Kroah-Hartman

From: Jia He <[email protected]>

commit b4223a510e2ab1bf0f971d50af7c1431014b25ad upstream.

When check_memblock_offlined_cb() returns failed rc(e.g. the memblock is online at that time), mem_hotplug_begin/done is unpaired in such case.

Therefore a warning:

 Call Trace:
  percpu_up_write+0x33/0x40
  try_remove_memory+0x66/0x120
  ? _cond_resched+0x19/0x30
  remove_memory+0x2b/0x40
  dev_dax_kmem_remove+0x36/0x72 [kmem]
  device_release_driver_internal+0xf0/0x1c0
  device_release_driver+0x12/0x20
  bus_remove_device+0xe1/0x150
  device_del+0x17b/0x3e0
  unregister_dev_dax+0x29/0x60
  devm_action_release+0x15/0x20
  release_nodes+0x19a/0x1e0
  devres_release_all+0x3f/0x50
  device_release_driver_internal+0x100/0x1c0
  driver_detach+0x4c/0x8f
  bus_remove_driver+0x5c/0xd0
  driver_unregister+0x31/0x50
  dax_pmem_exit+0x10/0xfe0 [dax_pmem]
Fixes f1037ec0cc8a ("mm/memory_hotplug: fix remove_memory() lockdep splat")
Cc <[email protected]> [5.6+]
Cc Andy Lutomirski <[email protected]>
Cc Baoquan He <[email protected]>
Cc Borislav Petkov <[email protected]>
Cc Catalin Marinas <[email protected]>
Cc Chuhong Yuan <[email protected]>
Cc Dave Hansen <[email protected]>
Cc Dave Jiang <[email protected]>
Cc Fenghua Yu <[email protected]>
Cc "H. Peter Anvin" <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Jonathan Cameron <[email protected]>
Cc Kaly Xin <[email protected]>
Cc Logan Gunthorpe <[email protected]>
Cc Masahiro Yamada <[email protected]>
Cc Mike Rapoport <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Rich Felker <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Tony Luck <[email protected]>
Cc Vishal Verma <[email protected]>
Cc Will Deacon <[email protected]>
Cc Yoshinori Sato <[email protected]>
---
 mm/memory_hotplug.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
mm/memory_hotplug.c +2 -3
@@ -1745,7 +1745,7 @@ static int __ref try_remove_memory(int n
 	 */
 	rc = walk_memory_blocks(start, size, NULL, check_memblock_offlined_cb);
 	if (rc)
-		goto done;
+		return rc;
 
 	/* remove memmap entry */
 	firmware_map_remove(start, start + size, "System RAM");
@@ -1765,9 +1765,8 @@ static int __ref try_remove_memory(int n
 
 	try_offline_node(nid);
 
-done:
 	mem_hotplug_done();
-	return rc;
+	return 0;
 }
 
 /**
author maintainer
#76 lore raw

PATCH 76/204 ftrace: Setup correct FTRACE_FL_REGS flags for module

In reply to Greg Kroah-Hartman

From: Chengming Zhou <[email protected]>

commit 8a224ffb3f52b0027f6b7279854c71a31c48fc97 upstream.

When module loaded and enabled, we will use __ftrace_replace_code for module if any ftrace_ops referenced it found. But we will get wrong ftrace_addr for module rec in ftrace_get_addr_new, because rec->flags has not been setup correctly. It can cause the callback function of a ftrace_ops has FTRACE_OPS_FL_SAVE_REGS to be called with pt_regs set to NULL.
So setup correct FTRACE_FL_REGS flags for rec when we call referenced_filters to find ftrace_ops references it.

Fixes 8c4f3c3fa9681 ("ftrace: Check module functions being traced on reload")
---
 kernel/trace/ftrace.c |   11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
kernel/trace/ftrace.c +7 -4
@@ -6198,8 +6198,11 @@ static int referenced_filters(struct dyn
 	int cnt = 0;
 
 	for (ops = ftrace_ops_list; ops != &ftrace_list_end; ops = ops->next) {
-		if (ops_references_rec(ops, rec))
-		    cnt++;
+		if (ops_references_rec(ops, rec)) {
+			cnt++;
+			if (ops->flags & FTRACE_OPS_FL_SAVE_REGS)
+				rec->flags |= FTRACE_FL_REGS;
+		}
 	}
 
 	return cnt;
@@ -6378,8 +6381,8 @@ void ftrace_module_enable(struct module
 		if (ftrace_start_up)
 			cnt += referenced_filters(rec);
 
-		/* This clears FTRACE_FL_DISABLED */
-		rec->flags = cnt;
+		rec->flags &= ~FTRACE_FL_DISABLED;
+		rec->flags += cnt;
 
 		if (ftrace_start_up && cnt) {
 			int failed = __ftrace_replace_code(rec, 1);
author maintainer
#77 lore raw

PATCH 77/204 kprobes: Fix NULL pointer dereference at kprobe_ftrace_handler

In reply to Greg Kroah-Hartman

From: Muchun Song <[email protected]>

commit 0cb2f1372baa60af8456388a574af6133edd7d80 upstream.

We found a case of kernel panic on our server. The stack trace is as follows(omit some irrelevant information):

  BUG: kernel NULL pointer dereference, address: 0000000000000080
  RIP: 0010:kprobe_ftrace_handler+0x5e/0xe0
  RSP: 0018:ffffb512c6550998 EFLAGS: 00010282
  RAX: 0000000000000000 RBX: ffff8e9d16eea018 RCX: 0000000000000000
  RDX: ffffffffbe1179c0 RSI: ffffffffc0535564 RDI: ffffffffc0534ec0
  RBP: ffffffffc0534ec1 R08: ffff8e9d1bbb0f00 R09: 0000000000000004
  R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
  R13: ffff8e9d1f797060 R14: 000000000000bacc R15: ffff8e9ce13eca00
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000080 CR3: 00000008453d0005 CR4: 00000000003606e0
  DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
  DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
  Call Trace:
   <IRQ>
   ftrace_ops_assist_func+0x56/0xe0
   ftrace_call+0x5/0x34
   tcpa_statistic_send+0x5/0x130 [ttcp_engine]

The tcpa_statistic_send is the function being kprobed. After analysis, the root cause is that the fourth parameter regs of kprobe_ftrace_handler is NULL. Why regs is NULL? We use the crash tool to analyze the kdump.

crash> dis tcpa_statistic_send -r

         <tcpa_statistic_send>: callq 0xffffffffbd8018c0 <ftrace_caller>

The tcpa_statistic_send calls ftrace_caller instead of ftrace_regs_caller. So it is reasonable that the fourth parameter regs of kprobe_ftrace_handler is NULL. In theory, we should call the ftrace_regs_caller instead of the ftrace_caller. After in-depth analysis, we found a reproducible path.

Writing a simple kernel module which starts a periodic timer. The timer's handler is named 'kprobe_test_timer_handler'. The module name is kprobe_test.ko.

1) insmod kprobe_test.ko
2) bpftrace -e 'kretprobe:kprobe_test_timer_handler {}'
3) echo 0 > /proc/sys/kernel/ftrace_enabled
4) rmmod kprobe_test
5) stop step 2) kprobe
6) insmod kprobe_test.ko
7) bpftrace -e 'kretprobe:kprobe_test_timer_handler {}'

We mark the kprobe as GONE but not disarm the kprobe in the step 4). The step 5) also do not disarm the kprobe when unregister kprobe. So we do not remove the ip from the filter. In this case, when the module loads again in the step 6), we will replace the code to ftrace_caller via the ftrace_module_enable(). When we register kprobe again, we will not replace ftrace_caller to ftrace_regs_caller because the ftrace is disabled in the step 3). So the step 7) will trigger kernel panic. Fix this problem by disarming the kprobe when the module is going away.

---
 kernel/kprobes.c |    7 +++++++
 1 file changed, 7 insertions(+)
kernel/kprobes.c +7 -0
@@ -2104,6 +2104,13 @@ static void kill_kprobe(struct kprobe *p
 	 * the original probed function (which will be freed soon) any more.
 	 */
 	arch_remove_kprobe(p);
+
+	/*
+	 * The module is going away. We should disarm the kprobe which
+	 * is using ftrace.
+	 */
+	if (kprobe_ftrace(p))
+		disarm_kprobe_ftrace(p);
 }
 
 /* Disable one kprobe */
author maintainer
#78 lore raw

PATCH 78/204 tracing/hwlat: Honor the tracing_cpumask

In reply to Greg Kroah-Hartman

From: Kevin Hao <[email protected]>

commit 96b4833b6827a62c295b149213c68b559514c929 upstream.

In calculation of the cpu mask for the hwlat kernel thread, the wrong cpu mask is used instead of the tracing_cpumask, this causes the tracing/tracing_cpumask useless for hwlat tracer. Fixes it.

Cc Ingo Molnar <[email protected]>
Fixes 0330f7aa8ee6 ("tracing: Have hwlat trace migrate across tracing_cpumask CPUs")
---
 kernel/trace/trace_hwlat.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
kernel/trace/trace_hwlat.c +3 -2
@@ -283,6 +283,7 @@ static bool disable_migrate;
 static void move_to_next_cpu(void)
 {
 	struct cpumask *current_mask = &save_cpumask;
+	struct trace_array *tr = hwlat_trace;
 	int next_cpu;
 
 	if (disable_migrate)
@@ -296,7 +297,7 @@ static void move_to_next_cpu(void)
 		goto disable;
 
 	get_online_cpus();
-	cpumask_and(current_mask, cpu_online_mask, tracing_buffer_mask);
+	cpumask_and(current_mask, cpu_online_mask, tr->tracing_cpumask);
 	next_cpu = cpumask_next(smp_processor_id(), current_mask);
 	put_online_cpus();
 
@@ -373,7 +374,7 @@ static int start_kthread(struct trace_ar
 	/* Just pick the first CPU on first iteration */
 	current_mask = &save_cpumask;
 	get_online_cpus();
-	cpumask_and(current_mask, cpu_online_mask, tracing_buffer_mask);
+	cpumask_and(current_mask, cpu_online_mask, tr->tracing_cpumask);
 	put_online_cpus();
 	next_cpu = cpumask_first(current_mask);
 
author maintainer
#79 lore raw

PATCH 79/204 tracing: Use trace_sched_process_free() instead of exit() for pid tracing

In reply to Greg Kroah-Hartman

From: Steven Rostedt (VMware) <[email protected]>

commit afcab636657421f7ebfa0783a91f90256bba0091 upstream.

On exit, if a process is preempted after the trace_sched_process_exit() tracepoint but before the process is done exiting, then when it gets scheduled in, the function tracers will not filter it properly against the function tracing pid filters.

That is because the function tracing pid filters hooks to the sched_process_exit() tracepoint to remove the exiting task's pid from the filter list. Because the filtering happens at the sched_switch tracepoint, when the exiting task schedules back in to finish up the exit, it will no longer be in the function pid filtering tables.

This was noticeable in the notrace self tests on a preemptable kernel, as the tests would fail as it exits and preempted after being taken off the notrace filter table and on scheduling back in it would not be in the notrace list, and then the ending of the exit function would trace. The test detected this and would fail.

Cc Namhyung Kim <[email protected]>
Fixes 1e10486ffee0a ("ftrace: Add 'function-fork' trace option")
Fixes c37775d57830a ("tracing: Add infrastructure to allow set_event_pid to follow children"
---
 kernel/trace/ftrace.c       |    4 ++--
 kernel/trace/trace_events.c |    4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)
kernel/trace/ftrace.c +2 -2
@@ -6980,12 +6980,12 @@ void ftrace_pid_follow_fork(struct trace
 	if (enable) {
 		register_trace_sched_process_fork(ftrace_pid_follow_sched_process_fork,
 						  tr);
-		register_trace_sched_process_exit(ftrace_pid_follow_sched_process_exit,
+		register_trace_sched_process_free(ftrace_pid_follow_sched_process_exit,
 						  tr);
 	} else {
 		unregister_trace_sched_process_fork(ftrace_pid_follow_sched_process_fork,
 						    tr);
-		unregister_trace_sched_process_exit(ftrace_pid_follow_sched_process_exit,
+		unregister_trace_sched_process_free(ftrace_pid_follow_sched_process_exit,
 						    tr);
 	}
 }
kernel/trace/trace_events.c +2 -2
@@ -538,12 +538,12 @@ void trace_event_follow_fork(struct trac
 	if (enable) {
 		register_trace_prio_sched_process_fork(event_filter_pid_sched_process_fork,
 						       tr, INT_MIN);
-		register_trace_prio_sched_process_exit(event_filter_pid_sched_process_exit,
+		register_trace_prio_sched_process_free(event_filter_pid_sched_process_exit,
 						       tr, INT_MAX);
 	} else {
 		unregister_trace_sched_process_fork(event_filter_pid_sched_process_fork,
 						    tr);
-		unregister_trace_sched_process_exit(event_filter_pid_sched_process_exit,
+		unregister_trace_sched_process_free(event_filter_pid_sched_process_exit,
 						    tr);
 	}
 }
author maintainer
#80 lore raw

PATCH 80/204 watchdog: f71808e_wdt: indicate WDIOF_CARDRESET support in watchdog_info.options

In reply to Greg Kroah-Hartman

From: Ahmad Fatoum <[email protected]>

commit e871e93fb08a619dfc015974a05768ed6880fd82 upstream.

The driver supports populating bootstatus with WDIOF_CARDRESET, but so far userspace couldn't portably determine whether absence of this flag meant no watchdog reset or no driver support. Or-in the bit to fix this.

---
 drivers/watchdog/f71808e_wdt.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/watchdog/f71808e_wdt.c +2 -1
@@ -692,7 +692,8 @@ static int __init watchdog_init(int sioa
 	watchdog.sioaddr = sioaddr;
 	watchdog.ident.options = WDIOC_SETTIMEOUT
 				| WDIOF_MAGICCLOSE
-				| WDIOF_KEEPALIVEPING;
+				| WDIOF_KEEPALIVEPING
+				| WDIOF_CARDRESET;
 
 	snprintf(watchdog.ident.identity,
 		sizeof(watchdog.ident.identity), "%s watchdog",
author maintainer
#81 lore raw

PATCH 81/204 watchdog: f71808e_wdt: remove use of wrong watchdog_info option

In reply to Greg Kroah-Hartman

From: Ahmad Fatoum <[email protected]>

commit 802141462d844f2e6a4d63a12260d79b7afc4c34 upstream.

The flags that should be or-ed into the watchdog_info.options by drivers all start with WDIOF_, e.g. WDIOF_SETTIMEOUT, which indicates that the driver's watchdog_ops has a usable set_timeout.

WDIOC_SETTIMEOUT was used instead, which expands to 0xc0045706, which equals:

WDIOF_FANFAULT | WDIOF_EXTERN1 | WDIOF_PRETIMEOUT | WDIOF_ALARMONLY | WDIOF_MAGICCLOSE | 0xc0045000

These were so far indicated to userspace on WDIOC_GETSUPPORT. As the driver has not yet been migrated to the new watchdog kernel API, the constant can just be dropped without substitute.

Fixes 96cb4eb019ce ("watchdog: f71808e_wdt: new watchdog driver for Fintek F71808E and F71882FG")
---
 drivers/watchdog/f71808e_wdt.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/watchdog/f71808e_wdt.c +1 -2
@@ -690,8 +690,7 @@ static int __init watchdog_init(int sioa
 	 * into the module have been registered yet.
 	 */
 	watchdog.sioaddr = sioaddr;
-	watchdog.ident.options = WDIOC_SETTIMEOUT
-				| WDIOF_MAGICCLOSE
+	watchdog.ident.options = WDIOF_MAGICCLOSE
 				| WDIOF_KEEPALIVEPING
 				| WDIOF_CARDRESET;
 
author maintainer
#82 lore raw

PATCH 82/204 watchdog: f71808e_wdt: clear watchdog timeout occurred flag

In reply to Greg Kroah-Hartman

From: Ahmad Fatoum <[email protected]>

commit 4f39d575844148fbf3081571a1f3b4ae04150958 upstream.

The flag indicating a watchdog timeout having occurred normally persists till Power-On Reset of the Fintek Super I/O chip. The user can clear it by writing a `1' to the bit.

The driver doesn't offer a restart method, so regular system reboot might not reset the Super I/O and if the watchdog isn't enabled, we won't touch the register containing the bit on the next boot. In this case all subsequent regular reboots will be wrongly flagged by the driver as being caused by the watchdog.

Fix this by having the flag cleared after read. This is also done by other drivers like those for the i6300esb and mpc8xxx_wdt.

---
 drivers/watchdog/f71808e_wdt.c |    7 +++++++
 1 file changed, 7 insertions(+)
drivers/watchdog/f71808e_wdt.c +7 -0
@@ -706,6 +706,13 @@ static int __init watchdog_init(int sioa
 	wdt_conf = superio_inb(sioaddr, F71808FG_REG_WDT_CONF);
 	watchdog.caused_reboot = wdt_conf & BIT(F71808FG_FLAG_WDTMOUT_STS);
 
+	/*
+	 * We don't want WDTMOUT_STS to stick around till regular reboot.
+	 * Write 1 to the bit to clear it to zero.
+	 */
+	superio_outb(sioaddr, F71808FG_REG_WDT_CONF,
+		     wdt_conf | BIT(F71808FG_FLAG_WDTMOUT_STS));
+
 	superio_exit(sioaddr);
 
 	err = watchdog_set_timeout(timeout);
author maintainer
#83 lore raw

PATCH 83/204 ceph: set sec_context xattr on symlink creation

In reply to Greg Kroah-Hartman

From: Jeff Layton <[email protected]>

commit b748fc7a8763a5b3f8149f12c45711cd73ef8176 upstream.

Symlink inodes should have the security context set in their xattrs on creation. We already set the context on creation, but we don't attach the pagelist. The effect is that symlink inodes don't get an SELinux context set on them at creation, so they end up unlabeled instead of inheriting the proper context. Make it do so.

---
 fs/ceph/dir.c |    4 ++++
 1 file changed, 4 insertions(+)
fs/ceph/dir.c +4 -0
@@ -924,6 +924,10 @@ static int ceph_symlink(struct inode *di
 	req->r_num_caps = 2;
 	req->r_dentry_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_AUTH_EXCL;
 	req->r_dentry_unless = CEPH_CAP_FILE_EXCL;
+	if (as_ctx.pagelist) {
+		req->r_pagelist = as_ctx.pagelist;
+		as_ctx.pagelist = NULL;
+	}
 	err = ceph_mdsc_do_request(mdsc, dir, req);
 	if (!err && !req->r_reply_info.head->is_dentry)
 		err = ceph_handle_notrace_create(dir, dentry);
author maintainer
#84 lore raw

PATCH 84/204 ceph: handle zero-length feature mask in session messages

In reply to Greg Kroah-Hartman

From: Jeff Layton <[email protected]>

commit 02e37571f9e79022498fd0525c073b07e9d9ac69 upstream.

Most session messages contain a feature mask, but the MDS will routinely send a REJECT message with one that is zero-length.

Commit 0fa8263367db ("ceph: fix endianness bug when handling MDS session feature bits") fixed the decoding of the feature mask, but failed to account for the MDS sending a zero-length feature mask. This causes REJECT message decoding to fail.

Skip trying to decode a feature mask if the word count is zero.

Cc: [email protected]
URL: https://tracker.ceph.com/issues/46823

Fixes 0fa8263367db ("ceph: fix endianness bug when handling MDS session feature bits")
---
 fs/ceph/mds_client.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
fs/ceph/mds_client.c +4 -2
@@ -3270,8 +3270,10 @@ static void handle_session(struct ceph_m
 			goto bad;
 		/* version >= 3, feature bits */
 		ceph_decode_32_safe(&p, end, len, bad);
-		ceph_decode_64_safe(&p, end, features, bad);
-		p += len - sizeof(features);
+		if (len) {
+			ceph_decode_64_safe(&p, end, features, bad);
+			p += len - sizeof(features);
+		}
 	}
 
 	mutex_lock(&mdsc->mutex);
author maintainer
#85 lore raw

PATCH 85/204 pseries: Fix 64 bit logical memory block panic

In reply to Greg Kroah-Hartman

From: Anton Blanchard <[email protected]>

commit 89c140bbaeee7a55ed0360a88f294ead2b95201b upstream.

Booting with a 4GB LMB size causes us to panic:

qemu-system-ppc64: OS terminated: OS panic:

      Memory block size not suitable: 0x0

Fix pseries_memory_block_size() to handle 64 bit LMBs.

---
 arch/powerpc/platforms/pseries/hotplug-memory.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/platforms/pseries/hotplug-memory.c +1 -1
@@ -27,7 +27,7 @@ static bool rtas_hp_event;
 unsigned long pseries_memory_block_size(void)
 {
 	struct device_node *np;
-	unsigned int memblock_size = MIN_MEMORY_BLOCK_SIZE;
+	u64 memblock_size = MIN_MEMORY_BLOCK_SIZE;
 	struct resource r;
 
 	np = of_find_node_by_path("/ibm,dynamic-reconfiguration-memory");
author maintainer
#86 lore raw

PATCH 86/204 dm: dont call report zones for more than the user requested

In reply to Greg Kroah-Hartman

From: Johannes Thumshirn <[email protected]>

commit a9cb9f4148ef6bb8fabbdaa85c42b2171fbd5a0d upstream.

Don't call report zones for more zones than the user actually requested, otherwise this can lead to out-of-bounds accesses in the callback functions.

Such a situation can happen if the target's ->report_zones() callback function returns 0 because we've reached the end of the target and then restart the report zones on the second target.

We're again calling into ->report_zones() and ultimately into the user supplied callback function but when we're not subtracting the number of zones already processed this may lead to out-of-bounds accesses in the user callbacks.

Fixes d41003513e61 ("block: rework zone reporting")
---
 drivers/md/dm.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/md/dm.c +2 -1
@@ -503,7 +503,8 @@ static int dm_blk_report_zones(struct ge
 		}
 
 		args.tgt = tgt;
-		ret = tgt->type->report_zones(tgt, &args, nr_zones);
+		ret = tgt->type->report_zones(tgt, &args,
+					      nr_zones - args.zone_idx);
 		if (ret < 0)
 			goto out;
 	} while (args.zone_idx < nr_zones &&
author maintainer
#87 lore raw

PATCH 87/204 module: Correctly truncate sysfs sections output

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

commit 11990a5bd7e558e9203c1070fc52fb6f0488e75b upstream.

The only-root-readable /sys/module/$module/sections/$section files did not truncate their output to the available buffer size. While most paths into the kernfs read handlers end up using PAGE_SIZE buffers, it's possible to get there through other paths (e.g. splice, sendfile). Actually limit the output to the "count" passed into the read function, and report it back correctly. *sigh*

---
 kernel/module.c |   22 +++++++++++++++++++---
 1 file changed, 19 insertions(+), 3 deletions(-)
kernel/module.c +19 -3
@@ -1517,18 +1517,34 @@ struct module_sect_attrs {
 	struct module_sect_attr attrs[];
 };
 
+#define MODULE_SECT_READ_SIZE (3 /* "0x", "\n" */ + (BITS_PER_LONG / 4))
 static ssize_t module_sect_read(struct file *file, struct kobject *kobj,
 				struct bin_attribute *battr,
 				char *buf, loff_t pos, size_t count)
 {
 	struct module_sect_attr *sattr =
 		container_of(battr, struct module_sect_attr, battr);
+	char bounce[MODULE_SECT_READ_SIZE + 1];
+	size_t wrote;
 
 	if (pos != 0)
 		return -EINVAL;
 
-	return sprintf(buf, "0x%px\n",
-		       kallsyms_show_value(file->f_cred) ? (void *)sattr->address : NULL);
+	/*
+	 * Since we're a binary read handler, we must account for the
+	 * trailing NUL byte that sprintf will write: if "buf" is
+	 * too small to hold the NUL, or the NUL is exactly the last
+	 * byte, the read will look like it got truncated by one byte.
+	 * Since there is no way to ask sprintf nicely to not write
+	 * the NUL, we have to use a bounce buffer.
+	 */
+	wrote = scnprintf(bounce, sizeof(bounce), "0x%px\n",
+			 kallsyms_show_value(file->f_cred)
+				? (void *)sattr->address : NULL);
+	count = min(count, wrote);
+	memcpy(buf, bounce, count);
+
+	return count;
 }
 
 static void free_sect_attrs(struct module_sect_attrs *sect_attrs)
@@ -1577,7 +1593,7 @@ static void add_sect_attrs(struct module
 			goto out;
 		sect_attrs->nsections++;
 		sattr->battr.read = module_sect_read;
-		sattr->battr.size = 3 /* "0x", "\n" */ + (BITS_PER_LONG / 4);
+		sattr->battr.size = MODULE_SECT_READ_SIZE;
 		sattr->battr.attr.mode = 0400;
 		*(gattr++) = &(sattr++)->battr;
 	}
author maintainer
#88 lore raw

PATCH 88/204 bootconfig: Fix to find the initargs correctly

In reply to Greg Kroah-Hartman

From: Masami Hiramatsu <[email protected]>

commit 477d08478170469d10b533624342d13701e24b34 upstream.

Since the parse_args() stops parsing at '--', bootconfig_params() will never get the '--' as param and initargs_found never be true. In the result, if we pass some init arguments via the bootconfig, those are always appended to the kernel command line with '--' even if the kernel command line already has '--'.

To fix this correctly, check the return value of parse_args() and set initargs_found true if the return value is not an error but a valid address.

Fixes f61872bb58a1 ("bootconfig: Use parse_args() to find bootconfig and '--'")
---
 init/main.c |   14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)
init/main.c +8 -6
@@ -385,8 +385,6 @@ static int __init bootconfig_params(char
 {
 	if (strcmp(param, "bootconfig") == 0) {
 		bootconfig_found = true;
-	} else if (strcmp(param, "--") == 0) {
-		initargs_found = true;
 	}
 	return 0;
 }
@@ -397,19 +395,23 @@ static void __init setup_boot_config(con
 	const char *msg;
 	int pos;
 	u32 size, csum;
-	char *data, *copy;
+	char *data, *copy, *err;
 	int ret;
 
 	/* Cut out the bootconfig data even if we have no bootconfig option */
 	data = get_boot_config_from_initrd(&size, &csum);
 
 	strlcpy(tmp_cmdline, boot_command_line, COMMAND_LINE_SIZE);
-	parse_args("bootconfig", tmp_cmdline, NULL, 0, 0, 0, NULL,
-		   bootconfig_params);
+	err = parse_args("bootconfig", tmp_cmdline, NULL, 0, 0, 0, NULL,
+			 bootconfig_params);
 
-	if (!bootconfig_found)
+	if (IS_ERR(err) || !bootconfig_found)
 		return;
 
+	/* parse_args() stops at '--' and returns an address */
+	if (err)
+		initargs_found = true;
+
 	if (!data) {
 		pr_err("'bootconfig' found on command line, but no bootconfig found\n");
 		return;
author maintainer
#89 lore raw

PATCH 89/204 perf probe: Fix wrong variable warning when the probe point is not found

In reply to Greg Kroah-Hartman

From: Masami Hiramatsu <[email protected]>

commit 11fd3eb874e73ee8069bcfd54e3c16fa7ce56fe6 upstream.

Fix a wrong "variable not found" warning when the probe point is not found in the debuginfo.

Since the debuginfo__find_probes() can return 0 even if it does not find given probe point in the debuginfo, fill_empty_trace_arg() can be called with tf.ntevs == 0 and it can emit a wrong warning. To fix this, reject ntevs == 0 in fill_empty_trace_arg().

E.g. without this patch;

  # perf probe -x /lib64/libc-2.30.so -a "memcpy arg1=%di"
  Failed to find the location of the '%di' variable at this address.
   Perhaps it has been optimized out.
   Use -V with the --range option to show '%di' location range.
  Added new events:
    probe_libc:memcpy    (on memcpy in /usr/lib64/libc-2.30.so with arg1=%di)
    probe_libc:memcpy    (on memcpy in /usr/lib64/libc-2.30.so with arg1=%di)

You can now use it in all perf tools, such as:

perf record -e probe_libc:memcpy -aR sleep 1

With this;

  # perf probe -x /lib64/libc-2.30.so -a "memcpy arg1=%di"
  Added new events:
    probe_libc:memcpy    (on memcpy in /usr/lib64/libc-2.30.so with arg1=%di)
    probe_libc:memcpy    (on memcpy in /usr/lib64/libc-2.30.so with arg1=%di)

You can now use it in all perf tools, such as:

perf record -e probe_libc:memcpy -aR sleep 1

---
 tools/perf/util/probe-finder.c |    3 +++
 1 file changed, 3 insertions(+)
tools/perf/util/probe-finder.c +3 -0
@@ -1408,6 +1408,9 @@ static int fill_empty_trace_arg(struct p
 	char *type;
 	int i, j, ret;
 
+	if (!ntevs)
+		return -ENOENT;
+
 	for (i = 0; i < pev->nargs; i++) {
 		type = NULL;
 		for (j = 0; j < ntevs; j++) {
author maintainer
#90 lore raw

PATCH 90/204 perf probe: Fix memory leakage when the probe point is not found

In reply to Greg Kroah-Hartman

From: Masami Hiramatsu <[email protected]>

commit 12d572e785b15bc764e956caaa8a4c846fd15694 upstream.

Fix the memory leakage in debuginfo__find_trace_events() when the probe point is not found in the debuginfo. If there is no probe point found in the debuginfo, debuginfo__find_probes() will NOT return -ENOENT, but 0.

Thus the caller of debuginfo__find_probes() must check the tf.ntevs and release the allocated memory for the array of struct probe_trace_event.

The current code releases the memory only if the debuginfo__find_probes() hits an error but not checks tf.ntevs. In the result, the memory allocated on *tevs are not released if tf.ntevs == 0.

This fixes the memory leakage by checking tf.ntevs == 0 in addition to ret < 0.

---
 tools/perf/util/probe-finder.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/util/probe-finder.c +1 -1
@@ -1467,7 +1467,7 @@ int debuginfo__find_trace_events(struct
 	if (ret >= 0 && tf.pf.skip_empty_arg)
 		ret = fill_empty_trace_arg(pev, tf.tevs, tf.ntevs);
 
-	if (ret < 0) {
+	if (ret < 0 || tf.ntevs == 0) {
 		for (i = 0; i < tf.ntevs; i++)
 			clear_probe_trace_event(&tf.tevs[i]);
 		zfree(tevs);
author maintainer
#91 lore raw

PATCH 91/204 perf intel-pt: Fix FUP packet state

In reply to Greg Kroah-Hartman

From: Adrian Hunter <[email protected]>

commit 401136bb084fd021acd9f8c51b52fe0a25e326b2 upstream.

While walking code towards a FUP ip, the packet state is INTEL_PT_STATE_FUP or INTEL_PT_STATE_FUP_NO_TIP. That was mishandled resulting in the state becoming INTEL_PT_STATE_IN_SYNC prematurely. The result was an occasional lost EXSTOP event.

---
 tools/perf/util/intel-pt-decoder/intel-pt-decoder.c |   21 ++++++--------------
 1 file changed, 7 insertions(+), 14 deletions(-)
tools/perf/util/intel-pt-decoder/intel-pt-decoder.c +7 -14
@@ -1164,6 +1164,7 @@ static int intel_pt_walk_fup(struct inte
 			return 0;
 		if (err == -EAGAIN ||
 		    intel_pt_fup_with_nlip(decoder, &intel_pt_insn, ip, err)) {
+			decoder->pkt_state = INTEL_PT_STATE_IN_SYNC;
 			if (intel_pt_fup_event(decoder))
 				return 0;
 			return -EAGAIN;
@@ -1942,17 +1943,13 @@ next:
 			}
 			if (decoder->set_fup_mwait)
 				no_tip = true;
+			if (no_tip)
+				decoder->pkt_state = INTEL_PT_STATE_FUP_NO_TIP;
+			else
+				decoder->pkt_state = INTEL_PT_STATE_FUP;
 			err = intel_pt_walk_fup(decoder);
-			if (err != -EAGAIN) {
-				if (err)
-					return err;
-				if (no_tip)
-					decoder->pkt_state =
-						INTEL_PT_STATE_FUP_NO_TIP;
-				else
-					decoder->pkt_state = INTEL_PT_STATE_FUP;
-				return 0;
-			}
+			if (err != -EAGAIN)
+				return err;
 			if (no_tip) {
 				no_tip = false;
 				break;
@@ -2599,15 +2596,11 @@ const struct intel_pt_state *intel_pt_de
 			err = intel_pt_walk_tip(decoder);
 			break;
 		case INTEL_PT_STATE_FUP:
-			decoder->pkt_state = INTEL_PT_STATE_IN_SYNC;
 			err = intel_pt_walk_fup(decoder);
 			if (err == -EAGAIN)
 				err = intel_pt_walk_fup_tip(decoder);
-			else if (!err)
-				decoder->pkt_state = INTEL_PT_STATE_FUP;
 			break;
 		case INTEL_PT_STATE_FUP_NO_TIP:
-			decoder->pkt_state = INTEL_PT_STATE_IN_SYNC;
 			err = intel_pt_walk_fup(decoder);
 			if (err == -EAGAIN)
 				err = intel_pt_walk_trace(decoder);
author maintainer
#92 lore raw

PATCH 92/204 perf intel-pt: Fix duplicate branch after CBR

In reply to Greg Kroah-Hartman

From: Adrian Hunter <[email protected]>

commit a58a057ce65b52125dd355b7d8b0d540ea267a5f upstream.

CBR events can result in a duplicate branch event, because the state type defaults to a branch. Fix by clearing the state type.

Example: trace 'sleep' and hope for a frequency change

Before:

   $ perf record -e intel_pt//u sleep 0.1
   [ perf record: Woken up 1 times to write data ]
   [ perf record: Captured and wrote 0.034 MB perf.data ]
   $ perf script --itrace=bpe > before.txt

After:

   $ perf script --itrace=bpe > after.txt
   $ diff -u before.txt after.txt
#  --- before.txt  2020-07-07 14:42:18.191508098 +0300
#  +++ after.txt   2020-07-07 14:42:36.587891753 +0300
   @@ -29673,7 +29673,6 @@
               sleep 93431 [007] 15411.619905:          1  branches:u:                 0 [unknown] ([unknown]) =>     7f0818abb2e0 clock_nanosleep@@GLIBC_2.17+0x0 (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
               sleep 93431 [007] 15411.619905:          1  branches:u:      7f0818abb30c clock_nanosleep@@GLIBC_2.17+0x2c (/usr/lib/x86_64-linux-gnu/libc-2.31.so) =>                0 [unknown] ([unknown])
               sleep 93431 [007] 15411.720069:         cbr:  cbr: 15 freq: 1507 MHz ( 56%)         7f0818abb30c clock_nanosleep@@GLIBC_2.17+0x2c (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
   -           sleep 93431 [007] 15411.720069:          1  branches:u:      7f0818abb30c clock_nanosleep@@GLIBC_2.17+0x2c (/usr/lib/x86_64-linux-gnu/libc-2.31.so) =>                0 [unknown] ([unknown])
               sleep 93431 [007] 15411.720076:          1  branches:u:                 0 [unknown] ([unknown]) =>     7f0818abb30e clock_nanosleep@@GLIBC_2.17+0x2e (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
               sleep 93431 [007] 15411.720077:          1  branches:u:      7f0818abb323 clock_nanosleep@@GLIBC_2.17+0x43 (/usr/lib/x86_64-linux-gnu/libc-2.31.so) =>     7f0818ac0eb7 __nanosleep+0x17 (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
               sleep 93431 [007] 15411.720077:          1  branches:u:      7f0818ac0ebf __nanosleep+0x1f (/usr/lib/x86_64-linux-gnu/libc-2.31.so) =>     55cb7e4c2827 rpl_nanosleep+0x97 (/usr/bin/sleep)
Fixes 91de8684f1cff ("perf intel-pt: Cater for CBR change in PSB+")
Fixes abe5a1d3e4bee ("perf intel-pt: Decoder to output CBR changes immediately")
Cc Jiri Olsa <[email protected]>
---
 tools/perf/util/intel-pt-decoder/intel-pt-decoder.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
tools/perf/util/intel-pt-decoder/intel-pt-decoder.c +6 -2
@@ -1977,8 +1977,10 @@ next:
 			 * possibility of another CBR change that gets caught up
 			 * in the PSB+.
 			 */
-			if (decoder->cbr != decoder->cbr_seen)
+			if (decoder->cbr != decoder->cbr_seen) {
+				decoder->state.type = 0;
 				return 0;
+			}
 			break;
 
 		case INTEL_PT_PIP:
@@ -2019,8 +2021,10 @@ next:
 
 		case INTEL_PT_CBR:
 			intel_pt_calc_cbr(decoder);
-			if (decoder->cbr != decoder->cbr_seen)
+			if (decoder->cbr != decoder->cbr_seen) {
+				decoder->state.type = 0;
 				return 0;
+			}
 			break;
 
 		case INTEL_PT_MODE_EXEC:
author maintainer
#93 lore raw

PATCH 93/204 gfs2: Never call gfs2_block_zero_range with an open transaction

In reply to Greg Kroah-Hartman

From: Bob Peterson <[email protected]>

commit 70499cdfeb3625c87eebe4f7a7ea06fa7447e5df upstream.

Before this patch, some functions started transactions then they called gfs2_block_zero_range. However, gfs2_block_zero_range, like writes, can start transactions, which results in a recursive transaction error. For example:

do_shrink
trunc_start

      gfs2_trans_begin <------------------------------------------------
         gfs2_block_zero_range
            iomap_zero_range(inode, from, length, NULL, &gfs2_iomap_ops);
               iomap_apply ... iomap_zero_range_actor
                  iomap_begin
                     gfs2_iomap_begin
                        gfs2_iomap_begin_write
                  actor (iomap_zero_range_actor)
		     iomap_zero
			iomap_write_begin
			   gfs2_iomap_page_prepare
			      gfs2_trans_begin <------------------------

This patch reorders the callers of gfs2_block_zero_range so that they only start their transactions after the call. It also adds a BUG_ON to ensure this doesn't happen again.

Fixes 2257e468a63b ("gfs2: implement gfs2_block_zero_range using iomap_zero_range")
---
 fs/gfs2/bmap.c |   69 ++++++++++++++++++++++++++++++++-------------------------
 1 file changed, 39 insertions(+), 30 deletions(-)
fs/gfs2/bmap.c +39 -30
@@ -1351,9 +1351,15 @@ int gfs2_extent_map(struct inode *inode,
 	return ret;
 }
 
+/*
+ * NOTE: Never call gfs2_block_zero_range with an open transaction because it
+ * uses iomap write to perform its actions, which begin their own transactions
+ * (iomap_begin, page_prepare, etc.)
+ */
 static int gfs2_block_zero_range(struct inode *inode, loff_t from,
 				 unsigned int length)
 {
+	BUG_ON(current->journal_info);
 	return iomap_zero_range(inode, from, length, NULL, &gfs2_iomap_ops);
 }
 
@@ -1414,6 +1420,16 @@ static int trunc_start(struct inode *ino
 	u64 oldsize = inode->i_size;
 	int error;
 
+	if (!gfs2_is_stuffed(ip)) {
+		unsigned int blocksize = i_blocksize(inode);
+		unsigned int offs = newsize & (blocksize - 1);
+		if (offs) {
+			error = gfs2_block_zero_range(inode, newsize,
+						      blocksize - offs);
+			if (error)
+				return error;
+		}
+	}
 	if (journaled)
 		error = gfs2_trans_begin(sdp, RES_DINODE + RES_JDATA, GFS2_JTRUNC_REVOKES);
 	else
@@ -1427,19 +1443,10 @@ static int trunc_start(struct inode *ino
 
 	gfs2_trans_add_meta(ip->i_gl, dibh);
 
-	if (gfs2_is_stuffed(ip)) {
+	if (gfs2_is_stuffed(ip))
 		gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode) + newsize);
-	} else {
-		unsigned int blocksize = i_blocksize(inode);
-		unsigned int offs = newsize & (blocksize - 1);
-		if (offs) {
-			error = gfs2_block_zero_range(inode, newsize,
-						      blocksize - offs);
-			if (error)
-				goto out;
-		}
+	else
 		ip->i_diskflags |= GFS2_DIF_TRUNC_IN_PROG;
-	}
 
 	i_size_write(inode, newsize);
 	ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
@@ -2448,25 +2455,7 @@ int __gfs2_punch_hole(struct file *file,
 	loff_t start, end;
 	int error;
 
-	start = round_down(offset, blocksize);
-	end = round_up(offset + length, blocksize) - 1;
-	error = filemap_write_and_wait_range(inode->i_mapping, start, end);
-	if (error)
-		return error;
-
-	if (gfs2_is_jdata(ip))
-		error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_JDATA,
-					 GFS2_JTRUNC_REVOKES);
-	else
-		error = gfs2_trans_begin(sdp, RES_DINODE, 0);
-	if (error)
-		return error;
-
-	if (gfs2_is_stuffed(ip)) {
-		error = stuffed_zero_range(inode, offset, length);
-		if (error)
-			goto out;
-	} else {
+	if (!gfs2_is_stuffed(ip)) {
 		unsigned int start_off, end_len;
 
 		start_off = offset & (blocksize - 1);
@@ -2489,6 +2478,26 @@ int __gfs2_punch_hole(struct file *file,
 		}
 	}
 
+	start = round_down(offset, blocksize);
+	end = round_up(offset + length, blocksize) - 1;
+	error = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (error)
+		return error;
+
+	if (gfs2_is_jdata(ip))
+		error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_JDATA,
+					 GFS2_JTRUNC_REVOKES);
+	else
+		error = gfs2_trans_begin(sdp, RES_DINODE, 0);
+	if (error)
+		return error;
+
+	if (gfs2_is_stuffed(ip)) {
+		error = stuffed_zero_range(inode, offset, length);
+		if (error)
+			goto out;
+	}
+
 	if (gfs2_is_jdata(ip)) {
 		BUG_ON(!current->journal_info);
 		gfs2_journaled_truncate_range(inode, offset, length);
author maintainer
#94 lore raw

PATCH 94/204 remoteproc: qcom: q6v5: Update running state before requesting stop

In reply to Greg Kroah-Hartman

From: Sibi Sankar <[email protected]>

commit 5b7be880074c73540948f8fc597e0407b98fabfa upstream.

Sometimes the stop triggers a watchdog rather than a stop-ack. Update the running state to false on requesting stop to skip the watchdog instead.

Error Logs:

$ echo stop > /sys/class/remoteproc/remoteproc0/state
ipa 1e40000.ipa: received modem stopping event
remoteproc-modem: watchdog received: sys_m_smsm_mpss.c:291:APPS force stop
qcom-q6v5-mss 4080000.remoteproc-modem: port failed halt
ipa 1e40000.ipa: received modem offline event
remoteproc0: stopped remote processor 4080000.remoteproc-modem
---
 drivers/remoteproc/qcom_q6v5.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/remoteproc/qcom_q6v5.c +2 -0
@@ -153,6 +153,8 @@ int qcom_q6v5_request_stop(struct qcom_q
 {
 	int ret;
 
+	q6v5->running = false;
+
 	qcom_smem_state_update_bits(q6v5->state,
 				    BIT(q6v5->stop_bit), BIT(q6v5->stop_bit));
 
author maintainer
#95 lore raw

PATCH 95/204 remoteproc: qcom_q6v5_mss: Validate MBA firmware size before load

In reply to Greg Kroah-Hartman

From: Sibi Sankar <[email protected]>

commit e013f455d95add874f310dc47c608e8c70692ae5 upstream.

The following mem abort is observed when the mba firmware size exceeds the allocated mba region. MBA firmware size is restricted to a maximum size of 1M and remaining memory region is used by modem debug policy firmware when available. Hence verify whether the MBA firmware size lies within the allocated memory region and is not greater than 1M before loading.

Err Logs:

Unable to handle kernel paging request at virtual address
Mem abort info:
...
Call trace:
  __memcpy+0x110/0x180
  rproc_start+0x40/0x218
  rproc_boot+0x5b4/0x608
  state_store+0x54/0xf8
  dev_attr_store+0x44/0x60
  sysfs_kf_write+0x58/0x80
  kernfs_fop_write+0x140/0x230
  vfs_write+0xc4/0x208
  ksys_write+0x74/0xf8
  __arm64_sys_write+0x24/0x30

...

Fixes 051fb70fd4ea4 ("remoteproc: qcom: Driver for the self-authenticating Hexagon v5")
---
 drivers/remoteproc/qcom_q6v5_mss.c |    6 ++++++
 1 file changed, 6 insertions(+)
drivers/remoteproc/qcom_q6v5_mss.c +6 -0
@@ -408,6 +408,12 @@ static int q6v5_load(struct rproc *rproc
 {
 	struct q6v5 *qproc = rproc->priv;
 
+	/* MBA is restricted to a maximum size of 1M */
+	if (fw->size > qproc->mba_size || fw->size > SZ_1M) {
+		dev_err(qproc->dev, "MBA firmware load failed\n");
+		return -EINVAL;
+	}
+
 	memcpy(qproc->mba_region, fw->data, fw->size);
 
 	return 0;
author maintainer
#96 lore raw

PATCH 96/204 remoteproc: qcom_q6v5_mss: Validate modem blob firmware size before load

In reply to Greg Kroah-Hartman

From: Sibi Sankar <[email protected]>

commit 135b9e8d1cd8ba5ac9ad9bcf24b464b7b052e5b8 upstream.

The following mem abort is observed when one of the modem blob firmware size exceeds the allocated mpss region. Fix this by restricting the copy size to segment size using request_firmware_into_buf before load.

Err Logs:

Unable to handle kernel paging request at virtual address
Mem abort info:
...
Call trace:
  __memcpy+0x110/0x180
  rproc_start+0xd0/0x190
  rproc_boot+0x404/0x550
  state_store+0x54/0xf8
  dev_attr_store+0x44/0x60
  sysfs_kf_write+0x58/0x80
  kernfs_fop_write+0x140/0x230
  vfs_write+0xc4/0x208
  ksys_write+0x74/0xf8

...

Fixes 051fb70fd4ea4 ("remoteproc: qcom: Driver for the self-authenticating Hexagon v5")
---
 drivers/remoteproc/qcom_q6v5_mss.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/remoteproc/qcom_q6v5_mss.c +2 -3
@@ -1145,15 +1145,14 @@ static int q6v5_mpss_load(struct q6v5 *q
 		} else if (phdr->p_filesz) {
 			/* Replace "xxx.xxx" with "xxx.bxx" */
 			sprintf(fw_name + fw_name_len - 3, "b%02d", i);
-			ret = request_firmware(&seg_fw, fw_name, qproc->dev);
+			ret = request_firmware_into_buf(&seg_fw, fw_name, qproc->dev,
+							ptr, phdr->p_filesz);
 			if (ret) {
 				dev_err(qproc->dev, "failed to load %s\n", fw_name);
 				iounmap(ptr);
 				goto release_firmware;
 			}
 
-			memcpy(ptr, seg_fw->data, seg_fw->size);
-
 			release_firmware(seg_fw);
 		}
 
author maintainer
#97 lore raw

PATCH 97/204 drm/imx: imx-ldb: Disable both channels for split mode in enc->disable()

In reply to Greg Kroah-Hartman

From: Liu Ying <[email protected]>

commit 3b2a999582c467d1883716b37ffcc00178a13713 upstream.

Both of the two LVDS channels should be disabled for split mode in the encoder's ->disable() callback, because they are enabled in the encoder's ->enable() callback.

Fixes 6556f7f82b9c ("drm: imx: Move imx-drm driver out of staging")
Cc Philipp Zabel <[email protected]>
Cc Sascha Hauer <[email protected]>
Cc Pengutronix Kernel Team <[email protected]>
Cc NXP Linux Team <[email protected]>
---
 drivers/gpu/drm/imx/imx-ldb.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/gpu/drm/imx/imx-ldb.c +4 -3
@@ -303,18 +303,19 @@ static void imx_ldb_encoder_disable(stru
 {
 	struct imx_ldb_channel *imx_ldb_ch = enc_to_imx_ldb_ch(encoder);
 	struct imx_ldb *ldb = imx_ldb_ch->ldb;
+	int dual = ldb->ldb_ctrl & LDB_SPLIT_MODE_EN;
 	int mux, ret;
 
 	drm_panel_disable(imx_ldb_ch->panel);
 
-	if (imx_ldb_ch == &ldb->channel[0])
+	if (imx_ldb_ch == &ldb->channel[0] || dual)
 		ldb->ldb_ctrl &= ~LDB_CH0_MODE_EN_MASK;
-	else if (imx_ldb_ch == &ldb->channel[1])
+	if (imx_ldb_ch == &ldb->channel[1] || dual)
 		ldb->ldb_ctrl &= ~LDB_CH1_MODE_EN_MASK;
 
 	regmap_write(ldb->regmap, IOMUXC_GPR2, ldb->ldb_ctrl);
 
-	if (ldb->ldb_ctrl & LDB_SPLIT_MODE_EN) {
+	if (dual) {
 		clk_disable_unprepare(ldb->clk[0]);
 		clk_disable_unprepare(ldb->clk[1]);
 	}
author maintainer
#98 lore raw

PATCH 98/204 drm/ingenic: Fix incorrect assumption about plane->index

In reply to Greg Kroah-Hartman

From: Paul Cercueil <[email protected]>

commit ca43f274e03f91c533643299ae4984965ce03205 upstream.

plane->index is NOT the index of the color plane in a YUV frame. Actually, a YUV frame is represented by a single drm_plane, even though it contains three Y, U, V planes.

v2-v3: No change

---
 drivers/gpu/drm/ingenic/ingenic-drm.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/ingenic/ingenic-drm.c +1 -1
@@ -384,7 +384,7 @@ static void ingenic_drm_plane_atomic_upd
 		addr = drm_fb_cma_get_gem_addr(state->fb, state, 0);
 		width = state->src_w >> 16;
 		height = state->src_h >> 16;
-		cpp = state->fb->format->cpp[plane->index];
+		cpp = state->fb->format->cpp[0];
 
 		priv->dma_hwdesc->addr = addr;
 		priv->dma_hwdesc->cmd = width * height * cpp / 4;
author maintainer
#99 lore raw

PATCH 99/204 crypto: algif_aead - Only wake up when ctx->more is zero

In reply to Greg Kroah-Hartman

From: Herbert Xu <[email protected]>

[ Upstream commit f3c802a1f30013f8f723b62d7fa49eb9e991da23 ]

AEAD does not support partial requests so we must not wake up while ctx->more is set. In order to distinguish between the case of no data sent yet and a zero-length request, a new init flag has been added to ctx.

SKCIPHER has also been modified to ensure that at least a block of data is available if there is more data to come.

Fixes 2d97591ef43d ("crypto: af_alg - consolidation of...")
---
 crypto/af_alg.c         | 11 ++++++++---
 crypto/algif_aead.c     |  4 ++--
 crypto/algif_skcipher.c |  4 ++--
 include/crypto/if_alg.h |  4 +++-
 4 files changed, 15 insertions(+), 8 deletions(-)
crypto/af_alg.c +8 -3
index 28fc323e3fe30..9fcb91ea10c41 100644
@@ -635,6 +635,7 @@ void af_alg_pull_tsgl(struct sock *sk, size_t used, struct scatterlist *dst,
 
 	if (!ctx->used)
 		ctx->merge = 0;
+	ctx->init = ctx->more;
 }
 EXPORT_SYMBOL_GPL(af_alg_pull_tsgl);
 
@@ -734,9 +735,10 @@ EXPORT_SYMBOL_GPL(af_alg_wmem_wakeup);
  *
  * @sk socket of connection to user space
  * @flags If MSG_DONTWAIT is set, then only report if function would sleep
+ * @min Set to minimum request size if partial requests are allowed.
  * @return 0 when writable memory is available, < 0 upon error
  */
-int af_alg_wait_for_data(struct sock *sk, unsigned flags)
+int af_alg_wait_for_data(struct sock *sk, unsigned flags, unsigned min)
 {
 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
 	struct alg_sock *ask = alg_sk(sk);
@@ -754,7 +756,9 @@ int af_alg_wait_for_data(struct sock *sk, unsigned flags)
 		if (signal_pending(current))
 			break;
 		timeout = MAX_SCHEDULE_TIMEOUT;
-		if (sk_wait_event(sk, &timeout, (ctx->used || !ctx->more),
+		if (sk_wait_event(sk, &timeout,
+				  ctx->init && (!ctx->more ||
+						(min && ctx->used >= min)),
 				  &wait)) {
 			err = 0;
 			break;
@@ -843,7 +847,7 @@ int af_alg_sendmsg(struct socket *sock, struct msghdr *msg, size_t size,
 	}
 
 	lock_sock(sk);
-	if (!ctx->more && ctx->used) {
+	if (ctx->init && (init || !ctx->more)) {
 		err = -EINVAL;
 		goto unlock;
 	}
@@ -854,6 +858,7 @@ int af_alg_sendmsg(struct socket *sock, struct msghdr *msg, size_t size,
 			memcpy(ctx->iv, con.iv->iv, ivsize);
 
 		ctx->aead_assoclen = con.aead_assoclen;
+		ctx->init = true;
 	}
 
 	while (size) {
crypto/algif_aead.c +2 -2
index 0ae000a61c7f5..d48d2156e6210 100644
@@ -106,8 +106,8 @@ static int _aead_recvmsg(struct socket *sock, struct msghdr *msg,
 	size_t usedpages = 0;		/* [in]  RX bufs to be used from user */
 	size_t processed = 0;		/* [in]  TX bufs to be consumed */
 
-	if (!ctx->used) {
-		err = af_alg_wait_for_data(sk, flags);
+	if (!ctx->init || ctx->more) {
+		err = af_alg_wait_for_data(sk, flags, 0);
 		if (err)
 			return err;
 	}
crypto/algif_skcipher.c +2 -2
index ec5567c87a6df..a51ba22fef58f 100644
@@ -61,8 +61,8 @@ static int _skcipher_recvmsg(struct socket *sock, struct msghdr *msg,
 	int err = 0;
 	size_t len = 0;
 
-	if (!ctx->used) {
-		err = af_alg_wait_for_data(sk, flags);
+	if (!ctx->init || (ctx->more && ctx->used < bs)) {
+		err = af_alg_wait_for_data(sk, flags, bs);
 		if (err)
 			return err;
 	}
include/crypto/if_alg.h +3 -1
index 088c1ded27148..ee6412314f8f3 100644
@@ -135,6 +135,7 @@ struct af_alg_async_req {
  *			SG?
  * @enc:		Cryptographic operation to be performed when
  *			recvmsg is invoked.
+ * @init:		True if metadata has been sent.
  * @len:		Length of memory allocated for this data structure.
  */
 struct af_alg_ctx {
@@ -151,6 +152,7 @@ struct af_alg_ctx {
 	bool more;
 	bool merge;
 	bool enc;
+	bool init;
 
 	unsigned int len;
 };
@@ -226,7 +228,7 @@ unsigned int af_alg_count_tsgl(struct sock *sk, size_t bytes, size_t offset);
 void af_alg_pull_tsgl(struct sock *sk, size_t used, struct scatterlist *dst,
 		      size_t dst_offset);
 void af_alg_wmem_wakeup(struct sock *sk);
-int af_alg_wait_for_data(struct sock *sk, unsigned flags);
+int af_alg_wait_for_data(struct sock *sk, unsigned flags, unsigned min);
 int af_alg_sendmsg(struct socket *sock, struct msghdr *msg, size_t size,
 		   unsigned int ivsize);
 ssize_t af_alg_sendpage(struct socket *sock, struct page *page,
Signature · 1 line
2.25.1
author maintainer
#100 lore raw

PATCH 100/204 mfd: arizona: Ensure 32k clock is put on driver unbind and error

In reply to Greg Kroah-Hartman

From: Charles Keepax <[email protected]>

[ Upstream commit ddff6c45b21d0437ce0c85f8ac35d7b5480513d7 ]

Whilst it doesn't matter if the internal 32k clock register settings are cleaned up on exit, as the part will be turned off losing any settings, hence the driver hasn't historially bothered. The external clock should however be cleaned up, as it could cause clocks to be left on, and will at best generate a warning on unbind.

Add clean up on both the probe error path and unbind for the 32k clock.

Fixes cdd8da8cc66b ("mfd: arizona: Add gating of external MCLKn clocks")
---
 drivers/mfd/arizona-core.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)
drivers/mfd/arizona-core.c +18 -0
index f73cf76d1373d..a5e443110fc3d 100644
@@ -1426,6 +1426,15 @@ int arizona_dev_init(struct arizona *arizona)
 	arizona_irq_exit(arizona);
 err_pm:
 	pm_runtime_disable(arizona->dev);
+
+	switch (arizona->pdata.clk32k_src) {
+	case ARIZONA_32KZ_MCLK1:
+	case ARIZONA_32KZ_MCLK2:
+		arizona_clk32k_disable(arizona);
+		break;
+	default:
+		break;
+	}
 err_reset:
 	arizona_enable_reset(arizona);
 	regulator_disable(arizona->dcvdd);
@@ -1448,6 +1457,15 @@ int arizona_dev_exit(struct arizona *arizona)
 	regulator_disable(arizona->dcvdd);
 	regulator_put(arizona->dcvdd);
 
+	switch (arizona->pdata.clk32k_src) {
+	case ARIZONA_32KZ_MCLK1:
+	case ARIZONA_32KZ_MCLK2:
+		arizona_clk32k_disable(arizona);
+		break;
+	default:
+		break;
+	}
+
 	mfd_remove_devices(arizona->dev);
 	arizona_free_irq(arizona, ARIZONA_IRQ_UNDERCLOCKED, arizona);
 	arizona_free_irq(arizona, ARIZONA_IRQ_OVERCLOCKED, arizona);
Signature · 1 line
2.25.1
author maintainer
#101 lore raw

PATCH 101/204 octeontx2-af: change (struct qmem)->entry_sz from u8 to u16

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit 393415203f5c916b5907e0a7c89f4c2c5a9c5505 ]

We need to increase TSO_HEADER_SIZE from 128 to 256.

Since otx2_sq_init() calls qmem_alloc() with TSO_HEADER_SIZE, we need to change (struct qmem)->entry_sz to avoid truncation to 0.

Fixes 7a37245ef23f ("octeontx2-af: NPA block admin queue init")
Cc Sunil Goutham <[email protected]>
---
 drivers/net/ethernet/marvell/octeontx2/af/common.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/marvell/octeontx2/af/common.h +1 -1
index cd33c2e6ca5fc..f48eb66ed021b 100644
@@ -43,7 +43,7 @@ struct qmem {
 	void            *base;
 	dma_addr_t	iova;
 	int		alloc_sz;
-	u8		entry_sz;
+	u16		entry_sz;
 	u8		align;
 	u32		qsize;
 };
Signature · 1 line
2.25.1
author maintainer
#102 lore raw

PATCH 102/204 mtd: rawnand: fsl_upm: Remove unused mtd var

In reply to Greg Kroah-Hartman

From: Boris Brezillon <[email protected]>

[ Upstream commit ccc49eff77bee2885447a032948959a134029fe3 ]

The mtd var in fun_wait_rnb() is now unused, let's get rid of it and fix the warning resulting from this unused var.

Fixes 50a487e7719c ("mtd: rawnand: Pass a nand_chip object to chip->dev_ready()")
---
 drivers/mtd/nand/raw/fsl_upm.c | 1 -
 1 file changed, 1 deletion(-)
drivers/mtd/nand/raw/fsl_upm.c +0 -1
index f31fae3a4c689..6b8ec72686e29 100644
@@ -62,7 +62,6 @@ static int fun_chip_ready(struct nand_chip *chip)
 static void fun_wait_rnb(struct fsl_upm_nand *fun)
 {
 	if (fun->rnb_gpio[fun->mchip_number] >= 0) {
-		struct mtd_info *mtd = nand_to_mtd(&fun->chip);
 		int cnt = 1000000;
 
 		while (--cnt && !fun_chip_ready(&fun->chip))
Signature · 1 line
2.25.1
author maintainer
#103 lore raw

PATCH 103/204 mtd: rawnand: brcmnand: ECC error handling on EDU transfers

In reply to Greg Kroah-Hartman

From: Kamal Dasu <[email protected]>

[ Upstream commit 4551e78ad98add1f16b70cf286d5aad3ce7bcd4c ]

Implement ECC correctable and uncorrectable error handling for EDU reads. If ECC correctable bitflips are encountered on EDU transfer, read page again using PIO. This is needed due to a NAND controller limitation where corrected data is not transferred to the DMA buffer on ECC error. This applies to ECC correctable errors that are reported by the controller hardware based on set number of bitflips threshold in the controller threshold register, bitflips below the threshold are corrected silently and are not reported by the controller hardware.

Fixes a5d53ad26a8b ("mtd: rawnand: brcmnand: Add support for flash-edu for dma transfers")
---
 drivers/mtd/nand/raw/brcmnand/brcmnand.c | 26 ++++++++++++++++++++++++
 1 file changed, 26 insertions(+)
drivers/mtd/nand/raw/brcmnand/brcmnand.c +26 -0
index cdae2311a3b69..0b1ea965cba08 100644
@@ -1859,6 +1859,22 @@ static int brcmnand_edu_trans(struct brcmnand_host *host, u64 addr, u32 *buf,
 	edu_writel(ctrl, EDU_STOP, 0); /* force stop */
 	edu_readl(ctrl, EDU_STOP);
 
+	if (!ret && edu_cmd == EDU_CMD_READ) {
+		u64 err_addr = 0;
+
+		/*
+		 * check for ECC errors here, subpage ECC errors are
+		 * retained in ECC error address register
+		 */
+		err_addr = brcmnand_get_uncorrecc_addr(ctrl);
+		if (!err_addr) {
+			err_addr = brcmnand_get_correcc_addr(ctrl);
+			if (err_addr)
+				ret = -EUCLEAN;
+		} else
+			ret = -EBADMSG;
+	}
+
 	return ret;
 }
 
@@ -2065,6 +2081,7 @@ static int brcmnand_read(struct mtd_info *mtd, struct nand_chip *chip,
 	u64 err_addr = 0;
 	int err;
 	bool retry = true;
+	bool edu_err = false;
 
 	dev_dbg(ctrl->dev, "read %llx -> %p\n", (unsigned long long)addr, buf);
 
@@ -2082,6 +2099,10 @@ static int brcmnand_read(struct mtd_info *mtd, struct nand_chip *chip,
 			else
 				return -EIO;
 		}
+
+		if (has_edu(ctrl) && err_addr)
+			edu_err = true;
+
 	} else {
 		if (oob)
 			memset(oob, 0x99, mtd->oobsize);
@@ -2129,6 +2150,11 @@ static int brcmnand_read(struct mtd_info *mtd, struct nand_chip *chip,
 	if (mtd_is_bitflip(err)) {
 		unsigned int corrected = brcmnand_count_corrected(ctrl);
 
+		/* in case of EDU correctable error we read again using PIO */
+		if (edu_err)
+			err = brcmnand_read_by_pio(mtd, chip, addr, trans, buf,
+						   oob, &err_addr);
+
 		dev_dbg(ctrl->dev, "corrected error at 0x%llx\n",
 			(unsigned long long)err_addr);
 		mtd->ecc_stats.corrected += corrected;
Signature · 1 line
2.25.1
author maintainer
#104 lore raw

PATCH 104/204 platform/chrome: cros_ec_ishtp: Fix a double-unlock issue

In reply to Greg Kroah-Hartman

From: Qiushi Wu <[email protected]>

[ Upstream commit aaa3cbbac326c95308e315f1ab964a3369c4d07d ]

In function cros_ec_ishtp_probe(), "up_write" is already called before function "cros_ec_dev_init". But "up_write" will be called again after the calling of the function "cros_ec_dev_init" failed. Thus add a call of the function “down_write” in this if branch for the completion of the exception handling.

Fixes 26a14267aff2 ("platform/chrome: Add ChromeOS EC ISHTP driver")
---
 drivers/platform/chrome/cros_ec_ishtp.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/platform/chrome/cros_ec_ishtp.c +3 -1
index 93a71e93a2f15..41d60af618c9d 100644
@@ -660,8 +660,10 @@ static int cros_ec_ishtp_probe(struct ishtp_cl_device *cl_device)
 
 	/* Register croc_ec_dev mfd */
 	rv = cros_ec_dev_init(client_data);
-	if (rv)
+	if (rv) {
+		down_write(&init_lock);
 		goto end_cros_ec_dev_init_error;
+	}
 
 	return 0;
 
Signature · 1 line
2.25.1
author maintainer
#105 lore raw

PATCH 105/204 RDMA/ipoib: Return void from ipoib_ib_dev_stop()

In reply to Greg Kroah-Hartman

From: Kamal Heib <[email protected]>

[ Upstream commit 95a5631f6c9f3045f26245e6045244652204dfdb ]

The return value from ipoib_ib_dev_stop() is always 0 - change it to be void.

---
 drivers/infiniband/ulp/ipoib/ipoib.h    | 2 +-
 drivers/infiniband/ulp/ipoib/ipoib_ib.c | 4 +---
 2 files changed, 2 insertions(+), 4 deletions(-)
drivers/infiniband/ulp/ipoib/ipoib.h +1 -1
index 9a3379c49541f..9ce6a36fe48ed 100644
@@ -515,7 +515,7 @@ void ipoib_ib_dev_cleanup(struct net_device *dev);
 
 int ipoib_ib_dev_open_default(struct net_device *dev);
 int ipoib_ib_dev_open(struct net_device *dev);
-int ipoib_ib_dev_stop(struct net_device *dev);
+void ipoib_ib_dev_stop(struct net_device *dev);
 void ipoib_ib_dev_up(struct net_device *dev);
 void ipoib_ib_dev_down(struct net_device *dev);
 int ipoib_ib_dev_stop_default(struct net_device *dev);
drivers/infiniband/ulp/ipoib/ipoib_ib.c +1 -3
index da3c5315bbb51..6ee64c25aaff4 100644
@@ -846,7 +846,7 @@ int ipoib_ib_dev_stop_default(struct net_device *dev)
 	return 0;
 }
 
-int ipoib_ib_dev_stop(struct net_device *dev)
+void ipoib_ib_dev_stop(struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = ipoib_priv(dev);
 
@@ -854,8 +854,6 @@ int ipoib_ib_dev_stop(struct net_device *dev)
 
 	clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
 	ipoib_flush_ah(dev);
-
-	return 0;
 }
 
 int ipoib_ib_dev_open_default(struct net_device *dev)
Signature · 1 line
2.25.1
author maintainer
#106 lore raw

PATCH 106/204 RDMA/ipoib: Fix ABBA deadlock with ipoib_reap_ah()

In reply to Greg Kroah-Hartman

From: Jason Gunthorpe <[email protected]>

[ Upstream commit 65936bf25f90fe440bb2d11624c7d10fab266639 ]

ipoib_mcast_carrier_on_task() insanely open codes a rtnl_lock() such that the only time flush_workqueue() can be called is if it also clears IPOIB_FLAG_OPER_UP.

Thus the flush inside ipoib_flush_ah() will deadlock if it gets unlucky enough, and lockdep doesn't help us to find it early:

CPU0 CPU1 CPU2
__ipoib_ib_dev_flush()

      down_read(vlan_rwsem)

                         ipoib_vlan_add()
                           rtnl_trylock()
                           down_write(vlan_rwsem)

				      ipoib_mcast_carrier_on_task()
					 while (!rtnl_trylock())
					      msleep(20);

      ipoib_flush_ah()
	flush_workqueue(priv->wq)

Clean up the ah_reaper related functions and lifecycle to make sense:

- Start/Stop of the reaper should only be done in open/stop NDOs, not in any other places

- cancel and flush of the reaper should only happen in the stop NDO. cancel is only functional when combined with IPOIB_STOP_REAPER.

- Non-stop places were flushing the AH's just need to flush out dead AH's synchronously and ignore the background task completely. It is fully locked and harmless to leave running.

Which ultimately fixes the ABBA deadlock by removing the unnecessary flush_workqueue() from the problematic place under the vlan_rwsem.

Fixes efc82eeeae4e ("IB/ipoib: No longer use flush as a parameter")
---
 drivers/infiniband/ulp/ipoib/ipoib_ib.c   | 65 ++++++++++-------------
 drivers/infiniband/ulp/ipoib/ipoib_main.c |  2 +
 2 files changed, 31 insertions(+), 36 deletions(-)
drivers/infiniband/ulp/ipoib/ipoib_ib.c +29 -36
index 6ee64c25aaff4..494f413dc3c6c 100644
@@ -670,13 +670,12 @@ int ipoib_send(struct net_device *dev, struct sk_buff *skb,
 	return rc;
 }
 
-static void __ipoib_reap_ah(struct net_device *dev)
+static void ipoib_reap_dead_ahs(struct ipoib_dev_priv *priv)
 {
-	struct ipoib_dev_priv *priv = ipoib_priv(dev);
 	struct ipoib_ah *ah, *tah;
 	unsigned long flags;
 
-	netif_tx_lock_bh(dev);
+	netif_tx_lock_bh(priv->dev);
 	spin_lock_irqsave(&priv->lock, flags);
 
 	list_for_each_entry_safe(ah, tah, &priv->dead_ahs, list)
@@ -687,37 +686,37 @@ static void __ipoib_reap_ah(struct net_device *dev)
 		}
 
 	spin_unlock_irqrestore(&priv->lock, flags);
-	netif_tx_unlock_bh(dev);
+	netif_tx_unlock_bh(priv->dev);
 }
 
 void ipoib_reap_ah(struct work_struct *work)
 {
 	struct ipoib_dev_priv *priv =
 		container_of(work, struct ipoib_dev_priv, ah_reap_task.work);
-	struct net_device *dev = priv->dev;
 
-	__ipoib_reap_ah(dev);
+	ipoib_reap_dead_ahs(priv);
 
 	if (!test_bit(IPOIB_STOP_REAPER, &priv->flags))
 		queue_delayed_work(priv->wq, &priv->ah_reap_task,
 				   round_jiffies_relative(HZ));
 }
 
-static void ipoib_flush_ah(struct net_device *dev)
+static void ipoib_start_ah_reaper(struct ipoib_dev_priv *priv)
 {
-	struct ipoib_dev_priv *priv = ipoib_priv(dev);
-
-	cancel_delayed_work(&priv->ah_reap_task);
-	flush_workqueue(priv->wq);
-	ipoib_reap_ah(&priv->ah_reap_task.work);
+	clear_bit(IPOIB_STOP_REAPER, &priv->flags);
+	queue_delayed_work(priv->wq, &priv->ah_reap_task,
+			   round_jiffies_relative(HZ));
 }
 
-static void ipoib_stop_ah(struct net_device *dev)
+static void ipoib_stop_ah_reaper(struct ipoib_dev_priv *priv)
 {
-	struct ipoib_dev_priv *priv = ipoib_priv(dev);
-
 	set_bit(IPOIB_STOP_REAPER, &priv->flags);
-	ipoib_flush_ah(dev);
+	cancel_delayed_work(&priv->ah_reap_task);
+	/*
+	 * After ipoib_stop_ah_reaper() we always go through
+	 * ipoib_reap_dead_ahs() which ensures the work is really stopped and
+	 * does a final flush out of the dead_ah's list
+	 */
 }
 
 static int recvs_pending(struct net_device *dev)
@@ -846,16 +845,6 @@ int ipoib_ib_dev_stop_default(struct net_device *dev)
 	return 0;
 }
 
-void ipoib_ib_dev_stop(struct net_device *dev)
-{
-	struct ipoib_dev_priv *priv = ipoib_priv(dev);
-
-	priv->rn_ops->ndo_stop(dev);
-
-	clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
-	ipoib_flush_ah(dev);
-}
-
 int ipoib_ib_dev_open_default(struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = ipoib_priv(dev);
@@ -899,10 +888,7 @@ int ipoib_ib_dev_open(struct net_device *dev)
 		return -1;
 	}
 
-	clear_bit(IPOIB_STOP_REAPER, &priv->flags);
-	queue_delayed_work(priv->wq, &priv->ah_reap_task,
-			   round_jiffies_relative(HZ));
-
+	ipoib_start_ah_reaper(priv);
 	if (priv->rn_ops->ndo_open(dev)) {
 		pr_warn("%s: Failed to open dev\n", dev->name);
 		goto dev_stop;
@@ -913,13 +899,20 @@ int ipoib_ib_dev_open(struct net_device *dev)
 	return 0;
 
 dev_stop:
-	set_bit(IPOIB_STOP_REAPER, &priv->flags);
-	cancel_delayed_work(&priv->ah_reap_task);
-	set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
-	ipoib_ib_dev_stop(dev);
+	ipoib_stop_ah_reaper(priv);
 	return -1;
 }
 
+void ipoib_ib_dev_stop(struct net_device *dev)
+{
+	struct ipoib_dev_priv *priv = ipoib_priv(dev);
+
+	priv->rn_ops->ndo_stop(dev);
+
+	clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
+	ipoib_stop_ah_reaper(priv);
+}
+
 void ipoib_pkey_dev_check_presence(struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = ipoib_priv(dev);
@@ -1230,7 +1223,7 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv,
 		ipoib_mcast_dev_flush(dev);
 		if (oper_up)
 			set_bit(IPOIB_FLAG_OPER_UP, &priv->flags);
-		ipoib_flush_ah(dev);
+		ipoib_reap_dead_ahs(priv);
 	}
 
 	if (level >= IPOIB_FLUSH_NORMAL)
@@ -1305,7 +1298,7 @@ void ipoib_ib_dev_cleanup(struct net_device *dev)
 	 * the neighbor garbage collection is stopped and reaped.
 	 * That should all be done now, so make a final ah flush.
 	 */
-	ipoib_stop_ah(dev);
+	ipoib_reap_dead_ahs(priv);
 
 	clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
 
drivers/infiniband/ulp/ipoib/ipoib_main.c +2 -0
index ceec24d451858..29ad4129d2f48 100644
@@ -1975,6 +1975,8 @@ static void ipoib_ndo_uninit(struct net_device *dev)
 
 	/* no more works over the priv->wq */
 	if (priv->wq) {
+		/* See ipoib_mcast_carrier_on_task() */
+		WARN_ON(test_bit(IPOIB_FLAG_OPER_UP, &priv->flags));
 		flush_workqueue(priv->wq);
 		destroy_workqueue(priv->wq);
 		priv->wq = NULL;
Signature · 1 line
2.25.1
author maintainer
#107 lore raw

PATCH 107/204 rtc: cpcap: fix range

In reply to Greg Kroah-Hartman

From: Sebastian Reichel <[email protected]>

[ Upstream commit 3180cfabf6fbf982ca6d1a6eb56334647cc1416b ]

Unbreak CPCAP driver, which has one more bit in the day counter increasing the max. range from 2014 to 2058. The original commit introducing the range limit was obviously wrong, since the driver has only been written in 2017 (3 years after 14 bits would have run out).

---
 drivers/rtc/rtc-cpcap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/rtc/rtc-cpcap.c +1 -1
index a603f1f211250..800667d73a6fb 100644
@@ -261,7 +261,7 @@ static int cpcap_rtc_probe(struct platform_device *pdev)
 		return PTR_ERR(rtc->rtc_dev);
 
 	rtc->rtc_dev->ops = &cpcap_rtc_ops;
-	rtc->rtc_dev->range_max = (1 << 14) * SECS_PER_DAY - 1;
+	rtc->rtc_dev->range_max = (timeu64_t) (DAY_MASK + 1) * SECS_PER_DAY - 1;
 
 	err = cpcap_get_vendor(dev, rtc->regmap, &rtc->vendor);
 	if (err)
Signature · 1 line
2.25.1
author maintainer
#108 lore raw

PATCH 108/204 media: staging: rkisp1: remove macro RKISP1_DIR_SINK_SRC

In reply to Greg Kroah-Hartman

From: Dafna Hirschfeld <[email protected]>

[ Upstream commit b861d139a36a4593498932bfec957bdcc7d98eb3 ]

The macro RKISP1_DIR_SINK_SRC is a mask of two flags.
The macro hides the fact that it's a mask and the code
is actually more clear if we replace it the with bitwise-or explicitly.

---
 drivers/staging/media/rkisp1/rkisp1-isp.c | 25 +++++++++++------------
 1 file changed, 12 insertions(+), 13 deletions(-)
drivers/staging/media/rkisp1/rkisp1-isp.c +12 -13
index fa53f05e37d81..31c5ae2aa29fb 100644
@@ -25,7 +25,6 @@
 
 #define RKISP1_DIR_SRC BIT(0)
 #define RKISP1_DIR_SINK BIT(1)
-#define RKISP1_DIR_SINK_SRC (RKISP1_DIR_SINK | RKISP1_DIR_SRC)
 
 /*
  * NOTE: MIPI controller and input MUX are also configured in this file.
@@ -69,84 +68,84 @@ static const struct rkisp1_isp_mbus_info rkisp1_isp_formats[] = {
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW10,
 		.bayer_pat	= RKISP1_RAW_RGGB,
 		.bus_width	= 10,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SBGGR10_1X10,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW10,
 		.bayer_pat	= RKISP1_RAW_BGGR,
 		.bus_width	= 10,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SGBRG10_1X10,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW10,
 		.bayer_pat	= RKISP1_RAW_GBRG,
 		.bus_width	= 10,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SGRBG10_1X10,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW10,
 		.bayer_pat	= RKISP1_RAW_GRBG,
 		.bus_width	= 10,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SRGGB12_1X12,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW12,
 		.bayer_pat	= RKISP1_RAW_RGGB,
 		.bus_width	= 12,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SBGGR12_1X12,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW12,
 		.bayer_pat	= RKISP1_RAW_BGGR,
 		.bus_width	= 12,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SGBRG12_1X12,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW12,
 		.bayer_pat	= RKISP1_RAW_GBRG,
 		.bus_width	= 12,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SGRBG12_1X12,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW12,
 		.bayer_pat	= RKISP1_RAW_GRBG,
 		.bus_width	= 12,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SRGGB8_1X8,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW8,
 		.bayer_pat	= RKISP1_RAW_RGGB,
 		.bus_width	= 8,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SBGGR8_1X8,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW8,
 		.bayer_pat	= RKISP1_RAW_BGGR,
 		.bus_width	= 8,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SGBRG8_1X8,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW8,
 		.bayer_pat	= RKISP1_RAW_GBRG,
 		.bus_width	= 8,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_SGRBG8_1X8,
 		.fmt_type	= RKISP1_FMT_BAYER,
 		.mipi_dt	= RKISP1_CIF_CSI2_DT_RAW8,
 		.bayer_pat	= RKISP1_RAW_GRBG,
 		.bus_width	= 8,
-		.direction	= RKISP1_DIR_SINK_SRC,
+		.direction	= RKISP1_DIR_SINK | RKISP1_DIR_SRC,
 	}, {
 		.mbus_code	= MEDIA_BUS_FMT_YUYV8_1X16,
 		.fmt_type	= RKISP1_FMT_YUV,
Signature · 1 line
2.25.1
author maintainer
#109 lore raw

PATCH 109/204 media: rockchip: rga: Introduce color fmt macros and refactor CSC mode logic

In reply to Greg Kroah-Hartman

From: Paul Kocialkowski <[email protected]>

[ Upstream commit ded874ece29d3fe2abd3775810a06056067eb68c ]

This introduces two macros: RGA_COLOR_FMT_IS_YUV and RGA_COLOR_FMT_IS_RGB which allow quick checking of the colorspace familily of a RGA color format.

These macros are then used to refactor the logic for CSC mode selection. The two nested tests for input colorspace are simplified into a single one, with a logical and, making the whole more readable.

---
 drivers/media/platform/rockchip/rga/rga-hw.c | 23 +++++++++-----------
 drivers/media/platform/rockchip/rga/rga-hw.h |  5 +++++
 2 files changed, 15 insertions(+), 13 deletions(-)
drivers/media/platform/rockchip/rga/rga-hw.c +10 -13
index 4be6dcf292fff..5607ee8d19176 100644
@@ -200,22 +200,19 @@ static void rga_cmd_set_trans_info(struct rga_ctx *ctx)
 	dst_info.data.format = ctx->out.fmt->hw_format;
 	dst_info.data.swap = ctx->out.fmt->color_swap;
 
-	if (ctx->in.fmt->hw_format >= RGA_COLOR_FMT_YUV422SP) {
-		if (ctx->out.fmt->hw_format < RGA_COLOR_FMT_YUV422SP) {
-			switch (ctx->in.colorspace) {
-			case V4L2_COLORSPACE_REC709:
-				src_info.data.csc_mode =
-					RGA_SRC_CSC_MODE_BT709_R0;
-				break;
-			default:
-				src_info.data.csc_mode =
-					RGA_SRC_CSC_MODE_BT601_R0;
-				break;
-			}
+	if (RGA_COLOR_FMT_IS_YUV(ctx->in.fmt->hw_format) &&
+	    RGA_COLOR_FMT_IS_RGB(ctx->out.fmt->hw_format)) {
+		switch (ctx->in.colorspace) {
+		case V4L2_COLORSPACE_REC709:
+			src_info.data.csc_mode = RGA_SRC_CSC_MODE_BT709_R0;
+			break;
+		default:
+			src_info.data.csc_mode = RGA_SRC_CSC_MODE_BT601_R0;
+			break;
 		}
 	}
 
-	if (ctx->out.fmt->hw_format >= RGA_COLOR_FMT_YUV422SP) {
+	if (RGA_COLOR_FMT_IS_YUV(ctx->out.fmt->hw_format)) {
 		switch (ctx->out.colorspace) {
 		case V4L2_COLORSPACE_REC709:
 			dst_info.data.csc_mode = RGA_SRC_CSC_MODE_BT709_R0;
drivers/media/platform/rockchip/rga/rga-hw.h +5 -0
index 96cb0314dfa70..e8917e5630a48 100644
@@ -95,6 +95,11 @@
 #define RGA_COLOR_FMT_CP_8BPP 15
 #define RGA_COLOR_FMT_MASK 15
 
+#define RGA_COLOR_FMT_IS_YUV(fmt) \
+	(((fmt) >= RGA_COLOR_FMT_YUV422SP) && ((fmt) < RGA_COLOR_FMT_CP_1BPP))
+#define RGA_COLOR_FMT_IS_RGB(fmt) \
+	((fmt) < RGA_COLOR_FMT_YUV422SP)
+
 #define RGA_COLOR_NONE_SWAP 0
 #define RGA_COLOR_RB_SWAP 1
 #define RGA_COLOR_ALPHA_SWAP 2
Signature · 1 line
2.25.1
author maintainer
#110 lore raw

PATCH 110/204 media: rockchip: rga: Only set output CSC mode for RGB input

In reply to Greg Kroah-Hartman

From: Paul Kocialkowski <[email protected]>

[ Upstream commit 0f879bab72f47e8ba2421a984e7acfa763d3e84e ]

Setting the output CSC mode is required for a YUV output, but must not be set when the input is also YUV. Doing this (as tested with a YUV420P to YUV420P conversion) results in wrong colors.

Adapt the logic to only set the output CSC mode when the output is YUV and the input is RGB. Also add a comment to clarify the rationale.

Fixes f7e7b48e6d79 ("[media] rockchip/rga: v4l2 m2m support")
---
 drivers/media/platform/rockchip/rga/rga-hw.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/media/platform/rockchip/rga/rga-hw.c +7 -1
index 5607ee8d19176..aaa96f256356b 100644
@@ -200,6 +200,11 @@ static void rga_cmd_set_trans_info(struct rga_ctx *ctx)
 	dst_info.data.format = ctx->out.fmt->hw_format;
 	dst_info.data.swap = ctx->out.fmt->color_swap;
 
+	/*
+	 * CSC mode must only be set when the colorspace families differ between
+	 * input and output. It must remain unset (zeroed) if both are the same.
+	 */
+
 	if (RGA_COLOR_FMT_IS_YUV(ctx->in.fmt->hw_format) &&
 	    RGA_COLOR_FMT_IS_RGB(ctx->out.fmt->hw_format)) {
 		switch (ctx->in.colorspace) {
@@ -212,7 +217,8 @@ static void rga_cmd_set_trans_info(struct rga_ctx *ctx)
 		}
 	}
 
-	if (RGA_COLOR_FMT_IS_YUV(ctx->out.fmt->hw_format)) {
+	if (RGA_COLOR_FMT_IS_RGB(ctx->in.fmt->hw_format) &&
+	    RGA_COLOR_FMT_IS_YUV(ctx->out.fmt->hw_format)) {
 		switch (ctx->out.colorspace) {
 		case V4L2_COLORSPACE_REC709:
 			dst_info.data.csc_mode = RGA_SRC_CSC_MODE_BT709_R0;
Signature · 1 line
2.25.1
author maintainer
#111 lore raw

PATCH 111/204 IB/uverbs: Set IOVA on IB MR in uverbs layer

In reply to Greg Kroah-Hartman

From: Yishai Hadas <[email protected]>

[ Upstream commit 04c0a5fcfcf65aade2fb238b6336445f1a99b646 ]

Set IOVA on IB MR in uverbs layer to let all drivers have it, this includes both reg/rereg MR flows.
As part of this change cleaned-up this setting from the drivers that already did it by themselves in their user flows.

Fixes e6f0330106f4 ("mlx4_ib: set user mr attributes in struct ib_mr")
---
 drivers/infiniband/core/uverbs_cmd.c | 4 ++++
 drivers/infiniband/hw/cxgb4/mem.c    | 1 -
 drivers/infiniband/hw/mlx4/mr.c      | 1 -
 3 files changed, 4 insertions(+), 2 deletions(-)
drivers/infiniband/core/uverbs_cmd.c +4 -0
index d6e9cc94dd900..b2eb87d18e602 100644
@@ -772,6 +772,7 @@ static int ib_uverbs_reg_mr(struct uverbs_attr_bundle *attrs)
 	mr->uobject = uobj;
 	atomic_inc(&pd->usecnt);
 	mr->res.type = RDMA_RESTRACK_MR;
+	mr->iova = cmd.hca_va;
 	rdma_restrack_uadd(&mr->res);
 
 	uobj->object = mr;
@@ -863,6 +864,9 @@ static int ib_uverbs_rereg_mr(struct uverbs_attr_bundle *attrs)
 		atomic_dec(&old_pd->usecnt);
 	}
 
+	if (cmd.flags & IB_MR_REREG_TRANS)
+		mr->iova = cmd.hca_va;
+
 	memset(&resp, 0, sizeof(resp));
 	resp.lkey      = mr->lkey;
 	resp.rkey      = mr->rkey;
drivers/infiniband/hw/cxgb4/mem.c +0 -1
index 962dc97a8ff2b..1e4f4e5255980 100644
@@ -399,7 +399,6 @@ static int finish_mem_reg(struct c4iw_mr *mhp, u32 stag)
 	mmid = stag >> 8;
 	mhp->ibmr.rkey = mhp->ibmr.lkey = stag;
 	mhp->ibmr.length = mhp->attr.len;
-	mhp->ibmr.iova = mhp->attr.va_fbo;
 	mhp->ibmr.page_size = 1U << (mhp->attr.page_size + 12);
 	pr_debug("mmid 0x%x mhp %p\n", mmid, mhp);
 	return xa_insert_irq(&mhp->rhp->mrs, mmid, mhp, GFP_KERNEL);
drivers/infiniband/hw/mlx4/mr.c +0 -1
index b0121c90c561f..184a281f89ec8 100644
@@ -439,7 +439,6 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 
 	mr->ibmr.rkey = mr->ibmr.lkey = mr->mmr.key;
 	mr->ibmr.length = length;
-	mr->ibmr.iova = virt_addr;
 	mr->ibmr.page_size = 1U << shift;
 
 	return &mr->ibmr;
Signature · 1 line
2.25.1
author maintainer
#112 lore raw

PATCH 112/204 sched/uclamp: Protect uclamp fast path code with static key

In reply to Greg Kroah-Hartman

From: Qais Yousef <[email protected]>

[ Upstream commit 46609ce227039fd192e0ecc7d940bed587fd2c78 ]

There is a report that when uclamp is enabled, a netperf UDP test regresses compared to a kernel compiled without uclamp.

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

While investigating the root cause, there were no sign that the uclamp code is doing anything particularly expensive but could suffer from bad cache behavior under certain circumstances that are yet to be understood.

https://lore.kernel.org/lkml/20200616110824.dgkkbyapn3io6wik@e107158-lin/

To reduce the pressure on the fast path anyway, add a static key that is by default will skip executing uclamp logic in the enqueue/dequeue_task() fast path until it's needed.

As soon as the user start using util clamp by:

1. Changing uclamp value of a task with sched_setattr()
2. Modifying the default sysctl_sched_util_clamp_{min, max}
3. Modifying the default cpu.uclamp.{min, max} value in cgroup

We flip the static key now that the user has opted to use util clamp. Effectively re-introducing uclamp logic in the enqueue/dequeue_task() fast path. It stays on from that point forward until the next reboot.

This should help minimize the effect of util clamp on workloads that don't need it but still allow distros to ship their kernels with uclamp compiled in by default.

SCHED_WARN_ON() in uclamp_rq_dec_id() was removed since now we can end up with unbalanced call to uclamp_rq_dec_id() if we flip the key while a task is running in the rq. Since we know it is harmless we just quietly return if we attempt a uclamp_rq_dec_id() when rq->uclamp[].bucket[].tasks is 0.

In schedutil, we introduce a new uclamp_is_enabled() helper which takes the static key into account to ensure RT boosting behavior is retained.

The following results demonstrates how this helps on 2 Sockets Xeon E5 2x10-Cores system.

                                   nouclamp                 uclamp      uclamp-static-key
Hmean     send-64         162.43 (   0.00%)      157.84 *  -2.82%*      163.39 *   0.59%*
Hmean     send-128        324.71 (   0.00%)      314.78 *  -3.06%*      326.18 *   0.45%*
Hmean     send-256        641.55 (   0.00%)      628.67 *  -2.01%*      648.12 *   1.02%*
Hmean     send-1024      2525.28 (   0.00%)     2448.26 *  -3.05%*     2543.73 *   0.73%*
Hmean     send-2048      4836.14 (   0.00%)     4712.08 *  -2.57%*     4867.69 *   0.65%*
Hmean     send-3312      7540.83 (   0.00%)     7425.45 *  -1.53%*     7621.06 *   1.06%*
Hmean     send-4096      9124.53 (   0.00%)     8948.82 *  -1.93%*     9276.25 *   1.66%*
Hmean     send-8192     15589.67 (   0.00%)    15486.35 *  -0.66%*    15819.98 *   1.48%*
Hmean     send-16384    26386.47 (   0.00%)    25752.25 *  -2.40%*    26773.74 *   1.47%*

The perf diff between nouclamp and uclamp-static-key when uclamp is disabled in the fast path:

     8.73%     -1.55%  [kernel.kallsyms]        [k] try_to_wake_up
     0.07%     +0.04%  [kernel.kallsyms]        [k] deactivate_task
     0.13%     -0.02%  [kernel.kallsyms]        [k] activate_task

The diff between nouclamp and uclamp-static-key when uclamp is enabled in the fast path:

     8.73%     -0.72%  [kernel.kallsyms]        [k] try_to_wake_up
     0.13%     +0.39%  [kernel.kallsyms]        [k] activate_task
     0.07%     +0.38%  [kernel.kallsyms]        [k] deactivate_task
---
 kernel/sched/core.c              | 74 +++++++++++++++++++++++++++++++-
 kernel/sched/cpufreq_schedutil.c |  2 +-
 kernel/sched/sched.h             | 47 +++++++++++++++++++-
 3 files changed, 119 insertions(+), 4 deletions(-)
kernel/sched/core.c +73 -1
index 1bae86fc128b2..49b5c5cf9271d 100644
@@ -793,6 +793,26 @@ unsigned int sysctl_sched_uclamp_util_max = SCHED_CAPACITY_SCALE;
 /* All clamps are required to be less or equal than these values */
 static struct uclamp_se uclamp_default[UCLAMP_CNT];
 
+/*
+ * This static key is used to reduce the uclamp overhead in the fast path. It
+ * primarily disables the call to uclamp_rq_{inc, dec}() in
+ * enqueue/dequeue_task().
+ *
+ * This allows users to continue to enable uclamp in their kernel config with
+ * minimum uclamp overhead in the fast path.
+ *
+ * As soon as userspace modifies any of the uclamp knobs, the static key is
+ * enabled, since we have an actual users that make use of uclamp
+ * functionality.
+ *
+ * The knobs that would enable this static key are:
+ *
+ *   * A task modifying its uclamp value with sched_setattr().
+ *   * An admin modifying the sysctl_sched_uclamp_{min, max} via procfs.
+ *   * An admin modifying the cgroup cpu.uclamp.{min, max}
+ */
+DEFINE_STATIC_KEY_FALSE(sched_uclamp_used);
+
 /* Integer rounded range for each bucket */
 #define UCLAMP_BUCKET_DELTA DIV_ROUND_CLOSEST(SCHED_CAPACITY_SCALE, UCLAMP_BUCKETS)
 
@@ -989,10 +1009,38 @@ static inline void uclamp_rq_dec_id(struct rq *rq, struct task_struct *p,
 
 	lockdep_assert_held(&rq->lock);
 
+	/*
+	 * If sched_uclamp_used was enabled after task @p was enqueued,
+	 * we could end up with unbalanced call to uclamp_rq_dec_id().
+	 *
+	 * In this case the uc_se->active flag should be false since no uclamp
+	 * accounting was performed at enqueue time and we can just return
+	 * here.
+	 *
+	 * Need to be careful of the following enqeueue/dequeue ordering
+	 * problem too
+	 *
+	 *	enqueue(taskA)
+	 *	// sched_uclamp_used gets enabled
+	 *	enqueue(taskB)
+	 *	dequeue(taskA)
+	 *	// Must not decrement bukcet->tasks here
+	 *	dequeue(taskB)
+	 *
+	 * where we could end up with stale data in uc_se and
+	 * bucket[uc_se->bucket_id].
+	 *
+	 * The following check here eliminates the possibility of such race.
+	 */
+	if (unlikely(!uc_se->active))
+		return;
+
 	bucket = &uc_rq->bucket[uc_se->bucket_id];
+
 	SCHED_WARN_ON(!bucket->tasks);
 	if (likely(bucket->tasks))
 		bucket->tasks--;
+
 	uc_se->active = false;
 
 	/*
@@ -1020,6 +1068,15 @@ static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p)
 {
 	enum uclamp_id clamp_id;
 
+	/*
+	 * Avoid any overhead until uclamp is actually used by the userspace.
+	 *
+	 * The condition is constructed such that a NOP is generated when
+	 * sched_uclamp_used is disabled.
+	 */
+	if (!static_branch_unlikely(&sched_uclamp_used))
+		return;
+
 	if (unlikely(!p->sched_class->uclamp_enabled))
 		return;
 
@@ -1035,6 +1092,15 @@ static inline void uclamp_rq_dec(struct rq *rq, struct task_struct *p)
 {
 	enum uclamp_id clamp_id;
 
+	/*
+	 * Avoid any overhead until uclamp is actually used by the userspace.
+	 *
+	 * The condition is constructed such that a NOP is generated when
+	 * sched_uclamp_used is disabled.
+	 */
+	if (!static_branch_unlikely(&sched_uclamp_used))
+		return;
+
 	if (unlikely(!p->sched_class->uclamp_enabled))
 		return;
 
@@ -1144,8 +1210,10 @@ int sysctl_sched_uclamp_handler(struct ctl_table *table, int write,
 		update_root_tg = true;
 	}
 
-	if (update_root_tg)
+	if (update_root_tg) {
+		static_branch_enable(&sched_uclamp_used);
 		uclamp_update_root_tg();
+	}
 
 	/*
 	 * We update all RUNNABLE tasks only when task groups are in use.
@@ -1210,6 +1278,8 @@ static void __setscheduler_uclamp(struct task_struct *p,
 	if (likely(!(attr->sched_flags & SCHED_FLAG_UTIL_CLAMP)))
 		return;
 
+	static_branch_enable(&sched_uclamp_used);
+
 	if (attr->sched_flags & SCHED_FLAG_UTIL_CLAMP_MIN) {
 		uclamp_se_set(&p->uclamp_req[UCLAMP_MIN],
 			      attr->sched_util_min, true);
@@ -7306,6 +7376,8 @@ static ssize_t cpu_uclamp_write(struct kernfs_open_file *of, char *buf,
 	if (req.ret)
 		return req.ret;
 
+	static_branch_enable(&sched_uclamp_used);
+
 	mutex_lock(&uclamp_mutex);
 	rcu_read_lock();
 
kernel/sched/cpufreq_schedutil.c +1 -1
index 7fbaee24c824f..dc6835bc64907 100644
@@ -210,7 +210,7 @@ unsigned long schedutil_cpu_util(int cpu, unsigned long util_cfs,
 	unsigned long dl_util, util, irq;
 	struct rq *rq = cpu_rq(cpu);
 
-	if (!IS_BUILTIN(CONFIG_UCLAMP_TASK) &&
+	if (!uclamp_is_used() &&
 	    type == FREQUENCY_UTIL && rt_rq_is_runnable(&rq->rt)) {
 		return max;
 	}
kernel/sched/sched.h +45 -2

75 lines not on this page · raw message

Signature · 1 line
2.25.1
author maintainer
#113 lore raw

PATCH 113/204 selftests/bpf: Test_progs indicate to shell on non-actions

In reply to Greg Kroah-Hartman

From: Jesper Dangaard Brouer <[email protected]>

[ Upstream commit 6c92bd5cd4650c39dd929565ee172984c680fead ]

When a user selects a non-existing test the summary is printed with indication 0 for all info types, and shell "success" (EXIT_SUCCESS) is indicated. This can be understood by a human end-user, but for shell scripting is it useful to indicate a shell failure (EXIT_FAILURE).

---
 tools/testing/selftests/bpf/test_progs.c | 3 +++
 1 file changed, 3 insertions(+)
tools/testing/selftests/bpf/test_progs.c +3 -0
index 93970ec1c9e94..ac40d9558b80e 100644
@@ -776,5 +776,8 @@ int main(int argc, char **argv)
 	free_str_set(&env.subtest_selector.whitelist);
 	free(env.subtest_selector.num_set);
 
+	if (env.succ_cnt + env.fail_cnt + env.skip_cnt == 0)
+		return EXIT_FAILURE;
+
 	return env.fail_cnt ? EXIT_FAILURE : EXIT_SUCCESS;
 }
Signature · 1 line
2.25.1
author maintainer
#114 lore raw

PATCH 114/204 bpf: selftests: Restore netns after each test

In reply to Greg Kroah-Hartman

From: Martin KaFai Lau <[email protected]>

[ Upstream commit 811d7e375d08312dba23f3b6bf7e58ec14aa5dcb ]

It is common for networking tests creating its netns and making its own setting under this new netns (e.g. changing tcp sysctl). If the test forgot to restore to the original netns, it would affect the result of other tests.

This patch saves the original netns at the beginning and then restores it after every test. Since the restore "setns()" is not expensive, it does it on all tests without tracking if a test has created a new netns or not.

The new restore_netns() could also be done in test__end_subtest() such that each subtest will get an automatic netns reset. However, the individual test would lose flexibility to have total control on netns for its own subtests. In some cases, forcing a test to do unnecessary netns re-configure for each subtest is time consuming. e.g. In my vm, forcing netns re-configure on each subtest in sk_assign.c increased the runtime from 1s to 8s. On top of that, test_progs.c is also doing per-test (instead of per-subtest) cleanup for cgroup. Thus, this patch also does per-test restore_netns(). The only existing per-subtest cleanup is reset_affinity() and no test is depending on this. Thus, it is removed from test__end_subtest() to give a consistent expectation to the individual tests. test_progs.c only ensures any affinity/netns/cgroup change made by an earlier test does not affect the following tests.

---
 tools/testing/selftests/bpf/test_progs.c | 23 +++++++++++++++++++++--
 tools/testing/selftests/bpf/test_progs.h |  2 ++
 2 files changed, 23 insertions(+), 2 deletions(-)
tools/testing/selftests/bpf/test_progs.c +21 -2
index ac40d9558b80e..64bb5728040f4 100644
@@ -121,6 +121,24 @@ static void reset_affinity() {
 	}
 }
 
+static void save_netns(void)
+{
+	env.saved_netns_fd = open("/proc/self/ns/net", O_RDONLY);
+	if (env.saved_netns_fd == -1) {
+		perror("open(/proc/self/ns/net)");
+		exit(-1);
+	}
+}
+
+static void restore_netns(void)
+{
+	if (setns(env.saved_netns_fd, CLONE_NEWNET) == -1) {
+		stdio_restore();
+		perror("setns(CLONE_NEWNS)");
+		exit(-1);
+	}
+}
+
 void test__end_subtest()
 {
 	struct prog_test_def *test = env.test;
@@ -138,8 +156,6 @@ void test__end_subtest()
 	       test->test_num, test->subtest_num,
 	       test->subtest_name, sub_error_cnt ? "FAIL" : "OK");
 
-	reset_affinity();
-
 	free(test->subtest_name);
 	test->subtest_name = NULL;
 }
@@ -732,6 +748,7 @@ int main(int argc, char **argv)
 		return -1;
 	}
 
+	save_netns();
 	stdio_hijack();
 	for (i = 0; i < prog_test_cnt; i++) {
 		struct prog_test_def *test = &prog_test_defs[i];
@@ -762,6 +779,7 @@ int main(int argc, char **argv)
 			test->error_cnt ? "FAIL" : "OK");
 
 		reset_affinity();
+		restore_netns();
 		if (test->need_cgroup_cleanup)
 			cleanup_cgroup_environment();
 	}
@@ -775,6 +793,7 @@ int main(int argc, char **argv)
 	free_str_set(&env.subtest_selector.blacklist);
 	free_str_set(&env.subtest_selector.whitelist);
 	free(env.subtest_selector.num_set);
+	close(env.saved_netns_fd);
 
 	if (env.succ_cnt + env.fail_cnt + env.skip_cnt == 0)
 		return EXIT_FAILURE;
tools/testing/selftests/bpf/test_progs.h +2 -0
index f4aff6b8284be..3817667deb103 100644
@@ -77,6 +77,8 @@ struct test_env {
 	int sub_succ_cnt; /* successful sub-tests */
 	int fail_cnt; /* total failed tests + sub-tests */
 	int skip_cnt; /* skipped tests */
+
+	int saved_netns_fd;
 };
 
 extern struct test_env env;
Signature · 1 line
2.25.1
author maintainer
#115 lore raw

PATCH 115/204 selftests/bpf: test_progs use another shell exit on non-actions

In reply to Greg Kroah-Hartman

From: Jesper Dangaard Brouer <[email protected]>

[ Upstream commit 3220fb667842a9725cbb71656f406eadb03c094b ]

This is a follow up adjustment to commit 6c92bd5cd465 ("selftests/bpf:
Test_progs indicate to shell on non-actions"), that returns shell exit indication EXIT_FAILURE (value 1) when user selects a non-existing test.

The problem with using EXIT_FAILURE is that a shell script cannot tell the difference between a non-existing test and the test failing.

This patch uses value 2 as shell exit indication. (Aside note unrecognized option parameters use value 64).

---
 tools/testing/selftests/bpf/test_progs.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
tools/testing/selftests/bpf/test_progs.c +3 -1
index 64bb5728040f4..a75a83155f6d0 100644
@@ -12,6 +12,8 @@
 #include <string.h>
 #include <execinfo.h> /* backtrace */
 
+#define EXIT_NO_TEST		2
+
 /* defined in test_progs.h */
 struct test_env env = {};
 
@@ -796,7 +798,7 @@ int main(int argc, char **argv)
 	close(env.saved_netns_fd);
 
 	if (env.succ_cnt + env.fail_cnt + env.skip_cnt == 0)
-		return EXIT_FAILURE;
+		return EXIT_NO_TEST;
 
 	return env.fail_cnt ? EXIT_FAILURE : EXIT_SUCCESS;
 }
Signature · 1 line
2.25.1
author maintainer
#116 lore raw

PATCH 116/204 selftests/bpf: test_progs avoid minus shell exit codes

In reply to Greg Kroah-Hartman

From: Jesper Dangaard Brouer <[email protected]>

[ Upstream commit b8c50df0cb3eb9008f8372e4ff0317eee993b8d1 ]

There are a number of places in test_progs that use minus-1 as the argument to exit(). This is confusing as a process exit status is masked to be a number between 0 and 255 as defined in man exit(3). Thus, users will see status 255 instead of minus-1.

This patch use positive exit code 3 instead of minus-1. These cases are put in the same group of infrastructure setup errors.

Fixes fd27b1835e70 ("selftests/bpf: Reset process and thread affinity after each test/sub-test")
Fixes 811d7e375d08 ("bpf: selftests: Restore netns after each test")
---
 tools/testing/selftests/bpf/test_progs.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
tools/testing/selftests/bpf/test_progs.c +5 -4
index a75a83155f6d0..c2eb58382113a 100644
@@ -13,6 +13,7 @@
 #include <execinfo.h> /* backtrace */
 
 #define EXIT_NO_TEST		2
+#define EXIT_ERR_SETUP_INFRA	3
 
 /* defined in test_progs.h */
 struct test_env env = {};
@@ -113,13 +114,13 @@ static void reset_affinity() {
 	if (err < 0) {
 		stdio_restore();
 		fprintf(stderr, "Failed to reset process affinity: %d!\n", err);
-		exit(-1);
+		exit(EXIT_ERR_SETUP_INFRA);
 	}
 	err = pthread_setaffinity_np(pthread_self(), sizeof(cpuset), &cpuset);
 	if (err < 0) {
 		stdio_restore();
 		fprintf(stderr, "Failed to reset thread affinity: %d!\n", err);
-		exit(-1);
+		exit(EXIT_ERR_SETUP_INFRA);
 	}
 }
 
@@ -128,7 +129,7 @@ static void save_netns(void)
 	env.saved_netns_fd = open("/proc/self/ns/net", O_RDONLY);
 	if (env.saved_netns_fd == -1) {
 		perror("open(/proc/self/ns/net)");
-		exit(-1);
+		exit(EXIT_ERR_SETUP_INFRA);
 	}
 }
 
@@ -137,7 +138,7 @@ static void restore_netns(void)
 	if (setns(env.saved_netns_fd, CLONE_NEWNET) == -1) {
 		stdio_restore();
 		perror("setns(CLONE_NEWNS)");
-		exit(-1);
+		exit(EXIT_ERR_SETUP_INFRA);
 	}
 }
 
Signature · 1 line
2.25.1
author maintainer
#117 lore raw

PATCH 117/204 USB: serial: ftdi_sio: make process-packet buffer unsigned

In reply to Greg Kroah-Hartman

From: Johan Hovold <[email protected]>

[ Upstream commit ab4cc4ef6724ea588e835fc1e764c4b4407a70b7 ]

Use an unsigned type for the process-packet buffer argument and give it a more apt name.

---
 drivers/usb/serial/ftdi_sio.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)
drivers/usb/serial/ftdi_sio.c +11 -11
index 9ad44a96dfe3a..96b9e2768ac5c 100644
@@ -2480,12 +2480,12 @@ static int ftdi_prepare_write_buffer(struct usb_serial_port *port,
 #define FTDI_RS_ERR_MASK (FTDI_RS_BI | FTDI_RS_PE | FTDI_RS_FE | FTDI_RS_OE)
 
 static int ftdi_process_packet(struct usb_serial_port *port,
-		struct ftdi_private *priv, char *packet, int len)
+		struct ftdi_private *priv, unsigned char *buf, int len)
 {
+	unsigned char status;
+	unsigned char *ch;
 	int i;
-	char status;
 	char flag;
-	char *ch;
 
 	if (len < 2) {
 		dev_dbg(&port->dev, "malformed packet\n");
@@ -2495,7 +2495,7 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 	/* Compare new line status to the old one, signal if different/
 	   N.B. packet may be processed more than once, but differences
 	   are only processed once.  */
-	status = packet[0] & FTDI_STATUS_B0_MASK;
+	status = buf[0] & FTDI_STATUS_B0_MASK;
 	if (status != priv->prev_status) {
 		char diff_status = status ^ priv->prev_status;
 
@@ -2521,7 +2521,7 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 	}
 
 	/* save if the transmitter is empty or not */
-	if (packet[1] & FTDI_RS_TEMT)
+	if (buf[1] & FTDI_RS_TEMT)
 		priv->transmit_empty = 1;
 	else
 		priv->transmit_empty = 0;
@@ -2535,29 +2535,29 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 	 * data payload to avoid over-reporting.
 	 */
 	flag = TTY_NORMAL;
-	if (packet[1] & FTDI_RS_ERR_MASK) {
+	if (buf[1] & FTDI_RS_ERR_MASK) {
 		/* Break takes precedence over parity, which takes precedence
 		 * over framing errors */
-		if (packet[1] & FTDI_RS_BI) {
+		if (buf[1] & FTDI_RS_BI) {
 			flag = TTY_BREAK;
 			port->icount.brk++;
 			usb_serial_handle_break(port);
-		} else if (packet[1] & FTDI_RS_PE) {
+		} else if (buf[1] & FTDI_RS_PE) {
 			flag = TTY_PARITY;
 			port->icount.parity++;
-		} else if (packet[1] & FTDI_RS_FE) {
+		} else if (buf[1] & FTDI_RS_FE) {
 			flag = TTY_FRAME;
 			port->icount.frame++;
 		}
 		/* Overrun is special, not associated with a char */
-		if (packet[1] & FTDI_RS_OE) {
+		if (buf[1] & FTDI_RS_OE) {
 			port->icount.overrun++;
 			tty_insert_flip_char(&port->port, 0, TTY_OVERRUN);
 		}
 	}
 
 	port->icount.rx += len;
-	ch = packet + 2;
+	ch = buf + 2;
 
 	if (port->port.console && port->sysrq) {
 		for (i = 0; i < len; i++, ch++) {
Signature · 1 line
2.25.1
author maintainer
#118 lore raw

PATCH 118/204 USB: serial: ftdi_sio: clean up receive processing

In reply to Greg Kroah-Hartman

From: Johan Hovold <[email protected]>

[ Upstream commit ce054039ba5e47b75a3be02a00274e52b06a6456 ]

Clean up receive processing by dropping the character pointer and keeping the length argument unchanged throughout the function.

Also make it more apparent that sysrq processing can consume a characters by adding an explicit continue.

---
 drivers/usb/serial/ftdi_sio.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)
drivers/usb/serial/ftdi_sio.c +9 -10
index 96b9e2768ac5c..33f1cca7eaa61 100644
@@ -2483,7 +2483,6 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 		struct ftdi_private *priv, unsigned char *buf, int len)
 {
 	unsigned char status;
-	unsigned char *ch;
 	int i;
 	char flag;
 
@@ -2526,8 +2525,7 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 	else
 		priv->transmit_empty = 0;
 
-	len -= 2;
-	if (!len)
+	if (len == 2)
 		return 0;	/* status only */
 
 	/*
@@ -2556,19 +2554,20 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 		}
 	}
 
-	port->icount.rx += len;
-	ch = buf + 2;
+	port->icount.rx += len - 2;
 
 	if (port->port.console && port->sysrq) {
-		for (i = 0; i < len; i++, ch++) {
-			if (!usb_serial_handle_sysrq_char(port, *ch))
-				tty_insert_flip_char(&port->port, *ch, flag);
+		for (i = 2; i < len; i++) {
+			if (usb_serial_handle_sysrq_char(port, buf[i]))
+				continue;
+			tty_insert_flip_char(&port->port, buf[i], flag);
 		}
 	} else {
-		tty_insert_flip_string_fixed_flag(&port->port, ch, flag, len);
+		tty_insert_flip_string_fixed_flag(&port->port, buf + 2, flag,
+				len - 2);
 	}
 
-	return len;
+	return len - 2;
 }
 
 static void ftdi_process_read_urb(struct urb *urb)
Signature · 1 line
2.25.1
author maintainer
#119 lore raw

PATCH 119/204 USB: serial: ftdi_sio: fix break and sysrq handling

In reply to Greg Kroah-Hartman

From: Johan Hovold <[email protected]>

[ Upstream commit 733fff67941dad64b8a630450b8372b1873edc41 ]

Only the last NUL in a packet should be flagged as a break character, for example, to avoid dropping unrelated characters when IGNBRK is set.

Also make sysrq work by consuming the break character instead of having it immediately cancel the sysrq request, and by not processing it prematurely to avoid triggering a sysrq based on an unrelated character received in the same packet (which was received *before* the break).

Note that the break flag can be left set also for a packet received immediately following a break and that and an ending NUL in such a packet will continue to be reported as a break as there's no good way to tell it apart from an actual break.

Tested on FT232R and FT232H.

Fixes 72fda3ca6fc1 ("USB: serial: ftd_sio: implement sysrq handling on break")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 drivers/usb/serial/ftdi_sio.c | 24 +++++++++++++++++-------
 1 file changed, 17 insertions(+), 7 deletions(-)
drivers/usb/serial/ftdi_sio.c +17 -7
index 33f1cca7eaa61..07b146d7033a6 100644
@@ -2483,6 +2483,7 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 		struct ftdi_private *priv, unsigned char *buf, int len)
 {
 	unsigned char status;
+	bool brkint = false;
 	int i;
 	char flag;
 
@@ -2534,13 +2535,17 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 	 */
 	flag = TTY_NORMAL;
 	if (buf[1] & FTDI_RS_ERR_MASK) {
-		/* Break takes precedence over parity, which takes precedence
-		 * over framing errors */
-		if (buf[1] & FTDI_RS_BI) {
-			flag = TTY_BREAK;
+		/*
+		 * Break takes precedence over parity, which takes precedence
+		 * over framing errors. Note that break is only associated
+		 * with the last character in the buffer and only when it's a
+		 * NUL.
+		 */
+		if (buf[1] & FTDI_RS_BI && buf[len - 1] == '\0') {
 			port->icount.brk++;
-			usb_serial_handle_break(port);
-		} else if (buf[1] & FTDI_RS_PE) {
+			brkint = true;
+		}
+		if (buf[1] & FTDI_RS_PE) {
 			flag = TTY_PARITY;
 			port->icount.parity++;
 		} else if (buf[1] & FTDI_RS_FE) {
@@ -2556,8 +2561,13 @@ static int ftdi_process_packet(struct usb_serial_port *port,
 
 	port->icount.rx += len - 2;
 
-	if (port->port.console && port->sysrq) {
+	if (brkint || (port->port.console && port->sysrq)) {
 		for (i = 2; i < len; i++) {
+			if (brkint && i == len - 1) {
+				if (usb_serial_handle_break(port))
+					return len - 3;
+				flag = TTY_BREAK;
+			}
 			if (usb_serial_handle_sysrq_char(port, buf[i]))
 				continue;
 			tty_insert_flip_char(&port->port, buf[i], flag);
Signature · 1 line
2.25.1
author maintainer
#120 lore raw

PATCH 120/204 crypto: af_alg - Fix regression on empty requests

In reply to Greg Kroah-Hartman

From: Herbert Xu <[email protected]>

[ Upstream commit 662bb52f50bca16a74fe92b487a14d7dccb85e1a ]

Some user-space programs rely on crypto requests that have no control metadata. This broke when a check was added to require the presence of control metadata with the ctx->init flag.

This patch fixes the regression by setting ctx->init as long as one sendmsg(2) has been made, with or without a control message.

Fixes f3c802a1f300 ("crypto: algif_aead - Only wake up when...")
---
 crypto/af_alg.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
crypto/af_alg.c +1 -1
index 9fcb91ea10c41..5882ed46f1adb 100644
@@ -851,6 +851,7 @@ int af_alg_sendmsg(struct socket *sock, struct msghdr *msg, size_t size,
 		err = -EINVAL;
 		goto unlock;
 	}
+	ctx->init = true;
 
 	if (init) {
 		ctx->enc = enc;
@@ -858,7 +859,6 @@ int af_alg_sendmsg(struct socket *sock, struct msghdr *msg, size_t size,
 			memcpy(ctx->iv, con.iv->iv, ivsize);
 
 		ctx->aead_assoclen = con.aead_assoclen;
-		ctx->init = true;
 	}
 
 	while (size) {
Signature · 1 line
2.25.1
author maintainer
#121 lore raw

PATCH 121/204 devres: keep both device name and resource name in pretty name

In reply to Greg Kroah-Hartman

From: Vladimir Oltean <[email protected]>

[ Upstream commit 35bd8c07db2ce8fd2834ef866240613a4ef982e7 ]

Sometimes debugging a device is easiest using devmem on its register map, and that can be seen with /proc/iomem. But some device drivers have many memory regions. Take for example a networking switch. Its memory map used to look like this in /proc/iomem:

1fc000000-1fc3fffff : pcie@1f0000000
1fc000000-1fc3fffff : 0000:00:00.5

    1fc010000-1fc01ffff : sys
    1fc030000-1fc03ffff : rew
    1fc060000-1fc0603ff : s2
    1fc070000-1fc0701ff : devcpu_gcb
    1fc080000-1fc0800ff : qs
    1fc090000-1fc0900cb : ptp
    1fc100000-1fc10ffff : port0
    1fc110000-1fc11ffff : port1
    1fc120000-1fc12ffff : port2
    1fc130000-1fc13ffff : port3
    1fc140000-1fc14ffff : port4
    1fc150000-1fc15ffff : port5
    1fc200000-1fc21ffff : qsys
    1fc280000-1fc28ffff : ana

But after the patch in Fixes: was applied, the information is now presented in a much more opaque way:

1fc000000-1fc3fffff : pcie@1f0000000
1fc000000-1fc3fffff : 0000:00:00.5

    1fc010000-1fc01ffff : 0000:00:00.5
    1fc030000-1fc03ffff : 0000:00:00.5
    1fc060000-1fc0603ff : 0000:00:00.5
    1fc070000-1fc0701ff : 0000:00:00.5
    1fc080000-1fc0800ff : 0000:00:00.5
    1fc090000-1fc0900cb : 0000:00:00.5
    1fc100000-1fc10ffff : 0000:00:00.5
    1fc110000-1fc11ffff : 0000:00:00.5
    1fc120000-1fc12ffff : 0000:00:00.5
    1fc130000-1fc13ffff : 0000:00:00.5
    1fc140000-1fc14ffff : 0000:00:00.5
    1fc150000-1fc15ffff : 0000:00:00.5
    1fc200000-1fc21ffff : 0000:00:00.5
    1fc280000-1fc28ffff : 0000:00:00.5

That patch made a fair comment that /proc/iomem might be confusing when it shows resources without an associated device, but we can do better than just hide the resource name altogether. Namely, we can print the device name _and_ the resource name. Like this:

1fc000000-1fc3fffff : pcie@1f0000000
1fc000000-1fc3fffff : 0000:00:00.5

    1fc010000-1fc01ffff : 0000:00:00.5 sys
    1fc030000-1fc03ffff : 0000:00:00.5 rew
    1fc060000-1fc0603ff : 0000:00:00.5 s2
    1fc070000-1fc0701ff : 0000:00:00.5 devcpu_gcb
    1fc080000-1fc0800ff : 0000:00:00.5 qs
    1fc090000-1fc0900cb : 0000:00:00.5 ptp
    1fc100000-1fc10ffff : 0000:00:00.5 port0
    1fc110000-1fc11ffff : 0000:00:00.5 port1
    1fc120000-1fc12ffff : 0000:00:00.5 port2
    1fc130000-1fc13ffff : 0000:00:00.5 port3
    1fc140000-1fc14ffff : 0000:00:00.5 port4
    1fc150000-1fc15ffff : 0000:00:00.5 port5
    1fc200000-1fc21ffff : 0000:00:00.5 qsys
    1fc280000-1fc28ffff : 0000:00:00.5 ana
Fixes 8d84b18f5678 ("devres: always use dev_name() in devm_ioremap_resource()")
---
 lib/devres.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
lib/devres.c +10 -1
index 6ef51f159c54b..ca0d28727ccef 100644
@@ -119,6 +119,7 @@ __devm_ioremap_resource(struct device *dev, const struct resource *res,
 {
 	resource_size_t size;
 	void __iomem *dest_ptr;
+	char *pretty_name;
 
 	BUG_ON(!dev);
 
@@ -129,7 +130,15 @@ __devm_ioremap_resource(struct device *dev, const struct resource *res,
 
 	size = resource_size(res);
 
-	if (!devm_request_mem_region(dev, res->start, size, dev_name(dev))) {
+	if (res->name)
+		pretty_name = devm_kasprintf(dev, GFP_KERNEL, "%s %s",
+					     dev_name(dev), res->name);
+	else
+		pretty_name = devm_kstrdup(dev, dev_name(dev), GFP_KERNEL);
+	if (!pretty_name)
+		return IOMEM_ERR_PTR(-ENOMEM);
+
+	if (!devm_request_mem_region(dev, res->start, size, pretty_name)) {
 		dev_err(dev, "can't request region for resource %pR\n", res);
 		return IOMEM_ERR_PTR(-EBUSY);
 	}
Signature · 1 line
2.25.1
author maintainer
#122 lore raw

PATCH 122/204 RDMA/counter: Only bind user QPs in auto mode

In reply to Greg Kroah-Hartman

From: Mark Zhang <[email protected]>

[ Upstream commit c9f557421e505f75da4234a6af8eff46bc08614b ]

In auto mode only bind user QPs to a dynamic counter, since this feature is mainly used for system statistic and diagnostic purpose, while there's no need to counter kernel QPs so far.

Fixes 99fa331dc862 ("RDMA/counter: Add "auto" configuration mode support")
---
 drivers/infiniband/core/counters.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/core/counters.c +1 -1
index 738d1faf4bba5..6deb1901fbd02 100644
@@ -288,7 +288,7 @@ int rdma_counter_bind_qp_auto(struct ib_qp *qp, u8 port)
 	struct rdma_counter *counter;
 	int ret;
 
-	if (!qp->res.valid)
+	if (!qp->res.valid || rdma_is_kernel_res(&qp->res))
 		return 0;
 
 	if (!rdma_is_port_valid(dev, port))
Signature · 1 line
2.25.1
author maintainer
#123 lore raw

PATCH 123/204 RDMA/counter: Allow manually bind QPs with different pids to same counter

In reply to Greg Kroah-Hartman

From: Mark Zhang <[email protected]>

[ Upstream commit cbeb7d896c0f296451ffa7b67e7706786b8364c8 ]

In manual mode allow bind user QPs with different pids to same counter, since this is allowed in auto mode.
Bind kernel QPs and user QPs to the same counter are not allowed.

Fixes 1bd8e0a9d0fd ("RDMA/counter: Allow manual mode configuration support")
---
 drivers/infiniband/core/counters.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/core/counters.c +1 -1
index 6deb1901fbd02..417ebf4d8ba9b 100644
@@ -483,7 +483,7 @@ int rdma_counter_bind_qpn(struct ib_device *dev, u8 port,
 		goto err;
 	}
 
-	if (counter->res.task != qp->res.task) {
+	if (rdma_is_kernel_res(&counter->res) != rdma_is_kernel_res(&qp->res)) {
 		ret = -EINVAL;
 		goto err_task;
 	}
Signature · 1 line
2.25.1
author maintainer
#124 lore raw

PATCH 124/204 mmc: renesas_sdhi_internal_dmac: clean up the code for dma complete

In reply to Greg Kroah-Hartman

From: Yoshihiro Shimoda <[email protected]>

[ Upstream commit 2b26e34e9af3fa24fa1266e9ea2d66a1f7d62dc0 ]

To add end() operation in the future, clean the code of renesas_sdhi_internal_dmac_complete_tasklet_fn(). No behavior change.

---
 drivers/mmc/host/renesas_sdhi_internal_dmac.c | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)
drivers/mmc/host/renesas_sdhi_internal_dmac.c +13 -5
index 47ac53e912411..201b8ed37f2e0 100644
@@ -229,15 +229,12 @@ static void renesas_sdhi_internal_dmac_issue_tasklet_fn(unsigned long arg)
 					    DTRAN_CTRL_DM_START);
 }
 
-static void renesas_sdhi_internal_dmac_complete_tasklet_fn(unsigned long arg)
+static bool renesas_sdhi_internal_dmac_complete(struct tmio_mmc_host *host)
 {
-	struct tmio_mmc_host *host = (struct tmio_mmc_host *)arg;
 	enum dma_data_direction dir;
 
-	spin_lock_irq(&host->lock);
-
 	if (!host->data)
-		goto out;
+		return false;
 
 	if (host->data->flags & MMC_DATA_READ)
 		dir = DMA_FROM_DEVICE;
@@ -250,6 +247,17 @@ static void renesas_sdhi_internal_dmac_complete_tasklet_fn(unsigned long arg)
 	if (dir == DMA_FROM_DEVICE)
 		clear_bit(SDHI_INTERNAL_DMAC_RX_IN_USE, &global_flags);
 
+	return true;
+}
+
+static void renesas_sdhi_internal_dmac_complete_tasklet_fn(unsigned long arg)
+{
+	struct tmio_mmc_host *host = (struct tmio_mmc_host *)arg;
+
+	spin_lock_irq(&host->lock);
+	if (!renesas_sdhi_internal_dmac_complete(host))
+		goto out;
+
 	tmio_mmc_do_data_irq(host);
 out:
 	spin_unlock_irq(&host->lock);
Signature · 1 line
2.25.1
author maintainer
#125 lore raw

PATCH 125/204 rtw88: pci: disable aspm for platform inter-op with module parameter

In reply to Greg Kroah-Hartman

From: Yan-Hsuan Chuang <[email protected]>

[ Upstream commit 68aa716b7dd36f55e080da9e27bc594346334c41 ]

Some platforms cannot read the DBI register successfully for the ASPM settings. After the read failed, the bus could be unstable, and the device just became unavailable [1]. For those platforms, the ASPM should be disabled. But as the ASPM can help the driver to save the power consumption in power save mode, the ASPM is still needed. So, add a module parameter for them to disable it, then the device can still work, while others can benefit from the less power consumption that brings by ASPM enabled.

[1] https://bugzilla.kernel.org/show_bug.cgi?id=206411
[2] Note that my lenovo T430 is the same.

Fixes 3dff7c6e3749 ("rtw88: allows to enable/disable HCI link PS mechanism")
---
 drivers/net/wireless/realtek/rtw88/pci.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/net/wireless/realtek/rtw88/pci.c +9 -0
index d735f3127fe8f..6c24ddc2a9751 100644
@@ -14,8 +14,11 @@
 #include "debug.h"
 
 static bool rtw_disable_msi;
+static bool rtw_pci_disable_aspm;
 module_param_named(disable_msi, rtw_disable_msi, bool, 0644);
+module_param_named(disable_aspm, rtw_pci_disable_aspm, bool, 0644);
 MODULE_PARM_DESC(disable_msi, "Set Y to disable MSI interrupt support");
+MODULE_PARM_DESC(disable_aspm, "Set Y to disable PCI ASPM support");
 
 static u32 rtw_pci_tx_queue_idx_addr[] = {
 	[RTW_TX_QUEUE_BK]	= RTK_PCI_TXBD_IDX_BKQ,
@@ -1189,6 +1192,9 @@ static void rtw_pci_clkreq_set(struct rtw_dev *rtwdev, bool enable)
 	u8 value;
 	int ret;
 
+	if (rtw_pci_disable_aspm)
+		return;
+
 	ret = rtw_dbi_read8(rtwdev, RTK_PCIE_LINK_CFG, &value);
 	if (ret) {
 		rtw_err(rtwdev, "failed to read CLKREQ_L1, ret=%d", ret);
@@ -1208,6 +1214,9 @@ static void rtw_pci_aspm_set(struct rtw_dev *rtwdev, bool enable)
 	u8 value;
 	int ret;
 
+	if (rtw_pci_disable_aspm)
+		return;
+
 	ret = rtw_dbi_read8(rtwdev, RTK_PCIE_LINK_CFG, &value);
 	if (ret) {
 		rtw_err(rtwdev, "failed to read ASPM, ret=%d", ret);
Signature · 1 line
2.25.1
author maintainer
#126 lore raw

PATCH 126/204 rtc: pl031: fix set_alarm by adding back call to alarm_irq_enable

In reply to Greg Kroah-Hartman

From: Sudeep Holla <[email protected]>

[ Upstream commit 4df2ef85f0efe44505f511ca5e4455585f53a2da ]

Commit c8ff5841a90b ("rtc: pl031: switch to rtc_time64_to_tm/rtc_tm_to_time64") seemed to have accidentally removed the call to pl031_alarm_irq_enable from pl031_set_alarm while switching to 64-bit apis.

Let us add back the same to get the set alarm functionality back.

Fixes c8ff5841a90b ("rtc: pl031: switch to rtc_time64_to_tm/rtc_tm_to_time64")
Cc Linus Walleij <[email protected]>
Cc Alexandre Belloni <[email protected]>
---
 drivers/rtc/rtc-pl031.c | 1 +
 1 file changed, 1 insertion(+)
drivers/rtc/rtc-pl031.c +1 -0
index 40d7450a1ce49..c6b89273feba8 100644
@@ -275,6 +275,7 @@ static int pl031_set_alarm(struct device *dev, struct rtc_wkalrm *alarm)
 	struct pl031_local *ldata = dev_get_drvdata(dev);
 
 	writel(rtc_tm_to_time64(&alarm->time), ldata->base + RTC_MR);
+	pl031_alarm_irq_enable(dev, alarm->enabled);
 
 	return 0;
 }
Signature · 1 line
2.25.1
author maintainer
#127 lore raw

PATCH 127/204 crypto: caam - Remove broken arc4 support

In reply to Greg Kroah-Hartman

From: Herbert Xu <[email protected]>

[ Upstream commit eeedb618378f8a09779546a3eeac16b000447d62 ]

The arc4 algorithm requires storing state in the request context in order to allow more than one encrypt/decrypt operation. As this driver does not seem to do that, it means that using it for more than one operation is broken.

---
 drivers/crypto/caam/caamalg.c | 29 -----------------------------
 drivers/crypto/caam/compat.h  |  1 -
 2 files changed, 30 deletions(-)
drivers/crypto/caam/caamalg.c +0 -29
index bf90a4fcabd1f..8149ac4d6ef22 100644
@@ -810,12 +810,6 @@ static int ctr_skcipher_setkey(struct crypto_skcipher *skcipher,
 	return skcipher_setkey(skcipher, key, keylen, ctx1_iv_off);
 }
 
-static int arc4_skcipher_setkey(struct crypto_skcipher *skcipher,
-				const u8 *key, unsigned int keylen)
-{
-	return skcipher_setkey(skcipher, key, keylen, 0);
-}
-
 static int des_skcipher_setkey(struct crypto_skcipher *skcipher,
 			       const u8 *key, unsigned int keylen)
 {
@@ -1967,21 +1961,6 @@ static struct caam_skcipher_alg driver_algs[] = {
 		},
 		.caam.class1_alg_type = OP_ALG_ALGSEL_3DES | OP_ALG_AAI_ECB,
 	},
-	{
-		.skcipher = {
-			.base = {
-				.cra_name = "ecb(arc4)",
-				.cra_driver_name = "ecb-arc4-caam",
-				.cra_blocksize = ARC4_BLOCK_SIZE,
-			},
-			.setkey = arc4_skcipher_setkey,
-			.encrypt = skcipher_encrypt,
-			.decrypt = skcipher_decrypt,
-			.min_keysize = ARC4_MIN_KEY_SIZE,
-			.max_keysize = ARC4_MAX_KEY_SIZE,
-		},
-		.caam.class1_alg_type = OP_ALG_ALGSEL_ARC4 | OP_ALG_AAI_ECB,
-	},
 };
 
 static struct caam_aead_alg driver_aeads[] = {
@@ -3457,7 +3436,6 @@ int caam_algapi_init(struct device *ctrldev)
 	struct caam_drv_private *priv = dev_get_drvdata(ctrldev);
 	int i = 0, err = 0;
 	u32 aes_vid, aes_inst, des_inst, md_vid, md_inst, ccha_inst, ptha_inst;
-	u32 arc4_inst;
 	unsigned int md_limit = SHA512_DIGEST_SIZE;
 	bool registered = false, gcm_support;
 
@@ -3477,8 +3455,6 @@ int caam_algapi_init(struct device *ctrldev)
 			   CHA_ID_LS_DES_SHIFT;
 		aes_inst = cha_inst & CHA_ID_LS_AES_MASK;
 		md_inst = (cha_inst & CHA_ID_LS_MD_MASK) >> CHA_ID_LS_MD_SHIFT;
-		arc4_inst = (cha_inst & CHA_ID_LS_ARC4_MASK) >>
-			    CHA_ID_LS_ARC4_SHIFT;
 		ccha_inst = 0;
 		ptha_inst = 0;
 
@@ -3499,7 +3475,6 @@ int caam_algapi_init(struct device *ctrldev)
 		md_inst = mdha & CHA_VER_NUM_MASK;
 		ccha_inst = rd_reg32(&priv->ctrl->vreg.ccha) & CHA_VER_NUM_MASK;
 		ptha_inst = rd_reg32(&priv->ctrl->vreg.ptha) & CHA_VER_NUM_MASK;
-		arc4_inst = rd_reg32(&priv->ctrl->vreg.afha) & CHA_VER_NUM_MASK;
 
 		gcm_support = aesa & CHA_VER_MISC_AES_GCM;
 	}
@@ -3522,10 +3497,6 @@ int caam_algapi_init(struct device *ctrldev)
 		if (!aes_inst && (alg_sel == OP_ALG_ALGSEL_AES))
 				continue;
 
-		/* Skip ARC4 algorithms if not supported by device */
-		if (!arc4_inst && alg_sel == OP_ALG_ALGSEL_ARC4)
-			continue;
-
 		/*
 		 * Check support for AES modes not available
 		 * on LP devices.
drivers/crypto/caam/compat.h +0 -1
index 60e2a54c19f11..c3c22a8de4c00 100644
@@ -43,7 +43,6 @@
 #include <crypto/akcipher.h>
 #include <crypto/scatterwalk.h>
 #include <crypto/skcipher.h>
-#include <crypto/arc4.h>
 #include <crypto/internal/skcipher.h>
 #include <crypto/internal/hash.h>
 #include <crypto/internal/rsa.h>
Signature · 1 line
2.25.1
author maintainer
#128 lore raw

PATCH 128/204 gpu: ipu-v3: image-convert: Combine rotate/no-rotate irq handlers

In reply to Greg Kroah-Hartman

From: Steve Longerbeam <[email protected]>

[ Upstream commit 0f6245f42ce9b7e4d20f2cda8d5f12b55a44d7d1 ]

Combine the rotate_irq() and norotate_irq() handlers into a single eof_irq() handler.

---
 drivers/gpu/ipu-v3/ipu-image-convert.c | 58 +++++++++-----------------
 1 file changed, 20 insertions(+), 38 deletions(-)
drivers/gpu/ipu-v3/ipu-image-convert.c +20 -38
index eeca50d9a1ee4..f8b031ded3cf2 100644
@@ -1709,9 +1709,10 @@ static irqreturn_t do_irq(struct ipu_image_convert_run *run)
 	return IRQ_WAKE_THREAD;
 }
 
-static irqreturn_t norotate_irq(int irq, void *data)
+static irqreturn_t eof_irq(int irq, void *data)
 {
 	struct ipu_image_convert_chan *chan = data;
+	struct ipu_image_convert_priv *priv = chan->priv;
 	struct ipu_image_convert_ctx *ctx;
 	struct ipu_image_convert_run *run;
 	unsigned long flags;
@@ -1728,45 +1729,26 @@ static irqreturn_t norotate_irq(int irq, void *data)
 
 	ctx = run->ctx;
 
-	if (ipu_rot_mode_is_irt(ctx->rot_mode)) {
-		/* this is a rotation operation, just ignore */
-		spin_unlock_irqrestore(&chan->irqlock, flags);
-		return IRQ_HANDLED;
-	}
-
-	ret = do_irq(run);
-out:
-	spin_unlock_irqrestore(&chan->irqlock, flags);
-	return ret;
-}
-
-static irqreturn_t rotate_irq(int irq, void *data)
-{
-	struct ipu_image_convert_chan *chan = data;
-	struct ipu_image_convert_priv *priv = chan->priv;
-	struct ipu_image_convert_ctx *ctx;
-	struct ipu_image_convert_run *run;
-	unsigned long flags;
-	irqreturn_t ret;
-
-	spin_lock_irqsave(&chan->irqlock, flags);
-
-	/* get current run and its context */
-	run = chan->current_run;
-	if (!run) {
+	if (irq == chan->out_eof_irq) {
+		if (ipu_rot_mode_is_irt(ctx->rot_mode)) {
+			/* this is a rotation op, just ignore */
+			ret = IRQ_HANDLED;
+			goto out;
+		}
+	} else if (irq == chan->rot_out_eof_irq) {
+		if (!ipu_rot_mode_is_irt(ctx->rot_mode)) {
+			/* this was NOT a rotation op, shouldn't happen */
+			dev_err(priv->ipu->dev,
+				"Unexpected rotation interrupt\n");
+			ret = IRQ_HANDLED;
+			goto out;
+		}
+	} else {
+		dev_err(priv->ipu->dev, "Received unknown irq %d\n", irq);
 		ret = IRQ_NONE;
 		goto out;
 	}
 
-	ctx = run->ctx;
-
-	if (!ipu_rot_mode_is_irt(ctx->rot_mode)) {
-		/* this was NOT a rotation operation, shouldn't happen */
-		dev_err(priv->ipu->dev, "Unexpected rotation interrupt\n");
-		spin_unlock_irqrestore(&chan->irqlock, flags);
-		return IRQ_HANDLED;
-	}
-
 	ret = do_irq(run);
 out:
 	spin_unlock_irqrestore(&chan->irqlock, flags);
@@ -1859,7 +1841,7 @@ static int get_ipu_resources(struct ipu_image_convert_chan *chan)
 						  chan->out_chan,
 						  IPU_IRQ_EOF);
 
-	ret = request_threaded_irq(chan->out_eof_irq, norotate_irq, do_bh,
+	ret = request_threaded_irq(chan->out_eof_irq, eof_irq, do_bh,
 				   0, "ipu-ic", chan);
 	if (ret < 0) {
 		dev_err(priv->ipu->dev, "could not acquire irq %d\n",
@@ -1872,7 +1854,7 @@ static int get_ipu_resources(struct ipu_image_convert_chan *chan)
 						     chan->rotation_out_chan,
 						     IPU_IRQ_EOF);
 
-	ret = request_threaded_irq(chan->rot_out_eof_irq, rotate_irq, do_bh,
+	ret = request_threaded_irq(chan->rot_out_eof_irq, eof_irq, do_bh,
 				   0, "ipu-ic", chan);
 	if (ret < 0) {
 		dev_err(priv->ipu->dev, "could not acquire irq %d\n",
Signature · 1 line
2.25.1
author maintainer
#129 lore raw

PATCH 129/204 gpu: ipu-v3: image-convert: Wait for all EOFs before completing a tile

In reply to Greg Kroah-Hartman

From: Steve Longerbeam <[email protected]>

[ Upstream commit dd81d821d0b3f77d949d0cac5c05c1f05b921d46 ]

Use a bit-mask of EOF irqs to determine when all required idmac channel EOFs have been received for a tile conversion, and only do tile completion processing after all EOFs have been received. Otherwise it was found that a conversion would stall after the completion of a tile and the start of the next tile, because the input/read idmac channel had not completed and entered idle state, thus locking up the channel when attempting to re-start it for the next tile.

Fixes 0537db801bb01 ("gpu: ipu-v3: image-convert: reconfigure IC per tile")
---
 drivers/gpu/ipu-v3/ipu-image-convert.c | 109 +++++++++++++++++++------
 1 file changed, 82 insertions(+), 27 deletions(-)
drivers/gpu/ipu-v3/ipu-image-convert.c +82 -27
index f8b031ded3cf2..aa1d4b6d278f7 100644
@@ -137,6 +137,17 @@ struct ipu_image_convert_ctx;
 struct ipu_image_convert_chan;
 struct ipu_image_convert_priv;
 
+enum eof_irq_mask {
+	EOF_IRQ_IN      = BIT(0),
+	EOF_IRQ_ROT_IN  = BIT(1),
+	EOF_IRQ_OUT     = BIT(2),
+	EOF_IRQ_ROT_OUT = BIT(3),
+};
+
+#define EOF_IRQ_COMPLETE (EOF_IRQ_IN | EOF_IRQ_OUT)
+#define EOF_IRQ_ROT_COMPLETE (EOF_IRQ_IN | EOF_IRQ_OUT |	\
+			      EOF_IRQ_ROT_IN | EOF_IRQ_ROT_OUT)
+
 struct ipu_image_convert_ctx {
 	struct ipu_image_convert_chan *chan;
 
@@ -173,6 +184,9 @@ struct ipu_image_convert_ctx {
 	/* where to place converted tile in dest image */
 	unsigned int out_tile_map[MAX_TILES];
 
+	/* mask of completed EOF irqs at every tile conversion */
+	enum eof_irq_mask eof_mask;
+
 	struct list_head list;
 };
 
@@ -189,6 +203,8 @@ struct ipu_image_convert_chan {
 	struct ipuv3_channel *rotation_out_chan;
 
 	/* the IPU end-of-frame irqs */
+	int in_eof_irq;
+	int rot_in_eof_irq;
 	int out_eof_irq;
 	int rot_out_eof_irq;
 
@@ -1380,6 +1396,9 @@ static int convert_start(struct ipu_image_convert_run *run, unsigned int tile)
 	dev_dbg(priv->ipu->dev, "%s: task %u: starting ctx %p run %p tile %u -> %u\n",
 		__func__, chan->ic_task, ctx, run, tile, dst_tile);
 
+	/* clear EOF irq mask */
+	ctx->eof_mask = 0;
+
 	if (ipu_rot_mode_is_irt(ctx->rot_mode)) {
 		/* swap width/height for resizer */
 		dest_width = d_image->tile[dst_tile].height;
@@ -1615,7 +1634,7 @@ static bool ic_settings_changed(struct ipu_image_convert_ctx *ctx)
 }
 
 /* hold irqlock when calling */
-static irqreturn_t do_irq(struct ipu_image_convert_run *run)
+static irqreturn_t do_tile_complete(struct ipu_image_convert_run *run)
 {
 	struct ipu_image_convert_ctx *ctx = run->ctx;
 	struct ipu_image_convert_chan *chan = ctx->chan;
@@ -1700,6 +1719,7 @@ static irqreturn_t do_irq(struct ipu_image_convert_run *run)
 		ctx->cur_buf_num ^= 1;
 	}
 
+	ctx->eof_mask = 0; /* clear EOF irq mask for next tile */
 	ctx->next_tile++;
 	return IRQ_HANDLED;
 done:
@@ -1715,8 +1735,9 @@ static irqreturn_t eof_irq(int irq, void *data)
 	struct ipu_image_convert_priv *priv = chan->priv;
 	struct ipu_image_convert_ctx *ctx;
 	struct ipu_image_convert_run *run;
+	irqreturn_t ret = IRQ_HANDLED;
+	bool tile_complete = false;
 	unsigned long flags;
-	irqreturn_t ret;
 
 	spin_lock_irqsave(&chan->irqlock, flags);
 
@@ -1729,27 +1750,33 @@ static irqreturn_t eof_irq(int irq, void *data)
 
 	ctx = run->ctx;
 
-	if (irq == chan->out_eof_irq) {
-		if (ipu_rot_mode_is_irt(ctx->rot_mode)) {
-			/* this is a rotation op, just ignore */
-			ret = IRQ_HANDLED;
-			goto out;
-		}
-	} else if (irq == chan->rot_out_eof_irq) {
+	if (irq == chan->in_eof_irq) {
+		ctx->eof_mask |= EOF_IRQ_IN;
+	} else if (irq == chan->out_eof_irq) {
+		ctx->eof_mask |= EOF_IRQ_OUT;
+	} else if (irq == chan->rot_in_eof_irq ||
+		   irq == chan->rot_out_eof_irq) {
 		if (!ipu_rot_mode_is_irt(ctx->rot_mode)) {
 			/* this was NOT a rotation op, shouldn't happen */
 			dev_err(priv->ipu->dev,
 				"Unexpected rotation interrupt\n");
-			ret = IRQ_HANDLED;
 			goto out;
 		}
+		ctx->eof_mask |= (irq == chan->rot_in_eof_irq) ?
+			EOF_IRQ_ROT_IN : EOF_IRQ_ROT_OUT;
 	} else {
 		dev_err(priv->ipu->dev, "Received unknown irq %d\n", irq);
 		ret = IRQ_NONE;
 		goto out;
 	}
 
-	ret = do_irq(run);
+	if (ipu_rot_mode_is_irt(ctx->rot_mode))
+		tile_complete =	(ctx->eof_mask == EOF_IRQ_ROT_COMPLETE);
+	else
+		tile_complete = (ctx->eof_mask == EOF_IRQ_COMPLETE);
+
+	if (tile_complete)
+		ret = do_tile_complete(run);
 out:
 	spin_unlock_irqrestore(&chan->irqlock, flags);
 	return ret;
@@ -1783,6 +1810,10 @@ static void force_abort(struct ipu_image_convert_ctx *ctx)
 
 static void release_ipu_resources(struct ipu_image_convert_chan *chan)
 {
+	if (chan->in_eof_irq >= 0)
+		free_irq(chan->in_eof_irq, chan);
+	if (chan->rot_in_eof_irq >= 0)
+		free_irq(chan->rot_in_eof_irq, chan);
 	if (chan->out_eof_irq >= 0)
 		free_irq(chan->out_eof_irq, chan);
 	if (chan->rot_out_eof_irq >= 0)
@@ -1801,7 +1832,27 @@ static void release_ipu_resources(struct ipu_image_convert_chan *chan)
 
 	chan->in_chan = chan->out_chan = chan->rotation_in_chan =
 		chan->rotation_out_chan = NULL;
-	chan->out_eof_irq = chan->rot_out_eof_irq = -1;
+	chan->in_eof_irq = -1;
+	chan->rot_in_eof_irq = -1;
+	chan->out_eof_irq = -1;
+	chan->rot_out_eof_irq = -1;
+}
+
+static int get_eof_irq(struct ipu_image_convert_chan *chan,
+		       struct ipuv3_channel *channel)
+{
+	struct ipu_image_convert_priv *priv = chan->priv;
+	int ret, irq;
+
+	irq = ipu_idmac_channel_irq(priv->ipu, channel, IPU_IRQ_EOF);
+
+	ret = request_threaded_irq(irq, eof_irq, do_bh, 0, "ipu-ic", chan);
+	if (ret < 0) {
+		dev_err(priv->ipu->dev, "could not acquire irq %d\n", irq);
+		return ret;
+	}
+
+	return irq;
 }
 
 static int get_ipu_resources(struct ipu_image_convert_chan *chan)
@@ -1837,31 +1888,33 @@ static int get_ipu_resources(struct ipu_image_convert_chan *chan)
 	}
 
 	/* acquire the EOF interrupts */
-	chan->out_eof_irq = ipu_idmac_channel_irq(priv->ipu,
-						  chan->out_chan,
-						  IPU_IRQ_EOF);
+	ret = get_eof_irq(chan, chan->in_chan);
+	if (ret < 0) {
+		chan->in_eof_irq = -1;
+		goto err;
+	}
+	chan->in_eof_irq = ret;
 
-	ret = request_threaded_irq(chan->out_eof_irq, eof_irq, do_bh,
-				   0, "ipu-ic", chan);
+	ret = get_eof_irq(chan, chan->rotation_in_chan);
 	if (ret < 0) {
-		dev_err(priv->ipu->dev, "could not acquire irq %d\n",
-			 chan->out_eof_irq);
-		chan->out_eof_irq = -1;
+		chan->rot_in_eof_irq = -1;
 		goto err;
 	}
+	chan->rot_in_eof_irq = ret;
 
-	chan->rot_out_eof_irq = ipu_idmac_channel_irq(priv->ipu,
-						     chan->rotation_out_chan,
-						     IPU_IRQ_EOF);
+	ret = get_eof_irq(chan, chan->out_chan);
+	if (ret < 0) {
+		chan->out_eof_irq = -1;
+		goto err;
+	}
+	chan->out_eof_irq = ret;
 
-	ret = request_threaded_irq(chan->rot_out_eof_irq, eof_irq, do_bh,
-				   0, "ipu-ic", chan);
+	ret = get_eof_irq(chan, chan->rotation_out_chan);
 	if (ret < 0) {
-		dev_err(priv->ipu->dev, "could not acquire irq %d\n",
-			chan->rot_out_eof_irq);
 		chan->rot_out_eof_irq = -1;
 		goto err;
 	}
+	chan->rot_out_eof_irq = ret;
 
 	return 0;
 err:
@@ -2440,6 +2493,8 @@ int ipu_image_convert_init(struct ipu_soc *ipu, struct device *dev)
 		chan->ic_task = i;
 		chan->priv = priv;
 		chan->dma_ch = &image_convert_dma_chan[i];
+		chan->in_eof_irq = -1;
+		chan->rot_in_eof_irq = -1;
 		chan->out_eof_irq = -1;
 		chan->rot_out_eof_irq = -1;
 
Signature · 1 line
2.25.1
author maintainer
#130 lore raw

PATCH 130/204 dm rq: dont call blk_mq_queue_stopped() in dm_stop_queue()

In reply to Greg Kroah-Hartman

From: Ming Lei <[email protected]>

[ Upstream commit e766668c6cd49d741cfb49eaeb38998ba34d27bc ]

dm_stop_queue() only uses blk_mq_quiesce_queue() so it doesn't formally stop the blk-mq queue; therefore there is no point making the blk_mq_queue_stopped() check -- it will never be stopped.

In addition, even though dm_stop_queue() actually tries to quiesce hw queues via blk_mq_quiesce_queue(), checking with blk_queue_quiesced() to avoid unnecessary queue quiesce isn't reliable because: the QUEUE_FLAG_QUIESCED flag is set before synchronize_rcu() and dm_stop_queue() may be called when synchronize_rcu() from another blk_mq_quiesce_queue() is in-progress.

Fixes 7b17c2f7292ba ("dm: Fix a race condition related to stopping and starting queues")
---
 drivers/md/dm-rq.c | 3 ---
 1 file changed, 3 deletions(-)
drivers/md/dm-rq.c +0 -3
index 3f8577e2c13be..2bd2444ad99c6 100644
@@ -70,9 +70,6 @@ void dm_start_queue(struct request_queue *q)
 
 void dm_stop_queue(struct request_queue *q)
 {
-	if (blk_mq_queue_stopped(q))
-		return;
-
 	blk_mq_quiesce_queue(q);
 }
 
Signature · 1 line
2.25.1
author maintainer
#131 lore raw

PATCH 131/204 ima: Fail rule parsing when appraise_flag=blacklist is unsupportable

In reply to Greg Kroah-Hartman

From: Tyler Hicks <[email protected]>

[ Upstream commit 5f3e92657bbfb63ad3109433d843c89996114b03 ]

Verifying that a file hash is not blacklisted is currently only supported for files with appended signatures (modsig). In the future, this might change.

For now, the "appraise_flag" option is only appropriate for appraise actions and its "blacklist" value is only appropriate when CONFIG_IMA_APPRAISE_MODSIG is enabled and "appraise_flag=blacklist" is only appropriate when "appraise_type=imasig|modsig" is also present. Make this clear at policy load so that IMA policy authors don't assume that other uses of "appraise_flag=blacklist" are supported.

Fixes 273df864cf74 ("ima: Check against blacklisted hashes for files with modsig")
---
 security/integrity/ima/ima_policy.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)
security/integrity/ima/ima_policy.c +14 -1
index 3e3e568c81309..a59bf2f5b2d4f 100644
@@ -1035,6 +1035,11 @@ static bool ima_validate_rule(struct ima_rule_entry *entry)
 		return false;
 	}
 
+	/* Ensure that combinations of flags are compatible with each other */
+	if (entry->flags & IMA_CHECK_BLACKLIST &&
+	    !(entry->flags & IMA_MODSIG_ALLOWED))
+		return false;
+
 	return true;
 }
 
@@ -1371,9 +1376,17 @@ static int ima_parse_rule(char *rule, struct ima_rule_entry *entry)
 				result = -EINVAL;
 			break;
 		case Opt_appraise_flag:
+			if (entry->action != APPRAISE) {
+				result = -EINVAL;
+				break;
+			}
+
 			ima_log_string(ab, "appraise_flag", args[0].from);
-			if (strstr(args[0].from, "blacklist"))
+			if (IS_ENABLED(CONFIG_IMA_APPRAISE_MODSIG) &&
+			    strstr(args[0].from, "blacklist"))
 				entry->flags |= IMA_CHECK_BLACKLIST;
+			else
+				result = -EINVAL;
 			break;
 		case Opt_permit_directio:
 			entry->flags |= IMA_PERMIT_DIRECTIO;
Signature · 1 line
2.25.1
author maintainer
#132 lore raw

PATCH 132/204 f2fs: compress: fix to avoid memory leak on cc->cpages

In reply to Greg Kroah-Hartman

From: Chao Yu <[email protected]>

[ Upstream commit 02772fbfcba8597eef9d5c5f7f94087132d0c1d4 ]

Memory allocated for storing compressed pages' poitner should be released after f2fs_write_compressed_pages(), otherwise it will cause memory leak issue.

Fixes 4c8ff7095bef ("f2fs: support data compression")
---
 fs/f2fs/compress.c | 2 ++
 1 file changed, 2 insertions(+)
fs/f2fs/compress.c +2 -0
index a5b2e72174bb1..527d50edcb956 100644
@@ -1250,6 +1250,8 @@ int f2fs_write_multi_pages(struct compress_ctx *cc,
 		err = f2fs_write_compressed_pages(cc, submitted,
 							wbc, io_type);
 		cops->destroy_compress_ctx(cc);
+		kfree(cc->cpages);
+		cc->cpages = NULL;
 		if (!err)
 			return 0;
 		f2fs_bug_on(F2FS_I_SB(cc->inode), err != -EAGAIN);
Signature · 1 line
2.25.1
author maintainer
#133 lore raw

PATCH 133/204 clk: actions: Fix h_clk for Actions S500 SoC

In reply to Greg Kroah-Hartman

From: Cristian Ciocaltea <[email protected]>

[ Upstream commit f47ee279d25fb0e010cae5d6e758e39b40eb6378 ]

The h_clk clock in the Actions Semi S500 SoC clock driver has an invalid parent. Replace with the correct one.

---
 drivers/clk/actions/owl-s500.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clk/actions/owl-s500.c +1 -1
index e2007ac4d235d..0eb83a0b70bcc 100644
@@ -183,7 +183,7 @@ static OWL_GATE(timer_clk, "timer_clk", "hosc", CMU_DEVCLKEN1, 27, 0, 0);
 static OWL_GATE(hdmi_clk, "hdmi_clk", "hosc", CMU_DEVCLKEN1, 3, 0, 0);
 
 /* divider clocks */
-static OWL_DIVIDER(h_clk, "h_clk", "ahbprevdiv_clk", CMU_BUSCLK1, 12, 2, NULL, 0, 0);
+static OWL_DIVIDER(h_clk, "h_clk", "ahbprediv_clk", CMU_BUSCLK1, 12, 2, NULL, 0, 0);
 static OWL_DIVIDER(rmii_ref_clk, "rmii_ref_clk", "ethernet_pll_clk", CMU_ETHERNETPLL, 1, 1, rmii_ref_div_table, 0, 0);
 
 /* factor clocks */
Signature · 1 line
2.25.1
author maintainer
#134 lore raw

PATCH 134/204 selftests/powerpc: ptrace-pkey: Rename variables to make it easier to follow code

In reply to Greg Kroah-Hartman

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

[ Upstream commit 9a11f12e0a6c374b3ef1ce81e32ce477d28eb1b8 ]

Rename variable to indicate that they are invalid values which we will use to test ptrace update of pkeys.

---
 .../selftests/powerpc/ptrace/ptrace-pkey.c    | 26 +++++++++----------
 1 file changed, 13 insertions(+), 13 deletions(-)
tools/testing/selftests/powerpc/ptrace/ptrace-pkey.c +13 -13
index bdbbbe8431e03..f9216c7a1829e 100644
@@ -44,7 +44,7 @@ struct shared_info {
 	unsigned long amr2;
 
 	/* AMR value that ptrace should refuse to write to the child. */
-	unsigned long amr3;
+	unsigned long invalid_amr;
 
 	/* IAMR value the parent expects to read from the child. */
 	unsigned long expected_iamr;
@@ -57,8 +57,8 @@ struct shared_info {
 	 * (even though they're valid ones) because userspace doesn't have
 	 * access to those registers.
 	 */
-	unsigned long new_iamr;
-	unsigned long new_uamor;
+	unsigned long invalid_iamr;
+	unsigned long invalid_uamor;
 };
 
 static int sys_pkey_alloc(unsigned long flags, unsigned long init_access_rights)
@@ -100,7 +100,7 @@ static int child(struct shared_info *info)
 
 	info->amr1 |= 3ul << pkeyshift(pkey1);
 	info->amr2 |= 3ul << pkeyshift(pkey2);
-	info->amr3 |= info->amr2 | 3ul << pkeyshift(pkey3);
+	info->invalid_amr |= info->amr2 | 3ul << pkeyshift(pkey3);
 
 	if (disable_execute)
 		info->expected_iamr |= 1ul << pkeyshift(pkey1);
@@ -111,8 +111,8 @@ static int child(struct shared_info *info)
 
 	info->expected_uamor |= 3ul << pkeyshift(pkey1) |
 				3ul << pkeyshift(pkey2);
-	info->new_iamr |= 1ul << pkeyshift(pkey1) | 1ul << pkeyshift(pkey2);
-	info->new_uamor |= 3ul << pkeyshift(pkey1);
+	info->invalid_iamr |= 1ul << pkeyshift(pkey1) | 1ul << pkeyshift(pkey2);
+	info->invalid_uamor |= 3ul << pkeyshift(pkey1);
 
 	/*
 	 * We won't use pkey3. We just want a plausible but invalid key to test
@@ -196,9 +196,9 @@ static int parent(struct shared_info *info, pid_t pid)
 	PARENT_SKIP_IF_UNSUPPORTED(ret, &info->child_sync);
 	PARENT_FAIL_IF(ret, &info->child_sync);
 
-	info->amr1 = info->amr2 = info->amr3 = regs[0];
-	info->expected_iamr = info->new_iamr = regs[1];
-	info->expected_uamor = info->new_uamor = regs[2];
+	info->amr1 = info->amr2 = info->invalid_amr = regs[0];
+	info->expected_iamr = info->invalid_iamr = regs[1];
+	info->expected_uamor = info->invalid_uamor = regs[2];
 
 	/* Wake up child so that it can set itself up. */
 	ret = prod_child(&info->child_sync);
@@ -234,10 +234,10 @@ static int parent(struct shared_info *info, pid_t pid)
 		return ret;
 
 	/* Write invalid AMR value in child. */
-	ret = ptrace_write_regs(pid, NT_PPC_PKEY, &info->amr3, 1);
+	ret = ptrace_write_regs(pid, NT_PPC_PKEY, &info->invalid_amr, 1);
 	PARENT_FAIL_IF(ret, &info->child_sync);
 
-	printf("%-30s AMR: %016lx\n", ptrace_write_running, info->amr3);
+	printf("%-30s AMR: %016lx\n", ptrace_write_running, info->invalid_amr);
 
 	/* Wake up child so that it can verify it didn't change. */
 	ret = prod_child(&info->child_sync);
@@ -249,7 +249,7 @@ static int parent(struct shared_info *info, pid_t pid)
 
 	/* Try to write to IAMR. */
 	regs[0] = info->amr1;
-	regs[1] = info->new_iamr;
+	regs[1] = info->invalid_iamr;
 	ret = ptrace_write_regs(pid, NT_PPC_PKEY, regs, 2);
 	PARENT_FAIL_IF(!ret, &info->child_sync);
 
@@ -257,7 +257,7 @@ static int parent(struct shared_info *info, pid_t pid)
 	       ptrace_write_running, regs[0], regs[1]);
 
 	/* Try to write to IAMR and UAMOR. */
-	regs[2] = info->new_uamor;
+	regs[2] = info->invalid_uamor;
 	ret = ptrace_write_regs(pid, NT_PPC_PKEY, regs, 3);
 	PARENT_FAIL_IF(!ret, &info->child_sync);
 
Signature · 1 line
2.25.1
author maintainer
#135 lore raw

PATCH 135/204 selftests/powerpc: ptrace-pkey: Update the test to mark an invalid pkey correctly

In reply to Greg Kroah-Hartman

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

[ Upstream commit 0eaa3b5ca7b5a76e3783639c828498343be66a01 ]

---
 .../selftests/powerpc/ptrace/ptrace-pkey.c    | 30 ++++++++-----------
 1 file changed, 12 insertions(+), 18 deletions(-)
tools/testing/selftests/powerpc/ptrace/ptrace-pkey.c +12 -18
index f9216c7a1829e..bc33d748d95b4 100644
@@ -66,11 +66,6 @@ static int sys_pkey_alloc(unsigned long flags, unsigned long init_access_rights)
 	return syscall(__NR_pkey_alloc, flags, init_access_rights);
 }
 
-static int sys_pkey_free(int pkey)
-{
-	return syscall(__NR_pkey_free, pkey);
-}
-
 static int child(struct shared_info *info)
 {
 	unsigned long reg;
@@ -100,7 +95,11 @@ static int child(struct shared_info *info)
 
 	info->amr1 |= 3ul << pkeyshift(pkey1);
 	info->amr2 |= 3ul << pkeyshift(pkey2);
-	info->invalid_amr |= info->amr2 | 3ul << pkeyshift(pkey3);
+	/*
+	 * invalid amr value where we try to force write
+	 * things which are deined by a uamor setting.
+	 */
+	info->invalid_amr = info->amr2 | (~0x0UL & ~info->expected_uamor);
 
 	if (disable_execute)
 		info->expected_iamr |= 1ul << pkeyshift(pkey1);
@@ -111,17 +110,12 @@ static int child(struct shared_info *info)
 
 	info->expected_uamor |= 3ul << pkeyshift(pkey1) |
 				3ul << pkeyshift(pkey2);
-	info->invalid_iamr |= 1ul << pkeyshift(pkey1) | 1ul << pkeyshift(pkey2);
-	info->invalid_uamor |= 3ul << pkeyshift(pkey1);
-
 	/*
-	 * We won't use pkey3. We just want a plausible but invalid key to test
-	 * whether ptrace will let us write to AMR bits we are not supposed to.
-	 *
-	 * This also tests whether the kernel restores the UAMOR permissions
-	 * after a key is freed.
+	 * Create an IAMR value different from expected value.
+	 * Kernel will reject an IAMR and UAMOR change.
 	 */
-	sys_pkey_free(pkey3);
+	info->invalid_iamr = info->expected_iamr | (1ul << pkeyshift(pkey1) | 1ul << pkeyshift(pkey2));
+	info->invalid_uamor = info->expected_uamor & ~(0x3ul << pkeyshift(pkey1));
 
 	printf("%-30s AMR: %016lx pkey1: %d pkey2: %d pkey3: %d\n",
 	       user_write, info->amr1, pkey1, pkey2, pkey3);
@@ -196,9 +190,9 @@ static int parent(struct shared_info *info, pid_t pid)
 	PARENT_SKIP_IF_UNSUPPORTED(ret, &info->child_sync);
 	PARENT_FAIL_IF(ret, &info->child_sync);
 
-	info->amr1 = info->amr2 = info->invalid_amr = regs[0];
-	info->expected_iamr = info->invalid_iamr = regs[1];
-	info->expected_uamor = info->invalid_uamor = regs[2];
+	info->amr1 = info->amr2 = regs[0];
+	info->expected_iamr = regs[1];
+	info->expected_uamor = regs[2];
 
 	/* Wake up child so that it can set itself up. */
 	ret = prod_child(&info->child_sync);
Signature · 1 line
2.25.1
author maintainer
#136 lore raw

PATCH 136/204 selftests/powerpc: ptrace-pkey: Dont update expected UAMOR value

In reply to Greg Kroah-Hartman

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

[ Upstream commit 3563b9bea0ca7f53e4218b5e268550341a49f333 ]

With commit 4a4a5e5d2aad ("powerpc/pkeys: key allocation/deallocation must not change pkey registers") we are not updating UAMOR on key allocation. So don't update the expected uamor value in the test.

Fixes 4a4a5e5d2aad ("powerpc/pkeys: key allocation/deallocation must not change pkey registers")
---
 tools/testing/selftests/powerpc/ptrace/ptrace-pkey.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)
tools/testing/selftests/powerpc/ptrace/ptrace-pkey.c +8 -3
index bc33d748d95b4..3694613f418f6 100644
@@ -101,15 +101,20 @@ static int child(struct shared_info *info)
 	 */
 	info->invalid_amr = info->amr2 | (~0x0UL & ~info->expected_uamor);
 
+	/*
+	 * if PKEY_DISABLE_EXECUTE succeeded we should update the expected_iamr
+	 */
 	if (disable_execute)
 		info->expected_iamr |= 1ul << pkeyshift(pkey1);
 	else
 		info->expected_iamr &= ~(1ul << pkeyshift(pkey1));
 
-	info->expected_iamr &= ~(1ul << pkeyshift(pkey2) | 1ul << pkeyshift(pkey3));
+	/*
+	 * We allocated pkey2 and pkey 3 above. Clear the IAMR bits.
+	 */
+	info->expected_iamr &= ~(1ul << pkeyshift(pkey2));
+	info->expected_iamr &= ~(1ul << pkeyshift(pkey3));
 
-	info->expected_uamor |= 3ul << pkeyshift(pkey1) |
-				3ul << pkeyshift(pkey2);
 	/*
 	 * Create an IAMR value different from expected value.
 	 * Kernel will reject an IAMR and UAMOR change.
Signature · 1 line
2.25.1
author maintainer
#137 lore raw

PATCH 137/204 iommu/omap: Check for failure of a call to omap_iommu_dump_ctx

In reply to Greg Kroah-Hartman

From: Colin Ian King <[email protected]>

[ Upstream commit dee9d154f40c58d02f69acdaa5cfd1eae6ebc28b ]

It is possible for the call to omap_iommu_dump_ctx to return a negative error number, so check for the failure and return the error number rather than pass the negative value to simple_read_from_buffer.

Fixes 14e0e6796a0d ("OMAP: iommu: add initial debugfs support")

Addresses-Coverity: ("Improper use of negative value")

---
 drivers/iommu/omap-iommu-debug.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/iommu/omap-iommu-debug.c +3 -0
index 8e19bfa94121e..a99afb5d9011c 100644
@@ -98,8 +98,11 @@ static ssize_t debug_read_regs(struct file *file, char __user *userbuf,
 	mutex_lock(&iommu_debug_lock);
 
 	bytes = omap_iommu_dump_ctx(obj, p, count);
+	if (bytes < 0)
+		goto err;
 	bytes = simple_read_from_buffer(userbuf, count, ppos, buf, bytes);
 
+err:
 	mutex_unlock(&iommu_debug_lock);
 	kfree(buf);
 
Signature · 1 line
2.25.1
author maintainer
#138 lore raw

PATCH 138/204 clk: qcom: gcc: fix sm8150 GPU and NPU clocks

In reply to Greg Kroah-Hartman

From: Jonathan Marek <[email protected]>

[ Upstream commit 667f39b59b494d96ae70f4217637db2ebbee3df0 ]

Fix the parents and set BRANCH_HALT_SKIP. From the downstream driver it should be a 500us delay and not skip, however this matches what was done for other clocks that had 500us delay in downstream.

Fixes f73a4230d5bb ("clk: qcom: gcc: Add GPU and NPU clocks for SM8150")
---
 drivers/clk/qcom/gcc-sm8150.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/clk/qcom/gcc-sm8150.c +6 -2
index 72524cf110487..55e9d6d75a0cd 100644
@@ -1617,6 +1617,7 @@ static struct clk_branch gcc_gpu_cfg_ahb_clk = {
 };
 
 static struct clk_branch gcc_gpu_gpll0_clk_src = {
+	.halt_check = BRANCH_HALT_SKIP,
 	.clkr = {
 		.enable_reg = 0x52004,
 		.enable_mask = BIT(15),
@@ -1632,13 +1633,14 @@ static struct clk_branch gcc_gpu_gpll0_clk_src = {
 };
 
 static struct clk_branch gcc_gpu_gpll0_div_clk_src = {
+	.halt_check = BRANCH_HALT_SKIP,
 	.clkr = {
 		.enable_reg = 0x52004,
 		.enable_mask = BIT(16),
 		.hw.init = &(struct clk_init_data){
 			.name = "gcc_gpu_gpll0_div_clk_src",
 			.parent_hws = (const struct clk_hw *[]){
-				&gcc_gpu_gpll0_clk_src.clkr.hw },
+				&gpll0_out_even.clkr.hw },
 			.num_parents = 1,
 			.flags = CLK_SET_RATE_PARENT,
 			.ops = &clk_branch2_ops,
@@ -1729,6 +1731,7 @@ static struct clk_branch gcc_npu_cfg_ahb_clk = {
 };
 
 static struct clk_branch gcc_npu_gpll0_clk_src = {
+	.halt_check = BRANCH_HALT_SKIP,
 	.clkr = {
 		.enable_reg = 0x52004,
 		.enable_mask = BIT(18),
@@ -1744,13 +1747,14 @@ static struct clk_branch gcc_npu_gpll0_clk_src = {
 };
 
 static struct clk_branch gcc_npu_gpll0_div_clk_src = {
+	.halt_check = BRANCH_HALT_SKIP,
 	.clkr = {
 		.enable_reg = 0x52004,
 		.enable_mask = BIT(19),
 		.hw.init = &(struct clk_init_data){
 			.name = "gcc_npu_gpll0_div_clk_src",
 			.parent_hws = (const struct clk_hw *[]){
-				&gcc_npu_gpll0_clk_src.clkr.hw },
+				&gpll0_out_even.clkr.hw },
 			.num_parents = 1,
 			.flags = CLK_SET_RATE_PARENT,
 			.ops = &clk_branch2_ops,
Signature · 1 line
2.25.1
author maintainer
#139 lore raw

PATCH 139/204 clk: qcom: clk-alpha-pll: remove unused/incorrect PLL_CAL_VAL

In reply to Greg Kroah-Hartman

From: Jonathan Marek <[email protected]>

[ Upstream commit c8b9002f44e4a1d2771b2f59f6de900864b1f9d7 ]

0x44 isn't a register offset, it is the value that goes into CAL_L_VAL.

Fixes 548a909597d5 ("clk: qcom: clk-alpha-pll: Add support for Trion PLLs")
---
 drivers/clk/qcom/clk-alpha-pll.c | 2 --
 1 file changed, 2 deletions(-)
drivers/clk/qcom/clk-alpha-pll.c +0 -2
index 9b2dfa08acb2a..1325139173c95 100644
@@ -56,7 +56,6 @@
 #define PLL_STATUS(p)		((p)->offset + (p)->regs[PLL_OFF_STATUS])
 #define PLL_OPMODE(p)		((p)->offset + (p)->regs[PLL_OFF_OPMODE])
 #define PLL_FRAC(p)		((p)->offset + (p)->regs[PLL_OFF_FRAC])
-#define PLL_CAL_VAL(p)		((p)->offset + (p)->regs[PLL_OFF_CAL_VAL])
 
 const u8 clk_alpha_pll_regs[][PLL_OFF_MAX_REGS] = {
 	[CLK_ALPHA_PLL_TYPE_DEFAULT] =  {
@@ -115,7 +114,6 @@ const u8 clk_alpha_pll_regs[][PLL_OFF_MAX_REGS] = {
 		[PLL_OFF_STATUS] = 0x30,
 		[PLL_OFF_OPMODE] = 0x38,
 		[PLL_OFF_ALPHA_VAL] = 0x40,
-		[PLL_OFF_CAL_VAL] = 0x44,
 	},
 	[CLK_ALPHA_PLL_TYPE_LUCID] =  {
 		[PLL_OFF_L_VAL] = 0x04,
Signature · 1 line
2.25.1
author maintainer
#140 lore raw

PATCH 140/204 iommu/vt-d: Enforce PASID devTLB field mask

In reply to Greg Kroah-Hartman

From: Liu Yi L <[email protected]>

[ Upstream commit 5f77d6ca5ca74e4b4a5e2e010f7ff50c45dea326 ]

Set proper masks to avoid invalid input spillover to reserved bits.

---
 include/linux/intel-iommu.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
include/linux/intel-iommu.h +2 -2
index 64a5335046b00..bc1abbc041092 100644
@@ -363,8 +363,8 @@ enum {
 
 #define QI_DEV_EIOTLB_ADDR(a)	((u64)(a) & VTD_PAGE_MASK)
 #define QI_DEV_EIOTLB_SIZE	(((u64)1) << 11)
-#define QI_DEV_EIOTLB_GLOB(g)	((u64)g)
-#define QI_DEV_EIOTLB_PASID(p)	(((u64)p) << 32)
+#define QI_DEV_EIOTLB_GLOB(g)	((u64)(g) & 0x1)
+#define QI_DEV_EIOTLB_PASID(p)	((u64)((p) & 0xfffff) << 32)
 #define QI_DEV_EIOTLB_SID(sid)	((u64)((sid) & 0xffff) << 16)
 #define QI_DEV_EIOTLB_QDEP(qd)	((u64)((qd) & 0x1f) << 4)
 #define QI_DEV_EIOTLB_PFSID(pfsid) (((u64)(pfsid & 0xf) << 12) | \
Signature · 1 line
2.25.1
author maintainer
#141 lore raw

PATCH 141/204 i2c: rcar: slave: only send STOP event when we have been addressed

In reply to Greg Kroah-Hartman

From: Wolfram Sang <[email protected]>

[ Upstream commit 314139f9f0abdba61ed9a8463bbcb0bf900ac5a2 ]

When the SSR interrupt is activated, it will detect every STOP condition on the bus, not only the ones after we have been addressed. So, enable this interrupt only after we have been addressed, and disable it otherwise.

Fixes de20d1857dd6 ("i2c: rcar: add slave support")
---
 drivers/i2c/busses/i2c-rcar.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/i2c/busses/i2c-rcar.c +4 -3
index 50dd98803ca0c..76bcdb27070e1 100644
@@ -583,13 +583,14 @@ static bool rcar_i2c_slave_irq(struct rcar_i2c_priv *priv)
 			rcar_i2c_write(priv, ICSIER, SDR | SSR | SAR);
 		}
 
-		rcar_i2c_write(priv, ICSSR, ~SAR & 0xff);
+		/* Clear SSR, too, because of old STOPs to other clients than us */
+		rcar_i2c_write(priv, ICSSR, ~(SAR | SSR) & 0xff);
 	}
 
 	/* master sent stop */
 	if (ssr_filtered & SSR) {
 		i2c_slave_event(priv->slave, I2C_SLAVE_STOP, &value);
-		rcar_i2c_write(priv, ICSIER, SAR | SSR);
+		rcar_i2c_write(priv, ICSIER, SAR);
 		rcar_i2c_write(priv, ICSSR, ~SSR & 0xff);
 	}
 
@@ -853,7 +854,7 @@ static int rcar_reg_slave(struct i2c_client *slave)
 	priv->slave = slave;
 	rcar_i2c_write(priv, ICSAR, slave->addr);
 	rcar_i2c_write(priv, ICSSR, 0);
-	rcar_i2c_write(priv, ICSIER, SAR | SSR);
+	rcar_i2c_write(priv, ICSIER, SAR);
 	rcar_i2c_write(priv, ICSCR, SIE | SDBS);
 
 	return 0;
Signature · 1 line
2.25.1
author maintainer
#142 lore raw

PATCH 142/204 f2fs: compress: fix to update isize when overwriting compressed file

In reply to Greg Kroah-Hartman

From: Chao Yu <[email protected]>

[ Upstream commit 944dd22ea4475bd11180fd2f431a4a547ca4d8f5 ]

We missed to update isize of compressed file in write_end() with below case:

cluster size is 16KB

- write 14KB data from offset 0
- overwrite 16KB data from offset 0

Fixes 4c8ff7095bef ("f2fs: support data compression")
---
 fs/f2fs/data.c | 4 ++++
 1 file changed, 4 insertions(+)
fs/f2fs/data.c +4 -0
index 10491ae1cb850..329afa55a581c 100644
@@ -3353,6 +3353,10 @@ static int f2fs_write_end(struct file *file,
 	if (f2fs_compressed_file(inode) && fsdata) {
 		f2fs_compress_write_end(inode, fsdata, page->index, copied);
 		f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
+
+		if (pos + copied > i_size_read(inode) &&
+				!f2fs_verity_in_progress(inode))
+			f2fs_i_size_write(inode, pos + copied);
 		return copied;
 	}
 #endif
Signature · 1 line
2.25.1
author maintainer
#143 lore raw

PATCH 143/204 clk: qcom: gcc-sdm660: Fix up gcc_mss_mnoc_bimc_axi_clk

In reply to Greg Kroah-Hartman

From: Konrad Dybcio <[email protected]>

[ Upstream commit 3386af51d3bcebcba3f7becdb1ef2e384abe90cf ]

Add missing halt_check, hwcg_reg and hwcg_bit properties. These were likely omitted when porting the driver upstream.

Fixes f2a76a2955c0 ("clk: qcom: Add Global Clock controller (GCC) driver for SDM660")
---
 drivers/clk/qcom/gcc-sdm660.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/clk/qcom/gcc-sdm660.c +3 -0
index bf5730832ef3d..c6fb57cd576f5 100644
@@ -1715,6 +1715,9 @@ static struct clk_branch gcc_mss_cfg_ahb_clk = {
 
 static struct clk_branch gcc_mss_mnoc_bimc_axi_clk = {
 	.halt_reg = 0x8a004,
+	.halt_check = BRANCH_HALT,
+	.hwcg_reg = 0x8a004,
+	.hwcg_bit = 1,
 	.clkr = {
 		.enable_reg = 0x8a004,
 		.enable_mask = BIT(0),
Signature · 1 line
2.25.1
author maintainer
#144 lore raw

PATCH 144/204 clk: clk-atlas6: fix return value check in atlas6_clk_init()

In reply to Greg Kroah-Hartman

From: Xu Wang <[email protected]>

[ Upstream commit 12b90b40854a8461a02ef19f6f4474cc88d64b66 ]

In case of error, the function clk_register() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR().

Fixes 7bf21bc81f28 ("clk: sirf: re-arch to make the codes support both prima2 and atlas6")
---
 drivers/clk/sirf/clk-atlas6.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clk/sirf/clk-atlas6.c +1 -1
index c84d5bab7ac28..b95483bb6a5ec 100644
@@ -135,7 +135,7 @@ static void __init atlas6_clk_init(struct device_node *np)
 
 	for (i = pll1; i < maxclk; i++) {
 		atlas6_clks[i] = clk_register(NULL, atlas6_clk_hw_array[i]);
-		BUG_ON(!atlas6_clks[i]);
+		BUG_ON(IS_ERR(atlas6_clks[i]));
 	}
 	clk_register_clkdev(atlas6_clks[cpu], NULL, "cpu");
 	clk_register_clkdev(atlas6_clks[io],  NULL, "io");
Signature · 1 line
2.25.1
author maintainer
#145 lore raw

PATCH 145/204 nvme: fix deadlock in disconnect during scan_work and/or ana_work

In reply to Greg Kroah-Hartman

From: Sagi Grimberg <[email protected]>

[ Upstream commit ecca390e80561debbfdb4dc96bf94595136889fa ]

A deadlock happens in the following scenario with multipath:
1) scan_work(nvme0) detects a new nsid while nvme0 is an optimized path to it, path nvme1 happens to be inaccessible.

2) Before scan_work is complete nvme0 disconnect is initiated nvme_delete_ctrl_sync() sets nvme0 state to NVME_CTRL_DELETING

3) scan_work(1) attempts to submit IO,
but nvme_path_is_optimized() observes nvme0 is not LIVE. Since nvme1 is a possible path IO is requeued and scan_work hangs.

--

Workqueue: nvme-wq nvme_scan_work [nvme_core]
kernel: Call Trace:
kernel:  __schedule+0x2b9/0x6c0
kernel:  schedule+0x42/0xb0
kernel:  io_schedule+0x16/0x40
kernel:  do_read_cache_page+0x438/0x830
kernel:  read_cache_page+0x12/0x20
kernel:  read_dev_sector+0x27/0xc0
kernel:  read_lba+0xc1/0x220
kernel:  efi_partition+0x1e6/0x708
kernel:  check_partition+0x154/0x244
kernel:  rescan_partitions+0xae/0x280
kernel:  __blkdev_get+0x40f/0x560
kernel:  blkdev_get+0x3d/0x140
kernel:  __device_add_disk+0x388/0x480
kernel:  device_add_disk+0x13/0x20
kernel:  nvme_mpath_set_live+0x119/0x140 [nvme_core]
kernel:  nvme_update_ns_ana_state+0x5c/0x60 [nvme_core]
kernel:  nvme_set_ns_ana_state+0x1e/0x30 [nvme_core]
kernel:  nvme_parse_ana_log+0xa1/0x180 [nvme_core]
kernel:  nvme_mpath_add_disk+0x47/0x90 [nvme_core]
kernel:  nvme_validate_ns+0x396/0x940 [nvme_core]
kernel:  nvme_scan_work+0x24f/0x380 [nvme_core]
kernel:  process_one_work+0x1db/0x380
kernel:  worker_thread+0x249/0x400
kernel:  kthread+0x104/0x140

--

4) Delete also hangs in flush_work(ctrl->scan_work) from nvme_remove_namespaces().

Similiarly a deadlock with ana_work may happen: if ana_work has started and calls nvme_mpath_set_live and device_add_disk, it will trigger I/O. When we trigger disconnect I/O will block because our accessible (optimized) path is disconnecting, but the alternate path is inaccessible, so I/O blocks. Then disconnect tries to flush the ana_work and hangs.

[  605.550896] Workqueue: nvme-wq nvme_ana_work [nvme_core]
[  605.552087] Call Trace:
[  605.552683]  __schedule+0x2b9/0x6c0
[  605.553507]  schedule+0x42/0xb0
[  605.554201]  io_schedule+0x16/0x40
[  605.555012]  do_read_cache_page+0x438/0x830
[  605.556925]  read_cache_page+0x12/0x20
[  605.557757]  read_dev_sector+0x27/0xc0
[  605.558587]  amiga_partition+0x4d/0x4c5
[  605.561278]  check_partition+0x154/0x244
[  605.562138]  rescan_partitions+0xae/0x280
[  605.563076]  __blkdev_get+0x40f/0x560
[  605.563830]  blkdev_get+0x3d/0x140
[  605.564500]  __device_add_disk+0x388/0x480
[  605.565316]  device_add_disk+0x13/0x20
[  605.566070]  nvme_mpath_set_live+0x5e/0x130 [nvme_core]
[  605.567114]  nvme_update_ns_ana_state+0x2c/0x30 [nvme_core]
[  605.568197]  nvme_update_ana_state+0xca/0xe0 [nvme_core]
[  605.569360]  nvme_parse_ana_log+0xa1/0x180 [nvme_core]
[  605.571385]  nvme_read_ana_log+0x76/0x100 [nvme_core]
[  605.572376]  nvme_ana_work+0x15/0x20 [nvme_core]
[  605.573330]  process_one_work+0x1db/0x380
[  605.574144]  worker_thread+0x4d/0x400
[  605.574896]  kthread+0x104/0x140
[  605.577205]  ret_from_fork+0x35/0x40
[  605.577955] INFO: task nvme:14044 blocked for more than 120 seconds.
[  605.579239]       Tainted: G           OE     5.3.5-050305-generic #201910071830
[  605.580712] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[  605.582320] nvme            D    0 14044  14043 0x00000000
[  605.583424] Call Trace:
[  605.583935]  __schedule+0x2b9/0x6c0
[  605.584625]  schedule+0x42/0xb0
[  605.585290]  schedule_timeout+0x203/0x2f0
[  605.588493]  wait_for_completion+0xb1/0x120
[  605.590066]  __flush_work+0x123/0x1d0
[  605.591758]  __cancel_work_timer+0x10e/0x190
[  605.593542]  cancel_work_sync+0x10/0x20
[  605.594347]  nvme_mpath_stop+0x2f/0x40 [nvme_core]
[  605.595328]  nvme_stop_ctrl+0x12/0x50 [nvme_core]
[  605.596262]  nvme_do_delete_ctrl+0x3f/0x90 [nvme_core]
[  605.597333]  nvme_sysfs_delete+0x5c/0x70 [nvme_core]
[  605.598320]  dev_attr_store+0x17/0x30

Fix this by introducing a new state: NVME_CTRL_DELETE_NOIO, which will indicate the phase of controller deletion where I/O cannot be allowed to access the namespace. NVME_CTRL_DELETING still allows mpath I/O to be issued to the bottom device, and only after we flush the ana_work and scan_work (after nvme_stop_ctrl and nvme_prep_remove_namespaces) we change the state to NVME_CTRL_DELETING_NOIO. Also we prevent ana_work from re-firing by aborting early if we are not LIVE, so we should be safe here.

In addition, change the transport drivers to follow the updated state machine.

Fixes 0d0b660f214d ("nvme: add ANA support")
---
 drivers/nvme/host/core.c      | 15 +++++++++++++++
 drivers/nvme/host/fabrics.c   |  2 +-
 drivers/nvme/host/fabrics.h   |  3 ++-
 drivers/nvme/host/fc.c        |  1 +
 drivers/nvme/host/multipath.c | 18 +++++++++++++++---
 drivers/nvme/host/nvme.h      |  1 +
 drivers/nvme/host/rdma.c      | 10 ++++++----
 drivers/nvme/host/tcp.c       | 15 +++++++++------
 8 files changed, 50 insertions(+), 15 deletions(-)
drivers/nvme/host/core.c +15 -0
index f7540a9e54fd2..ee67113d96b1b 100644
@@ -368,6 +368,16 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl,
 			break;
 		}
 		break;
+	case NVME_CTRL_DELETING_NOIO:
+		switch (old_state) {
+		case NVME_CTRL_DELETING:
+		case NVME_CTRL_DEAD:
+			changed = true;
+			/* FALLTHRU */
+		default:
+			break;
+		}
+		break;
 	case NVME_CTRL_DEAD:
 		switch (old_state) {
 		case NVME_CTRL_DELETING:
@@ -405,6 +415,7 @@ static bool nvme_state_terminal(struct nvme_ctrl *ctrl)
 	case NVME_CTRL_CONNECTING:
 		return false;
 	case NVME_CTRL_DELETING:
+	case NVME_CTRL_DELETING_NOIO:
 	case NVME_CTRL_DEAD:
 		return true;
 	default:
@@ -3280,6 +3291,7 @@ static ssize_t nvme_sysfs_show_state(struct device *dev,
 		[NVME_CTRL_RESETTING]	= "resetting",
 		[NVME_CTRL_CONNECTING]	= "connecting",
 		[NVME_CTRL_DELETING]	= "deleting",
+		[NVME_CTRL_DELETING_NOIO]= "deleting (no IO)",
 		[NVME_CTRL_DEAD]	= "dead",
 	};
 
@@ -3860,6 +3872,9 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl)
 	if (ctrl->state == NVME_CTRL_DEAD)
 		nvme_kill_queues(ctrl);
 
+	/* this is a no-op when called from the controller reset handler */
+	nvme_change_ctrl_state(ctrl, NVME_CTRL_DELETING_NOIO);
+
 	down_write(&ctrl->namespaces_rwsem);
 	list_splice_init(&ctrl->namespaces, &ns_list);
 	up_write(&ctrl->namespaces_rwsem);
drivers/nvme/host/fabrics.c +1 -1
index 2a6c8190eeb76..4ec4829d62334 100644
@@ -547,7 +547,7 @@ static struct nvmf_transport_ops *nvmf_lookup_transport(
 blk_status_t nvmf_fail_nonready_command(struct nvme_ctrl *ctrl,
 		struct request *rq)
 {
-	if (ctrl->state != NVME_CTRL_DELETING &&
+	if (ctrl->state != NVME_CTRL_DELETING_NOIO &&
 	    ctrl->state != NVME_CTRL_DEAD &&
 	    !blk_noretry_request(rq) && !(rq->cmd_flags & REQ_NVME_MPATH))
 		return BLK_STS_RESOURCE;
drivers/nvme/host/fabrics.h +2 -1
index a0ec40ab62eeb..a9c1e3b4585ec 100644
@@ -182,7 +182,8 @@ bool nvmf_ip_options_match(struct nvme_ctrl *ctrl,
 static inline bool nvmf_check_ready(struct nvme_ctrl *ctrl, struct request *rq,
 		bool queue_live)
 {
-	if (likely(ctrl->state == NVME_CTRL_LIVE))
+	if (likely(ctrl->state == NVME_CTRL_LIVE ||
+		   ctrl->state == NVME_CTRL_DELETING))
 		return true;
 	return __nvmf_check_ready(ctrl, rq, queue_live);
 }
drivers/nvme/host/fc.c +1 -0
index 564e3f220ac79..a70220df1f570 100644
@@ -800,6 +800,7 @@ nvme_fc_ctrl_connectivity_loss(struct nvme_fc_ctrl *ctrl)
 		break;
 
 	case NVME_CTRL_DELETING:
+	case NVME_CTRL_DELETING_NOIO:
 	default:
 		/* no action to take - let it delete */
 		break;
drivers/nvme/host/multipath.c +15 -3
index d3914b7e8f52c..8f235fbfe44ee 100644
@@ -167,9 +167,18 @@ void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl)
 
 static bool nvme_path_is_disabled(struct nvme_ns *ns)
 {
-	return ns->ctrl->state != NVME_CTRL_LIVE ||
-		test_bit(NVME_NS_ANA_PENDING, &ns->flags) ||
-		test_bit(NVME_NS_REMOVING, &ns->flags);
+	/*
+	 * We don't treat NVME_CTRL_DELETING as a disabled path as I/O should
+	 * still be able to complete assuming that the controller is connected.
+	 * Otherwise it will fail immediately and return to the requeue list.
+	 */
+	if (ns->ctrl->state != NVME_CTRL_LIVE &&
+	    ns->ctrl->state != NVME_CTRL_DELETING)
+		return true;
+	if (test_bit(NVME_NS_ANA_PENDING, &ns->flags) ||
+	    test_bit(NVME_NS_REMOVING, &ns->flags))
+		return true;
+	return false;
 }
 
 static struct nvme_ns *__nvme_find_path(struct nvme_ns_head *head, int node)
@@ -575,6 +584,9 @@ static void nvme_ana_work(struct work_struct *work)
 {
 	struct nvme_ctrl *ctrl = container_of(work, struct nvme_ctrl, ana_work);
 
+	if (ctrl->state != NVME_CTRL_LIVE)
+		return;
+
 	nvme_read_ana_log(ctrl);
 }
 
drivers/nvme/host/nvme.h +1 -0
index 8f1b0a30fd2a6..ff0b4079e8d6d 100644
@@ -183,6 +183,7 @@ enum nvme_ctrl_state {
 	NVME_CTRL_RESETTING,
 	NVME_CTRL_CONNECTING,
 	NVME_CTRL_DELETING,
+	NVME_CTRL_DELETING_NOIO,
 	NVME_CTRL_DEAD,
 };
 
drivers/nvme/host/rdma.c +6 -4
index 19c94080512cf..fdab0054cd809 100644
@@ -1023,11 +1023,12 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new)
 	changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE);
 	if (!changed) {
 		/*
-		 * state change failure is ok if we're in DELETING state,
+		 * state change failure is ok if we started ctrl delete,
 		 * unless we're during creation of a new controller to
 		 * avoid races with teardown flow.
 		 */
-		WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING);
+		WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING &&
+			     ctrl->ctrl.state != NVME_CTRL_DELETING_NOIO);
 		WARN_ON_ONCE(new);
 		ret = -EINVAL;
 		goto destroy_io;
@@ -1080,8 +1081,9 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
 
 	if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) {
-		/* state change failure is ok if we're in DELETING state */
-		WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING);
+		/* state change failure is ok if we started ctrl delete */
+		WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING &&
+			     ctrl->ctrl.state != NVME_CTRL_DELETING_NOIO);
 		return;
 	}
 
drivers/nvme/host/tcp.c +9 -6
index 99eaa0474e10b..06d6c1c6de35b 100644
@@ -1938,11 +1938,12 @@ static int nvme_tcp_setup_ctrl(struct nvme_ctrl *ctrl, bool new)
 
 	if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE)) {
 		/*
-		 * state change failure is ok if we're in DELETING state,
+		 * state change failure is ok if we started ctrl delete,
 		 * unless we're during creation of a new controller to
 		 * avoid races with teardown flow.
 		 */
-		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING);
+		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING &&
+			     ctrl->state != NVME_CTRL_DELETING_NOIO);
 		WARN_ON_ONCE(new);
 		ret = -EINVAL;
 		goto destroy_io;
@@ -1998,8 +1999,9 @@ static void nvme_tcp_error_recovery_work(struct work_struct *work)
 	blk_mq_unquiesce_queue(ctrl->admin_q);
 
 	if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING)) {
-		/* state change failure is ok if we're in DELETING state */
-		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING);
+		/* state change failure is ok if we started ctrl delete */
+		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING &&
+			     ctrl->state != NVME_CTRL_DELETING_NOIO);
 		return;
 	}
 
@@ -2034,8 +2036,9 @@ static void nvme_reset_ctrl_work(struct work_struct *work)
 	nvme_tcp_teardown_ctrl(ctrl, false);
 
 	if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING)) {
-		/* state change failure is ok if we're in DELETING state */
-		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING);
+		/* state change failure is ok if we started ctrl delete */
+		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING &&
+			     ctrl->state != NVME_CTRL_DELETING_NOIO);
 		return;
 	}
 
Signature · 1 line
2.25.1
author maintainer
#146 lore raw

PATCH 146/204 sched/uclamp: Fix a deadlock when enabling uclamp static key

In reply to Greg Kroah-Hartman

From: Qais Yousef <[email protected]>

[ Upstream commit e65855a52b479f98674998cb23b21ef5a8144b04 ]

The following splat was caught when setting uclamp value of a task:

  BUG: sleeping function called from invalid context at ./include/linux/percpu-rwsem.h:49

   cpus_read_lock+0x68/0x130
   static_key_enable+0x1c/0x38
   __sched_setscheduler+0x900/0xad8

Fix by ensuring we enable the key outside of the critical section in __sched_setscheduler()

Fixes 46609ce22703 ("sched/uclamp: Protect uclamp fast path code with static key")
---
 kernel/sched/core.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
kernel/sched/core.c +9 -2
index 49b5c5cf9271d..ebecf1cc3b788 100644
@@ -1248,6 +1248,15 @@ static int uclamp_validate(struct task_struct *p,
 	if (upper_bound > SCHED_CAPACITY_SCALE)
 		return -EINVAL;
 
+	/*
+	 * We have valid uclamp attributes; make sure uclamp is enabled.
+	 *
+	 * We need to do that here, because enabling static branches is a
+	 * blocking operation which obviously cannot be done while holding
+	 * scheduler locks.
+	 */
+	static_branch_enable(&sched_uclamp_used);
+
 	return 0;
 }
 
@@ -1278,8 +1287,6 @@ static void __setscheduler_uclamp(struct task_struct *p,
 	if (likely(!(attr->sched_flags & SCHED_FLAG_UTIL_CLAMP)))
 		return;
 
-	static_branch_enable(&sched_uclamp_used);
-
 	if (attr->sched_flags & SCHED_FLAG_UTIL_CLAMP_MIN) {
 		uclamp_se_set(&p->uclamp_req[UCLAMP_MIN],
 			      attr->sched_util_min, true);
Signature · 1 line
2.25.1
author maintainer
#147 lore raw

PATCH 147/204 pwm: bcm-iproc: handle clk_get_rate() return

In reply to Greg Kroah-Hartman

From: Rayagonda Kokatanur <[email protected]>

[ Upstream commit 6ced5ff0be8e94871ba846dfbddf69d21363f3d7 ]

Handle clk_get_rate() returning 0 to avoid possible division by zero.

Fixes daa5abc41c80 ("pwm: Add support for Broadcom iProc PWM controller")
---
 drivers/pwm/pwm-bcm-iproc.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/pwm/pwm-bcm-iproc.c +7 -2
index 1f829edd8ee70..d392a828fc493 100644
@@ -85,8 +85,6 @@ static void iproc_pwmc_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
 	u64 tmp, multi, rate;
 	u32 value, prescale;
 
-	rate = clk_get_rate(ip->clk);
-
 	value = readl(ip->base + IPROC_PWM_CTRL_OFFSET);
 
 	if (value & BIT(IPROC_PWM_CTRL_EN_SHIFT(pwm->hwpwm)))
@@ -99,6 +97,13 @@ static void iproc_pwmc_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
 	else
 		state->polarity = PWM_POLARITY_INVERSED;
 
+	rate = clk_get_rate(ip->clk);
+	if (rate == 0) {
+		state->period = 0;
+		state->duty_cycle = 0;
+		return;
+	}
+
 	value = readl(ip->base + IPROC_PWM_PRESCALE_OFFSET);
 	prescale = value >> IPROC_PWM_PRESCALE_SHIFT(pwm->hwpwm);
 	prescale &= IPROC_PWM_PRESCALE_MAX;
Signature · 1 line
2.25.1
author maintainer
#148 lore raw

PATCH 148/204 tools build feature: Use CC and CXX from parent

In reply to Greg Kroah-Hartman

From: Thomas Hebb <[email protected]>

[ Upstream commit e3232c2f39acafd5a29128425bc30b9884642cfa ]

commit c8c188679ccf ("tools build: Use the same CC for feature detection and actual build") changed these assignments from unconditional (:=) to conditional (?=) so that they wouldn't clobber values from the environment. However, conditional assignment does not work properly for variables that Make implicitly sets, among which are CC and CXX. To quote tools/scripts/Makefile.include, which handles this properly:

# Makefiles suck: This macro sets a default value of $(2) for the

  # variable named by $(1), unless the variable has been set by
  # environment or command line. This is necessary for CC and AR
  # because make sets default values, so the simpler ?= approach
  # won't work as expected.

In other words, the conditional assignments will not run even if the variables are not overridden in the environment; Make will set CC to "cc" and CXX to "g++" when it starts[1], meaning the variables are not empty by the time the conditional assignments are evaluated. This breaks cross-compilation when CROSS_COMPILE is set but CC isn't, since "cc" gets used for feature detection instead of the cross compiler (and likewise for CXX).

To fix the issue, just pass down the values of CC and CXX computed by the parent Makefile, which gets included by the Makefile that actually builds whatever we're detecting features for and so is guaranteed to have good values. This is a better solution anyway, since it means we aren't trying to replicate the logic of the parent build system and so don't risk it getting out of sync.

Leave PKG_CONFIG alone, since 1) there's no common logic to compute it in Makefile.include, and 2) it's not an implicit variable, so conditional assignment works properly.

[1] https://www.gnu.org/software/make/manual/html_node/Implicit-Variables.html

Fixes c8c188679ccf ("tools build: Use the same CC for feature detection and actual build")
Cc David Carrillo-Cisneros <[email protected]>
Cc Ian Rogers <[email protected]>
Cc Igor Lubashev <[email protected]>
Cc Namhyung Kim <[email protected]>
Cc Quentin Monnet <[email protected]>
Cc Song Liu <[email protected]>
Cc Stephane Eranian <[email protected]>
Cc thomas hebb <[email protected]>
---
 tools/build/Makefile.feature | 2 +-
 tools/build/feature/Makefile | 2 --
 2 files changed, 1 insertion(+), 3 deletions(-)
tools/build/Makefile.feature +1 -1
index 3e0c019ef2971..941cddd5c00cd 100644
@@ -8,7 +8,7 @@ endif
 
 feature_check = $(eval $(feature_check_code))
 define feature_check_code
-  feature-$(1) := $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) CFLAGS="$(EXTRA_CFLAGS) $(FEATURE_CHECK_CFLAGS-$(1))" CXXFLAGS="$(EXTRA_CXXFLAGS) $(FEATURE_CHECK_CXXFLAGS-$(1))" LDFLAGS="$(LDFLAGS) $(FEATURE_CHECK_LDFLAGS-$(1))" -C $(feature_dir) $(OUTPUT_FEATURES)test-$1.bin >/dev/null 2>/dev/null && echo 1 || echo 0)
+  feature-$(1) := $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) CC=$(CC) CXX=$(CXX) CFLAGS="$(EXTRA_CFLAGS) $(FEATURE_CHECK_CFLAGS-$(1))" CXXFLAGS="$(EXTRA_CXXFLAGS) $(FEATURE_CHECK_CXXFLAGS-$(1))" LDFLAGS="$(LDFLAGS) $(FEATURE_CHECK_LDFLAGS-$(1))" -C $(feature_dir) $(OUTPUT_FEATURES)test-$1.bin >/dev/null 2>/dev/null && echo 1 || echo 0)
 endef
 
 feature_set = $(eval $(feature_set_code))
tools/build/feature/Makefile +0 -2
index 92012381393ad..ef4ca6e408427 100644
@@ -73,8 +73,6 @@ FILES=                                          \
 
 FILES := $(addprefix $(OUTPUT),$(FILES))
 
-CC ?= $(CROSS_COMPILE)gcc
-CXX ?= $(CROSS_COMPILE)g++
 PKG_CONFIG ?= $(CROSS_COMPILE)pkg-config
 LLVM_CONFIG ?= llvm-config
 CLANG ?= clang
Signature · 1 line
2.25.1
author maintainer
#149 lore raw

PATCH 149/204 i2c: rcar: avoid race when unregistering slave

In reply to Greg Kroah-Hartman

From: Wolfram Sang <[email protected]>

[ Upstream commit c7c9e914f9a0478fba4dc6f227cfd69cf84a4063 ]

Due to the lockless design of the driver, it is theoretically possible to access a NULL pointer, if a slave interrupt was running while we were unregistering the slave. To make this rock solid, disable the interrupt for a short time while we are clearing the interrupt_enable register. This patch is purely based on code inspection. The OOPS is super-hard to trigger because clearing SAR (the address) makes interrupts even more unlikely to happen as well. While here, reinit SCR to SDBS because this bit should always be set according to documentation. There is no effect, though, because the interface is disabled.

Fixes 7b814d852af6 ("i2c: rcar: avoid race when unregistering slave client")
---
 drivers/i2c/busses/i2c-rcar.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/i2c/busses/i2c-rcar.c +5 -3
index 76bcdb27070e1..5615e7c43b436 100644
@@ -866,12 +866,14 @@ static int rcar_unreg_slave(struct i2c_client *slave)
 
 	WARN_ON(!priv->slave);
 
-	/* disable irqs and ensure none is running before clearing ptr */
+	/* ensure no irq is running before clearing ptr */
+	disable_irq(priv->irq);
 	rcar_i2c_write(priv, ICSIER, 0);
-	rcar_i2c_write(priv, ICSCR, 0);
+	rcar_i2c_write(priv, ICSSR, 0);
+	enable_irq(priv->irq);
+	rcar_i2c_write(priv, ICSCR, SDBS);
 	rcar_i2c_write(priv, ICSAR, 0); /* Gen2: must be 0 if not using slave */
 
-	synchronize_irq(priv->irq);
 	priv->slave = NULL;
 
 	pm_runtime_put(rcar_i2c_priv_to_dev(priv));
Signature · 1 line
2.25.1