6.8.7-rc1 review
Subsystems (4)
Greg Kroah-Hartman
maintainercover 6.8.7-rc1 review
This is the start of the stable review cycle for the 6.8.7 release. There are 172 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Wed, 17 Apr 2024 14:19:30 +0000. Anything received after that time might be too late.
The whole patch series can be found in one patch at:
https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.8.7-rc1.gz
or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.8.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 6.8.7-rc1
Fudongwang <[email protected]>
drm/amd/display: fix disable otg wa logic in DCN316
Wenjing Liu <[email protected]>
drm/amd/display: always reset ODM mode in context when adding first plane
Alex Hung <[email protected]>
drm/amd/display: Return max resolution supported by DWB
Dillon Varone <[email protected]>
drm/amd/display: Do not recursively call manual trigger programming
Harry Wentland <[email protected]>
drm/amd/display: Set VSC SDP Colorimetry same way for MST and SST
Harry Wentland <[email protected]>
drm/amd/display: Program VSC SDP colorimetry for all DP sinks >= 1.4
Yifan Zhang <[email protected]>
drm/amdgpu: differentiate external rev id for gfx 11.5.0
Tim Huang <[email protected]>
drm/amdgpu: fix incorrect number of active RBs for gfx11
Alex Deucher <[email protected]>
drm/amdgpu: always force full reset for SOC21
Lijo Lazar <[email protected]>
drm/amdgpu: Reset dGPU if suspend got aborted
Ville Syrjälä <[email protected]>
drm/i915: Disable live M/N updates when using bigjoiner
Ville Syrjälä <[email protected]>
drm/i915: Disable port sync when bigjoiner is used
Ville Syrjälä <[email protected]>
drm/i915/psr: Disable PSR when bigjoiner is used
Ville Syrjälä <[email protected]>
drm/i915/cdclk: Fix CDCLK programming order when pipes are active
Josh Poimboeuf <[email protected]>
x86/bugs: Replace CONFIG_SPECTRE_BHI_{ON,OFF} with CONFIG_MITIGATION_SPECTRE_BHI
Josh Poimboeuf <[email protected]>
x86/bugs: Remove CONFIG_BHI_MITIGATION_AUTO and spectre_bhi=auto
Josh Poimboeuf <[email protected]>
x86/bugs: Clarify that syscall hardening isn't a BHI mitigation
Josh Poimboeuf <[email protected]>
x86/bugs: Fix BHI handling of RRSBA
Ingo Molnar <[email protected]>
x86/bugs: Rename various 'ia32_cap' variables to 'x86_arch_cap_msr'
Josh Poimboeuf <[email protected]>
x86/bugs: Cache the value of MSR_IA32_ARCH_CAPABILITIES
Josh Poimboeuf <[email protected]>
x86/bugs: Fix BHI documentation
Daniel Sneddon <[email protected]>
x86/bugs: Fix return type of spectre_bhi_state()
Amir Goldstein <[email protected]>
kernfs: annotate different lockdep class for of->mutex of writable files
Oleg Nesterov <[email protected]>
selftests: kselftest: Fix build failure with NOLIBC
Arnd Bergmann <[email protected]>
irqflags: Explicitly ignore lockdep_hrtimer_exit() argument
Adam Dunlap <[email protected]>
x86/apic: Force native_apic_mem_read() to use the MOV instruction
Nathan Chancellor <[email protected]>
selftests: kselftest: Mark functions that unconditionally call exit() as __noreturn
John Stultz <[email protected]>
selftests: timers: Fix abs() warning in posix_timers test
John Stultz <[email protected]>
selftests: timers: Fix posix_timers ksft_print_msg() warning
Oleg Nesterov <[email protected]>
selftests/timers/posix_timers: Reimplement check_timer_distribution()
Sean Christopherson <[email protected]>
x86/cpu: Actually turn off mitigations by default for SPECULATION_MITIGATIONS=n
Namhyung Kim <[email protected]>
perf/x86: Fix out of range data
Gavin Shan <[email protected]>
vhost: Add smp_rmb() in vhost_enable_notify()
Gavin Shan <[email protected]>
vhost: Add smp_rmb() in vhost_vq_avail_empty()
Frank Li <[email protected]>
arm64: dts: imx8-ss-dma: fix spi lpcg indices
Frank Li <[email protected]>
arm64: dts: imx8-ss-lsio: fix pwm lpcg indices
Frank Li <[email protected]>
arm64: dts: imx8-ss-dma: fix pwm lpcg indices
Frank Li <[email protected]>
arm64: dts: imx8-ss-conn: fix usb lpcg indices
Frank Li <[email protected]>
arm64: dts: imx8-ss-dma: fix adc lpcg indices
Frank Li <[email protected]>
arm64: dts: imx8-ss-dma: fix can lpcg indices
Frank Li <[email protected]>
arm64: dts: imx8qm-ss-dma: fix can lpcg indices
Lang Yu <[email protected]>
drm/amdgpu/umsch: reinitialize write pointer in hw init
Johan Hovold <[email protected]>
drm/msm/dp: fix runtime PM leak on connect failure
Johan Hovold <[email protected]>
drm/msm/dp: fix runtime PM leak on disconnect
Ville Syrjälä <[email protected]>
drm/client: Fully protect modes[] with dev->mode_config.mutex
Boris Brezillon <[email protected]>
drm/panfrost: Fix the error path in panfrost_mmu_map_fault_addr()
Jammy Huang <[email protected]>
drm/ast: Fix soft lockup
Harish Kasiviswanathan <[email protected]>
drm/amdkfd: Reset GPU on queue preemption failure
Ville Syrjälä <[email protected]>
drm/i915/vrr: Disable VRR when using bigjoiner
Zack Rusin <[email protected]>
drm/vmwgfx: Enable DMA mappings with SEV
Jacek Lawrynowicz <[email protected]>
accel/ivpu: Fix deadlock in context_xa
Jacek Lawrynowicz <[email protected]>
accel/ivpu: Return max freq for DRM_IVPU_PARAM_CORE_CLOCK_RATE
Jacek Lawrynowicz <[email protected]>
accel/ivpu: Put NPU back to D3hot after failed resume
Wachowski, Karol <[email protected]>
accel/ivpu: Fix PCI D0 state entry in resume
Wachowski, Karol <[email protected]>
accel/ivpu: Check return code of ipc->lock init
Alexander Wetzel <[email protected]>
scsi: sg: Avoid race in error handling & drop bogus warn
Alexander Wetzel <[email protected]>
scsi: sg: Avoid sg device teardown race
Masami Hiramatsu <[email protected]>
fs/proc: Skip bootloader comment if no embedded kernel parameters
Zhenhua Huang <[email protected]>
fs/proc: remove redundant comments from /proc/bootconfig
Zheng Yejian <[email protected]>
kprobes: Fix possible use-after-free issue on kprobe registration
Pavel Begunkov <[email protected]>
io_uring/net: restore msg_control on sendzc retry
Boris Burkov <[email protected]>
btrfs: qgroup: convert PREALLOC to PERTRANS after record_root_in_trans
Boris Burkov <[email protected]>
btrfs: record delayed inode root in transaction
Boris Burkov <[email protected]>
btrfs: qgroup: fix qgroup prealloc rsv leak in subvolume operations
Boris Burkov <[email protected]>
btrfs: qgroup: correctly model root qgroup rsv in convert
Jens Axboe <[email protected]>
io_uring: disable io-wq execution of multishot NOWAIT requests
Pavel Begunkov <[email protected]>
io_uring: refactor DEFER_TASKRUN multishot checks
Lu Baolu <[email protected]>
iommu/vt-d: Fix WARN_ON in iommu probe path
Jacob Pan <[email protected]>
iommu/vt-d: Allocate local memory for page request queue
Xuchun Shang <[email protected]>
iommu/vt-d: Fix wrong use of pasid config
Arnd Bergmann <[email protected]>
tracing: hide unused ftrace_event_id_fops
Karthik Poosa <[email protected]>
drm/xe/hwmon: Cast result to output precision on left shift of operand
Lucas De Marchi <[email protected]>
drm/xe/display: Fix double mutex initialization
David Arinzon <[email protected]>
net: ena: Set tx_info->xdpf value to NULL
David Arinzon <[email protected]>
net: ena: Fix incorrect descriptor free behavior
David Arinzon <[email protected]>
net: ena: Wrong missing IO completions check order
David Arinzon <[email protected]>
net: ena: Fix potential sign extension issue
Michal Luczaj <[email protected]>
af_unix: Fix garbage collector racing against connect()
Kuniyuki Iwashima <[email protected]>
af_unix: Do not use atomic ops for unix_sk(sk)->inflight.
Arınç ÜNAL <[email protected]>
net: dsa: mt7530: trap link-local frames regardless of ST Port State
Gerd Bayer <[email protected]>
Revert "s390/ism: fix receive message buffer allocation"
Daniel Machon <[email protected]>
net: sparx5: fix wrong config being used when reconfiguring PCS
Rahul Rameshbabu <[email protected]>
net/mlx5e: Do not produce metadata freelist entries in Tx port ts WQE xmit
Carolina Jubran <[email protected]>
net/mlx5e: HTB, Fix inconsistencies with QoS SQs number
Carolina Jubran <[email protected]>
net/mlx5e: Fix mlx5e_priv_init() cleanup flow
Carolina Jubran <[email protected]>
net/mlx5e: RSS, Block changing channels number when RXFH is configured
Cosmin Ratiu <[email protected]>
net/mlx5: Correctly compare pkt reformat ids
Cosmin Ratiu <[email protected]>
net/mlx5: Properly link new fs rules into the tree
Michael Liang <[email protected]>
net/mlx5: offset comp irq index in name by one
Shay Drory <[email protected]>
net/mlx5: Register devlink first under devlink lock
Moshe Shemesh <[email protected]>
net/mlx5: SF, Stop waiting for FW as teardown was called
Eric Dumazet <[email protected]>
netfilter: complete validation of user input
Archie Pusaka <[email protected]>
Bluetooth: l2cap: Don't double set the HCI_CONN_MGMT_CONNECTED bit
Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sock: Fix not validating setsockopt user input
Luiz Augusto von Dentz <[email protected]>
Bluetooth: ISO: Fix not validating setsockopt user input
Luiz Augusto von Dentz <[email protected]>
Bluetooth: L2CAP: Fix not validating setsockopt user input
Luiz Augusto von Dentz <[email protected]>
Bluetooth: RFCOMM: Fix not validating setsockopt user input
Luiz Augusto von Dentz <[email protected]>
Bluetooth: SCO: Fix not validating setsockopt user input
Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sync: Fix using the same interval and window for Coded PHY
Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sync: Use QoS to determine which PHY to scan
Luiz Augusto von Dentz <[email protected]>
Bluetooth: ISO: Don't reject BT_ISO_QOS if parameters are unset
Luiz Augusto von Dentz <[email protected]>
Bluetooth: ISO: Align broadcast sync_timeout with connection timeout
Brett Creeley <[email protected]>
pds_core: Fix pdsc_check_pci_health function to use work thread
Shannon Nelson <[email protected]>
pds_core: use pci_reset_function for health reset
Jiri Benc <[email protected]>
ipv6: fix race condition between ipv6_get_ifaddr and ipv6_del_addr
Arnd Bergmann <[email protected]>
ipv4/route: avoid unused-but-set-variable warning
Arnd Bergmann <[email protected]>
ipv6: fib: hide unused 'pn' variable
Geetha sowjanya <[email protected]>
octeontx2-af: Fix NIX SQ mode and BP config
Kuniyuki Iwashima <[email protected]>
af_unix: Clear stale u->oob_skb.
Marek Vasut <[email protected]>
net: ks8851: Handle softirqs at the end of IRQ thread to fix hang
Marek Vasut <[email protected]>
net: ks8851: Inline ks8851_rx_skb()
Dave Jiang <[email protected]>
cxl: Fix retrieving of access_coordinates in PCIe path
Dave Jiang <[email protected]>
cxl: Remove checking of iter in cxl_endpoint_get_perf_coordinates()
Dave Jiang <[email protected]>
cxl: Split out host bridge access coordinates
Dave Jiang <[email protected]>
cxl: Split out combine_coordinates() for common shared usage
Dave Jiang <[email protected]>
ACPI: HMAT / cxl: Add retrieval of generic port coordinates for both access classes
Dave Jiang <[email protected]>
ACPI: HMAT: Introduce 2 levels of generic port access class
Dave Jiang <[email protected]>
base/node / ACPI: Enumerate node access class for 'struct access_coordinate'
Raag Jadav <[email protected]>
ACPI: bus: allow _UID matching for integer zero
Pavan Chebbi <[email protected]>
bnxt_en: Reset PTP tx_avail after possible firmware reset
Vikas Gupta <[email protected]>
bnxt_en: Fix error recovery for RoCE ulp client
Vikas Gupta <[email protected]>
bnxt_en: Fix possible memory leak in bnxt_rdma_aux_device_init()
Gerd Bayer <[email protected]>
s390/ism: fix receive message buffer allocation
Eric Dumazet <[email protected]>
geneve: fix header validation in geneve[6]_xmit_skb
Arnd Bergmann <[email protected]>
lib: checksum: hide unused expected_csum_ipv6_magic[]
Ming Lei <[email protected]>
block: fix q->blkg_list corruption during disk rebind
Hariprasad Kelam <[email protected]>
octeontx2-pf: Fix transmit scheduler resource leak
Eric Dumazet <[email protected]>
xsk: validate user input for XDP_{UMEM|COMPLETION}_FILL_RING
Petr Tesarik <[email protected]>
u64_stats: fix u64_stats_init() for lockdep when used repeatedly in one file
Ilya Maximets <[email protected]>
net: openvswitch: fix unwanted error log on timeout policy probing
Dan Carpenter <[email protected]>
scsi: qla2xxx: Fix off by one in qla_edif_app_getstats()
Xiang Chen <[email protected]>
scsi: hisi_sas: Modify the deadline for ata_wait_after_reset()
Luca Weiss <[email protected]>
drm/msm/adreno: Set highest_bank_bit for A619
Arnd Bergmann <[email protected]>
nouveau: fix function cast warning
Alex Constantino <[email protected]>
Revert "drm/qxl: simplify qxl_fence_wait"
Kwangjin Ko <[email protected]>
cxl/core: Fix initialization of mbox_cmd.size_out in get event
Frank Li <[email protected]>
arm64: dts: imx8-ss-conn: fix usdhc wrong lpcg clock order
Dmitry Baryshkov <[email protected]>
dt-bindings: display/msm: sm8150-mdss: add DP node
Dmitry Baryshkov <[email protected]>
drm/msm/dpu: make error messages at dpu_core_irq_register_callback() more sensible
Dmitry Baryshkov <[email protected]>
drm/msm/dpu: don't allow overriding data from catalog
Stephen Boyd <[email protected]>
drm/msm: Add newlines to some debug prints
Tim Harvey <[email protected]>
arm64: dts: freescale: imx8mp-venice-gw73xx-2x: fix USB vbus regulator
Tim Harvey <[email protected]>
arm64: dts: freescale: imx8mp-venice-gw72xx-2x: fix USB vbus regulator
Dave Jiang <[email protected]>
cxl/core/regs: Fix usage of map->reg_type in cxl_decode_regblock() before assigned
Yuquan Wang <[email protected]>
cxl/mem: Fix for the index of Clear Event Record Handle
Cristian Marussi <[email protected]>
firmware: arm_scmi: Make raw debugfs entries non-seekable
Jens Wiklander <[email protected]>
firmware: arm_ffa: Fix the partition ID check in ffa_notification_info_get()
Aaro Koskinen <[email protected]>
ARM: OMAP2+: fix USB regression on Nokia N8x0
Aaro Koskinen <[email protected]>
mmc: omap: restore original power up/down steps
Aaro Koskinen <[email protected]>
mmc: omap: fix deferred probe
Aaro Koskinen <[email protected]>
mmc: omap: fix broken slot switch lookup
Aaro Koskinen <[email protected]>
ARM: OMAP2+: fix N810 MMC gpiod table
Aaro Koskinen <[email protected]>
ARM: OMAP2+: fix bogus MMC GPIO labels on Nokia N8x0
David Sterba <[email protected]>
btrfs: tests: allocate dummy fs_info and root in test_find_delalloc()
Nini Song <[email protected]>
media: cec: core: remove length check of Timer Status
Anna-Maria Behnsen <[email protected]>
PM: s2idle: Make sure CPUs will wakeup directly on resume
Hans de Goede <[email protected]>
ACPI: scan: Do not increase dep_unmet for already met dependencies
Noah Loomans <[email protected]>
platform/chrome: cros_ec_uart: properly fix race condition
Tim Huang <[email protected]>
drm/amd/pm: fixes a random hang in S4 for SMU v13.0.4/11
Dmitry Antipov <[email protected]>
Bluetooth: Fix memory leak in hci_req_sync_complete()
Steven Rostedt (Google) <[email protected]>
ring-buffer: Only update pages_touched when a new page is touched
Yu Kuai <[email protected]>
raid1: fix use-after-free for original bio in raid1_write_request()
Fabio Estevam <[email protected]>
ARM: dts: imx7s-warp: Pass OV2680 link-frequencies
Gavin Shan <[email protected]>
arm64: tlb: Fix TLBI RANGE operand
Breno Leitao <[email protected]>
virtio_net: Do not send RSS key if it is not supported
Xiubo Li <[email protected]>
ceph: switch to use cap_delay_lock for the unlink delay list
NeilBrown <[email protected]>
ceph: redirty page before returning AOP_WRITEPAGE_ACTIVATE
Sven Eckelmann <[email protected]>
batman-adv: Avoid infinite loop trying to resize local TT
Peyton Lee <[email protected]>
drm/amdgpu/vpe: power on vpe when hw_init
Damien Le Moal <[email protected]>
ata: libata-scsi: Fix ata_scsi_dev_rescan() error path
Igor Pylypiv <[email protected]>
ata: libata-core: Allow command duration limits detection for ACS-4 drives
Steve French <[email protected]>
smb3: fix Open files on server counter going negative
-------------
Diffstat:
Documentation/admin-guide/hw-vuln/spectre.rst | 22 +- Documentation/admin-guide/kernel-parameters.txt | 12 +- .../bindings/display/msm/qcom,sm8150-mdss.yaml | 9 + Makefile | 4 +- arch/arm/boot/dts/nxp/imx/imx7s-warp.dts | 1 + arch/arm/mach-omap2/board-n8x0.c | 23 +-- arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi | 16 +- arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi | 40 ++-- arch/arm64/boot/dts/freescale/imx8-ss-lsio.dtsi | 16 +- .../boot/dts/freescale/imx8mp-venice-gw72xx.dtsi | 2 +- .../boot/dts/freescale/imx8mp-venice-gw73xx.dtsi | 2 +- arch/arm64/boot/dts/freescale/imx8qm-ss-dma.dtsi | 8 +- arch/arm64/include/asm/tlbflush.h | 20 +- arch/x86/Kconfig | 21 +- arch/x86/events/core.c | 1 + arch/x86/include/asm/apic.h | 3 +- arch/x86/kernel/apic/apic.c | 6 +- arch/x86/kernel/cpu/bugs.c | 82 ++++---- arch/x86/kernel/cpu/common.c | 48 ++--- block/blk-cgroup.c | 9 +- block/blk-cgroup.h | 2 + block/blk-core.c | 2 + drivers/accel/ivpu/ivpu_drv.c | 20 +- drivers/accel/ivpu/ivpu_hw.h | 6 + drivers/accel/ivpu/ivpu_hw_37xx.c | 7 +- drivers/accel/ivpu/ivpu_hw_40xx.c | 6 + drivers/accel/ivpu/ivpu_ipc.c | 8 +- drivers/accel/ivpu/ivpu_pm.c | 7 +- drivers/acpi/numa/hmat.c | 43 ++-- drivers/acpi/scan.c | 3 +- drivers/ata/libata-core.c | 2 +- drivers/ata/libata-scsi.c | 9 +- drivers/base/node.c | 6 +- drivers/cxl/acpi.c | 8 +- drivers/cxl/core/cdat.c | 58 ++++-- drivers/cxl/core/mbox.c | 5 +- drivers/cxl/core/port.c | 76 ++++--- drivers/cxl/core/regs.c | 5 +- drivers/cxl/cxl.h | 8 +- drivers/firmware/arm_ffa/driver.c | 2 +- drivers/firmware/arm_scmi/raw_mode.c | 7 +- drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c | 6 + drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 2 +- drivers/gpu/drm/amd/amdgpu/soc21.c | 32 ++- drivers/gpu/drm/amd/amdgpu/umsch_mm_v4_0.c | 2 + .../gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 1 + drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 15 +- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c | 6 +- .../amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c | 19 +- drivers/gpu/drm/amd/display/dc/core/dc_state.c | 9 + .../gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c | 3 - .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c | 12 +- drivers/gpu/drm/ast/ast_dp.c | 3 + drivers/gpu/drm/drm_client_modeset.c | 3 +- drivers/gpu/drm/i915/display/intel_cdclk.c | 7 +- drivers/gpu/drm/i915/display/intel_cdclk.h | 3 + drivers/gpu/drm/i915/display/intel_ddi.c | 5 + drivers/gpu/drm/i915/display/intel_dp.c | 6 +- drivers/gpu/drm/i915/display/intel_psr.c | 11 + drivers/gpu/drm/i915/display/intel_vrr.c | 7 + drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 4 + drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c | 10 +- drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c | 8 +- drivers/gpu/drm/msm/dp/dp_display.c | 2 + drivers/gpu/drm/msm/msm_fb.c | 6 +- drivers/gpu/drm/msm/msm_kms.c | 4 +- .../gpu/drm/nouveau/nvkm/subdev/bios/shadowof.c | 7 +- drivers/gpu/drm/panfrost/panfrost_mmu.c | 13 +- drivers/gpu/drm/qxl/qxl_release.c | 50 ++++- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 11 +- drivers/gpu/drm/xe/xe_display.c | 5 - drivers/gpu/drm/xe/xe_hwmon.c | 4 +- drivers/iommu/intel/iommu.c | 11 +- drivers/iommu/intel/perfmon.c | 2 +- drivers/iommu/intel/svm.c | 2 +- drivers/md/raid1.c | 2 +- drivers/media/cec/core/cec-adap.c | 14 -- drivers/mmc/host/omap.c | 48 +++-- drivers/net/dsa/mt7530.c | 229 ++++++++++++++++++--- drivers/net/dsa/mt7530.h | 5 + drivers/net/ethernet/amazon/ena/ena_com.c | 2 +- drivers/net/ethernet/amazon/ena/ena_netdev.c | 35 ++-- drivers/net/ethernet/amazon/ena/ena_xdp.c | 4 +- drivers/net/ethernet/amd/pds_core/core.c | 14 +- drivers/net/ethernet/amd/pds_core/core.h | 5 +- drivers/net/ethernet/amd/pds_core/dev.c | 3 + drivers/net/ethernet/amd/pds_core/main.c | 8 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 + drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 6 +- .../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 22 +- drivers/net/ethernet/marvell/octeontx2/nic/qos.c | 1 + drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h | 8 +- drivers/net/ethernet/mellanox/mlx5/core/en/qos.c | 33 +-- drivers/net/ethernet/mellanox/mlx5/core/en/selq.c | 2 + .../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 17 ++ drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 - drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 7 +- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 17 +- drivers/net/ethernet/mellanox/mlx5/core/main.c | 37 ++-- drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 4 +- .../ethernet/mellanox/mlx5/core/sf/dev/driver.c | 22 +- drivers/net/ethernet/micrel/ks8851.h | 3 - drivers/net/ethernet/micrel/ks8851_common.c | 16 +- drivers/net/ethernet/micrel/ks8851_par.c | 11 - drivers/net/ethernet/micrel/ks8851_spi.c | 11 - .../net/ethernet/microchip/sparx5/sparx5_port.c | 4 +- drivers/net/geneve.c | 4 +- drivers/net/virtio_net.c | 28 ++- drivers/platform/chrome/cros_ec_uart.c | 28 +-- drivers/scsi/hisi_sas/hisi_sas_main.c | 2 +- drivers/scsi/qla2xxx/qla_edif.c | 2 +- drivers/scsi/sg.c | 20 +- drivers/vhost/vhost.c | 28 ++- fs/btrfs/delayed-inode.c | 3 + fs/btrfs/inode.c | 13 +- fs/btrfs/ioctl.c | 37 +++- fs/btrfs/qgroup.c | 2 + fs/btrfs/root-tree.c | 10 - fs/btrfs/root-tree.h | 2 - fs/btrfs/tests/extent-io-tests.c | 28 ++- fs/btrfs/transaction.c | 17 +- fs/ceph/addr.c | 4 +- fs/ceph/caps.c | 4 +- fs/ceph/mds_client.c | 9 +- fs/ceph/mds_client.h | 3 +- fs/kernfs/file.c | 9 +- fs/proc/bootconfig.c | 12 +- fs/smb/client/cached_dir.c | 4 +- include/acpi/acpi_bus.h | 8 +- include/linux/bootconfig.h | 1 + include/linux/dma-fence.h | 7 + include/linux/irqflags.h | 2 +- include/linux/node.h | 18 +- include/linux/u64_stats_sync.h | 9 +- include/net/addrconf.h | 4 + include/net/af_unix.h | 2 +- include/net/bluetooth/bluetooth.h | 11 + include/net/ip_tunnels.h | 33 +++ init/main.c | 5 + io_uring/io_uring.c | 25 +++ io_uring/net.c | 22 +- io_uring/rw.c | 2 - kernel/cpu.c | 3 +- kernel/kprobes.c | 18 +- kernel/power/suspend.c | 6 + kernel/trace/ring_buffer.c | 6 +- kernel/trace/trace_events.c | 4 + lib/checksum_kunit.c | 5 +- net/batman-adv/translation-table.c | 2 +- net/bluetooth/hci_request.c | 4 +- net/bluetooth/hci_sock.c | 21 +- net/bluetooth/hci_sync.c | 66 +++++- net/bluetooth/iso.c | 50 ++--- net/bluetooth/l2cap_core.c | 3 +- net/bluetooth/l2cap_sock.c | 52 ++--- net/bluetooth/rfcomm/sock.c | 14 +- net/bluetooth/sco.c | 23 +-- net/ipv4/netfilter/arp_tables.c | 4 + net/ipv4/netfilter/ip_tables.c | 4 + net/ipv4/route.c | 4 +- net/ipv6/addrconf.c | 7 +- net/ipv6/ip6_fib.c | 7 +- net/ipv6/netfilter/ip6_tables.c | 4 + net/openvswitch/conntrack.c | 5 +- net/unix/af_unix.c | 8 +- net/unix/garbage.c | 35 +++- net/unix/scm.c | 8 +- net/xdp/xsk.c | 2 + tools/testing/selftests/kselftest.h | 33 ++- tools/testing/selftests/timers/posix_timers.c | 105 +++++----- 170 files changed, 1559 insertions(+), 882 deletions(-)
Greg Kroah-Hartman
maintainerPATCH 1/172 smb3: fix Open files on server counter going negative
6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steve French <[email protected]>
commit 28e0947651ce6a2200b9a7eceb93282e97d7e51a upstream.
We were decrementing the count of open files on server twice for the case where we were closing cached directories.
--- fs/smb/client/cached_dir.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
fs/smb/client/cached_dir.c +2 -2
@@ -433,8 +433,8 @@ smb2_close_cached_fid(struct kref *ref)if (cfid->is_open) { rc =SMB2_close (0 , cfid->tcon, cfid->fid.persistent_fid, cfid->fid.volatile_fid); -if (rc != -EBUSY && rc != -EAGAIN ) -atomic_dec (&cfid->tcon->num_remote_opens); +if (rc)/* should we retry on -EBUSY or -EAGAIN? */ +cifs_dbg (VFS ,"close cached dir rc %d\n" , rc); }free_cached_dir (cfid);
Greg Kroah-Hartman
maintainerPATCH 2/172 ata: libata-core: Allow command duration limits detection for ACS-4 drives
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Igor Pylypiv <[email protected]>
commit c0297e7dd50795d559f3534887a6de1756b35d0f upstream.
Even though the command duration limits (CDL) feature was first added in ACS-5 (major version 12), there are some ACS-4 (major version 11) drives that implement CDL as well.
IDENTIFY_DEVICE, SUPPORTED_CAPABILITIES, and CURRENT_SETTINGS log pages are mandatory in the ACS-4 standard so it should be safe to read these log pages on older drives implementing the ACS-4 standard.
--- drivers/ata/libata-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/ata/libata-core.c +1 -1
@@ -2539,7 +2539,7 @@ static void ata_dev_config_cdl(struct atbool cdl_enabled;u64 val; -if (ata_id_major_version (dev->id) <12 ) +if (ata_id_major_version (dev->id) <11 ) goto not_supported; if (!ata_log_supported(dev, ATA_LOG_IDENTIFY_DEVICE) ||
Greg Kroah-Hartman
maintainerPATCH 3/172 ata: libata-scsi: Fix ata_scsi_dev_rescan() error path
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Damien Le Moal <[email protected]>
commit 79336504781e7fee5ddaf046dcc186c8dfdf60b1 upstream.
Commit 0c76106cb975 ("scsi: sd: Fix TCG OPAL unlock on system resume") incorrectly handles failures of scsi_resume_device() in ata_scsi_dev_rescan(), leading to a double call to spin_unlock_irqrestore() to unlock a device port. Fix this by redefining the goto labels used in case of errors and only unlock the port scsi_scan_mutex when scsi_resume_device() fails.
Bug found with the Smatch static checker warning:
drivers/ata/libata-scsi.c:4774 ata_scsi_dev_rescan() error: double unlocked 'ap->lock' (orig line 4757)
--- drivers/ata/libata-scsi.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
drivers/ata/libata-scsi.c +5 -4
@@ -4745,7 +4745,7 @@ void ata_scsi_dev_rescan(struct work_str* bail out. */ if (ap->pflags &ATA_PFLAG_SUSPENDED ) - goto unlock; + goto unlock_ap;if (!sdev)continue ;
@@ -4758,7 +4758,7 @@ void ata_scsi_dev_rescan(struct work_strif (do_resume) { ret =scsi_resume_device (sdev);if (ret == -EWOULDBLOCK ) - goto unlock; + goto unlock_scan; dev->flags &= ~ATA_DFLAG_RESUMING ; } ret =scsi_rescan_device (sdev);
@@ -4766,12 +4766,13 @@ void ata_scsi_dev_rescan(struct work_strspin_lock_irqsave (ap->lock, flags);if (ret) - goto unlock; + goto unlock_ap; } } -unlock: +unlock_ap:spin_unlock_irqrestore (ap->lock, flags); +unlock_scan:mutex_unlock (&ap->scsi_scan_mutex);/* Reschedule with a delay if scsi_rescan_device() returned an error */
Greg Kroah-Hartman
maintainerPATCH 4/172 drm/amdgpu/vpe: power on vpe when hw_init
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Peyton Lee <[email protected]>
commit eed14eb48ee176fe0144c6a999d00c855d0b199b upstream.
To fix mode2 reset failure.
Should power on VPE when hw_init.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c | 6 ++++++ 1 file changed, 6 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c +6 -0
@@ -390,6 +390,12 @@ static int vpe_hw_init(void *handle)struct amdgpu_vpe *vpe = &adev->vpe;int ret; +/* Power on VPE */ + ret =amdgpu_device_ip_set_powergating_state (adev,AMD_IP_BLOCK_TYPE_VPE , +AMD_PG_STATE_UNGATE ); +if (ret) +return ret; + ret =vpe_load_microcode (vpe);if (ret)return ret;
Greg Kroah-Hartman
maintainerPATCH 5/172 batman-adv: Avoid infinite loop trying to resize local TT
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit b1f532a3b1e6d2e5559c7ace49322922637a28aa upstream.
If the MTU of one of an attached interface becomes too small to transmit the local translation table then it must be resized to fit inside all fragments (when enabled) or a single packet.
But if the MTU becomes too low to transmit even the header + the VLAN specific part then the resizing of the local TT will never succeed. This can for example happen when the usable space is 110 bytes and 11 VLANs are on top of batman-adv. In this case, at least 116 byte would be needed. There will just be an endless spam of
batman_adv: batadv0: Forced to purge local tt entries to fit new maximum fragment MTU (110)
in the log but the function will never finish. Problem here is that the timeout will be halved all the time and will then stagnate at 0 and therefore never be able to reduce the table even more.
There are other scenarios possible with a similar result. The number of BATADV_TT_CLIENT_NOPURGE entries in the local TT can for example be too high to fit inside a packet. Such a scenario can therefore happen also with only a single VLAN + 7 non-purgable addresses - requiring at least 120 bytes.
While this should be handled proactively when:
* interface with too low MTU is added
* VLAN is added
* non-purgeable local mac is added
* MTU of an attached interface is reduced
* fragmentation setting gets disabled (which most likely requires dropping attached interfaces)
not all of these scenarios can be prevented because batman-adv is only consuming events without the the possibility to prevent these actions (non-purgable MAC address added, MTU of an attached interface is reduced). It is therefore necessary to also make sure that the code is able to handle also the situations when there were already incompatible system configuration are present.
--- net/batman-adv/translation-table.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/batman-adv/translation-table.c +1 -1
@@ -3948,7 +3948,7 @@ void batadv_tt_local_resize_to_mtu(strucspin_lock_bh (&bat_priv->tt.commit_lock); -while (true) { +while (timeout) { table_size =batadv_tt_local_table_transmit_size (bat_priv);if (packet_size_max >= table_size)break ;
Greg Kroah-Hartman
maintainerPATCH 6/172 ceph: redirty page before returning AOP_WRITEPAGE_ACTIVATE
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: NeilBrown <[email protected]>
commit b372e96bd0a32729d55d27f613c8bc80708a82e1 upstream.
The page has been marked clean before writepage is called. If we don't redirty it before postponing the write, it might never get written.
--- fs/ceph/addr.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
fs/ceph/addr.c +3 -1
@@ -795,8 +795,10 @@ static int ceph_writepage(struct page *pihold (inode);if (wbc->sync_mode ==WB_SYNC_NONE && -ceph_inode_to_fs_client (inode)->write_congested) +ceph_inode_to_fs_client (inode)->write_congested) { +redirty_page_for_writepage (wbc, page);return AOP_WRITEPAGE_ACTIVATE ; + }wait_on_page_fscache (page);
Greg Kroah-Hartman
maintainerPATCH 7/172 ceph: switch to use cap_delay_lock for the unlink delay list
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xiubo Li <[email protected]>
commit 17f8dc2db52185460f212052f3a692c1fdc167ba upstream.
The same list item will be used in both cap_delay_list and cap_unlink_delay_list, so it's buggy to use two different locks to protect them.
--- fs/ceph/caps.c | 4 ++-- fs/ceph/mds_client.c | 9 ++++----- fs/ceph/mds_client.h | 3 +-- 3 files changed, 7 insertions(+), 9 deletions(-)
fs/ceph/caps.c +2 -2
@@ -4775,13 +4775,13 @@ int ceph_drop_caps_for_unlink(struct inodoutc (mdsc->fsc->client,"%p %llx.%llx\n" , inode,ceph_vinop (inode)); -spin_lock (&mdsc->cap_unlink_delay_lock); +spin_lock (&mdsc->cap_delay_lock); ci->i_ceph_flags |=CEPH_I_FLUSH ;if (!list_empty (&ci->i_cap_delay_list))list_del_init (&ci->i_cap_delay_list);list_add_tail (&ci->i_cap_delay_list, &mdsc->cap_unlink_delay_list); -spin_unlock (&mdsc->cap_unlink_delay_lock); +spin_unlock (&mdsc->cap_delay_lock);/* * Fire the work immediately, because the MDS maybe
fs/ceph/mds_client.c +4 -5
@@ -2504,7 +2504,7 @@ static void ceph_cap_unlink_work(structstruct ceph_client *cl = mdsc->fsc->client;doutc (cl,"begin\n" ); -spin_lock (&mdsc->cap_unlink_delay_lock); +spin_lock (&mdsc->cap_delay_lock);while (!list_empty (&mdsc->cap_unlink_delay_list)) {struct ceph_inode_info *ci;struct inode *inode;
@@ -2516,15 +2516,15 @@ static void ceph_cap_unlink_work(struct inode =igrab (&ci->netfs.inode);if (inode) { -spin_unlock (&mdsc->cap_unlink_delay_lock); +spin_unlock (&mdsc->cap_delay_lock);doutc (cl,"on %p %llx.%llx\n" , inode,ceph_vinop (inode));ceph_check_caps (ci,CHECK_CAPS_FLUSH );iput (inode); -spin_lock (&mdsc->cap_unlink_delay_lock); +spin_lock (&mdsc->cap_delay_lock); } } -spin_unlock (&mdsc->cap_unlink_delay_lock); +spin_unlock (&mdsc->cap_delay_lock);doutc (cl,"done\n" ); }
@@ -5404,7 +5404,6 @@ int ceph_mdsc_init(struct ceph_fs_clientINIT_LIST_HEAD (&mdsc->cap_wait_list);spin_lock_init (&mdsc->cap_delay_lock);INIT_LIST_HEAD (&mdsc->cap_unlink_delay_list); -spin_lock_init (&mdsc->cap_unlink_delay_lock);INIT_LIST_HEAD (&mdsc->snap_flush_list);spin_lock_init (&mdsc->snap_flush_lock); mdsc->last_cap_flush_tid =1 ;
fs/ceph/mds_client.h +1 -2
@@ -461,9 +461,8 @@ struct ceph_mds_client {struct delayed_work delayed_work;/* delayed work */ unsigned long last_renew_caps;/* last time we renewed our caps */ struct list_head cap_delay_list;/* caps with delayed release */ -spinlock_t cap_delay_lock;/* protects cap_delay_list */ struct list_head cap_unlink_delay_list;/* caps with delayed release for unlink */ -spinlock_t cap_unlink_delay_lock;/* protects cap_unlink_delay_list */ +spinlock_t cap_delay_lock;/* protects cap_delay_list and cap_unlink_delay_list */ struct list_head snap_flush_list;/* cap_snaps ready to flush */ spinlock_t snap_flush_lock;
Greg Kroah-Hartman
maintainerPATCH 8/172 virtio_net: Do not send RSS key if it is not supported
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Breno Leitao <[email protected]>
commit 059a49aa2e25c58f90b50151f109dd3c4cdb3a47 upstream.
There is a bug when setting the RSS options in virtio_net that can break the whole machine, getting the kernel into an infinite loop.
Running the following command in any QEMU virtual machine with virtionet will reproduce this problem:
# ethtool -X eth0 hfunc toeplitz
This is how the problem happens:
1) ethtool_set_rxfh() calls virtnet_set_rxfh()
2) virtnet_set_rxfh() calls virtnet_commit_rss_command()
3) virtnet_commit_rss_command() populates 4 entries for the rss scatter-gather
4) Since the command above does not have a key, then the last scatter-gatter entry will be zeroed, since rss_key_size == 0. sg_buf_size = vi->rss_key_size;
5) This buffer is passed to qemu, but qemu is not happy with a buffer with zero length, and do the following in virtqueue_map_desc() (QEMU function):
if (!sz) {
virtio_error(vdev, "virtio: zero sized buffers are not allowed");6) virtio_error() (also QEMU function) set the device as broken
vdev->broken = true;
7) Qemu bails out, and do not repond this crazy kernel.
8) The kernel is waiting for the response to come back (function virtnet_send_command())
9) The kernel is waiting doing the following :
while (!virtqueue_get_buf(vi->cvq, &tmp) && !virtqueue_is_broken(vi->cvq)) cpu_relax();
10) None of the following functions above is true, thus, the kernel loops here forever. Keeping in mind that virtqueue_is_broken() does not look at the qemu vdev->broken, so, it never realizes that the vitio is broken at QEMU side.
Fix it by not sending RSS commands if the feature is not available in the device.
--- drivers/net/virtio_net.c | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-)
drivers/net/virtio_net.c +22 -4
@@ -3768,6 +3768,7 @@ static int virtnet_set_rxfh(struct net_d struct netlink_ext_ack *extack) {struct virtnet_info *vi =netdev_priv (dev); +bool update = false;int i; if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
@@ -3775,13 +3776,28 @@ static int virtnet_set_rxfh(struct net_dreturn -EOPNOTSUPP ;if (rxfh->indir) { +if (!vi->has_rss) +return -EOPNOTSUPP ; +for (i =0 ; i < vi->rss_indir_table_size; ++i) vi->ctrl->rss.indirection_table[i] = rxfh->indir[i]; + update = true; } -if (rxfh->key) + +if (rxfh->key) { +/* If either _F_HASH_REPORT or _F_RSS are negotiated, the +* device provides hash calculation capabilities, that is, +* hash_key is configured. +*/ +if (!vi->has_rss && !vi->has_rss_hash_report) +return -EOPNOTSUPP ; +memcpy (vi->ctrl->rss.key, rxfh->key, vi->rss_key_size); + update = true; + } -virtnet_commit_rss_command (vi); +if (update) +virtnet_commit_rss_command (vi);return 0 ; }
@@ -4686,13 +4702,15 @@ static int virtnet_probe(struct virtio_dif (virtio_has_feature (vdev,VIRTIO_NET_F_HASH_REPORT )) vi->has_rss_hash_report = true; -if (virtio_has_feature (vdev,VIRTIO_NET_F_RSS )) +if (virtio_has_feature (vdev,VIRTIO_NET_F_RSS )) { vi->has_rss = true; -if (vi->has_rss || vi->has_rss_hash_report) { vi->rss_indir_table_size =virtio_cread16 (vdev, offsetof(struct virtio_net_config , rss_max_indirection_table_length)); + } + +if (vi->has_rss || vi->has_rss_hash_report) { vi->rss_key_size =virtio_cread8 (vdev, offsetof(struct virtio_net_config , rss_max_key_size));
Greg Kroah-Hartman
maintainerPATCH 9/172 arm64: tlb: Fix TLBI RANGE operand
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gavin Shan <[email protected]>
commit e3ba51ab24fddef79fc212f9840de54db8fd1685 upstream.
KVM/arm64 relies on TLBI RANGE feature to flush TLBs when the dirty pages are collected by VMM and the page table entries become write protected during live migration. Unfortunately, the operand passed to the TLBI RANGE instruction isn't correctly sorted out due to the commit 117940aa6e5f ("KVM: arm64: Define kvm_tlb_flush_vmid_range()"). It leads to crash on the destination VM after live migration because TLBs aren't flushed completely and some of the dirty pages are missed.
For example, I have a VM where 8GB memory is assigned, starting from 0x40000000 (1GB). Note that the host has 4KB as the base page size. In the middile of migration, kvm_tlb_flush_vmid_range() is executed to flush TLBs. It passes MAX_TLBI_RANGE_PAGES as the argument to __kvm_tlb_flush_vmid_range() and __flush_s2_tlb_range_op(). SCALE#3 and NUM#31, corresponding to MAX_TLBI_RANGE_PAGES, isn't supported by __TLBI_RANGE_NUM(). In this specific case, -1 has been returned from __TLBI_RANGE_NUM() for SCALE#3/2/1/0 and rejected by the loop in the __flush_tlb_range_op() until the variable @scale underflows and becomes -9, 0xffff708000040000 is set as the operand. The operand is wrong since it's sorted out by __TLBI_VADDR_RANGE() according to invalid @scale and @num.
Fix it by extending __TLBI_RANGE_NUM() to support the combination of SCALE#3 and NUM#31. With the changes, [-1 31] instead of [-1 30] can be returned from the macro, meaning the TLBs for 0x200000 pages in the above example can be flushed in one shoot with SCALE#3 and NUM#31. The macro TLBI_RANGE_MASK is dropped since no one uses it any more. The comments are also adjusted accordingly.
--- arch/arm64/include/asm/tlbflush.h | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-)
arch/arm64/include/asm/tlbflush.h +12 -10
@@ -161,12 +161,18 @@ static inline unsigned long get_trans_gr#define MAX_TLBI_RANGE_PAGES __TLBI_RANGE_PAGES(31, 3)/* -* Generate 'num' values from -1 to 30 with -1 rejected by the -* __flush_tlb_range() loop below. -*/ -#define TLBI_RANGE_MASK GENMASK_ULL(4, 0) -#define __TLBI_RANGE_NUM (pages, scale) \ - ((((pages) >> (5 * (scale) + 1)) & TLBI_RANGE_MASK) - 1) +* Generate 'num' values from -1 to 31 with -1 rejected by the +* __flush_tlb_range() loop below. Its return value is only +* significant for a maximum of MAX_TLBI_RANGE_PAGES pages. If +* 'pages' is more than that, you must iterate over the overall +* range. +*/ +#define __TLBI_RANGE_NUM (pages, scale) \ + ({ \ + int __pages = min((pages), \ + __TLBI_RANGE_PAGES(31, (scale))); \ + (__pages >> (5 * (scale) + 1)) - 1; \ + })/* * TLB Invalidation
@@ -379,10 +385,6 @@ static inline void arch_tlbbatch_flush(s* 3. If there is 1 page remaining, flush it through non-range operations. Range * operations can only span an even number of pages. We save this for last to * ensure 64KB start alignment is maintained for the LPA2 case. -* -* Note that certain ranges can be represented by either num = 31 and -* scale or num = 0 and scale + 1. The loop below favours the latter -* since num is limited to 30 by the __TLBI_RANGE_NUM() macro. */ #define __flush_tlb_range_op (op, start, pages, stride, \ asid, tlb_level, tlbi_user, lpa2) \
Greg Kroah-Hartman
maintainerPATCH 10/172 ARM: dts: imx7s-warp: Pass OV2680 link-frequencies
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Fabio Estevam <[email protected]>
commit 135f218255b28c5bbf71e9e32a49e5c734cabbe5 upstream.
Since commit 63b0cd30b78e ("media: ov2680: Add bus-cfg / endpoint property verification") the ov2680 no longer probes on a imx7s-warp7:
ov2680 1-0036: error -EINVAL: supported link freq 330000000 not found ov2680 1-0036: probe with driver ov2680 failed with error -22
Fix it by passing the required 'link-frequencies' property as recommended by:
https://www.kernel.org/doc/html/v6.9-rc1/driver-api/media/camera-sensor.html#handling-clocks
--- arch/arm/boot/dts/nxp/imx/imx7s-warp.dts | 1 + 1 file changed, 1 insertion(+)
arch/arm/boot/dts/nxp/imx/imx7s-warp.dts +1 -0
@@ -210,6 +210,7 @@ remote-endpoint = <&mipi_from_sensor>; clock-lanes = <0>; data-lanes = <1>; + link-frequencies = /bits/ 64 <330000000>; }; }; };
Greg Kroah-Hartman
maintainerPATCH 11/172 raid1: fix use-after-free for original bio in raid1_write_request()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yu Kuai <[email protected]>
commit fcf3f7e2fc8a53a6140beee46ec782a4c88e4744 upstream.
r1_bio->bios[] is used to record new bios that will be issued to underlying disks, however, in raid1_write_request(), r1_bio->bios[] will set to the original bio temporarily. Meanwhile, if blocked rdev is set, free_r1bio() will be called causing that all r1_bio->bios[] to be freed:
raid1_write_request()
r1_bio = alloc_r1bio(mddev, bio); -> r1_bio->bios[] is NULL for (i = 0; i < disks; i++) -> for each rdev in conf
// first rdev is normal
r1_bio->bios[0] = bio; -> set to original bio
// second rdev is blocked
if (test_bit(Blocked, &rdev->flags))
break
if (blocked_rdev)
free_r1bio()
put_all_bios()
bio_put(r1_bio->bios[0]) -> original bio is freed
Test scripts:
mdadm -CR /dev/md0 -l1 -n4 /dev/sd[abcd] --assume-clean
fio -filename=/dev/md0 -ioengine=libaio -rw=write -bs=4k -numjobs=1 \
-iodepth=128 -name=test -direct=1echo blocked > /sys/block/md0/md/rd2/state
Test result:
BUG bio-264 (Not tainted): Object already free
-----------------------------------------------------------------------------
Allocated in mempool_alloc_slab+0x24/0x50 age=1 cpu=1 pid=869
kmem_cache_alloc+0x324/0x480 mempool_alloc_slab+0x24/0x50 mempool_alloc+0x6e/0x220 bio_alloc_bioset+0x1af/0x4d0 blkdev_direct_IO+0x164/0x8a0 blkdev_write_iter+0x309/0x440 aio_write+0x139/0x2f0 io_submit_one+0x5ca/0xb70 __do_sys_io_submit+0x86/0x270 __x64_sys_io_submit+0x22/0x30 do_syscall_64+0xb1/0x210 entry_SYSCALL_64_after_hwframe+0x6c/0x74 Freed in mempool_free_slab+0x1f/0x30 age=1 cpu=1 pid=869 kmem_cache_free+0x28c/0x550 mempool_free_slab+0x1f/0x30 mempool_free+0x40/0x100 bio_free+0x59/0x80 bio_put+0xf0/0x220 free_r1bio+0x74/0xb0 raid1_make_request+0xadf/0x1150 md_handle_request+0xc7/0x3b0 md_submit_bio+0x76/0x130 __submit_bio+0xd8/0x1d0 submit_bio_noacct_nocheck+0x1eb/0x5c0 submit_bio_noacct+0x169/0xd40 submit_bio+0xee/0x1d0 blkdev_direct_IO+0x322/0x8a0 blkdev_write_iter+0x309/0x440 aio_write+0x139/0x2f0
Since that bios for underlying disks are not allocated yet, fix this problem by using mempool_free() directly to free the r1_bio.
--- drivers/md/raid1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/md/raid1.c +1 -1
@@ -1474,7 +1474,7 @@ static void raid1_write_request(struct mfor (j =0 ; j < i; j++)if (r1_bio->bios[j])rdev_dec_pending (conf->mirrors[j].rdev, mddev); -free_r1bio (r1_bio); +mempool_free (r1_bio, &conf->r1bio_pool);allow_barrier (conf, bio->bi_iter.bi_sector);if (bio->bi_opf &REQ_NOWAIT ) {
Greg Kroah-Hartman
maintainerPATCH 12/172 ring-buffer: Only update pages_touched when a new page is touched
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt (Google) <[email protected]>
commit ffe3986fece696cf65e0ef99e74c75f848be8e30 upstream.
The "buffer_percent" logic that is used by the ring buffer splice code to only wake up the tasks when there's no data after the buffer is filled to the percentage of the "buffer_percent" file is dependent on three variables that determine the amount of data that is in the ring buffer:
1) pages_read - incremented whenever a new sub-buffer is consumed
2) pages_lost - incremented every time a writer overwrites a sub-buffer
3) pages_touched - incremented when a write goes to a new sub-buffer
The percentage is the calculation of:
(pages_touched - (pages_lost + pages_read)) / nr_pages
Basically, the amount of data is the total number of sub-bufs that have been touched, minus the number of sub-bufs lost and sub-bufs consumed. This is divided by the total count to give the buffer percentage. When the percentage is greater than the value in the "buffer_percent" file, it wakes up splice readers waiting for that amount.
It was observed that over time, the amount read from the splice was constantly decreasing the longer the trace was running. That is, if one asked for 60%, it would read over 60% when it first starts tracing, but then it would be woken up at under 60% and would slowly decrease the amount of data read after being woken up, where the amount becomes much less than the buffer percent.
This was due to an accounting of the pages_touched incrementation. This value is incremented whenever a writer transfers to a new sub-buffer. But the place where it was incremented was incorrect. If a writer overflowed the current sub-buffer it would go to the next one. If it gets preempted by an interrupt at that time, and the interrupt performs a trace, it too will end up going to the next sub-buffer. But only one should increment the counter. Unfortunately, that was not the case.
Change the cmpxchg() that does the real switch of the tail-page into a try_cmpxchg(), and on success, perform the increment of pages_touched. This will only increment the counter once for when the writer moves to a new sub-buffer, and not when there's a race and is incremented for when a writer and its preempting writer both move to the same new sub-buffer.
--- kernel/trace/ring_buffer.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
kernel/trace/ring_buffer.c +3 -3
@@ -1400,7 +1400,6 @@ static void rb_tail_page_update(struct r old_write =local_add_return (RB_WRITE_INTCNT , &next_page->write); old_entries =local_add_return (RB_WRITE_INTCNT , &next_page->entries); -local_inc (&cpu_buffer->pages_touched);/* * Just make sure we have seen our old_write and synchronize * with any interrupts that come in.
@@ -1437,8 +1436,9 @@ static void rb_tail_page_update(struct r*/ local_set (&next_page->page->commit,0 ); -/* Again, either we update tail_page or an interrupt does */ - (void )cmpxchg (&cpu_buffer->tail_page, tail_page, next_page); +/* Either we update tail_page or an interrupt does */ +if (try_cmpxchg (&cpu_buffer->tail_page, &tail_page, next_page)) +local_inc (&cpu_buffer->pages_touched); } }
Greg Kroah-Hartman
maintainerPATCH 13/172 Bluetooth: Fix memory leak in hci_req_sync_complete()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dmitry Antipov <[email protected]>
commit 45d355a926ab40f3ae7bc0b0a00cb0e3e8a5a810 upstream.
In 'hci_req_sync_complete()', always free the previous sync request state before assigning reference to a new one.
--- net/bluetooth/hci_request.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
net/bluetooth/hci_request.c +3 -1
@@ -105,8 +105,10 @@ void hci_req_sync_complete(struct hci_deif (hdev->req_status ==HCI_REQ_PEND ) { hdev->req_result = result; hdev->req_status =HCI_REQ_DONE ; -if (skb) +if (skb) { +kfree_skb (hdev->req_skb); hdev->req_skb =skb_get (skb); + }wake_up_interruptible (&hdev->req_wait_q); } }
Greg Kroah-Hartman
maintainerPATCH 14/172 drm/amd/pm: fixes a random hang in S4 for SMU v13.0.4/11
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tim Huang <[email protected]>
commit 31729e8c21ecfd671458e02b6511eb68c2225113 upstream.
While doing multiple S4 stress tests, GC/RLC/PMFW get into an invalid state resulting into hard hangs.
Adding a GFX reset as workaround just before sending the MP1_UNLOAD message avoids this failure.
--- drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c +11 -1
@@ -226,8 +226,18 @@ static int smu_v13_0_4_system_features_cstruct amdgpu_device *adev = smu->adev;int ret =0 ; -if (!en && !adev->in_s0ix) +if (!en && !adev->in_s0ix) { +/* Adds a GFX reset as workaround just before sending the +* MP1_UNLOAD message to prevent GC/RLC/PMFW from entering +* an invalid state. +*/ + ret =smu_cmn_send_smc_msg_with_param (smu, SMU_MSG_GfxDeviceDriverReset, +SMU_RESET_MODE_2 ,NULL ); +if (ret) +return ret; + ret =smu_cmn_send_smc_msg (smu, SMU_MSG_PrepareMp1ForUnload,NULL ); + }return ret; }
Greg Kroah-Hartman
maintainerPATCH 15/172 platform/chrome: cros_ec_uart: properly fix race condition
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Noah Loomans <[email protected]>
commit 5e700b384ec13f5bcac9855cb28fcc674f1d3593 upstream.
The cros_ec_uart_probe() function calls devm_serdev_device_open() before it calls serdev_device_set_client_ops(). This can trigger a NULL pointer dereference:
BUG: kernel NULL pointer dereference, address: 0000000000000000
...
Call Trace:
<TASK>
...
? ttyport_receive_bufA simplified version of crashing code is as follows:
static inline size_t serdev_controller_receive_buf(struct serdev_controller *ctrl,
const u8 *data,
size_t count)
{
struct serdev_device *serdev = ctrl->serdev;
if (!serdev || !serdev->ops->receive_buf) // CRASH!
return 0;
return serdev->ops->receive_buf(serdev, data, count);
}It assumes that if SERPORT_ACTIVE is set and serdev exists, serdev->ops will also exist. This conflicts with the existing cros_ec_uart_probe() logic, as it first calls devm_serdev_device_open() (which sets SERPORT_ACTIVE), and only later sets serdev->ops via serdev_device_set_client_ops().
Commit 01f95d42b8f4 ("platform/chrome: cros_ec_uart: fix race condition") attempted to fix a similar race condition, but while doing so, made the window of error for this race condition to happen much wider.
Attempt to fix the race condition again, making sure we fully setup before calling devm_serdev_device_open().
--- drivers/platform/chrome/cros_ec_uart.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-)
drivers/platform/chrome/cros_ec_uart.c +14 -14
@@ -263,12 +263,6 @@ static int cros_ec_uart_probe(struct serif (!ec_dev)return -ENOMEM ; - ret =devm_serdev_device_open (dev, serdev); -if (ret) { -dev_err (dev,"Unable to open UART device" ); -return ret; - } -serdev_device_set_drvdata (serdev, ec_dev);init_waitqueue_head (&ec_uart->response.wait_queue);
@@ -280,14 +274,6 @@ static int cros_ec_uart_probe(struct serreturn ret; } - ret =serdev_device_set_baudrate (serdev, ec_uart->baudrate); -if (ret <0 ) { -dev_err (dev,"Failed to set up host baud rate (%d)" , ret); -return ret; - } - -serdev_device_set_flow_control (serdev, ec_uart->flowcontrol); -/* Initialize ec_dev for cros_ec */ ec_dev->phys_name =dev_name (dev); ec_dev->dev = dev;
@@ -301,6 +287,20 @@ static int cros_ec_uart_probe(struct serserdev_device_set_client_ops (serdev, &cros_ec_uart_client_ops); + ret =devm_serdev_device_open (dev, serdev); +if (ret) { +dev_err (dev,"Unable to open UART device" ); +return ret; + } + + ret =serdev_device_set_baudrate (serdev, ec_uart->baudrate); +if (ret <0 ) { +dev_err (dev,"Failed to set up host baud rate (%d)" , ret); +return ret; + } + +serdev_device_set_flow_control (serdev, ec_uart->flowcontrol); +return cros_ec_register (ec_dev); }
Greg Kroah-Hartman
maintainerPATCH 16/172 ACPI: scan: Do not increase dep_unmet for already met dependencies
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hans de Goede <[email protected]>
commit d730192ff0246356a2d7e63ff5bd501060670eec upstream.
On the Toshiba Encore WT10-A tablet the BATC battery ACPI device depends on 3 other devices:
Name (_DEP, Package (0x03) // _DEP: Dependencies
{
I2C1,
GPO2,
GPO0
})acpi_scan_check_dep() adds all 3 of these to the acpi_dep_list and then before an acpi_device is created for the BATC handle (and thus before acpi_scan_dep_init() runs) acpi_scan_clear_dep() gets called for both GPIO depenencies, with free_when_met not set for the dependencies.
Since there is no adev for BATC yet, there also is no dep_unmet to decrement. The only result of acpi_scan_clear_dep() in this case is dep->met getting set.
Soon after acpi_scan_clear_dep() has been called for the GPIO dependencies the acpi_device gets created for the BATC handle and acpi_scan_dep_init() runs, this sees 3 dependencies on the acpi_dep_list and initializes unmet_dep to 3. Later when the dependency for I2C1 is met unmet_dep becomes 2, but since the 2 GPIO deps where already met it never becomes 0 causing battery monitoring to not work.
Fix this by modifying acpi_scan_dep_init() to not increase dep_met for dependencies which have already been marked as being met.
--- drivers/acpi/scan.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/acpi/scan.c +2 -1
@@ -1802,7 +1802,8 @@ static void acpi_scan_dep_init(struct acif (dep->honor_dep) adev->flags.honor_deps =1 ; - adev->dep_unmet++; +if (!dep->met) + adev->dep_unmet++; } } }
Greg Kroah-Hartman
maintainerPATCH 17/172 PM: s2idle: Make sure CPUs will wakeup directly on resume
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anna-Maria Behnsen <[email protected]>
commit 3c89a068bfd0698a5478f4cf39493595ef757d5e upstream.
s2idle works like a regular suspend with freezing processes and freezing devices. All CPUs except the control CPU go into idle. Once this is completed the control CPU kicks all other CPUs out of idle, so that they reenter the idle loop and then enter s2idle state. The control CPU then issues an swait() on the suspend state and therefore enters the idle loop as well.
Due to being kicked out of idle, the other CPUs leave their NOHZ states, which means the tick is active and the corresponding hrtimer is programmed to the next jiffie.
On entering s2idle the CPUs shut down their local clockevent device to prevent wakeups. The last CPU which enters s2idle shuts down its local clockevent and freezes timekeeping.
On resume, one of the CPUs receives the wakeup interrupt, unfreezes timekeeping and its local clockevent and starts the resume process. At that point all other CPUs are still in s2idle with their clockevents switched off. They only resume when they are kicked by another CPU or after resuming devices and then receiving a device interrupt.
That means there is no guarantee that all CPUs will wakeup directly on resume. As a consequence there is no guarantee that timers which are queued on those CPUs and should expire directly after resume, are handled. Also timer list timers which are remotely queued to one of those CPUs after resume will not result in a reprogramming IPI as the tick is active. Queueing a hrtimer will also not result in a reprogramming IPI because the first hrtimer event is already in the past.
The recent introduction of the timer pull model (7ee988770326 ("timers:
Implement the hierarchical pull model")) amplifies this problem, if the current migrator is one of the non woken up CPUs. When a non pinned timer list timer is queued and the queuing CPU goes idle, it relies on the still suspended migrator CPU to expire the timer which will happen by chance.
The problem exists since commit 8d89835b0467 ("PM: suspend: Do not pause cpuidle in the suspend-to-idle path"). There the cpuidle_pause() call which in turn invoked a wakeup for all idle CPUs was moved to a later point in the resume process. This might not be reached or reached very late because it waits on a timer of a still suspended CPU.
Address this by kicking all CPUs out of idle after the control CPU returns from swait() so that they resume their timers and restore consistent system state.
--- kernel/power/suspend.c | 6 ++++++ 1 file changed, 6 insertions(+)
kernel/power/suspend.c +6 -0
@@ -106,6 +106,12 @@ static void s2idle_enter(void)swait_event_exclusive (s2idle_wait_head, s2idle_state ==S2IDLE_STATE_WAKE ); +/* +* Kick all CPUs to ensure that they resume their timers and restore +* consistent system state. +*/ +wake_up_all_idle_cpus (); +cpus_read_unlock ();raw_spin_lock_irq (&s2idle_lock);
Greg Kroah-Hartman
maintainerPATCH 18/172 media: cec: core: remove length check of Timer Status
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Nini Song <[email protected]>
commit ce5d241c3ad4568c12842168288993234345c0eb upstream.
The valid_la is used to check the length requirements, including special cases of Timer Status. If the length is shorter than 5, that means no Duration Available is returned, the message will be forced to be invalid.
However, the description of Duration Available in the spec is that this parameter may be returned when these cases, or that it can be optionally return when these cases. The key words in the spec description are flexible choices.
Remove the special length check of Timer Status to fit the spec which is not compulsory about that.
--- drivers/media/cec/core/cec-adap.c | 14 -------------- 1 file changed, 14 deletions(-)
drivers/media/cec/core/cec-adap.c +0 -14
@@ -1151,20 +1151,6 @@ void cec_received_msg_ts(struct cec_adapif (valid_la && min_len) {/* These messages have special length requirements */ switch (cmd) { -case CEC_MSG_TIMER_STATUS : -if (msg->msg[2 ] &0x10 ) { -switch (msg->msg[2 ] &0xf ) { -case CEC_OP_PROG_INFO_NOT_ENOUGH_SPACE : -case CEC_OP_PROG_INFO_MIGHT_NOT_BE_ENOUGH_SPACE : -if (msg->len <5 ) - valid_la = false; -break ; - } - }else if ((msg->msg[2 ] &0xf ) ==CEC_OP_PROG_ERROR_DUPLICATE ) { -if (msg->len <5 ) - valid_la = false; - } -break ;case CEC_MSG_RECORD_ON :switch (msg->msg[2 ]) {case CEC_OP_RECORD_SRC_OWN :
Greg Kroah-Hartman
maintainerPATCH 19/172 btrfs: tests: allocate dummy fs_info and root in test_find_delalloc()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Sterba <[email protected]>
commit b2136cc288fce2f24a92f3d656531b2d50ebec5a upstream.
Allocate fs_info and root to have a valid fs_info pointer in case it's dereferenced by a helper outside of tests, like find_lock_delalloc_range().
--- fs/btrfs/tests/extent-io-tests.c | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-)
fs/btrfs/tests/extent-io-tests.c +24 -4
@@ -11,6 +11,7 @@#include "btrfs-tests.h" #include "../ctree.h" #include "../extent_io.h" +#include "../disk-io.h" #include "../btrfs_inode.h" #define PROCESS_UNLOCK (1 << 0)
@@ -105,9 +106,11 @@ static void dump_extent_io_tree(const st } } -static int test_find_delalloc (u32 sectorsize) +static int test_find_delalloc (u32 sectorsize,u32 nodesize) { -struct inode *inode; +struct btrfs_fs_info *fs_info; +struct btrfs_root *root =NULL ; +struct inode *inode =NULL ;struct extent_io_tree *tmp;struct page *page;struct page *locked_page =NULL ;
@@ -121,12 +124,27 @@ static int test_find_delalloc(u32 sectortest_msg ("running find delalloc tests" ); + fs_info =btrfs_alloc_dummy_fs_info (nodesize, sectorsize); +if (!fs_info) { +test_std_err (TEST_ALLOC_FS_INFO ); +return -ENOMEM ; + } + + root =btrfs_alloc_dummy_root (fs_info); +if (IS_ERR (root)) { +test_std_err (TEST_ALLOC_ROOT ); + ret =PTR_ERR (root); + goto out; + } + inode =btrfs_new_test_inode ();if (!inode) {test_std_err (TEST_ALLOC_INODE ); -return -ENOMEM ; + ret = -ENOMEM ; + goto out; } tmp = &BTRFS_I (inode)->io_tree; +BTRFS_I (inode)->root = root;/* * Passing NULL as we don't have fs_info but tracepoints are not used
@@ -316,6 +334,8 @@ out:process_page_range (inode,0 , total_dirty -1 ,PROCESS_UNLOCK |PROCESS_RELEASE );iput (inode); +btrfs_free_dummy_root (root); +btrfs_free_dummy_fs_info (fs_info);return ret; }
@@ -794,7 +814,7 @@ int btrfs_test_extent_io(u32 sectorsize,test_msg ("running extent I/O tests" ); - ret =test_find_delalloc (sectorsize); + ret =test_find_delalloc (sectorsize, nodesize);if (ret) goto out;
Greg Kroah-Hartman
maintainerPATCH 20/172 ARM: OMAP2+: fix bogus MMC GPIO labels on Nokia N8x0
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aaro Koskinen <[email protected]>
[ Upstream commit 95f37eb52e18879a1b16e51b972d992b39e50a81 ]
The GPIO bank width is 32 on OMAP2, so all labels are incorrect.
Message-ID: <[email protected]>
--- arch/arm/mach-omap2/board-n8x0.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-)
arch/arm/mach-omap2/board-n8x0.c +3 -6
@@ -144,8 +144,7 @@ static struct gpiod_lookup_table nokia8xx_mmc_gpio_table = { .dev_id = "mmci-omap.0", .table = {/* Slot switch, GPIO 96 */ - GPIO_LOOKUP("gpio-80-111", 16, - "switch", GPIO_ACTIVE_HIGH), + GPIO_LOOKUP("gpio-96-127", 0, "switch", GPIO_ACTIVE_HIGH), { } }, };
@@ -154,11 +153,9 @@ static struct gpiod_lookup_table nokia810_mmc_gpio_table = { .dev_id = "mmci-omap.0", .table = {/* Slot index 1, VSD power, GPIO 23 */ -GPIO_LOOKUP_IDX ("gpio-16-31" ,7 , -"vsd" ,1 ,GPIO_ACTIVE_HIGH ), +GPIO_LOOKUP_IDX ("gpio-0-31" ,23 ,"vsd" ,1 ,GPIO_ACTIVE_HIGH ),/* Slot index 1, VIO power, GPIO 9 */ -GPIO_LOOKUP_IDX ("gpio-0-15" ,9 , - "vio", 1, GPIO_ACTIVE_HIGH), + GPIO_LOOKUP_IDX("gpio-0-31", 9, "vio", 1, GPIO_ACTIVE_HIGH), { } }, };
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 21/172 ARM: OMAP2+: fix N810 MMC gpiod table
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aaro Koskinen <[email protected]>
[ Upstream commit 480d44d0820dd5ae043dc97c0b46dabbe53cb1cf ]
Trying to append a second table for the same dev_id doesn't seem to work. The second table is just silently ignored. As a result eMMC GPIOs are not present.
Fix by using separate tables for N800 and N810.
Message-ID: <[email protected]>
--- arch/arm/mach-omap2/board-n8x0.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
arch/arm/mach-omap2/board-n8x0.c +5 -3
@@ -140,7 +140,7 @@ static int slot1_cover_open;static int slot2_cover_open;static struct device *mmc_device; -static struct gpiod_lookup_table nokia8xx_mmc_gpio_table = { +static struct gpiod_lookup_table nokia800_mmc_gpio_table = { .dev_id ="mmci-omap.0" , .table = {/* Slot switch, GPIO 96 */
@@ -152,6 +152,8 @@ static struct gpiod_lookup_table nokia8xx_mmc_gpio_table = {static struct gpiod_lookup_table nokia810_mmc_gpio_table = { .dev_id ="mmci-omap.0" , .table = { +/* Slot switch, GPIO 96 */ +GPIO_LOOKUP ("gpio-96-127" ,0 ,"switch" ,GPIO_ACTIVE_HIGH ),/* Slot index 1, VSD power, GPIO 23 */ GPIO_LOOKUP_IDX ("gpio-0-31" ,23 ,"vsd" ,1 ,GPIO_ACTIVE_HIGH ),/* Slot index 1, VIO power, GPIO 9 */
@@ -412,8 +414,6 @@ static struct omap_mmc_platform_data *mmc_data[OMAP24XX_NR_MMC];static void __init n8x0_mmc_init (void ) { -gpiod_add_lookup_table (&nokia8xx_mmc_gpio_table); -if (board_is_n810 ()) { mmc1_data.slots[0 ].name ="external" ;
@@ -426,6 +426,8 @@ static void __init n8x0_mmc_init(void) mmc1_data.slots[1 ].name ="internal" ; mmc1_data.slots[1 ].ban_openended =1 ;gpiod_add_lookup_table (&nokia810_mmc_gpio_table); + }else { +gpiod_add_lookup_table (&nokia800_mmc_gpio_table); } mmc1_data.nr_slots =2 ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 22/172 mmc: omap: fix broken slot switch lookup
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aaro Koskinen <[email protected]>
[ Upstream commit d4debbcbffa45c3de5df0040af2eea74a9e794a3 ]
The lookup is done before host->dev is initialized. It will always just fail silently, and the MMC behaviour is totally unpredictable as the switch is left in an undefined state. Fix that.
Message-ID: <[email protected]>
--- drivers/mmc/host/omap.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-)
drivers/mmc/host/omap.c +6 -7
@@ -1384,13 +1384,6 @@ static int mmc_omap_probe(struct platform_device *pdev)if (IS_ERR (host->virt_base))return PTR_ERR (host->virt_base); - host->slot_switch =gpiod_get_optional (host->dev,"switch" , -GPIOD_OUT_LOW ); -if (IS_ERR (host->slot_switch)) -return dev_err_probe (host->dev,PTR_ERR (host->slot_switch), -"error looking up slot switch GPIO\n" ); - -INIT_WORK (&host->slot_release_work, mmc_omap_slot_release_work);INIT_WORK (&host->send_stop_work, mmc_omap_send_stop_work);
@@ -1409,6 +1402,12 @@ static int mmc_omap_probe(struct platform_device *pdev) host->dev = &pdev->dev;platform_set_drvdata (pdev, host); + host->slot_switch =gpiod_get_optional (host->dev,"switch" , +GPIOD_OUT_LOW ); +if (IS_ERR (host->slot_switch)) +return dev_err_probe (host->dev,PTR_ERR (host->slot_switch), +"error looking up slot switch GPIO\n" ); + host->id = pdev->id; host->irq = irq; host->phys_base = res->start;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 23/172 mmc: omap: fix deferred probe
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aaro Koskinen <[email protected]>
[ Upstream commit f6862c7f156d04f81c38467e1c304b7e9517e810 ]
After a deferred probe, GPIO descriptor lookup will fail with EBUSY. Fix by using managed descriptors.
Message-ID: <[email protected]>
--- drivers/mmc/host/omap.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/mmc/host/omap.c +8 -8
@@ -1259,18 +1259,18 @@ static int mmc_omap_new_slot(struct mmc_omap_host *host, int id) slot->pdata = &host->pdata->slots[id];/* Check for some optional GPIO controls */ - slot->vsd =gpiod_get_index_optional (host->dev,"vsd" , - id,GPIOD_OUT_LOW ); + slot->vsd =devm_gpiod_get_index_optional (host->dev,"vsd" , + id,GPIOD_OUT_LOW );if (IS_ERR (slot->vsd))return dev_err_probe (host->dev,PTR_ERR (slot->vsd),"error looking up VSD GPIO\n" ); - slot->vio =gpiod_get_index_optional (host->dev,"vio" , - id,GPIOD_OUT_LOW ); + slot->vio =devm_gpiod_get_index_optional (host->dev,"vio" , + id,GPIOD_OUT_LOW );if (IS_ERR (slot->vio))return dev_err_probe (host->dev,PTR_ERR (slot->vio),"error looking up VIO GPIO\n" ); - slot->cover =gpiod_get_index_optional (host->dev,"cover" , - id,GPIOD_IN ); + slot->cover =devm_gpiod_get_index_optional (host->dev,"cover" , + id,GPIOD_IN );if (IS_ERR (slot->cover))return dev_err_probe (host->dev,PTR_ERR (slot->cover),"error looking up cover switch GPIO\n" );
@@ -1402,8 +1402,8 @@ static int mmc_omap_probe(struct platform_device *pdev) host->dev = &pdev->dev;platform_set_drvdata (pdev, host); - host->slot_switch =gpiod_get_optional (host->dev,"switch" , -GPIOD_OUT_LOW ); + host->slot_switch =devm_gpiod_get_optional (host->dev,"switch" , +GPIOD_OUT_LOW );if (IS_ERR (host->slot_switch))return dev_err_probe (host->dev,PTR_ERR (host->slot_switch),"error looking up slot switch GPIO\n" );
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 24/172 mmc: omap: restore original power up/down steps
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aaro Koskinen <[email protected]>
[ Upstream commit 894ad61b85d6ba8efd4274aa8719d9ff1c89ea54 ]
Commit e519f0bb64ef ("ARM/mmc: Convert old mmci-omap to GPIO descriptors") moved Nokia N810 MMC power up/down from the board file into the MMC driver.
The change removed some delays, and ordering without a valid reason. Restore power up/down to match the original code. This matters only on N810 where the 2nd GPIO is in use. Other boards will see an additional delay but that should be a lesser concern than omitting delays altogether.
Message-ID: <[email protected]>
--- drivers/mmc/host/omap.c | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-)
drivers/mmc/host/omap.c +19 -4
@@ -1119,10 +1119,25 @@ static void mmc_omap_set_power(struct mmc_omap_slot *slot, int power_on, host = slot->host; -if (slot->vsd) -gpiod_set_value (slot->vsd, power_on); -if (slot->vio) -gpiod_set_value (slot->vio, power_on); +if (power_on) { +if (slot->vsd) { +gpiod_set_value (slot->vsd, power_on); +msleep (1 ); + } +if (slot->vio) { +gpiod_set_value (slot->vio, power_on); +msleep (1 ); + } + }else { +if (slot->vio) { +gpiod_set_value (slot->vio, power_on); +msleep (50 ); + } +if (slot->vsd) { +gpiod_set_value (slot->vsd, power_on); +msleep (50 ); + } + }if (slot->pdata->set_power !=NULL ) slot->pdata->set_power(mmc_dev(slot->mmc), slot->id, power_on,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 25/172 ARM: OMAP2+: fix USB regression on Nokia N8x0
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aaro Koskinen <[email protected]>
[ Upstream commit 4421405e3634a3189b541cf1e34598e44260720d ]
GPIO chip labels are wrong for OMAP2, so the USB does not work. Fix.
Message-ID: <[email protected]>
--- arch/arm/mach-omap2/board-n8x0.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
arch/arm/mach-omap2/board-n8x0.c +2 -4
@@ -79,10 +79,8 @@ static struct musb_hdrc_platform_data tusb_data = {static struct gpiod_lookup_table tusb_gpio_table = { .dev_id ="musb-tusb" , .table = { -GPIO_LOOKUP ("gpio-0-15" ,0 ,"enable" , -GPIO_ACTIVE_HIGH ), -GPIO_LOOKUP ("gpio-48-63" ,10 ,"int" , -GPIO_ACTIVE_HIGH ), +GPIO_LOOKUP ("gpio-0-31" ,0 ,"enable" ,GPIO_ACTIVE_HIGH ), +GPIO_LOOKUP ("gpio-32-63" ,26 ,"int" ,GPIO_ACTIVE_HIGH ), { } }, };
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 26/172 firmware: arm_ffa: Fix the partition ID check in ffa_notification_info_get()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Wiklander <[email protected]>
[ Upstream commit 1a4bd2b128fb5ca62e4d1c5ca298d3d06b9c1e8e ]
FFA_NOTIFICATION_INFO_GET retrieves information about pending notifications. Notifications can be either global or per VCPU. Global notifications are reported with the partition ID only in the list of endpoints with pending notifications. ffa_notification_info_get() incorrectly expect no ID at all for global notifications. Fix this by checking for ID = 1 instead of ID = 0.
--- drivers/firmware/arm_ffa/driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/firmware/arm_ffa/driver.c +1 -1
@@ -790,7 +790,7 @@ static void ffa_notification_info_get(void) part_id = packed_id_list[ids_processed++]; -if (!ids_count[list]) {/* Global Notification */ +if (ids_count[list] ==1 ) {/* Global Notification */ __do_sched_recv_cb (part_id,0 , false);continue ; }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 27/172 firmware: arm_scmi: Make raw debugfs entries non-seekable
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Cristian Marussi <[email protected]>
[ Upstream commit b70c7996d4ffb2e02895132e8a79a37cee66504f ]
SCMI raw debugfs entries are used to inject and snoop messages out of the SCMI core and, as such, the underlying virtual files have no reason to support seeking.
Modify the related file_operations descriptors to be non-seekable.
--- drivers/firmware/arm_scmi/raw_mode.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/firmware/arm_scmi/raw_mode.c +6 -1
@@ -921,7 +921,7 @@ static int scmi_dbg_raw_mode_open(struct inode *inode, struct file *filp) rd->raw = raw; filp->private_data = rd; -return 0 ; +return nonseekable_open (inode, filp); }static int scmi_dbg_raw_mode_release (struct inode *inode,struct file *filp)
@@ -950,6 +950,7 @@ static const struct file_operations scmi_dbg_raw_mode_reset_fops = { .open = scmi_dbg_raw_mode_open, .release = scmi_dbg_raw_mode_release, .write = scmi_dbg_raw_mode_reset_write, + .llseek = no_llseek, .owner = THIS_MODULE, };
@@ -959,6 +960,7 @@ static const struct file_operations scmi_dbg_raw_mode_message_fops = { .read = scmi_dbg_raw_mode_message_read, .write = scmi_dbg_raw_mode_message_write, .poll = scmi_dbg_raw_mode_message_poll, + .llseek = no_llseek, .owner = THIS_MODULE, };
@@ -975,6 +977,7 @@ static const struct file_operations scmi_dbg_raw_mode_message_async_fops = { .read = scmi_dbg_raw_mode_message_read, .write = scmi_dbg_raw_mode_message_async_write, .poll = scmi_dbg_raw_mode_message_poll, + .llseek = no_llseek, .owner = THIS_MODULE, };
@@ -998,6 +1001,7 @@ static const struct file_operations scmi_dbg_raw_mode_notification_fops = { .release = scmi_dbg_raw_mode_release, .read = scmi_test_dbg_raw_mode_notif_read, .poll = scmi_test_dbg_raw_mode_notif_poll, + .llseek = no_llseek, .owner = THIS_MODULE, };
@@ -1021,6 +1025,7 @@ static const struct file_operations scmi_dbg_raw_mode_errors_fops = { .release = scmi_dbg_raw_mode_release, .read = scmi_test_dbg_raw_mode_errors_read, .poll = scmi_test_dbg_raw_mode_errors_poll, + .llseek = no_llseek, .owner = THIS_MODULE, };
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 28/172 cxl/mem: Fix for the index of Clear Event Record Handle
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yuquan Wang <[email protected]>
[ Upstream commit b7c59b038c656214f56432867056997c2e0fc268 ]
The dev_dbg info for Clear Event Records mailbox command would report the handle of the next record to clear not the current one.
This was because the index 'i' had incremented before printing the current handle value.
--- drivers/cxl/core/mbox.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/cxl/core/mbox.c +1 -1
@@ -915,7 +915,7 @@ static int cxl_clear_event_record(struct cxl_memdev_state *mds, payload->handles[i++] = gen->hdr.handle;dev_dbg (mds->cxlds.dev,"Event log '%d': Clearing %u\n" , log, -le16_to_cpu (payload->handles[i])); +le16_to_cpu (payload->handles[i -1 ]));if (i == max_handles) { payload->nr_recs = i;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 29/172 cxl/core/regs: Fix usage of map->reg_type in cxl_decode_regblock() before assigned
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dave Jiang <[email protected]>
[ Upstream commit 5c88a9ccd4c431d58b532e4158b6999a8350062c ]
In the error path, map->reg_type is being used for kernel warning before its value is setup. Found by code inspection. Exposure to user is wrong reg_type being emitted via kernel log. Use a local var for reg_type and retrieve value for usage.
--- drivers/cxl/core/regs.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/cxl/core/regs.c +3 -2
@@ -271,6 +271,7 @@ EXPORT_SYMBOL_NS_GPL(cxl_map_device_regs, CXL);static bool cxl_decode_regblock (struct pci_dev *pdev,u32 reg_lo,u32 reg_hi,struct cxl_register_map *map) { +u8 reg_type =FIELD_GET (CXL_DVSEC_REG_LOCATOR_BLOCK_ID_MASK , reg_lo);int bar =FIELD_GET (CXL_DVSEC_REG_LOCATOR_BIR_MASK , reg_lo);u64 offset = ((u64 )reg_hi <<32 ) | (reg_lo &CXL_DVSEC_REG_LOCATOR_BLOCK_OFF_LOW_MASK );
@@ -278,11 +279,11 @@ static bool cxl_decode_regblock(struct pci_dev *pdev, u32 reg_lo, u32 reg_hi,if (offset >pci_resource_len (pdev, bar)) {dev_warn (&pdev->dev,"BAR%d: %pr: too small (offset: %pa, type: %d)\n" , bar, - &pdev->resource[bar], &offset, map->reg_type); + &pdev->resource[bar], &offset, reg_type);return false; } - map->reg_type =FIELD_GET (CXL_DVSEC_REG_LOCATOR_BLOCK_ID_MASK , reg_lo); + map->reg_type = reg_type; map->resource =pci_resource_start (pdev, bar) + offset; map->max_size =pci_resource_len (pdev, bar) - offset;return true;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 30/172 arm64: dts: freescale: imx8mp-venice-gw72xx-2x: fix USB vbus regulator
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tim Harvey <[email protected]>
[ Upstream commit 8cb10cba124c4798b6cb333245ecdc8dde78aeae ]
When using usb-conn-gpio to control USB role and VBUS, the vbus-supply property must be present in the usb-conn-gpio node. Additionally it should not be present in the phy node as that isn't what controls vbus and will upset the use count.
This resolves an issue where VBUS is enabled with OTG in peripheral mode.
--- arch/arm64/boot/dts/freescale/imx8mp-venice-gw72xx.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/freescale/imx8mp-venice-gw72xx.dtsi +1 -1
@@ -14,6 +14,7 @@ connector { pinctrl-0 = <&pinctrl_usbcon1>; type = "micro"; label = "otg"; + vbus-supply = <®_usb1_vbus>; id-gpios = <&gpio3 21 GPIO_ACTIVE_HIGH>; port {
@@ -183,7 +184,6 @@ &usb3_0 { }; &usb3_phy0 { - vbus-supply = <®_usb1_vbus>; status = "okay"; };
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 31/172 arm64: dts: freescale: imx8mp-venice-gw73xx-2x: fix USB vbus regulator
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tim Harvey <[email protected]>
[ Upstream commit 6f8e0aca838e163e81fde176e945161d50679339 ]
When using usb-conn-gpio to control USB role and VBUS, the vbus-supply property must be present in the usb-conn-gpio node. Additionally it should not be present in the phy node as that isn't what controls vbus and will upset the use count.
This resolves an issue where VBUS is enabled with OTG in peripheral mode.
--- arch/arm64/boot/dts/freescale/imx8mp-venice-gw73xx.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/freescale/imx8mp-venice-gw73xx.dtsi +1 -1
@@ -14,6 +14,7 @@ connector { pinctrl-0 = <&pinctrl_usbcon1>; type = "micro"; label = "otg"; + vbus-supply = <®_usb1_vbus>; id-gpios = <&gpio3 21 GPIO_ACTIVE_HIGH>; port {
@@ -202,7 +203,6 @@ &usb3_0 { }; &usb3_phy0 { - vbus-supply = <®_usb1_vbus>; status = "okay"; };
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 32/172 drm/msm: Add newlines to some debug prints
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stephen Boyd <[email protected]>
[ Upstream commit c588f7d67044d6d59ef92d75a970b64929984d89 ]
These debug prints are missing newlines, leading to multiple messages being printed on one line and hard to read logs. Add newlines to have the debug prints on separate lines. The DBG macro used to add a newline, but I missed that while migrating to drm_dbg wrappers.
Patchwork: https://patchwork.freedesktop.org/patch/584769/
--- drivers/gpu/drm/msm/msm_fb.c | 6 +++--- drivers/gpu/drm/msm/msm_kms.c | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-)
drivers/gpu/drm/msm/msm_fb.c +3 -3
@@ -89,7 +89,7 @@ int msm_framebuffer_prepare(struct drm_framebuffer *fb,for (i =0 ; i < n; i++) { ret =msm_gem_get_and_pin_iova (fb->obj[i], aspace, &msm_fb->iova[i]); -drm_dbg_state (fb->dev,"FB[%u]: iova[%d]: %08llx (%d)" , +drm_dbg_state (fb->dev,"FB[%u]: iova[%d]: %08llx (%d)\n" , fb->base.id, i, msm_fb->iova[i], ret);if (ret)return ret;
@@ -176,7 +176,7 @@ static struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev,const struct msm_format *format;int ret, i, n; -drm_dbg_state (dev,"create framebuffer: mode_cmd=%p (%dx%d@%4.4s)" , +drm_dbg_state (dev,"create framebuffer: mode_cmd=%p (%dx%d@%4.4s)\n" , mode_cmd, mode_cmd->width, mode_cmd->height, (char *)&mode_cmd->pixel_format);
@@ -232,7 +232,7 @@ static struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev,refcount_set (&msm_fb->dirtyfb,1 ); -drm_dbg_state (dev,"create: FB ID: %d (%p)" , fb->base.id, fb); +drm_dbg_state (dev,"create: FB ID: %d (%p)\n" , fb->base.id, fb);return fb;
drivers/gpu/drm/msm/msm_kms.c +2 -2
@@ -149,7 +149,7 @@ int msm_crtc_enable_vblank(struct drm_crtc *crtc)struct msm_kms *kms = priv->kms;if (!kms)return -ENXIO ; -drm_dbg_vbl (dev,"crtc=%u" , crtc->base.id); +drm_dbg_vbl (dev,"crtc=%u\n" , crtc->base.id);return vblank_ctrl_queue_work (priv, crtc, true); }
@@ -160,7 +160,7 @@ void msm_crtc_disable_vblank(struct drm_crtc *crtc)struct msm_kms *kms = priv->kms;if (!kms)return ; -drm_dbg_vbl (dev,"crtc=%u" , crtc->base.id); +drm_dbg_vbl (dev,"crtc=%u\n" , crtc->base.id);vblank_ctrl_queue_work (priv, crtc, false); }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 33/172 drm/msm/dpu: dont allow overriding data from catalog
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dmitry Baryshkov <[email protected]>
[ Upstream commit 4f3b77ae5ff5b5ba9d99c5d5450db388dbee5107 ]
The data from catalog is marked as const, so it is a part of the RO segment. Allowing userspace to write to it through debugfs can cause protection faults. Set debugfs file mode to read-only for debug entries corresponding to perf_cfg coming from catalog.
Patchwork: https://patchwork.freedesktop.org/patch/582844/
--- drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c +5 -5
@@ -459,15 +459,15 @@ int dpu_core_perf_debugfs_init(struct dpu_kms *dpu_kms, struct dentry *parent) &perf->core_clk_rate);debugfs_create_u32 ("enable_bw_release" ,0600 , entry, (u32 *)&perf->enable_bw_release); -debugfs_create_u32 ("threshold_low" ,0600 , entry, +debugfs_create_u32 ("threshold_low" ,0400 , entry, (u32 *)&perf->perf_cfg->max_bw_low); -debugfs_create_u32 ("threshold_high" ,0600 , entry, +debugfs_create_u32 ("threshold_high" ,0400 , entry, (u32 *)&perf->perf_cfg->max_bw_high); -debugfs_create_u32 ("min_core_ib" ,0600 , entry, +debugfs_create_u32 ("min_core_ib" ,0400 , entry, (u32 *)&perf->perf_cfg->min_core_ib); -debugfs_create_u32 ("min_llcc_ib" ,0600 , entry, +debugfs_create_u32 ("min_llcc_ib" ,0400 , entry, (u32 *)&perf->perf_cfg->min_llcc_ib); -debugfs_create_u32 ("min_dram_ib" ,0600 , entry, +debugfs_create_u32 ("min_dram_ib" ,0400 , entry, (u32 *)&perf->perf_cfg->min_dram_ib);debugfs_create_file ("perf_mode" ,0600 , entry, (u32 *)perf, &dpu_core_perf_mode_fops);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 34/172 drm/msm/dpu: make error messages at dpu_core_irq_register_callback() more sensible
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dmitry Baryshkov <[email protected]>
[ Upstream commit 8844f467d6a58dc915f241e81c46e0c126f8c070 ]
There is little point in using %ps to print a value known to be NULL. On the other hand it makes sense to print the callback symbol in the 'invalid IRQ' message. Correct those two error messages to make more sense.
Patchwork: https://patchwork.freedesktop.org/patch/585565/
--- drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c +4 -4
@@ -525,14 +525,14 @@ int dpu_core_irq_register_callback(struct dpu_kms *dpu_kms,int ret;if (!irq_cb) { -DPU_ERROR ("invalid IRQ=[%d, %d] irq_cb:%ps\n" , -DPU_IRQ_REG (irq_idx),DPU_IRQ_BIT (irq_idx), irq_cb); +DPU_ERROR ("IRQ=[%d, %d] NULL callback\n" , +DPU_IRQ_REG (irq_idx),DPU_IRQ_BIT (irq_idx));return -EINVAL ; }if (!dpu_core_irq_is_valid (irq_idx)) { -DPU_ERROR ("invalid IRQ=[%d, %d]\n" , -DPU_IRQ_REG (irq_idx),DPU_IRQ_BIT (irq_idx)); +DPU_ERROR ("invalid IRQ=[%d, %d] irq_cb:%ps\n" , +DPU_IRQ_REG (irq_idx),DPU_IRQ_BIT (irq_idx), irq_cb);return -EINVAL ; }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 35/172 dt-bindings: display/msm: sm8150-mdss: add DP node
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dmitry Baryshkov <[email protected]>
[ Upstream commit be1b7acb929137e3943fe380671242beb485190c ]
As Qualcomm SM8150 got support for the DisplayPort, add displayport@ node as a valid child to the MDSS node.
Patchwork: https://patchwork.freedesktop.org/patch/586156/
--- .../bindings/display/msm/qcom,sm8150-mdss.yaml | 9 +++++++++ 1 file changed, 9 insertions(+)
Documentation/devicetree/bindings/display/msm/qcom,sm8150-mdss.yaml +9 -0
@@ -53,6 +53,15 @@ patternProperties: compatible: const: qcom,sm8150-dpu + "^displayport-controller@[0-9a-f]+$": + type: object + additionalProperties: true + + properties: + compatible: + contains: + const: qcom,sm8150-dp + "^dsi@[0-9a-f]+$": type: object additionalProperties: true
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 36/172 arm64: dts: imx8-ss-conn: fix usdhc wrong lpcg clock order
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Frank Li <[email protected]>
[ Upstream commit c6ddd6e7b166532a0816825442ff60f70aed9647 ]
The actual clock show wrong frequency:
echo on >/sys/devices/platform/bus\@5b000000/5b010000.mmc/power/control cat /sys/kernel/debug/mmc0/ios
clock: 200000000 Hz
actual clock: 166000000 Hz
^^^^^^^^^
.....
According to
sdhc0_lpcg: clock-controller@5b200000 {
compatible = "fsl,imx8qxp-lpcg";
reg = <0x5b200000 0x10000>;
#clock-cells = <1>;
clocks = <&clk IMX_SC_R_SDHC_0 IMX_SC_PM_CLK_PER>,
<&conn_ipg_clk>, <&conn_axi_clk>;
clock-indices = <IMX_LPCG_CLK_0>, <IMX_LPCG_CLK_4>,
<IMX_LPCG_CLK_5>;
clock-output-names = "sdhc0_lpcg_per_clk",
"sdhc0_lpcg_ipg_clk",
"sdhc0_lpcg_ahb_clk";
power-domains = <&pd IMX_SC_R_SDHC_0>;
}"per_clk" should be IMX_LPCG_CLK_0 instead of IMX_LPCG_CLK_5.
After correct clocks order:
echo on >/sys/devices/platform/bus\@5b000000/5b010000.mmc/power/control cat /sys/kernel/debug/mmc0/ios
clock: 200000000 Hz
actual clock: 198000000 Hz
^^^^^^^^
...
--- arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi +6 -6
@@ -67,8 +67,8 @@ usdhc1: mmc@5b010000 { interrupts = <GIC_SPI 232 IRQ_TYPE_LEVEL_HIGH>; reg = <0x5b010000 0x10000>; clocks = <&sdhc0_lpcg IMX_LPCG_CLK_4>, - <&sdhc0_lpcg IMX_LPCG_CLK_0>, - <&sdhc0_lpcg IMX_LPCG_CLK_5>; + <&sdhc0_lpcg IMX_LPCG_CLK_5>, + <&sdhc0_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "ahb", "per"; power-domains = <&pd IMX_SC_R_SDHC_0>; status = "disabled";
@@ -78,8 +78,8 @@ usdhc2: mmc@5b020000 { interrupts = <GIC_SPI 233 IRQ_TYPE_LEVEL_HIGH>; reg = <0x5b020000 0x10000>; clocks = <&sdhc1_lpcg IMX_LPCG_CLK_4>, - <&sdhc1_lpcg IMX_LPCG_CLK_0>, - <&sdhc1_lpcg IMX_LPCG_CLK_5>; + <&sdhc1_lpcg IMX_LPCG_CLK_5>, + <&sdhc1_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "ahb", "per"; power-domains = <&pd IMX_SC_R_SDHC_1>; fsl,tuning-start-tap = <20>;
@@ -91,8 +91,8 @@ usdhc3: mmc@5b030000 { interrupts = <GIC_SPI 234 IRQ_TYPE_LEVEL_HIGH>; reg = <0x5b030000 0x10000>; clocks = <&sdhc2_lpcg IMX_LPCG_CLK_4>, - <&sdhc2_lpcg IMX_LPCG_CLK_0>, - <&sdhc2_lpcg IMX_LPCG_CLK_5>; + <&sdhc2_lpcg IMX_LPCG_CLK_5>, + <&sdhc2_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "ahb", "per"; power-domains = <&pd IMX_SC_R_SDHC_2>; status = "disabled";
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 37/172 cxl/core: Fix initialization of mbox_cmd.size_out in get event
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kwangjin Ko <[email protected]>
[ Upstream commit f7c52345ccc96343c0a05bdea3121c8ac7b67d5f ]
Since mbox_cmd.size_out is overwritten with the actual output size in the function below, it needs to be initialized every time.
cxl_internal_send_cmd -> __cxl_pci_mbox_send_cmd
Problem scenario:
1) The size_out variable is initially set to the size of the mailbox.
2) Read an event.
- size_out is set to 160 bytes(header 32B + one event 128B).
- Two event are created while reading.
3) Read the new *two* events.
- size_out is still set to 160 bytes.
- Although the value of out_len is 288 bytes, only 160 bytes are copied from the mailbox register to the local variable.
- record_count is set to 2.
- Accessing records[1] will result in reading incorrect data.
--- drivers/cxl/core/mbox.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/cxl/core/mbox.c +2 -1
@@ -958,13 +958,14 @@ static void cxl_mem_get_records_log(struct cxl_memdev_state *mds, .payload_in = &log_type, .size_in = sizeof(log_type), .payload_out = payload, - .size_out = mds->payload_size, .min_out = struct_size(payload, records, 0), }; do { int rc, i; + mbox_cmd.size_out = mds->payload_size; + rc = cxl_internal_send_cmd(mds, &mbox_cmd); if (rc) { dev_err_ratelimited(dev,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 38/172 Revert "drm/qxl: simplify qxl_fence_wait"
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Constantino <[email protected]>
[ Upstream commit 07ed11afb68d94eadd4ffc082b97c2331307c5ea ]
This reverts commit 5a838e5d5825c85556011478abde708251cc0776.
Changes from commit 5a838e5d5825 ("drm/qxl: simplify qxl_fence_wait") would result in a '[TTM] Buffer eviction failed' exception whenever it reached a timeout.
Due to a dependency to DMA_FENCE_WARN this also restores some code deleted by commit d72277b6c37d ("dma-buf: nuke DMA_FENCE_TRACE macros v2").
--- drivers/gpu/drm/qxl/qxl_release.c | 50 +++++++++++++++++++++++++++---- include/linux/dma-fence.h | 7 +++++ 2 files changed, 52 insertions(+), 5 deletions(-)
drivers/gpu/drm/qxl/qxl_release.c +45 -5
@@ -58,16 +58,56 @@ static long qxl_fence_wait(struct dma_fence *fence, bool intr, signed long timeout) {struct qxl_device *qdev; +struct qxl_release *release; +int count =0 , sc =0 ; +bool have_drawable_releases;unsigned long cur, end = jiffies + timeout; qdev = container_of(fence->lock, struct qxl_device, release_lock); + release = container_of(fence, struct qxl_release, base); + have_drawable_releases = release->type ==QXL_RELEASE_DRAWABLE ; -if (!wait_event_timeout (qdev->release_event, - (dma_fence_is_signaled (fence) || - (qxl_io_notify_oom (qdev),0 )), - timeout)) -return 0 ; +retry: + sc++; + +if (dma_fence_is_signaled (fence)) + goto signaled; + +qxl_io_notify_oom (qdev); + +for (count =0 ; count <11 ; count++) { +if (!qxl_queue_garbage_collect (qdev, true)) +break ; + +if (dma_fence_is_signaled (fence)) + goto signaled; + } + +if (dma_fence_is_signaled (fence)) + goto signaled; + +if (have_drawable_releases || sc <4 ) { +if (sc >2 ) +/* back off */ +usleep_range (500 ,1000 ); + +if (time_after (jiffies, end)) +return 0 ; + +if (have_drawable_releases && sc >300 ) { +DMA_FENCE_WARN (fence, +"failed to wait on release %llu after spincount %d\n" , + fence->context & ~0xf0000000 , sc); + goto signaled; + } + goto retry; + } +/* +* yeah, original sync_obj_wait gave up after 3 spins when +* have_drawable_releases is not set. +*/ +signaled: cur = jiffies;if (time_after (cur, end))return 0 ;
include/linux/dma-fence.h +7 -0
@@ -682,4 +682,11 @@ static inline bool dma_fence_is_container(struct dma_fence *fence)return dma_fence_is_array (fence) ||dma_fence_is_chain (fence); } +#define DMA_FENCE_WARN(f, fmt, args...) \ + do { \ + struct dma_fence *__ff = (f); \ + pr_warn("f %llu#%llu: " fmt, __ff->context, __ff->seqno,\ + ##args); \ + } while (0) +#endif /* __LINUX_DMA_FENCE_H */
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 39/172 nouveau: fix function cast warning
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit 185fdb4697cc9684a02f2fab0530ecdd0c2f15d4 ]
Calling a function through an incompatible pointer type causes breaks kcfi, so clang warns about the assignment:
drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowof.c:73:10: error: cast from 'void (*)(const void *)' to 'void (*)(void *)' converts to incompatible function type [-Werror,-Wcast-function-type-strict]
73 | .fini = (void(*)(void *))kfree,
Avoid this with a trivial wrapper.
--- drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowof.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/gpu/drm/nouveau/nvkm/subdev/bios/shadowof.c +6 -1
@@ -66,11 +66,16 @@ of_init(struct nvkm_bios *bios, const char *name)return ERR_PTR (-EINVAL ); } +static void of_fini (void *p) +{ +kfree (p); +} +const struct nvbios_source nvbios_of = { .name ="OpenFirmware" , .init = of_init, - .fini = (void (*)(void *))kfree, + .fini = of_fini, .read = of_read, .size = of_size, .rw = false,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 40/172 drm/msm/adreno: Set highest_bank_bit for A619
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luca Weiss <[email protected]>
[ Upstream commit 9dc23cba0927d09cb481da064c8413eb9df42e2b ]
The default highest_bank_bit of 15 didn't seem to cause issues so far but downstream defines it to be 14. But similar to [0] leaving it on 14 (or 15 for that matter) causes some corruption issues with some resolutions with DisplayPort, like 1920x1200.
So set it to 13 for now so that there's no screen corruption.
[0] commit 6a0dbcd20ef2 ("drm/msm/a6xx: set highest_bank_bit to 13 for a610")
Patchwork: https://patchwork.freedesktop.org/patch/585215/
--- drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/gpu/drm/msm/adreno/a6xx_gpu.c +4 -0
@@ -1295,6 +1295,10 @@ static void a6xx_calc_ubwc_config(struct adreno_gpu *gpu)if (adreno_is_a618 (gpu)) gpu->ubwc_config.highest_bank_bit =14 ; +if (adreno_is_a619 (gpu)) +/* TODO: Should be 14 but causes corruption at e.g. 1920x1200 on DP */ + gpu->ubwc_config.highest_bank_bit =13 ; +if (adreno_is_a619_holi (gpu)) gpu->ubwc_config.highest_bank_bit =13 ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 41/172 scsi: hisi_sas: Modify the deadline for ata_wait_after_reset()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xiang Chen <[email protected]>
[ Upstream commit 0098c55e0881f0b32591f2110410d5c8b7f9bd5a ]
We found that the second parameter of function ata_wait_after_reset() is incorrectly used. We call smp_ata_check_ready_type() to poll the device type until the 30s timeout, so the correct deadline should be (jiffies + 30000).
--- drivers/scsi/hisi_sas/hisi_sas_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/hisi_sas/hisi_sas_main.c +1 -1
@@ -1792,7 +1792,7 @@ static int hisi_sas_debug_I_T_nexus_reset(struct domain_device *device)if (dev_is_sata (device)) {struct ata_link *link = &device->sata_dev.ap->link; - rc =ata_wait_after_reset (link,HISI_SAS_WAIT_PHYUP_TIMEOUT , + rc =ata_wait_after_reset (link, jiffies +HISI_SAS_WAIT_PHYUP_TIMEOUT , smp_ata_check_ready_type); }else {msleep (2000 );
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 42/172 scsi: qla2xxx: Fix off by one in qla_edif_app_getstats()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dan Carpenter <[email protected]>
[ Upstream commit 4406e4176f47177f5e51b4cc7e6a7a2ff3dbfbbd ]
The app_reply->elem[] array is allocated earlier in this function and it has app_req.num_ports elements. Thus this > comparison needs to be >= to prevent memory corruption.
--- drivers/scsi/qla2xxx/qla_edif.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/qla2xxx/qla_edif.c +1 -1
@@ -1100,7 +1100,7 @@ qla_edif_app_getstats(scsi_qla_host_t *vha, struct bsg_job *bsg_job)list_for_each_entry_safe (fcport, tf, &vha->vp_fcports, list) {if (fcport->edif.enable) { -if (pcnt > app_req.num_ports) +if (pcnt >= app_req.num_ports)break ; app_reply->elem[pcnt].rekey_count =
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 43/172 net: openvswitch: fix unwanted error log on timeout policy probing
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ilya Maximets <[email protected]>
[ Upstream commit 4539f91f2a801c0c028c252bffae56030cfb2cae ]
On startup, ovs-vswitchd probes different datapath features including support for timeout policies. While probing, it tries to execute certain operations with OVS_PACKET_ATTR_PROBE or OVS_FLOW_ATTR_PROBE attributes set. These attributes tell the openvswitch module to not log any errors when they occur as it is expected that some of the probes will fail.
For some reason, setting the timeout policy ignores the PROBE attribute and logs a failure anyway. This is causing the following kernel log on each re-start of ovs-vswitchd:
kernel: Failed to associated timeout policy `ovs_test_tp'
Fix that by using the same logging macro that all other messages are using. The message will still be printed at info level when needed and will be rate limited, but with a net rate limiter instead of generic printk one.
The nf_ct_set_timeout() itself will still print some info messages, but at least this change makes logging in openvswitch module more consistent.
--- net/openvswitch/conntrack.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
net/openvswitch/conntrack.c +3 -2
@@ -1380,8 +1380,9 @@ int ovs_ct_copy_action(struct net *net, const struct nlattr *attr,if (ct_info.timeout[0 ]) {if (nf_ct_set_timeout (net, ct_info.ct, family, key->ip.proto, ct_info.timeout)) -pr_info_ratelimited ("Failed to associated timeout " -"policy `%s'\n" , ct_info.timeout); +OVS_NLERR (log, +"Failed to associated timeout policy '%s'" , + ct_info.timeout);else ct_info.nf_ct_timeout =rcu_dereference (nf_ct_timeout_find (ct_info.ct)->timeout);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 44/172 u64_stats: fix u64_stats_init() for lockdep when used repeatedly in one file
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Petr Tesarik <[email protected]>
[ Upstream commit 38a15d0a50e0a43778561a5861403851f0b0194c ]
Fix bogus lockdep warnings if multiple u64_stats_sync variables are initialized in the same file.
With CONFIG_LOCKDEP, seqcount_init() is a macro which declares:
static struct lock_class_key __key;
Since u64_stats_init() is a function (albeit an inline one), all calls within the same file end up using the same instance, effectively treating them all as a single lock-class.
--- include/linux/u64_stats_sync.h | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
include/linux/u64_stats_sync.h +5 -4
@@ -135,10 +135,11 @@ static inline void u64_stats_inc(u64_stats_t *p) p->v++; } -static inline void u64_stats_init (struct u64_stats_sync *syncp) -{ -seqcount_init (&syncp->seq); -} +#define u64_stats_init (syncp) \ + do { \ + struct u64_stats_sync *__s = (syncp); \ + seqcount_init(&__s->seq); \ + } while (0)static inline void __u64_stats_update_begin (struct u64_stats_sync *syncp) {
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 45/172 xsk: validate user input for XDP_{UMEM|COMPLETION}_FILL_RING
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit 237f3cf13b20db183d3706d997eedc3c49eacd44 ]
syzbot reported an illegal copy in xsk_setsockopt() [1]
Make sure to validate setsockopt() @optlen parameter.
[1]
BUG: KASAN: slab-out-of-bounds in copy_from_sockptr_offset include/linux/sockptr.h:49 [inline] BUG: KASAN: slab-out-of-bounds in copy_from_sockptr include/linux/sockptr.h:55 [inline] BUG: KASAN: slab-out-of-bounds in xsk_setsockopt+0x909/0xa40 net/xdp/xsk.c:1420 Read of size 4 at addr ffff888028c6cde3 by task syz-executor.0/7549 CPU: 0 PID: 7549 Comm: syz-executor.0 Not tainted 6.8.0-syzkaller-08951-gfe46a7dd189e #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 Call Trace: <TASK> __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114 print_address_description mm/kasan/report.c:377 [inline] print_report+0x169/0x550 mm/kasan/report.c:488 kasan_report+0x143/0x180 mm/kasan/report.c:601 copy_from_sockptr_offset include/linux/sockptr.h:49 [inline] copy_from_sockptr include/linux/sockptr.h:55 [inline] xsk_setsockopt+0x909/0xa40 net/xdp/xsk.c:1420 do_sock_setsockopt+0x3af/0x720 net/socket.c:2311 __sys_setsockopt+0x1ae/0x250 net/socket.c:2334 __do_sys_setsockopt net/socket.c:2343 [inline] __se_sys_setsockopt net/socket.c:2340 [inline] __x64_sys_setsockopt+0xb5/0xd0 net/socket.c:2340 do_syscall_64+0xfb/0x240 entry_SYSCALL_64_after_hwframe+0x6d/0x75 RIP: 0033:0x7fb40587de69 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 e1 20 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b0 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007fb40665a0c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000036 RAX: ffffffffffffffda RBX: 00007fb4059abf80 RCX: 00007fb40587de69 RDX: 0000000000000005 RSI: 000000000000011b RDI: 0000000000000006 RBP: 00007fb4058ca47a R08: 0000000000000002 R09: 0000000000000000 R10: 0000000020001980 R11: 0000000000000246 R12: 0000000000000000 R13: 000000000000000b R14: 00007fb4059abf80 R15: 00007fff57ee4d08 </TASK> Allocated by task 7549: kasan_save_stack mm/kasan/common.c:47 [inline] kasan_save_track+0x3f/0x80 mm/kasan/common.c:68 poison_kmalloc_redzone mm/kasan/common.c:370 [inline] __kasan_kmalloc+0x98/0xb0 mm/kasan/common.c:387 kasan_kmalloc include/linux/kasan.h:211 [inline] __do_kmalloc_node mm/slub.c:3966 [inline] __kmalloc+0x233/0x4a0 mm/slub.c:3979 kmalloc include/linux/slab.h:632 [inline] __cgroup_bpf_run_filter_setsockopt+0xd2f/0x1040 kernel/bpf/cgroup.c:1869 do_sock_setsockopt+0x6b4/0x720 net/socket.c:2293 __sys_setsockopt+0x1ae/0x250 net/socket.c:2334 __do_sys_setsockopt net/socket.c:2343 [inline] __se_sys_setsockopt net/socket.c:2340 [inline] __x64_sys_setsockopt+0xb5/0xd0 net/socket.c:2340 do_syscall_64+0xfb/0x240 entry_SYSCALL_64_after_hwframe+0x6d/0x75 The buggy address belongs to the object at ffff888028c6cde0 which belongs to the cache kmalloc-8 of size 8
The buggy address is located 1 bytes to the right of
allocated 2-byte region [ffff888028c6cde0, ffff888028c6cde2)
The buggy address belongs to the physical page:
page:ffffea0000a31b00 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff888028c6c9c0 pfn:0x28c6c
anon flags: 0xfff00000000800(slab|node=0|zone=1|lastcpupid=0x7ff)
page_type: 0xffffffff()
raw: 00fff00000000800 ffff888014c41280 0000000000000000 dead000000000001
raw: ffff888028c6c9c0 0000000080800057 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 0, migratetype Unmovable, gfp_mask 0x112cc0(GFP_USER|__GFP_NOWARN|__GFP_NORETRY), pid 6648, tgid 6644 (syz-executor.0), ts 133906047828, free_ts 133859922223
set_page_owner include/linux/page_owner.h:31 [inline]
post_alloc_hook+0x1ea/0x210 mm/page_alloc.c:1533
prep_new_page mm/page_alloc.c:1540 [inline]
get_page_from_freelist+0x33ea/0x3580 mm/page_alloc.c:3311
__alloc_pages+0x256/0x680 mm/page_alloc.c:4569
__alloc_pages_node include/linux/gfp.h:238 [inline]
alloc_pages_node include/linux/gfp.h:261 [inline]
alloc_slab_page+0x5f/0x160 mm/slub.c:2175
allocate_slab mm/slub.c:2338 [inline]
new_slab+0x84/0x2f0 mm/slub.c:2391
___slab_alloc+0xc73/0x1260 mm/slub.c:3525
__slab_alloc mm/slub.c:3610 [inline]
__slab_alloc_node mm/slub.c:3663 [inline]
slab_alloc_node mm/slub.c:3835 [inline]
__do_kmalloc_node mm/slub.c:3965 [inline]
__kmalloc_node+0x2db/0x4e0 mm/slub.c:3973
kmalloc_node include/linux/slab.h:648 [inline]
__vmalloc_area_node mm/vmalloc.c:3197 [inline]
__vmalloc_node_range+0x5f9/0x14a0 mm/vmalloc.c:3392
__vmalloc_node mm/vmalloc.c:3457 [inline]
vzalloc+0x79/0x90 mm/vmalloc.c:3530
bpf_check+0x260/0x19010 kernel/bpf/verifier.c:21162
bpf_prog_load+0x1667/0x20f0 kernel/bpf/syscall.c:2895
__sys_bpf+0x4ee/0x810 kernel/bpf/syscall.c:5631
__do_sys_bpf kernel/bpf/syscall.c:5738 [inline]
__se_sys_bpf kernel/bpf/syscall.c:5736 [inline]
__x64_sys_bpf+0x7c/0x90 kernel/bpf/syscall.c:5736
do_syscall_64+0xfb/0x240
entry_SYSCALL_64_after_hwframe+0x6d/0x75
page last free pid 6650 tgid 6647 stack trace:
reset_page_owner include/linux/page_owner.h:24 [inline]
free_pages_prepare mm/page_alloc.c:1140 [inline]
free_unref_page_prepare+0x95d/0xa80 mm/page_alloc.c:2346
free_unref_page_list+0x5a3/0x850 mm/page_alloc.c:2532
release_pages+0x2117/0x2400 mm/swap.c:1042
tlb_batch_pages_flush mm/mmu_gather.c:98 [inline]
tlb_flush_mmu_free mm/mmu_gather.c:293 [inline]
tlb_flush_mmu+0x34d/0x4e0 mm/mmu_gather.c:300
tlb_finish_mmu+0xd4/0x200 mm/mmu_gather.c:392
exit_mmap+0x4b6/0xd40 mm/mmap.c:3300
__mmput+0x115/0x3c0 kernel/fork.c:1345
exit_mm+0x220/0x310 kernel/exit.c:569
do_exit+0x99e/0x27e0 kernel/exit.c:865
do_group_exit+0x207/0x2c0 kernel/exit.c:1027
get_signal+0x176e/0x1850 kernel/signal.c:2907
arch_do_signal_or_restart+0x96/0x860 arch/x86/kernel/signal.c:310
exit_to_user_mode_loop kernel/entry/common.c:105 [inline]
exit_to_user_mode_prepare include/linux/entry-common.h:328 [inline]
__syscall_exit_to_user_mode_work kernel/entry/common.c:201 [inline]
syscall_exit_to_user_mode+0xc9/0x360 kernel/entry/common.c:212
do_syscall_64+0x10a/0x240 arch/x86/entry/common.c:89
entry_SYSCALL_64_after_hwframe+0x6d/0x75
Memory state around the buggy address:
ffff888028c6cc80: fa fc fc fc fa fc fc fc fa fc fc fc fa fc fc fc
ffff888028c6cd00: fa fc fc fc fa fc fc fc 00 fc fc fc 06 fc fc fc
>ffff888028c6cd80: fa fc fc fc fa fc fc fc fa fc fc fc 02 fc fc fc
^
ffff888028c6ce00: fa fc fc fc fa fc fc fc fa fc fc fc fa fc fc fc
ffff888028c6ce80: fa fc fc fc fa fc fc fc fa fc fc fc fa fc fc fc--- net/xdp/xsk.c | 2 ++ 1 file changed, 2 insertions(+)
net/xdp/xsk.c +2 -0
@@ -1414,6 +1414,8 @@ static int xsk_setsockopt(struct socket *sock, int level, int optname,struct xsk_queue **q;int entries; +if (optlen <sizeof (entries)) +return -EINVAL ;if (copy_from_sockptr (&entries, optval,sizeof (entries)))return -EFAULT ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 46/172 octeontx2-pf: Fix transmit scheduler resource leak
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hariprasad Kelam <[email protected]>
[ Upstream commit bccb798e07f8bb8b91212fe8ed1e421685449076 ]
Inorder to support shaping and scheduling, Upon class creation Netdev driver allocates trasmit schedulers.
The previous patch which added support for Round robin scheduling has a bug due to which driver is not freeing transmit schedulers post class deletion.
This patch fixes the same.
--- drivers/net/ethernet/marvell/octeontx2/nic/qos.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/ethernet/marvell/octeontx2/nic/qos.c +1 -0
@@ -382,6 +382,7 @@ static void otx2_qos_read_txschq_cfg_tl(struct otx2_qos_node *parent,otx2_qos_read_txschq_cfg_tl (node, cfg); cnt = cfg->static_node_pos[node->level]; cfg->schq_contig_list[node->level][cnt] = node->schq; + cfg->schq_index_used[node->level][cnt] = true; cfg->schq_contig[node->level]++; cfg->static_node_pos[node->level]++;otx2_qos_read_txschq_cfg_schq (node, cfg);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 47/172 block: fix q->blkg_list corruption during disk rebind
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ming Lei <[email protected]>
[ Upstream commit 8b8ace080319a866f5dfe9da8e665ae51d971c54 ]
Multiple gendisk instances can allocated/added for single request queue in case of disk rebind. blkg may still stay in q->blkg_list when calling blkcg_init_disk() for rebind, then q->blkg_list becomes corrupted.
Fix the list corruption issue by:
- add blkg_init_queue() to initialize q->blkg_list & q->blkcg_mutex only
- move calling blkg_init_queue() into blk_alloc_queue()
The list corruption should be started since commit f1c006f1c685 ("blk-cgroup:
synchronize pd_free_fn() from blkg_free_workfn() and blkcg_deactivate_policy()") which delays removing blkg from q->blkg_list into blkg_free_workfn().
--- block/blk-cgroup.c | 9 ++++++--- block/blk-cgroup.h | 2 ++ block/blk-core.c | 2 ++ 3 files changed, 10 insertions(+), 3 deletions(-)
block/blk-cgroup.c +6 -3
@@ -1409,6 +1409,12 @@ static int blkcg_css_online(struct cgroup_subsys_state *css)return 0 ; } +void blkg_init_queue (struct request_queue *q) +{ +INIT_LIST_HEAD (&q->blkg_list); +mutex_init (&q->blkcg_mutex); +} +int blkcg_init_disk (struct gendisk *disk) {struct request_queue *q = disk->queue;
@@ -1416,9 +1422,6 @@ int blkcg_init_disk(struct gendisk *disk)bool preloaded;int ret; -INIT_LIST_HEAD (&q->blkg_list); -mutex_init (&q->blkcg_mutex); - new_blkg =blkg_alloc (&blkcg_root, disk,GFP_KERNEL );if (!new_blkg)return -ENOMEM ;
block/blk-cgroup.h +2 -0
@@ -188,6 +188,7 @@ struct blkcg_policy {extern struct blkcg blkcg_root;extern bool blkcg_debug_stats; +void blkg_init_queue (struct request_queue *q);int blkcg_init_disk (struct gendisk *disk);void blkcg_exit_disk (struct gendisk *disk);
@@ -481,6 +482,7 @@ struct blkcg { };static inline struct blkcg_gq *blkg_lookup (struct blkcg *blkcg,void *key) {return NULL ; } +static inline void blkg_init_queue (struct request_queue *q) { }static inline int blkcg_init_disk (struct gendisk *disk) {return 0 ; }static inline void blkcg_exit_disk (struct gendisk *disk) { }static inline int blkcg_policy_register (struct blkcg_policy *pol) {return 0 ; }
block/blk-core.c +2 -0
@@ -431,6 +431,8 @@ struct request_queue *blk_alloc_queue(int node_id)init_waitqueue_head (&q->mq_freeze_wq);mutex_init (&q->mq_freeze_lock); +blkg_init_queue (q); +/* * Init percpu_ref in atomic mode so that it's faster to shutdown. * See blk_register_queue() for details.
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 48/172 lib: checksum: hide unused expected_csum_ipv6_magic[]
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit e9d47b7b31563a6524b9f64ea70ed0289cc4d9c4 ]
When CONFIG_NET is disabled, an extra warning shows up for this unused variable:
lib/checksum_kunit.c:218:18: error: 'expected_csum_ipv6_magic' defined but not used [-Werror=unused-const-variable=]
Replace the #ifdef with an IS_ENABLED() check that makes the compiler's dead-code-elimination take care of the link failure.
--- lib/checksum_kunit.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
lib/checksum_kunit.c +3 -2
@@ -594,13 +594,15 @@ static void test_ip_fast_csum(struct kunit *test)static void test_csum_ipv6_magic (struct kunit *test) { -#if defined(CONFIG_NET )const struct in6_addr *saddr;const struct in6_addr *daddr;unsigned int len;unsigned char proto;__wsum csum; +if (!IS_ENABLED (CONFIG_NET )) +return ; +const int daddr_offset =sizeof (struct in6_addr );const int len_offset =sizeof (struct in6_addr ) +sizeof (struct in6_addr ); const int proto_offset = sizeof(struct in6_addr) + sizeof(struct in6_addr) +
@@ -618,7 +620,6 @@ static void test_csum_ipv6_magic(struct kunit *test)CHECK_EQ (to_sum16 (expected_csum_ipv6_magic[i]),csum_ipv6_magic (saddr, daddr, len, proto, csum)); } -#endif /* !CONFIG_NET */ }static struct kunit_case __refdata checksum_test_cases[] = {
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 49/172 geneve: fix header validation in geneve[6]_xmit_skb
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit d8a6213d70accb403b82924a1c229e733433a5ef ]
syzbot is able to trigger an uninit-value in geneve_xmit() [1]
Problem : While most ip tunnel helpers (like ip_tunnel_get_dsfield()) uses skb_protocol(skb, true), pskb_inet_may_pull() is only using skb->protocol.
If anything else than ETH_P_IPV6 or ETH_P_IP is found in skb->protocol, pskb_inet_may_pull() does nothing at all.
If a vlan tag was provided by the caller (af_packet in the syzbot case), the network header might not point to the correct location, and skb linear part could be smaller than expected.
Add skb_vlan_inet_prepare() to perform a complete mac validation.
Use this in geneve for the moment, I suspect we need to adopt this more broadly.
v4 - Jakub reported v3 broke l2_tos_ttl_inherit.sh selftest
- Only call __vlan_get_protocol() for vlan types.
v2,v3 - Addressed Sabrina comments on v1 and v2
[1]
BUG: KMSAN: uninit-value in geneve_xmit_skb drivers/net/geneve.c:910 [inline] BUG: KMSAN: uninit-value in geneve_xmit+0x302d/0x5420 drivers/net/geneve.c:1030 geneve_xmit_skb drivers/net/geneve.c:910 [inline] geneve_xmit+0x302d/0x5420 drivers/net/geneve.c:1030 __netdev_start_xmit include/linux/netdevice.h:4903 [inline] netdev_start_xmit include/linux/netdevice.h:4917 [inline] xmit_one net/core/dev.c:3531 [inline] dev_hard_start_xmit+0x247/0xa20 net/core/dev.c:3547 __dev_queue_xmit+0x348d/0x52c0 net/core/dev.c:4335 dev_queue_xmit include/linux/netdevice.h:3091 [inline] packet_xmit+0x9c/0x6c0 net/packet/af_packet.c:276 packet_snd net/packet/af_packet.c:3081 [inline] packet_sendmsg+0x8bb0/0x9ef0 net/packet/af_packet.c:3113 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x30f/0x380 net/socket.c:745 __sys_sendto+0x685/0x830 net/socket.c:2191 __do_sys_sendto net/socket.c:2203 [inline] __se_sys_sendto net/socket.c:2199 [inline] __x64_sys_sendto+0x125/0x1d0 net/socket.c:2199 do_syscall_64+0xd5/0x1f0 entry_SYSCALL_64_after_hwframe+0x6d/0x75
Uninit was created at:
slab_post_alloc_hook mm/slub.c:3804 [inline] slab_alloc_node mm/slub.c:3845 [inline] kmem_cache_alloc_node+0x613/0xc50 mm/slub.c:3888 kmalloc_reserve+0x13d/0x4a0 net/core/skbuff.c:577 __alloc_skb+0x35b/0x7a0 net/core/skbuff.c:668 alloc_skb include/linux/skbuff.h:1318 [inline] alloc_skb_with_frags+0xc8/0xbf0 net/core/skbuff.c:6504 sock_alloc_send_pskb+0xa81/0xbf0 net/core/sock.c:2795 packet_alloc_skb net/packet/af_packet.c:2930 [inline] packet_snd net/packet/af_packet.c:3024 [inline] packet_sendmsg+0x722d/0x9ef0 net/packet/af_packet.c:3113 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x30f/0x380 net/socket.c:745 __sys_sendto+0x685/0x830 net/socket.c:2191 __do_sys_sendto net/socket.c:2203 [inline] __se_sys_sendto net/socket.c:2199 [inline] __x64_sys_sendto+0x125/0x1d0 net/socket.c:2199 do_syscall_64+0xd5/0x1f0 entry_SYSCALL_64_after_hwframe+0x6d/0x75 CPU: 0 PID: 5033 Comm: syz-executor346 Not tainted 6.9.0-rc1-syzkaller-00005-g928a87efa423 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 02/29/2024
--- drivers/net/geneve.c | 4 ++-- include/net/ip_tunnels.h | 33 +++++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+), 2 deletions(-)
drivers/net/geneve.c +2 -2
@@ -830,7 +830,7 @@ static int geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev,__be16 sport;int err; -if (!pskb_inet_may_pull (skb)) +if (!skb_vlan_inet_prepare (skb))return -EINVAL ;if (!gs4)
@@ -937,7 +937,7 @@ static int geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev,__be16 sport;int err; -if (!pskb_inet_may_pull (skb)) +if (!skb_vlan_inet_prepare (skb))return -EINVAL ;if (!gs6)
include/net/ip_tunnels.h +33 -0
@@ -360,6 +360,39 @@ static inline bool pskb_inet_may_pull(struct sk_buff *skb)return pskb_network_may_pull (skb, nhlen); } +/* Variant of pskb_inet_may_pull(). +*/ +static inline bool skb_vlan_inet_prepare (struct sk_buff *skb) +{ +int nhlen =0 , maclen =ETH_HLEN ; +__be16 type = skb->protocol; + +/* Essentially this is skb_protocol(skb, true) +* And we get MAC len. +*/ +if (eth_type_vlan (type)) + type =__vlan_get_protocol (skb, type, &maclen); + +switch (type) { +#if IS_ENABLED (CONFIG_IPV6 ) +case htons (ETH_P_IPV6 ): + nhlen =sizeof (struct ipv6hdr ); +break ; +#endif +case htons (ETH_P_IP ): + nhlen =sizeof (struct iphdr ); +break ; + } +/* For ETH_P_IPV6/ETH_P_IP we make sure to pull +* a base network header in skb->head. +*/ +if (!pskb_may_pull (skb, maclen + nhlen)) +return false; + +skb_set_network_header (skb, maclen); +return true; +} +static inline int ip_encap_hlen (struct ip_tunnel_encap *e) {const struct ip_tunnel_encap_ops *ops;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 50/172 s390/ism: fix receive message buffer allocation
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gerd Bayer <[email protected]>
[ Upstream commit 58effa3476536215530c9ec4910ffc981613b413 ]
Since [1], dma_alloc_coherent() does not accept requests for GFP_COMP anymore, even on archs that may be able to fulfill this. Functionality that relied on the receive buffer being a compound page broke at that point:
The SMC-D protocol, that utilizes the ism device driver, passes receive buffers to the splice processor in a struct splice_pipe_desc with a single entry list of struct pages. As the buffer is no longer a compound page, the splice processor now rejects requests to handle more than a page worth of data.
Replace dma_alloc_coherent() and allocate a buffer with folio_alloc and create a DMA map for it with dma_map_page(). Since only receive buffers on ISM devices use DMA, qualify the mapping as FROM_DEVICE. Since ISM devices are available on arch s390, only and on that arch all DMA is coherent, there is no need to introduce and export some kind of dma_sync_to_cpu() method to be called by the SMC-D protocol layer.
Analogously, replace dma_free_coherent by a two step dma_unmap_page, then folio_put to free the receive buffer.
[1] https://lore.kernel.org/all/[email protected]/
--- drivers/s390/net/ism_drv.c | 38 +++++++++++++++++++++++++++++--------- 1 file changed, 29 insertions(+), 9 deletions(-)
drivers/s390/net/ism_drv.c +29 -9
@@ -14,6 +14,8 @@#include <linux/err.h> #include <linux/ctype.h> #include <linux/processor.h> +#include <linux/dma-mapping.h> +#include <linux/mm.h> #include "ism.h"
@@ -292,13 +294,15 @@ static int ism_read_local_gid(struct ism_dev *ism)static void ism_free_dmb (struct ism_dev *ism,struct ism_dmb *dmb) {clear_bit (dmb->sba_idx, ism->sba_bitmap); -dma_free_coherent (&ism->pdev->dev, dmb->dmb_len, - dmb->cpu_addr, dmb->dma_addr); +dma_unmap_page (&ism->pdev->dev, dmb->dma_addr, dmb->dmb_len, +DMA_FROM_DEVICE ); +folio_put (virt_to_folio (dmb->cpu_addr)); }static int ism_alloc_dmb (struct ism_dev *ism,struct ism_dmb *dmb) {unsigned long bit; +int rc;if (PAGE_ALIGN (dmb->dmb_len) >dma_get_max_seg_size (&ism->pdev->dev))return -EINVAL ;
@@ -315,14 +319,30 @@ static int ism_alloc_dmb(struct ism_dev *ism, struct ism_dmb *dmb) test_and_set_bit(dmb->sba_idx, ism->sba_bitmap)) return -EINVAL; - dmb->cpu_addr =dma_alloc_coherent (&ism->pdev->dev, dmb->dmb_len, - &dmb->dma_addr, -GFP_KERNEL | __GFP_NOWARN | - __GFP_NOMEMALLOC | __GFP_NORETRY); -if (!dmb->cpu_addr) -clear_bit (dmb->sba_idx, ism->sba_bitmap); + dmb->cpu_addr = +folio_address (folio_alloc (GFP_KERNEL | __GFP_NOWARN | + __GFP_NOMEMALLOC | __GFP_NORETRY, +get_order (dmb->dmb_len))); -return dmb->cpu_addr ?0 : -ENOMEM ; +if (!dmb->cpu_addr) { + rc = -ENOMEM ; + goto out_bit; + } + dmb->dma_addr =dma_map_page (&ism->pdev->dev, +virt_to_page (dmb->cpu_addr),0 , + dmb->dmb_len,DMA_FROM_DEVICE ); +if (dma_mapping_error (&ism->pdev->dev, dmb->dma_addr)) { + rc = -ENOMEM ; + goto out_free; + } + +return 0 ; + +out_free: +kfree (dmb->cpu_addr); +out_bit: +clear_bit (dmb->sba_idx, ism->sba_bitmap); +return rc; } int ism_register_dmb(struct ism_dev *ism, struct ism_dmb *dmb,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 51/172 bnxt_en: Fix possible memory leak in bnxt_rdma_aux_device_init()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vikas Gupta <[email protected]>
[ Upstream commit 7ac10c7d728d75bc9daaa8fade3c7a3273b9a9ff ]
If ulp = kzalloc() fails, the allocated edev will leak because it is not properly assigned and the cleanup path will not be able to free it. Fix it by assigning it properly immediately after allocation.
--- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +2 -1
@@ -392,12 +392,13 @@ void bnxt_rdma_aux_device_init(struct bnxt *bp)if (!edev) goto aux_dev_uninit; + aux_priv->edev = edev; + ulp =kzalloc (sizeof (*ulp),GFP_KERNEL );if (!ulp) goto aux_dev_uninit; edev->ulp_tbl = ulp; - aux_priv->edev = edev; bp->edev = edev;bnxt_set_edev_info (edev, bp);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 52/172 bnxt_en: Fix error recovery for RoCE ulp client
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vikas Gupta <[email protected]>
[ Upstream commit b5ea7d33ba2a42b95b4298d08d2af9cdeeaf0090 ]
Since runtime MSIXs vector allocation/free has been removed, the L2 driver needs to repopulate the MSIX entries for the ulp client as the irq table may change during the recovery process.
--- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +3 -0
@@ -210,6 +210,9 @@ void bnxt_ulp_start(struct bnxt *bp, int err)if (err)return ; +if (edev->ulp_tbl->msix_requested) +bnxt_fill_msix_vecs (bp, edev->msix_entries); +if (aux_priv) {struct auxiliary_device *adev;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 53/172 bnxt_en: Reset PTP tx_avail after possible firmware reset
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Pavan Chebbi <[email protected]>
[ Upstream commit faa12ca245585379d612736a4b5e98e88481ea59 ]
It is possible that during error recovery and firmware reset, there is a pending TX PTP packet waiting for the timestamp. We need to reset this condition so that after recovery, the tx_avail count for PTP is reset back to the initial value. Otherwise, we may not accept any PTP TX timestamps after recovery.
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +2 -0
@@ -11526,6 +11526,8 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)/* VF-reps may need to be re-opened after the PF is re-opened */ if (BNXT_PF (bp))bnxt_vf_reps_open (bp); +if (bp->ptp_cfg) +atomic_set (&bp->ptp_cfg->tx_avail,BNXT_MAX_TX_TS );bnxt_ptp_init_rtc (bp, true);bnxt_ptp_cfg_tstamp_filters (bp);return 0 ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 54/172 ACPI: bus: allow _UID matching for integer zero
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Raag Jadav <[email protected]>
[ Upstream commit aca1a5287ea328fd1f7e2bfa6806646486d86a70 ]
Commit b2b32a173881 ("ACPI: bus: update acpi_dev_hid_uid_match() to support multiple types") added _UID matching support for both integer and string types, which satisfies NULL @uid2 argument for string types using inversion, but this logic prevents _UID comparision in case the argument is integer 0, which may result in false positives.
Fix this using _Generic(), which will allow NULL @uid2 argument for string types as well as _UID matching for all possible integer values.
--- include/acpi/acpi_bus.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
include/acpi/acpi_bus.h +5 -3
@@ -911,17 +911,19 @@ static inline bool acpi_int_uid_match(struct acpi_device *adev, u64 uid2)* acpi_dev_hid_uid_match - Match device by supplied HID and UID * @adev: ACPI device to match. * @hid2: Hardware ID of the device. -* @uid2: Unique ID of the device, pass 0 or NULL to not check _UID. +* @uid2: Unique ID of the device, pass NULL to not check _UID. * * Matches HID and UID in @adev with given @hid2 and @uid2. Absence of @uid2 * will be treated as a match. If user wants to validate @uid2, it should be * done before calling this function. * -* Returns: %true if matches or @uid2 is 0 or NULL, %false otherwise. +* Returns: %true if matches or @uid2 is NULL, %false otherwise. */ #define acpi_dev_hid_uid_match (adev, hid2, uid2) \ (acpi_dev_hid_match(adev, hid2) && \ - (!(uid2) || acpi_dev_uid_match(adev, uid2))) +/* Distinguish integer 0 from NULL @uid2 */ \ + (_Generic(uid2, ACPI_STR_TYPES(!(uid2)), default: 0) || \ + acpi_dev_uid_match(adev, uid2)))void acpi_dev_clear_dependencies (struct acpi_device *supplier);bool acpi_dev_ready_for_enumeration (const struct acpi_device *device);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 55/172 base/node / ACPI: Enumerate node access class for struct access_coordinate
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dave Jiang <[email protected]>
[ Upstream commit 11270e526276ffad4c4237acb393da82a3287487 ]
Both generic node and HMAT handling code have been using magic numbers to indicate access classes for 'struct access_coordinate'. Introduce enums to enumerate the access0 and access1 classes shared by the two subsystems. Update the function parameters and callers as appropriate to utilize the new enum.
Access0 is named to ACCESS_COORDINATE_LOCAL in order to indicate that the access class is for 'struct access_coordinate' between a target node and the nearest initiator node.
Access1 is named to ACCESS_COORDINATE_CPU in order to indicate that the access class is for 'struct access_coordinate' between a target node and the nearest CPU node.
Stable-dep-of: 592780b8391f ("cxl: Fix retrieving of access_coordinates in PCIe path")
--- drivers/acpi/numa/hmat.c | 26 ++++++++++++++------------ drivers/base/node.c | 6 +++--- include/linux/node.h | 18 +++++++++++++++--- 3 files changed, 32 insertions(+), 18 deletions(-)
drivers/acpi/numa/hmat.c +14 -12
@@ -59,9 +59,7 @@ struct target_cache { };enum { -NODE_ACCESS_CLASS_0 =0 , -NODE_ACCESS_CLASS_1 , -NODE_ACCESS_CLASS_GENPORT_SINK , +NODE_ACCESS_CLASS_GENPORT_SINK =ACCESS_COORDINATE_MAX ,NODE_ACCESS_CLASS_MAX , };
@@ -374,11 +372,11 @@ static __init void hmat_update_target(unsigned int tgt_pxm, unsigned int init_pxif (target && target->processor_pxm == init_pxm) {hmat_update_target_access (target, type, value, -NODE_ACCESS_CLASS_0 ); +ACCESS_COORDINATE_LOCAL );/* If the node has a CPU, update access 1 */ if (node_state (pxm_to_node (init_pxm),N_CPU ))hmat_update_target_access (target, type, value, -NODE_ACCESS_CLASS_1 ); +ACCESS_COORDINATE_CPU ); } }
@@ -709,7 +707,8 @@ static void hmat_update_target_attrs(struct memory_target *target,*/ if (target->processor_pxm !=PXM_INVAL ) { cpu_nid =pxm_to_node (target->processor_pxm); -if (access ==0 ||node_state (cpu_nid,N_CPU )) { +if (access ==ACCESS_COORDINATE_LOCAL || +node_state (cpu_nid,N_CPU )) {set_bit (target->processor_pxm, p_nodes);return ; }
@@ -737,7 +736,8 @@ static void hmat_update_target_attrs(struct memory_target *target,list_for_each_entry (initiator, &initiators, node) {u32 value; -if (access ==1 && !initiator->has_cpu) { +if (access ==ACCESS_COORDINATE_CPU && + !initiator->has_cpu) {clear_bit (initiator->processor_pxm, p_nodes);continue ; }
@@ -782,8 +782,10 @@ static void hmat_register_target_initiators(struct memory_target *target) { static DECLARE_BITMAP(p_nodes, MAX_NUMNODES); -__hmat_register_target_initiators (target, p_nodes,0 ); -__hmat_register_target_initiators (target, p_nodes,1 ); +__hmat_register_target_initiators (target, p_nodes, + ACCESS_COORDINATE_LOCAL); +__hmat_register_target_initiators (target, p_nodes, +ACCESS_COORDINATE_CPU ); }static void hmat_register_target_cache (struct memory_target *target)
@@ -854,8 +856,8 @@ static void hmat_register_target(struct memory_target *target)if (!target->registered) {hmat_register_target_initiators (target);hmat_register_target_cache (target); -hmat_register_target_perf (target,NODE_ACCESS_CLASS_0 ); -hmat_register_target_perf (target,NODE_ACCESS_CLASS_1 ); +hmat_register_target_perf (target,ACCESS_COORDINATE_LOCAL ); +hmat_register_target_perf (target,ACCESS_COORDINATE_CPU ); target->registered = true; }mutex_unlock (&target_lock);
@@ -927,7 +929,7 @@ static int hmat_calculate_adistance(struct notifier_block *self,return NOTIFY_OK ;mutex_lock (&target_lock); -hmat_update_target_attrs (target, p_nodes,1 ); +hmat_update_target_attrs (target, p_nodes,ACCESS_COORDINATE_CPU );mutex_unlock (&target_lock); perf = &target->coord[1 ];
drivers/base/node.c +3 -3
@@ -126,7 +126,7 @@ static void node_access_release(struct device *dev) }static struct node_access_nodes *node_init_node_access (struct node *node, -unsigned int access) +enum access_coordinate_class access) {struct node_access_nodes *access_node;struct device *dev;
@@ -191,7 +191,7 @@ static struct attribute *access_attrs[] = {* @access: The access class the for the given attributes */ void node_set_perf_attrs (unsigned int nid,struct access_coordinate *coord, -unsigned int access) +enum access_coordinate_class access) {struct node_access_nodes *c;struct node *node;
@@ -689,7 +689,7 @@ int register_cpu_under_node(unsigned int cpu, unsigned int nid)*/ int register_memory_node_under_compute_node (unsigned int mem_nid,unsigned int cpu_nid, -unsigned int access) +enum access_coordinate_class access) {struct node *init_node, *targ_node;struct node_access_nodes *initiator, *target;
include/linux/node.h +15 -3
@@ -34,6 +34,18 @@ struct access_coordinate {unsigned int write_latency; }; +/* +* ACCESS_COORDINATE_LOCAL correlates to ACCESS CLASS 0 +* - access_coordinate between target node and nearest initiator node +* ACCESS_COORDINATE_CPU correlates to ACCESS CLASS 1 +* - access_coordinate between target node and nearest CPU node +*/ +enum access_coordinate_class { +ACCESS_COORDINATE_LOCAL , +ACCESS_COORDINATE_CPU , +ACCESS_COORDINATE_MAX +}; +enum cache_indexing {NODE_CACHE_DIRECT_MAP ,NODE_CACHE_INDEXED ,
@@ -66,7 +78,7 @@ struct node_cache_attrs {#ifdef CONFIG_HMEM_REPORTING void node_add_cache (unsigned int nid,struct node_cache_attrs *cache_attrs);void node_set_perf_attrs (unsigned int nid,struct access_coordinate *coord, -unsigned access); +enum access_coordinate_class access);#else static inline void node_add_cache (unsigned int nid,struct node_cache_attrs *cache_attrs)
@@ -75,7 +87,7 @@ static inline void node_add_cache(unsigned int nid,static inline void node_set_perf_attrs (unsigned int nid,struct access_coordinate *coord, -unsigned access) +enum access_coordinate_class access) { }#endif
@@ -137,7 +149,7 @@ extern void unregister_memory_block_under_nodes(struct memory_block *mem_blk);extern int register_memory_node_under_compute_node (unsigned int mem_nid,unsigned int cpu_nid, -unsigned access); +enum access_coordinate_class access);#else static inline void node_dev_init (void ) {
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 56/172 ACPI: HMAT: Introduce 2 levels of generic port access class
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dave Jiang <[email protected]>
[ Upstream commit 1745a7b364dfd339ab2696b7d51d7ed950ed2598 ]
In order to compute access0 and access1 classes for CXL memory, 2 levels of generic port information must be stored. Access0 will indicate the generic port access coordinates to the closest initiator and access1 will indicate the generic port access coordinates to the cloest CPU.
Stable-dep-of: 592780b8391f ("cxl: Fix retrieving of access_coordinates in PCIe path")
--- drivers/acpi/numa/hmat.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-)
drivers/acpi/numa/hmat.c +10 -5
@@ -59,7 +59,8 @@ struct target_cache { };enum { -NODE_ACCESS_CLASS_GENPORT_SINK =ACCESS_COORDINATE_MAX , +NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL =ACCESS_COORDINATE_MAX , +NODE_ACCESS_CLASS_GENPORT_SINK_CPU ,NODE_ACCESS_CLASS_MAX , };
@@ -141,7 +142,7 @@ int acpi_get_genport_coordinates(u32 uid,if (!target)return -ENOENT ; - *coord = target->coord[NODE_ACCESS_CLASS_GENPORT_SINK ]; + *coord = target->coord[NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL ];return 0 ; }
@@ -695,7 +696,8 @@ static void hmat_update_target_attrs(struct memory_target *target,int i;/* Don't update for generic port if there's no device handle */ -if (access ==NODE_ACCESS_CLASS_GENPORT_SINK && +if ((access ==NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL || + access ==NODE_ACCESS_CLASS_GENPORT_SINK_CPU ) && !(*(u16 *)target->gen_port_device_handle))return ;
@@ -736,7 +738,8 @@ static void hmat_update_target_attrs(struct memory_target *target,list_for_each_entry (initiator, &initiators, node) {u32 value; -if (access ==ACCESS_COORDINATE_CPU && +if ((access ==ACCESS_COORDINATE_CPU || + access ==NODE_ACCESS_CLASS_GENPORT_SINK_CPU ) && !initiator->has_cpu) {clear_bit (initiator->processor_pxm, p_nodes);continue ;
@@ -775,7 +778,9 @@ static void hmat_update_generic_target(struct memory_target *target) static DECLARE_BITMAP(p_nodes, MAX_NUMNODES);hmat_update_target_attrs (target, p_nodes, - NODE_ACCESS_CLASS_GENPORT_SINK); + NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL); +hmat_update_target_attrs (target, p_nodes, +NODE_ACCESS_CLASS_GENPORT_SINK_CPU ); }static void hmat_register_target_initiators (struct memory_target *target)
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 57/172 ACPI: HMAT / cxl: Add retrieval of generic port coordinates for both access classes
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dave Jiang <[email protected]>
[ Upstream commit bd98cbbbf82a3086423865816e1b5ab4bb4b6c60 ]
Update acpi_get_genport_coordinates() to allow retrieval of both access classes of the 'struct access_coordinate' for a generic target. The update will allow CXL code to compute access coordinates for both access class.
Stable-dep-of: 592780b8391f ("cxl: Fix retrieving of access_coordinates in PCIe path")
--- drivers/acpi/numa/hmat.c | 8 ++++++-- drivers/cxl/acpi.c | 8 +++++--- drivers/cxl/core/port.c | 2 +- drivers/cxl/cxl.h | 2 +- 4 files changed, 13 insertions(+), 7 deletions(-)
drivers/acpi/numa/hmat.c +6 -2
@@ -126,7 +126,8 @@ static struct memory_target *acpi_find_genport_target(u32 uid)/** * acpi_get_genport_coordinates - Retrieve the access coordinates for a generic port * @uid: ACPI unique id -* @coord: The access coordinates written back out for the generic port +* @coord: The access coordinates written back out for the generic port. +* Expect 2 levels array. * * Return: 0 on success. Errno on failure. *
@@ -142,7 +143,10 @@ int acpi_get_genport_coordinates(u32 uid,if (!target)return -ENOENT ; - *coord = target->coord[NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL ]; + coord[ACCESS_COORDINATE_LOCAL ] = + target->coord[NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL ]; + coord[ACCESS_COORDINATE_CPU ] = + target->coord[NODE_ACCESS_CLASS_GENPORT_SINK_CPU ];return 0 ; }
drivers/cxl/acpi.c +5 -3
@@ -530,13 +530,15 @@ static int get_genport_coordinates(struct device *dev, struct cxl_dport *dport)if (kstrtou32 (acpi_device_uid (hb),0 , &uid))return -EINVAL ; - rc =acpi_get_genport_coordinates (uid, &dport->hb_coord); + rc =acpi_get_genport_coordinates (uid, dport->hb_coord);if (rc <0 )return rc;/* Adjust back to picoseconds from nanoseconds */ - dport->hb_coord.read_latency *=1000 ; - dport->hb_coord.write_latency *=1000 ; +for (int i =0 ; i <ACCESS_COORDINATE_MAX ; i++) { + dport->hb_coord[i].read_latency *=1000 ; + dport->hb_coord[i].write_latency *=1000 ; + }return 0 ; }
drivers/cxl/core/port.c +1 -1
@@ -2152,7 +2152,7 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port, }/* Augment with the generic port (host bridge) perf data */ -combine_coordinates (&c, &dport->hb_coord); +combine_coordinates (&c, &dport->hb_coord[ACCESS_COORDINATE_LOCAL ]);/* Get the calculated PCI paths bandwidth */ pdev =to_pci_dev (port->uport_dev->parent);
drivers/cxl/cxl.h +1 -1
@@ -671,7 +671,7 @@ struct cxl_dport {struct cxl_port *port;struct cxl_regs regs;struct access_coordinate sw_coord; -struct access_coordinate hb_coord; +struct access_coordinate hb_coord[ACCESS_COORDINATE_MAX ];long link_latency; };
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 58/172 cxl: Split out combine_coordinates() for common shared usage
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dave Jiang <[email protected]>
[ Upstream commit 032f7b37adff6985e22516053698b77131c2ce96 ]
Refactor the common code of combining coordinates in order to reduce code. Create a new function cxl_cooordinates_combine() it combine two 'struct access_coordinate'.
Stable-dep-of: 592780b8391f ("cxl: Fix retrieving of access_coordinates in PCIe path")
--- drivers/cxl/core/cdat.c | 32 +++++++++++++++++++++++--------- drivers/cxl/core/port.c | 18 ++---------------- drivers/cxl/cxl.h | 4 ++++ 3 files changed, 29 insertions(+), 25 deletions(-)
drivers/cxl/core/cdat.c +23 -9
@@ -185,15 +185,7 @@ static int cxl_port_perf_data_calculate(struct cxl_port *port,xa_for_each (dsmas_xa, index, dent) {int qos_class; - dent->coord.read_latency = dent->coord.read_latency + - c.read_latency; - dent->coord.write_latency = dent->coord.write_latency + - c.write_latency; - dent->coord.read_bandwidth =min_t (int, c.read_bandwidth, - dent->coord.read_bandwidth); - dent->coord.write_bandwidth =min_t (int, c.write_bandwidth, - dent->coord.write_bandwidth); - +cxl_coordinates_combine (&dent->coord, &dent->coord, &c); dent->entries =1 ; rc = cxl_root->ops->qos_class (cxl_root, &dent->coord,1 , &qos_class);
@@ -484,4 +476,26 @@ void cxl_switch_parse_cdat(struct cxl_port *port) }EXPORT_SYMBOL_NS_GPL (cxl_switch_parse_cdat,CXL ); +/** +* cxl_coordinates_combine - Combine the two input coordinates +* +* @out: Output coordinate of c1 and c2 combined +* @c1: input coordinates +* @c2: input coordinates +*/ +void cxl_coordinates_combine (struct access_coordinate *out, +struct access_coordinate *c1, +struct access_coordinate *c2) +{ +if (c1->write_bandwidth && c2->write_bandwidth) + out->write_bandwidth =min (c1->write_bandwidth, + c2->write_bandwidth); + out->write_latency = c1->write_latency + c2->write_latency; + +if (c1->read_bandwidth && c2->read_bandwidth) + out->read_bandwidth =min (c1->read_bandwidth, + c2->read_bandwidth); + out->read_latency = c1->read_latency + c2->read_latency; +} +MODULE_IMPORT_NS (CXL );
drivers/cxl/core/port.c +2 -16
@@ -2096,20 +2096,6 @@ bool schedule_cxl_memdev_detach(struct cxl_memdev *cxlmd) }EXPORT_SYMBOL_NS_GPL (schedule_cxl_memdev_detach,CXL ); -static void combine_coordinates (struct access_coordinate *c1, -struct access_coordinate *c2) -{ -if (c2->write_bandwidth) - c1->write_bandwidth =min (c1->write_bandwidth, - c2->write_bandwidth); - c1->write_latency += c2->write_latency; - -if (c2->read_bandwidth) - c1->read_bandwidth =min (c1->read_bandwidth, - c2->read_bandwidth); - c1->read_latency += c2->read_latency; -} -/** * cxl_endpoint_get_perf_coordinates - Retrieve performance numbers stored in dports * of CXL path
@@ -2143,7 +2129,7 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,* nothing to gather. */ while (iter && !is_cxl_root (to_cxl_port (iter->dev.parent))) { -combine_coordinates (&c, &dport->sw_coord); +cxl_coordinates_combine (&c, &c, &dport->sw_coord); c.write_latency += dport->link_latency; c.read_latency += dport->link_latency;
@@ -2152,7 +2138,7 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port, }/* Augment with the generic port (host bridge) perf data */ -combine_coordinates (&c, &dport->hb_coord[ACCESS_COORDINATE_LOCAL ]); +cxl_coordinates_combine (&c, &c, &dport->hb_coord[ACCESS_COORDINATE_LOCAL ]);/* Get the calculated PCI paths bandwidth */ pdev =to_pci_dev (port->uport_dev->parent);
drivers/cxl/cxl.h +4 -0
@@ -882,6 +882,10 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,void cxl_memdev_update_perf (struct cxl_memdev *cxlmd); +void cxl_coordinates_combine (struct access_coordinate *out, +struct access_coordinate *c1, +struct access_coordinate *c2); +/* * Unit test builds overrides this to __weak, find the 'strong' version * of these symbols in tools/testing/cxl/.
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 59/172 cxl: Split out host bridge access coordinates
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dave Jiang <[email protected]>
[ Upstream commit 863027d40993f13155451bd898bfe4c4e9b7002f ]
The difference between access class 0 and access class 1 for 'struct access_coordinate', if any, is that class 0 is for the distance from the target to the closest initiator and that class 1 is for the distance from the target to the closest CPU. For CXL memory, the nearest initiator may not necessarily be a CPU node. The performance path from the CXL endpoint to the host bridge should remain the same. However, the numbers extracted and stored from HMAT is the difference for the two access classes. Split out the performance numbers for the host bridge (generic target) from the calculation of the entire path in order to allow calculation of both access classes for a CXL region.
Stable-dep-of: 592780b8391f ("cxl: Fix retrieving of access_coordinates in PCIe path")
--- drivers/cxl/core/cdat.c | 28 ++++++++++++++++++++++------ drivers/cxl/core/port.c | 35 ++++++++++++++++++++++++++++++++--- drivers/cxl/cxl.h | 2 ++ 3 files changed, 56 insertions(+), 9 deletions(-)
drivers/cxl/core/cdat.c +22 -6
@@ -162,15 +162,22 @@ static int cxl_cdat_endpoint_process(struct cxl_port *port,static int cxl_port_perf_data_calculate (struct cxl_port *port,struct xarray *dsmas_xa) { -struct access_coordinate c; +struct access_coordinate ep_c; +struct access_coordinate coord[ACCESS_COORDINATE_MAX ];struct dsmas_entry *dent;int valid_entries =0 ;unsigned long index;int rc; - rc =cxl_endpoint_get_perf_coordinates (port, &c); + rc =cxl_endpoint_get_perf_coordinates (port, &ep_c);if (rc) { -dev_dbg (&port->dev,"Failed to retrieve perf coordinates.\n" ); +dev_dbg (&port->dev,"Failed to retrieve ep perf coordinates.\n" ); +return rc; + } + + rc =cxl_hb_get_perf_coordinates (port, coord); +if (rc) { +dev_dbg (&port->dev,"Failed to retrieve hb perf coordinates.\n" );return rc; }
@@ -185,10 +192,19 @@ static int cxl_port_perf_data_calculate(struct cxl_port *port,xa_for_each (dsmas_xa, index, dent) {int qos_class; -cxl_coordinates_combine (&dent->coord, &dent->coord, &c); +cxl_coordinates_combine (&dent->coord, &dent->coord, &ep_c); +/* +* Keeping the host bridge coordinates separate from the dsmas +* coordinates in order to allow calculation of access class +* 0 and 1 for region later. +*/ +cxl_coordinates_combine (&coord[ACCESS_COORDINATE_LOCAL ], + &coord[ACCESS_COORDINATE_LOCAL ], + &dent->coord); dent->entries =1 ; - rc = cxl_root->ops->qos_class (cxl_root, &dent->coord,1 , - &qos_class); + rc = cxl_root->ops->qos_class (cxl_root, + &coord[ACCESS_COORDINATE_LOCAL ], +1 , &qos_class);if (rc !=1 )continue ;
drivers/cxl/core/port.c +32 -3
@@ -2096,6 +2096,38 @@ bool schedule_cxl_memdev_detach(struct cxl_memdev *cxlmd) }EXPORT_SYMBOL_NS_GPL (schedule_cxl_memdev_detach,CXL ); +/** +* cxl_hb_get_perf_coordinates - Retrieve performance numbers between initiator +* and host bridge +* +* @port: endpoint cxl_port +* @coord: output access coordinates +* +* Return: errno on failure, 0 on success. +*/ +int cxl_hb_get_perf_coordinates (struct cxl_port *port, +struct access_coordinate *coord) +{ +struct cxl_port *iter = port; +struct cxl_dport *dport; + +if (!is_cxl_endpoint (port)) +return -EINVAL ; + + dport = iter->parent_dport; +while (iter && !is_cxl_root (to_cxl_port (iter->dev.parent))) { + iter =to_cxl_port (iter->dev.parent); + dport = iter->parent_dport; + } + + coord[ACCESS_COORDINATE_LOCAL ] = + dport->hb_coord[ACCESS_COORDINATE_LOCAL ]; + coord[ACCESS_COORDINATE_CPU ] = + dport->hb_coord[ACCESS_COORDINATE_CPU ]; + +return 0 ; +} +/** * cxl_endpoint_get_perf_coordinates - Retrieve performance numbers stored in dports * of CXL path
@@ -2137,9 +2169,6 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port, dport = iter->parent_dport; } -/* Augment with the generic port (host bridge) perf data */ -cxl_coordinates_combine (&c, &c, &dport->hb_coord[ACCESS_COORDINATE_LOCAL ]); -/* Get the calculated PCI paths bandwidth */ pdev =to_pci_dev (port->uport_dev->parent); bw =pcie_bandwidth_available (pdev,NULL ,NULL ,NULL );
drivers/cxl/cxl.h +2 -0
@@ -879,6 +879,8 @@ void cxl_switch_parse_cdat(struct cxl_port *port);int cxl_endpoint_get_perf_coordinates (struct cxl_port *port,struct access_coordinate *coord); +int cxl_hb_get_perf_coordinates (struct cxl_port *port, +struct access_coordinate *coord);void cxl_memdev_update_perf (struct cxl_memdev *cxlmd);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 60/172 cxl: Remove checking of iter in cxl_endpoint_get_perf_coordinates()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dave Jiang <[email protected]>
[ Upstream commit 648dae58a830ecceea3b1bebf68432435980f137 ]
The while() loop in cxl_endpoint_get_perf_coordinates() checks to see if 'iter' is valid as part of the condition breaking out of the loop. is_cxl_root() will stop the loop before the next iteration could go NULL. Remove the iter check.
The presence of the iter or removing the iter does not impact the behavior of the code. This is a code clean up and not a bug fix.
Stable-dep-of: 592780b8391f ("cxl: Fix retrieving of access_coordinates in PCIe path")
--- drivers/cxl/core/port.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/cxl/core/port.c +1 -1
@@ -2160,7 +2160,7 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,* port each iteration. If the parent is cxl root then there is * nothing to gather. */ -while (iter && !is_cxl_root (to_cxl_port (iter->dev.parent))) { +while (!is_cxl_root (to_cxl_port (iter->dev.parent))) {cxl_coordinates_combine (&c, &c, &dport->sw_coord); c.write_latency += dport->link_latency; c.read_latency += dport->link_latency;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 61/172 cxl: Fix retrieving of access_coordinates in PCIe path
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dave Jiang <[email protected]>
[ Upstream commit 592780b8391fe31f129ef4823c1513528f4dcb76 ]
Current loop in cxl_endpoint_get_perf_coordinates() incorrectly assumes the Root Port (RP) dport is the one with generic port access_coordinate. However those coordinates are one level up in the Host Bridge (HB). Current code causes the computation code to pick up 0s as the coordinates and cause minimal bandwidth to result in 0.
Add check to skip RP when combining coordinates.
--- drivers/cxl/core/port.c | 35 ++++++++++++++++++++++------------- 1 file changed, 22 insertions(+), 13 deletions(-)
drivers/cxl/core/port.c +22 -13
@@ -2128,6 +2128,11 @@ int cxl_hb_get_perf_coordinates(struct cxl_port *port,return 0 ; } +static bool parent_port_is_cxl_root (struct cxl_port *port) +{ +return is_cxl_root (to_cxl_port (port->dev.parent)); +} +/** * cxl_endpoint_get_perf_coordinates - Retrieve performance numbers stored in dports * of CXL path
@@ -2147,27 +2152,31 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,struct cxl_dport *dport;struct pci_dev *pdev;unsigned int bw; +bool is_cxl_root;if (!is_cxl_endpoint (port))return -EINVAL ; - dport = iter->parent_dport; -/* -* Exit the loop when the parent port of the current port is cxl root. -* The iterative loop starts at the endpoint and gathers the -* latency of the CXL link from the current iter to the next downstream -* port each iteration. If the parent is cxl root then there is -* nothing to gather. +* Exit the loop when the parent port of the current iter port is cxl +* root. The iterative loop starts at the endpoint and gathers the +* latency of the CXL link from the current device/port to the connected +* downstream port each iteration. */ -while (!is_cxl_root (to_cxl_port (iter->dev.parent))) { -cxl_coordinates_combine (&c, &c, &dport->sw_coord); +do { + dport = iter->parent_dport; + iter =to_cxl_port (iter->dev.parent); + is_cxl_root =parent_port_is_cxl_root (iter); + +/* +* There's no valid access_coordinate for a root port since RPs do not +* have CDAT and therefore needs to be skipped. +*/ +if (!is_cxl_root) +cxl_coordinates_combine (&c, &c, &dport->sw_coord); c.write_latency += dport->link_latency; c.read_latency += dport->link_latency; - - iter =to_cxl_port (iter->dev.parent); - dport = iter->parent_dport; - } + }while (!is_cxl_root);/* Get the calculated PCI paths bandwidth */ pdev =to_pci_dev (port->uport_dev->parent);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 62/172 net: ks8851: Inline ks8851_rx_skb()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marek Vasut <[email protected]>
[ Upstream commit f96f700449b6d190e06272f1cf732ae8e45b73df ]
Both ks8851_rx_skb_par() and ks8851_rx_skb_spi() call netif_rx(skb), inline the netif_rx(skb) call directly into ks8851_common.c and drop the .rx_skb callback and ks8851_rx_skb() wrapper. This removes one indirect call from the driver, no functional change otherwise.
Stable-dep-of: be0384bf599c ("net: ks8851: Handle softirqs at the end of IRQ thread to fix hang")
--- drivers/net/ethernet/micrel/ks8851.h | 3 --- drivers/net/ethernet/micrel/ks8851_common.c | 12 +----------- drivers/net/ethernet/micrel/ks8851_par.c | 11 ----------- drivers/net/ethernet/micrel/ks8851_spi.c | 11 ----------- 4 files changed, 1 insertion(+), 36 deletions(-)
drivers/net/ethernet/micrel/ks8851.h +0 -3
@@ -368,7 +368,6 @@ union ks8851_tx_hdr {* @rdfifo: FIFO read callback * @wrfifo: FIFO write callback * @start_xmit: start_xmit() implementation callback -* @rx_skb: rx_skb() implementation callback * @flush_tx_work: flush_tx_work() implementation callback * * The @statelock is used to protect information in the structure which may
@@ -423,8 +422,6 @@ struct ks8851_net { struct sk_buff *txp, bool irq);netdev_tx_t (*start_xmit)(struct sk_buff *skb,struct net_device *dev); -void (*rx_skb)(struct ks8851_net *ks, -struct sk_buff *skb);void (*flush_tx_work)(struct ks8851_net *ks); };
drivers/net/ethernet/micrel/ks8851_common.c +1 -11
@@ -231,16 +231,6 @@ static void ks8851_dbg_dumpkkt(struct ks8851_net *ks, u8 *rxpkt) rxpkt[12], rxpkt[13], rxpkt[14], rxpkt[15]); } -/** -* ks8851_rx_skb - receive skbuff -* @ks: The device state. -* @skb: The skbuff -*/ -static void ks8851_rx_skb (struct ks8851_net *ks,struct sk_buff *skb) -{ - ks->rx_skb (ks, skb); -} -/** * ks8851_rx_pkts - receive packets from the host * @ks: The device information.
@@ -309,7 +299,7 @@ static void ks8851_rx_pkts(struct ks8851_net *ks)ks8851_dbg_dumpkkt (ks, rxpkt); skb->protocol =eth_type_trans (skb, ks->netdev); -ks8851_rx_skb (ks, skb); +netif_rx (skb); ks->netdev->stats.rx_packets++; ks->netdev->stats.rx_bytes += rxlen;
drivers/net/ethernet/micrel/ks8851_par.c +0 -11
@@ -210,16 +210,6 @@ static void ks8851_wrfifo_par(struct ks8851_net *ks, struct sk_buff *txp,iowrite16_rep (ksp->hw_addr, txp->data, len /2 ); } -/** -* ks8851_rx_skb_par - receive skbuff -* @ks: The device state. -* @skb: The skbuff -*/ -static void ks8851_rx_skb_par (struct ks8851_net *ks,struct sk_buff *skb) -{ -netif_rx (skb); -} -static unsigned int ks8851_rdreg16_par_txqcr (struct ks8851_net *ks) {return ks8851_rdreg16_par (ks,KS_TXQCR );
@@ -298,7 +288,6 @@ static int ks8851_probe_par(struct platform_device *pdev) ks->rdfifo = ks8851_rdfifo_par; ks->wrfifo = ks8851_wrfifo_par; ks->start_xmit = ks8851_start_xmit_par; - ks->rx_skb = ks8851_rx_skb_par;#define STD_IRQ (IRQ_LCI |/* Link Change */ \ IRQ_RXI | /* RX done */ \
drivers/net/ethernet/micrel/ks8851_spi.c +0 -11
@@ -298,16 +298,6 @@ static unsigned int calc_txlen(unsigned int len)return ALIGN (len +4 ,4 ); } -/** -* ks8851_rx_skb_spi - receive skbuff -* @ks: The device state -* @skb: The skbuff -*/ -static void ks8851_rx_skb_spi (struct ks8851_net *ks,struct sk_buff *skb) -{ -netif_rx (skb); -} -/** * ks8851_tx_work - process tx packet(s) * @work: The work strucutre what was scheduled.
@@ -435,7 +425,6 @@ static int ks8851_probe_spi(struct spi_device *spi) ks->rdfifo = ks8851_rdfifo_spi; ks->wrfifo = ks8851_wrfifo_spi; ks->start_xmit = ks8851_start_xmit_spi; - ks->rx_skb = ks8851_rx_skb_spi; ks->flush_tx_work = ks8851_flush_tx_work_spi;#define STD_IRQ (IRQ_LCI |/* Link Change */ \
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 63/172 net: ks8851: Handle softirqs at the end of IRQ thread to fix hang
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marek Vasut <[email protected]>
[ Upstream commit be0384bf599cf1eb8d337517feeb732d71f75a6f ]
The ks8851_irq() thread may call ks8851_rx_pkts() in case there are any packets in the MAC FIFO, which calls netif_rx(). This netif_rx() implementation is guarded by local_bh_disable() and local_bh_enable(). The local_bh_enable() may call do_softirq() to run softirqs in case any are pending. One of the softirqs is net_rx_action, which ultimately reaches the driver .start_xmit callback. If that happens, the system hangs. The entire call chain is below:
ks8851_start_xmit_par from netdev_start_xmit netdev_start_xmit from dev_hard_start_xmit dev_hard_start_xmit from sch_direct_xmit sch_direct_xmit from __dev_queue_xmit __dev_queue_xmit from __neigh_update __neigh_update from neigh_updateneigh_update from arp_process.constprop.0 arp_process.constprop.0 from __netif_receive_skb_one_core __netif_receive_skb_one_core from process_backlog process_backlog from __napi_poll.constprop.0 __napi_poll.constprop.0 from net_rx_action net_rx_action from __do_softirq__do_softirq from call_with_stack call_with_stack from do_softirqdo_softirq from __local_bh_enable_ip __local_bh_enable_ip from netif_rxnetif_rx from ks8851_irqks8851_irq from irq_thread_fnirq_thread_fn from irq_threadirq_thread from kthread
kthread from ret_from_fork
The hang happens because ks8851_irq() first locks a spinlock in ks8851_par.c ks8851_lock_par() spin_lock_irqsave(&ksp->lock, ...) and with that spinlock locked, calls netif_rx(). Once the execution reaches ks8851_start_xmit_par(), it calls ks8851_lock_par() again which attempts to claim the already locked spinlock again, and the hang happens.
Move the do_softirq() call outside of the spinlock protected section of ks8851_irq() by disabling BHs around the entire spinlock protected section of ks8851_irq() handler. Place local_bh_enable() outside of the spinlock protected section, so that it can trigger do_softirq() without the ks8851_par.c ks8851_lock_par() spinlock being held, and safely call ks8851_start_xmit_par() without attempting to lock the already locked spinlock.
Since ks8851_irq() is protected by local_bh_disable()/local_bh_enable() now, replace netif_rx() with __netif_rx() which is not duplicating the local_bh_disable()/local_bh_enable() calls.
--- drivers/net/ethernet/micrel/ks8851_common.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/ethernet/micrel/ks8851_common.c +5 -1
@@ -299,7 +299,7 @@ static void ks8851_rx_pkts(struct ks8851_net *ks)ks8851_dbg_dumpkkt (ks, rxpkt); skb->protocol =eth_type_trans (skb, ks->netdev); -netif_rx (skb); +__netif_rx (skb); ks->netdev->stats.rx_packets++; ks->netdev->stats.rx_bytes += rxlen;
@@ -330,6 +330,8 @@ static irqreturn_t ks8851_irq(int irq, void *_ks)unsigned long flags;unsigned int status; +local_bh_disable (); +ks8851_lock (ks, &flags); status =ks8851_rdreg16 (ks,KS_ISR );
@@ -406,6 +408,8 @@ static irqreturn_t ks8851_irq(int irq, void *_ks)if (status &IRQ_LCI )mii_check_link (&ks->mii); +local_bh_enable (); +return IRQ_HANDLED ; }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 64/172 af_unix: Clear stale u->oob_skb.
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit b46f4eaa4f0ec38909fb0072eea3aeddb32f954e ]
syzkaller started to report deadlock of unix_gc_lock after commit 4090fa373f0e ("af_unix: Replace garbage collection algorithm."), but it just uncovers the bug that has been there since commit 314001f0bf92 ("af_unix: Add OOB support").
The repro basically does the following.
from socket import *
from array import array
c1, c2 = socketpair(AF_UNIX, SOCK_STREAM)
c1.sendmsg([b'a'], [(SOL_SOCKET, SCM_RIGHTS, array("i", [c2.fileno()]))], MSG_OOB) c2.recv(1) # blocked as no normal data in recv queue
c2.close() # done async and unblock recv()
c1.close() # done async and trigger GC
A socket sends its file descriptor to itself as OOB data and tries to receive normal data, but finally recv() fails due to async close().
The problem here is wrong handling of OOB skb in manage_oob(). When recvmsg() is called without MSG_OOB, manage_oob() is called to check if the peeked skb is OOB skb. In such a case, manage_oob() pops it out of the receive queue but does not clear unix_sock(sk)->oob_skb. This is wrong in terms of uAPI.
Let's say we send "hello" with MSG_OOB, and "world" without MSG_OOB. The 'o' is handled as OOB data. When recv() is called twice without MSG_OOB, the OOB data should be lost.
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
from socket import *
c1, c2 =socketpair(AF_UNIX, SOCK_STREAM, 0)c1.send(b'hello', MSG_OOB)# 'o' is OOB data
5
Quoted 1 line
Quoted 1 line
Quoted 1 line
c1.send(b'world')
5
Quoted 1 line
Quoted 1 line
Quoted 1 line
c2.recv(5)# OOB data is not received
b'hell'
Quoted 1 line
Quoted 1 line
Quoted 1 line
c2.recv(5)# OOB date is skipped
b'world'
Quoted 1 line
Quoted 1 line
Quoted 1 line
c2.recv(5, MSG_OOB)# This should return an error
b'o'
In the same situation, TCP actually returns -EINVAL for the last recv().
Also, if we do not clear unix_sk(sk)->oob_skb, unix_poll() always set EPOLLPRI even though the data has passed through by previous recv().
To avoid these issues, we must clear unix_sk(sk)->oob_skb when dequeuing it from recv queue.
The reason why the old GC did not trigger the deadlock is because the old GC relied on the receive queue to detect the loop.
When it is triggered, the socket with OOB data is marked as GC candidate because file refcount == inflight count (1). However, after traversing all inflight sockets, the socket still has a positive inflight count (1), thus the socket is excluded from candidates. Then, the old GC lose the chance to garbage-collect the socket.
With the old GC, the repro continues to create true garbage that will never be freed nor detected by kmemleak as it's linked to the global inflight list. That's why we couldn't even notice the issue.
--- net/unix/af_unix.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
net/unix/af_unix.c +3 -1
@@ -2604,7 +2604,9 @@ static struct sk_buff *manage_oob(struct sk_buff *skb, struct sock *sk, } }else if (!(flags &MSG_PEEK )) {skb_unlink (skb, &sk->sk_receive_queue); -consume_skb (skb); +WRITE_ONCE (u->oob_skb,NULL ); +if (!WARN_ON_ONCE (skb_unref (skb))) +kfree_skb (skb); skb =skb_peek (&sk->sk_receive_queue); } }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 65/172 octeontx2-af: Fix NIX SQ mode and BP config
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Geetha sowjanya <[email protected]>
[ Upstream commit faf23006185e777db18912685922c5ddb2df383f ]
NIX SQ mode and link backpressure configuration is required for all platforms. But in current driver this code is wrongly placed under specific platform check. This patch fixes the issue by moving the code out of platform check.
--- .../ethernet/marvell/octeontx2/af/rvu_nix.c | 20 +++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-)
drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c +10 -10
@@ -4721,18 +4721,18 @@ static int rvu_nix_block_init(struct rvu *rvu, struct nix_hw *nix_hw)*/ rvu_write64 (rvu, blkaddr,NIX_AF_CFG ,rvu_read64 (rvu, blkaddr,NIX_AF_CFG ) |0x40ULL ); + } -/* Set chan/link to backpressure TL3 instead of TL2 */ -rvu_write64 (rvu, blkaddr,NIX_AF_PSE_CHANNEL_LEVEL ,0x01 ); +/* Set chan/link to backpressure TL3 instead of TL2 */ +rvu_write64 (rvu, blkaddr,NIX_AF_PSE_CHANNEL_LEVEL ,0x01 ); -/* Disable SQ manager's sticky mode operation (set TM6 = 0) -* This sticky mode is known to cause SQ stalls when multiple -* SQs are mapped to same SMQ and transmitting pkts at a time. -*/ - cfg =rvu_read64 (rvu, blkaddr,NIX_AF_SQM_DBG_CTL_STATUS ); - cfg &= ~BIT_ULL (15 ); -rvu_write64 (rvu, blkaddr,NIX_AF_SQM_DBG_CTL_STATUS , cfg); - } +/* Disable SQ manager's sticky mode operation (set TM6 = 0) +* This sticky mode is known to cause SQ stalls when multiple +* SQs are mapped to same SMQ and transmitting pkts at a time. +*/ + cfg =rvu_read64 (rvu, blkaddr,NIX_AF_SQM_DBG_CTL_STATUS ); + cfg &= ~BIT_ULL (15 ); +rvu_write64 (rvu, blkaddr,NIX_AF_SQM_DBG_CTL_STATUS , cfg); ltdefs = rvu->kpu.lt_def;/* Calibrate X2P bus to check if CGX/LBK links are fine */
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 66/172 ipv6: fib: hide unused pn variable
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit 74043489fcb5e5ca4074133582b5b8011b67f9e7 ]
When CONFIG_IPV6_SUBTREES is disabled, the only user is hidden, causing a 'make W=1' warning:
net/ipv6/ip6_fib.c: In function 'fib6_add':net/ipv6/ip6_fib.c:1388:32: error: variable 'pn' set but not used [-Werror=unused-but-set-variable]
Add another #ifdef around the variable declaration, matching the other uses in this file.
--- net/ipv6/ip6_fib.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
net/ipv6/ip6_fib.c +5 -2
@@ -1375,7 +1375,10 @@ int fib6_add(struct fib6_node *root, struct fib6_info *rt, struct nl_info *info, struct netlink_ext_ack *extack) { struct fib6_table *table = rt->fib6_table; -struct fib6_node *fn, *pn =NULL ; +struct fib6_node *fn; +#ifdef CONFIG_IPV6_SUBTREES +struct fib6_node *pn =NULL ; +#endif int err = -ENOMEM ;int allow_create =1 ;int replace_required =0 ;
@@ -1399,9 +1402,9 @@ int fib6_add(struct fib6_node *root, struct fib6_info *rt, goto out; } +#ifdef CONFIG_IPV6_SUBTREES pn = fn; -#ifdef CONFIG_IPV6_SUBTREES if (rt->fib6_src.plen) {struct fib6_node *sn;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 67/172 ipv4/route: avoid unused-but-set-variable warning
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit cf1b7201df59fb936f40f4a807433fe3f2ce310a ]
The log_martians variable is only used in an #ifdef, causing a 'make W=1' warning with gcc:
net/ipv4/route.c: In function 'ip_rt_send_redirect':net/ipv4/route.c:880:13: error: variable 'log_martians' set but not used [-Werror=unused-but-set-variable]
Change the #ifdef to an equivalent IS_ENABLED() to let the compiler see where the variable is used.
--- net/ipv4/route.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
net/ipv4/route.c +1 -3
@@ -926,13 +926,11 @@ void ip_rt_send_redirect(struct sk_buff *skb)icmp_send (skb,ICMP_REDIRECT ,ICMP_REDIR_HOST , gw); peer->rate_last = jiffies; ++peer->n_redirects; -#ifdef CONFIG_IP_ROUTE_VERBOSE -if (log_martians && +if (IS_ENABLED (CONFIG_IP_ROUTE_VERBOSE ) && log_martians && peer->n_redirects == ip_rt_redirect_number)net_warn_ratelimited ("host %pI4/if%d ignores redirects for %pI4 to %pI4\n" , &ip_hdr (skb)->saddr,inet_iif (skb), &ip_hdr (skb)->daddr, &gw); -#endif } out_put_peer:inet_putpeer (peer);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 68/172 ipv6: fix race condition between ipv6_get_ifaddr and ipv6_del_addr
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jiri Benc <[email protected]>
[ Upstream commit 7633c4da919ad51164acbf1aa322cc1a3ead6129 ]
Although ipv6_get_ifaddr walks inet6_addr_lst under the RCU lock, it still means hlist_for_each_entry_rcu can return an item that got removed from the list. The memory itself of such item is not freed thanks to RCU but nothing guarantees the actual content of the memory is sane.
In particular, the reference count can be zero. This can happen if ipv6_del_addr is called in parallel. ipv6_del_addr removes the entry from inet6_addr_lst (hlist_del_init_rcu(&ifp->addr_lst)) and drops all references (__in6_ifa_put(ifp) + in6_ifa_put(ifp)). With bad enough timing, this can happen:
1. In ipv6_get_ifaddr, hlist_for_each_entry_rcu returns an entry.
2. Then, the whole ipv6_del_addr is executed for the given entry. The reference count drops to zero and kfree_rcu is scheduled.
3. ipv6_get_ifaddr continues and tries to increments the reference count (in6_ifa_hold).
4. The rcu is unlocked and the entry is freed.
5. The freed entry is returned.
Prevent increasing of the reference count in such case. The name in6_ifa_hold_safe is chosen to mimic the existing fib6_info_hold_safe.
[ 41.506330] refcount_t: addition on 0; use-after-free. [ 41.506760] WARNING: CPU: 0 PID: 595 at lib/refcount.c:25 refcount_warn_saturate+0xa5/0x130 [ 41.507413] Modules linked in: veth bridge stp llc [ 41.507821] CPU: 0 PID: 595 Comm: python3 Not tainted 6.9.0-rc2.main-00208-g49563be82afa #14 [ 41.508479] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) [ 41.509163] RIP: 0010:refcount_warn_saturate+0xa5/0x130 [ 41.509586] Code: ad ff 90 0f 0b 90 90 c3 cc cc cc cc 80 3d c0 30 ad 01 00 75 a0 c6 05 b7 30 ad 01 01 90 48 c7 c7 38 cc 7a 8c e8 cc 18 ad ff 90 <0f> 0b 90 90 c3 cc cc cc cc 80 3d 98 30 ad 01 00 0f 85 75 ff ff ff [ 41.510956] RSP: 0018:ffffbda3c026baf0 EFLAGS: 00010282 [ 41.511368] RAX: 0000000000000000 RBX: ffff9e9c46914800 RCX: 0000000000000000 [ 41.511910] RDX: ffff9e9c7ec29c00 RSI: ffff9e9c7ec1c900 RDI: ffff9e9c7ec1c900 [ 41.512445] RBP: ffff9e9c43660c9c R08: 0000000000009ffb R09: 00000000ffffdfff [ 41.512998] R10: 00000000ffffdfff R11: ffffffff8ca58a40 R12: ffff9e9c4339a000 [ 41.513534] R13: 0000000000000001 R14: ffff9e9c438a0000 R15: ffffbda3c026bb48 [ 41.514086] FS: 00007fbc4cda1740(0000) GS:ffff9e9c7ec00000(0000) knlGS:0000000000000000 [ 41.514726] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 41.515176] CR2: 000056233b337d88 CR3: 000000000376e006 CR4: 0000000000370ef0 [ 41.515713] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 41.516252] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 41.516799] Call Trace: [ 41.517037] <TASK> [ 41.517249] ? __warn+0x7b/0x120 [ 41.517535] ? refcount_warn_saturate+0xa5/0x130 [ 41.517923] ? report_bug+0x164/0x190 [ 41.518240] ? handle_bug+0x3d/0x70 [ 41.518541] ? exc_invalid_op+0x17/0x70 [ 41.520972] ? asm_exc_invalid_op+0x1a/0x20 [ 41.521325] ? refcount_warn_saturate+0xa5/0x130 [ 41.521708] ipv6_get_ifaddr+0xda/0xe0 [ 41.522035] inet6_rtm_getaddr+0x342/0x3f0 [ 41.522376] ? __pfx_inet6_rtm_getaddr+0x10/0x10 [ 41.522758] rtnetlink_rcv_msg+0x334/0x3d0 [ 41.523102] ? netlink_unicast+0x30f/0x390 [ 41.523445] ? __pfx_rtnetlink_rcv_msg+0x10/0x10 [ 41.523832] netlink_rcv_skb+0x53/0x100 [ 41.524157] netlink_unicast+0x23b/0x390 [ 41.524484] netlink_sendmsg+0x1f2/0x440 [ 41.524826] __sys_sendto+0x1d8/0x1f0 [ 41.525145] __x64_sys_sendto+0x1f/0x30 [ 41.525467] do_syscall_64+0xa5/0x1b0 [ 41.525794] entry_SYSCALL_64_after_hwframe+0x72/0x7a [ 41.526213] RIP: 0033:0x7fbc4cfcea9a [ 41.526528] Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f 1e fa 41 89 ca 64 8b 04 25 18 00 00 00 85 c0 75 15 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 7e c3 0f 1f 44 00 00 41 54 48 83 ec 30 44 89 [ 41.527942] RSP: 002b:00007ffcf54012a8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c [ 41.528593] RAX: ffffffffffffffda RBX: 00007ffcf5401368 RCX: 00007fbc4cfcea9a [ 41.529173] RDX: 000000000000002c RSI: 00007fbc4b9d9bd0 RDI: 0000000000000005 [ 41.529786] RBP: 00007fbc4bafb040 R08: 00007ffcf54013e0 R09: 000000000000000c [ 41.530375] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 [ 41.530977] R13: ffffffffc4653600 R14: 0000000000000001 R15: 00007fbc4ca85d1b [ 41.531573] </TASK>
--- include/net/addrconf.h | 4 ++++ net/ipv6/addrconf.c | 7 ++++--- 2 files changed, 8 insertions(+), 3 deletions(-)
include/net/addrconf.h +4 -0
@@ -437,6 +437,10 @@ static inline void in6_ifa_hold(struct inet6_ifaddr *ifp)refcount_inc (&ifp->refcnt); } +static inline bool in6_ifa_hold_safe (struct inet6_ifaddr *ifp) +{ +return refcount_inc_not_zero (&ifp->refcnt); +}/* * compute link-local solicited-node multicast address
net/ipv6/addrconf.c +4 -3
@@ -2061,9 +2061,10 @@ struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, const struct in6_addr *addif (ipv6_addr_equal (&ifp->addr, addr)) {if (!dev || ifp->idev->dev == dev || !(ifp->scope&(IFA_LINK |IFA_HOST ) || strict)) { - result = ifp; -in6_ifa_hold (ifp); -break ; +if (in6_ifa_hold_safe (ifp)) { + result = ifp; +break ; + } } } }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 69/172 pds_core: use pci_reset_function for health reset
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Shannon Nelson <[email protected]>
[ Upstream commit 2cbab3c296f1addd73b40549a2271b30f960df8b ]
We get the benefit of all the PCI reset locking and recovery if we use the existing pci_reset_function() that will call our local reset handlers.
Stable-dep-of: 81665adf25d2 ("pds_core: Fix pdsc_check_pci_health function to use work thread")
--- drivers/net/ethernet/amd/pds_core/core.c | 3 +-- drivers/net/ethernet/amd/pds_core/core.h | 3 --- drivers/net/ethernet/amd/pds_core/main.c | 7 ++++--- 3 files changed, 5 insertions(+), 8 deletions(-)
drivers/net/ethernet/amd/pds_core/core.c +1 -2
@@ -609,8 +609,7 @@ static void pdsc_check_pci_health(struct pdsc *pdsc)if (fw_status !=PDS_RC_BAD_PCI )return ; -pdsc_reset_prepare (pdsc->pdev); -pdsc_reset_done (pdsc->pdev); +pci_reset_function (pdsc->pdev); }void pdsc_health_thread (struct work_struct *work)
drivers/net/ethernet/amd/pds_core/core.h +0 -3
@@ -283,9 +283,6 @@ int pdsc_devcmd_init(struct pdsc *pdsc);int pdsc_devcmd_reset (struct pdsc *pdsc);int pdsc_dev_init (struct pdsc *pdsc); -void pdsc_reset_prepare (struct pci_dev *pdev); -void pdsc_reset_done (struct pci_dev *pdev); -int pdsc_intr_alloc (struct pdsc *pdsc,char *name,irq_handler_t handler,void *data);void pdsc_intr_free (struct pdsc *pdsc,int index);
drivers/net/ethernet/amd/pds_core/main.c +4 -3
@@ -468,7 +468,7 @@ static void pdsc_restart_health_thread(struct pdsc *pdsc)mod_timer (&pdsc->wdtimer, jiffies +1 ); } -void pdsc_reset_prepare (struct pci_dev *pdev) +static void pdsc_reset_prepare (struct pci_dev *pdev) {struct pdsc *pdsc =pci_get_drvdata (pdev);
@@ -477,10 +477,11 @@ void pdsc_reset_prepare(struct pci_dev *pdev)pdsc_unmap_bars (pdsc);pci_release_regions (pdev); -pci_disable_device (pdev); +if (pci_is_enabled (pdev)) +pci_disable_device (pdev); } -void pdsc_reset_done (struct pci_dev *pdev) +static void pdsc_reset_done (struct pci_dev *pdev) {struct pdsc *pdsc =pci_get_drvdata (pdev);struct device *dev = pdsc->dev;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 70/172 pds_core: Fix pdsc_check_pci_health function to use work thread
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Brett Creeley <[email protected]>
[ Upstream commit 81665adf25d28a00a986533f1d3a5df76b79cad9 ]
When the driver notices fw_status == 0xff it tries to perform a PCI reset on itself via pci_reset_function() in the context of the driver's health thread. However, pdsc_reset_prepare calls pdsc_stop_health_thread(), which attempts to stop/flush the health thread. This results in a deadlock because the stop/flush will never complete since the driver called pci_reset_function() from the health thread context. Fix by changing the pdsc_check_pci_health_function() to queue a newly introduced pdsc_pci_reset_thread() on the pdsc's work queue.
Unloading the driver in the fw_down/dead state uncovered another issue, which can be seen in the following trace:
WARNING: CPU: 51 PID: 6914 at kernel/workqueue.c:1450 __queue_work+0x358/0x440 [...] RIP: 0010:__queue_work+0x358/0x440 [...] Call Trace: <TASK> ? __warn+0x85/0x140 ? __queue_work+0x358/0x440 ? report_bug+0xfc/0x1e0 ? handle_bug+0x3f/0x70 ? exc_invalid_op+0x17/0x70 ? asm_exc_invalid_op+0x1a/0x20 ? __queue_work+0x358/0x440 queue_work_on+0x28/0x30 pdsc_devcmd_locked+0x96/0xe0 [pds_core] pdsc_devcmd_reset+0x71/0xb0 [pds_core] pdsc_teardown+0x51/0xe0 [pds_core] pdsc_remove+0x106/0x200 [pds_core] pci_device_remove+0x37/0xc0 device_release_driver_internal+0xae/0x140 driver_detach+0x48/0x90 bus_remove_driver+0x6d/0xf0 pci_unregister_driver+0x2e/0xa0 pdsc_cleanup_module+0x10/0x780 [pds_core] __x64_sys_delete_module+0x142/0x2b0 ? syscall_trace_enter.isra.18+0x126/0x1a0 do_syscall_64+0x3b/0x90 entry_SYSCALL_64_after_hwframe+0x72/0xdc RIP: 0033:0x7fbd9d03a14b
[...]
Fix this by preventing the devcmd reset if the FW is not running.
--- drivers/net/ethernet/amd/pds_core/core.c | 13 ++++++++++++- drivers/net/ethernet/amd/pds_core/core.h | 2 ++ drivers/net/ethernet/amd/pds_core/dev.c | 3 +++ drivers/net/ethernet/amd/pds_core/main.c | 1 + 4 files changed, 18 insertions(+), 1 deletion(-)
drivers/net/ethernet/amd/pds_core/core.c +12 -1
@@ -595,6 +595,16 @@ void pdsc_fw_up(struct pdsc *pdsc)pdsc_teardown (pdsc,PDSC_TEARDOWN_RECOVERY ); } +void pdsc_pci_reset_thread (struct work_struct *work) +{ + struct pdsc *pdsc = container_of(work, struct pdsc, pci_reset_work); +struct pci_dev *pdev = pdsc->pdev; + +pci_dev_get (pdev); +pci_reset_function (pdev); +pci_dev_put (pdev); +} +static void pdsc_check_pci_health (struct pdsc *pdsc) {u8 fw_status;
@@ -609,7 +619,8 @@ static void pdsc_check_pci_health(struct pdsc *pdsc)if (fw_status !=PDS_RC_BAD_PCI )return ; -pci_reset_function (pdsc->pdev); +/* prevent deadlock between pdsc_reset_prepare and pdsc_health_thread */ +queue_work (pdsc->wq, &pdsc->pci_reset_work); }void pdsc_health_thread (struct work_struct *work)
drivers/net/ethernet/amd/pds_core/core.h +2 -0
@@ -197,6 +197,7 @@ struct pdsc {struct pdsc_qcq notifyqcq;u64 last_eid;struct pdsc_viftype *viftype_status; +struct work_struct pci_reset_work; };/** enum pds_core_dbell_bits - bitwise composition of dbell values.
@@ -312,5 +313,6 @@ int pdsc_firmware_update(struct pdsc *pdsc, const struct firmware *fw,void pdsc_fw_down (struct pdsc *pdsc);void pdsc_fw_up (struct pdsc *pdsc); +void pdsc_pci_reset_thread (struct work_struct *work);#endif /* _PDSC_H_ */
drivers/net/ethernet/amd/pds_core/dev.c +3 -0
@@ -229,6 +229,9 @@ int pdsc_devcmd_reset(struct pdsc *pdsc) .reset.opcode = PDS_CORE_CMD_RESET, }; +if (!pdsc_is_fw_running (pdsc)) +return 0 ; +return pdsc_devcmd (pdsc, &cmd, &comp, pdsc->devcmd_timeout); }
drivers/net/ethernet/amd/pds_core/main.c +1 -0
@@ -238,6 +238,7 @@ static int pdsc_init_pf(struct pdsc *pdsc)snprintf (wq_name,sizeof (wq_name),"%s.%d" ,PDS_CORE_DRV_NAME , pdsc->uid); pdsc->wq =create_singlethread_workqueue (wq_name);INIT_WORK (&pdsc->health_work, pdsc_health_thread); +INIT_WORK (&pdsc->pci_reset_work, pdsc_pci_reset_thread);timer_setup (&pdsc->wdtimer, pdsc_wdtimer_cb,0 ); pdsc->wdtimer_period =PDSC_WATCHDOG_SECS *HZ ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 71/172 Bluetooth: ISO: Align broadcast sync_timeout with connection timeout
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit 42ed95de82c01184a88945d3ca274be6a7ea607d ]
This aligns broadcast sync_timeout with existing connection timeouts which are 20 seconds long.
Stable-dep-of: b37cab587aa3 ("Bluetooth: ISO: Don't reject BT_ISO_QOS if parameters are unset")
--- include/net/bluetooth/bluetooth.h | 2 ++ net/bluetooth/iso.c | 4 ++-- 2 files changed, 4 insertions(+), 2 deletions(-)
include/net/bluetooth/bluetooth.h +2 -0
@@ -164,6 +164,8 @@ struct bt_voice {#define BT_ISO_QOS_BIG_UNSET 0xff#define BT_ISO_QOS_BIS_UNSET 0xff +#define BT_ISO_SYNC_TIMEOUT 0x07d0/* 20 secs */ +struct bt_iso_io_qos {__u32 interval;__u16 latency;
net/bluetooth/iso.c +2 -2
@@ -837,10 +837,10 @@ static struct bt_iso_qos default_qos = { .bcode = {0x00}, .options = 0x00, .skip = 0x0000, - .sync_timeout = 0x4000, + .sync_timeout = BT_ISO_SYNC_TIMEOUT, .sync_cte_type = 0x00, .mse = 0x00, - .timeout = 0x4000, + .timeout = BT_ISO_SYNC_TIMEOUT, }, };
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 72/172 Bluetooth: ISO: Dont reject BT_ISO_QOS if parameters are unset
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit b37cab587aa3c9ab29c6b10aa55627dad713011f ]
Consider certain values (0x00) as unset and load proper default if an application has not set them properly.
--- net/bluetooth/iso.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
net/bluetooth/iso.c +8 -2
@@ -1430,8 +1430,8 @@ static bool check_ucast_qos(struct bt_iso_qos *qos)static bool check_bcast_qos (struct bt_iso_qos *qos) { -if (qos->bcast.sync_factor ==0x00 ) -return false; +if (!qos->bcast.sync_factor) + qos->bcast.sync_factor =0x01 ;if (qos->bcast.packing >0x01 )return false;
@@ -1454,6 +1454,9 @@ static bool check_bcast_qos(struct bt_iso_qos *qos)if (qos->bcast.skip >0x01f3 )return false; +if (!qos->bcast.sync_timeout) + qos->bcast.sync_timeout =BT_ISO_SYNC_TIMEOUT ; +if (qos->bcast.sync_timeout <0x000a || qos->bcast.sync_timeout >0x4000 )return false;
@@ -1463,6 +1466,9 @@ static bool check_bcast_qos(struct bt_iso_qos *qos)if (qos->bcast.mse >0x1f )return false; +if (!qos->bcast.timeout) + qos->bcast.sync_timeout =BT_ISO_SYNC_TIMEOUT ; +if (qos->bcast.timeout <0x000a || qos->bcast.timeout >0x4000 )return false;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 73/172 Bluetooth: hci_sync: Use QoS to determine which PHY to scan
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit 22cbf4f84c00da64196eb15034feee868e63eef0 ]
This used the hci_conn QoS to determine which PHY to scan when creating a PA Sync.
Stable-dep-of: 53cb4197e63a ("Bluetooth: hci_sync: Fix using the same interval and window for Coded PHY")
--- net/bluetooth/hci_sync.c | 66 +++++++++++++++++++++++++++++++++------- 1 file changed, 55 insertions(+), 11 deletions(-)
net/bluetooth/hci_sync.c +55 -11
@@ -2611,6 +2611,14 @@ static u8 hci_update_accept_list_sync(struct hci_dev *hdev)return filter_policy; } +static void hci_le_scan_phy_params (struct hci_cp_le_scan_phy_params *cp, +u8 type,u16 interval,u16 window) +{ + cp->type = type; + cp->interval =cpu_to_le16 (interval); + cp->window =cpu_to_le16 (window); +} +static int hci_le_set_ext_scan_param_sync (struct hci_dev *hdev,u8 type,u16 interval,u16 window,u8 own_addr_type,u8 filter_policy)
@@ -2618,7 +2626,7 @@ static int hci_le_set_ext_scan_param_sync(struct hci_dev *hdev, u8 type,struct hci_cp_le_set_ext_scan_params *cp;struct hci_cp_le_scan_phy_params *phy;u8 data[sizeof (*cp) +sizeof (*phy) *2 ]; -u8 num_phy =0 ; +u8 num_phy =0x00 ; cp = (void *)data; phy = (void *)cp->data;
@@ -2628,28 +2636,64 @@ static int hci_le_set_ext_scan_param_sync(struct hci_dev *hdev, u8 type, cp->own_addr_type = own_addr_type; cp->filter_policy = filter_policy; +/* Check if PA Sync is in progress then select the PHY based on the +* hci_conn.iso_qos. +*/ +if (hci_dev_test_flag (hdev,HCI_PA_SYNC )) { +struct hci_cp_le_add_to_accept_list *sent; + + sent =hci_sent_cmd_data (hdev,HCI_OP_LE_ADD_TO_ACCEPT_LIST ); +if (sent) { +struct hci_conn *conn; + + conn =hci_conn_hash_lookup_ba (hdev,ISO_LINK , + &sent->bdaddr); +if (conn) { +struct bt_iso_qos *qos = &conn->iso_qos; + +if (qos->bcast.in.phy &BT_ISO_PHY_1M || + qos->bcast.in.phy &BT_ISO_PHY_2M ) { + cp->scanning_phys |=LE_SCAN_PHY_1M ; +hci_le_scan_phy_params (phy, type, + interval, + window); + num_phy++; + phy++; + } + +if (qos->bcast.in.phy &BT_ISO_PHY_CODED ) { + cp->scanning_phys |=LE_SCAN_PHY_CODED ; +hci_le_scan_phy_params (phy, type, + interval, + window); + num_phy++; + phy++; + } + +if (num_phy) + goto done; + } + } + } +if (scan_1m (hdev) ||scan_2m (hdev)) { cp->scanning_phys |=LE_SCAN_PHY_1M ; - - phy->type = type; - phy->interval =cpu_to_le16 (interval); - phy->window =cpu_to_le16 (window); - +hci_le_scan_phy_params (phy, type, interval, window); num_phy++; phy++; }if (scan_coded (hdev)) { cp->scanning_phys |=LE_SCAN_PHY_CODED ; - - phy->type = type; - phy->interval =cpu_to_le16 (interval); - phy->window =cpu_to_le16 (window); - +hci_le_scan_phy_params (phy, type, interval, window); num_phy++; phy++; } +done: +if (!num_phy) +return -EINVAL ; +return __hci_cmd_sync_status (hdev,HCI_OP_LE_SET_EXT_SCAN_PARAMS ,sizeof (*cp) +sizeof (*phy) * num_phy, data,HCI_CMD_TIMEOUT );
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 74/172 Bluetooth: hci_sync: Fix using the same interval and window for Coded PHY
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit 53cb4197e63ab2363aa28c3029061e4d516e7626 ]
Coded PHY recommended intervals are 3 time bigger than the 1M PHY so this aligns with that by multiplying by 3 the values given to 1M PHY since the code already used recommended values for that.
--- net/bluetooth/hci_sync.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
net/bluetooth/hci_sync.c +3 -3
@@ -2664,8 +2664,8 @@ static int hci_le_set_ext_scan_param_sync(struct hci_dev *hdev, u8 type,if (qos->bcast.in.phy &BT_ISO_PHY_CODED ) { cp->scanning_phys |=LE_SCAN_PHY_CODED ;hci_le_scan_phy_params (phy, type, - interval, - window); + interval *3 , + window *3 ); num_phy++; phy++; }
@@ -2685,7 +2685,7 @@ static int hci_le_set_ext_scan_param_sync(struct hci_dev *hdev, u8 type,if (scan_coded (hdev)) { cp->scanning_phys |=LE_SCAN_PHY_CODED ; -hci_le_scan_phy_params (phy, type, interval, window); +hci_le_scan_phy_params (phy, type, interval *3 , window *3 ); num_phy++; phy++; }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 75/172 Bluetooth: SCO: Fix not validating setsockopt user input
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit 51eda36d33e43201e7a4fd35232e069b2c850b01 ]
syzbot reported sco_sock_setsockopt() is copying data without checking user input length.
BUG: KASAN: slab-out-of-bounds in copy_from_sockptr_offset include/linux/sockptr.h:49 [inline] BUG: KASAN: slab-out-of-bounds in copy_from_sockptr include/linux/sockptr.h:55 [inline] BUG: KASAN: slab-out-of-bounds in sco_sock_setsockopt+0xc0b/0xf90 net/bluetooth/sco.c:893 Read of size 4 at addr ffff88805f7b15a3 by task syz-executor.5/12578
--- include/net/bluetooth/bluetooth.h | 9 +++++++++ net/bluetooth/sco.c | 23 ++++++++++------------- 2 files changed, 19 insertions(+), 13 deletions(-)
include/net/bluetooth/bluetooth.h +9 -0
@@ -585,6 +585,15 @@ static inline struct sk_buff *bt_skb_sendmmsg(struct sock *sk,return skb; } +static inline int bt_copy_from_sockptr (void *dst,size_t dst_size, +sockptr_t src,size_t src_size) +{ +if (dst_size > src_size) +return -EINVAL ; + +return copy_from_sockptr (dst, src, dst_size); +} +int bt_to_errno (u16 code);__u8 bt_status (int err);
net/bluetooth/sco.c +10 -13
@@ -823,7 +823,7 @@ static int sco_sock_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval, unsigned int optlen) {struct sock *sk = sock->sk; -int len, err =0 ; +int err =0 ;struct bt_voice voice;u32 opt;struct bt_codecs *codecs;
@@ -842,10 +842,9 @@ static int sco_sock_setsockopt(struct socket *sock, int level, int optname,break ; } -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - }if (opt)set_bit (BT_SK_DEFER_SETUP , &bt_sk (sk)->flags);
@@ -862,11 +861,10 @@ static int sco_sock_setsockopt(struct socket *sock, int level, int optname, voice.setting =sco_pi (sk)->setting; - len = min_t(unsigned int, sizeof(voice), optlen); -if (copy_from_sockptr (&voice, optval, len)) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&voice,sizeof (voice), optval, + optlen); +if (err)break ; - }/* Explicitly check for these values */ if (voice.setting != BT_VOICE_TRANSPARENT &&
@@ -889,10 +887,9 @@ static int sco_sock_setsockopt(struct socket *sock, int level, int optname,break ;case BT_PKT_STATUS : -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - }if (opt)set_bit (BT_SK_PKT_STATUS , &bt_sk (sk)->flags);
@@ -933,9 +930,9 @@ static int sco_sock_setsockopt(struct socket *sock, int level, int optname,break ; } -if (copy_from_sockptr (buffer, optval, optlen)) { + err =bt_copy_from_sockptr (buffer, optlen, optval, optlen); +if (err) {hci_dev_put (hdev); - err = -EFAULT ;break ; }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 76/172 Bluetooth: RFCOMM: Fix not validating setsockopt user input
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit a97de7bff13b1cc825c1b1344eaed8d6c2d3e695 ]
syzbot reported rfcomm_sock_setsockopt_old() is copying data without checking user input length.
BUG: KASAN: slab-out-of-bounds in copy_from_sockptr_offset include/linux/sockptr.h:49 [inline] BUG: KASAN: slab-out-of-bounds in copy_from_sockptr include/linux/sockptr.h:55 [inline] BUG: KASAN: slab-out-of-bounds in rfcomm_sock_setsockopt_old net/bluetooth/rfcomm/sock.c:632 [inline] BUG: KASAN: slab-out-of-bounds in rfcomm_sock_setsockopt+0x893/0xa70 net/bluetooth/rfcomm/sock.c:673 Read of size 4 at addr ffff8880209a8bc3 by task syz-executor632/5064
--- net/bluetooth/rfcomm/sock.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-)
net/bluetooth/rfcomm/sock.c +5 -9
@@ -629,7 +629,7 @@ static int rfcomm_sock_setsockopt_old(struct socket *sock, int optname,switch (optname) {case RFCOMM_LM : -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { +if (bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen)) { err = -EFAULT ;break ; }
@@ -664,7 +664,6 @@ static int rfcomm_sock_setsockopt(struct socket *sock, int level, int optname,struct sock *sk = sock->sk;struct bt_security sec;int err =0 ; -size_t len;u32 opt;BT_DBG ("sk %p" , sk);
@@ -686,11 +685,9 @@ static int rfcomm_sock_setsockopt(struct socket *sock, int level, int optname, sec.level =BT_SECURITY_LOW ; - len = min_t(unsigned int, sizeof(sec), optlen); -if (copy_from_sockptr (&sec, optval, len)) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&sec,sizeof (sec), optval, optlen); +if (err)break ; - }if (sec.level >BT_SECURITY_HIGH ) { err = -EINVAL ;
@@ -706,10 +703,9 @@ static int rfcomm_sock_setsockopt(struct socket *sock, int level, int optname,break ; } -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - }if (opt)set_bit (BT_SK_DEFER_SETUP , &bt_sk (sk)->flags);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 77/172 Bluetooth: L2CAP: Fix not validating setsockopt user input
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit 4f3951242ace5efc7131932e2e01e6ac6baed846 ]
Check user input length before copying data.
--- net/bluetooth/l2cap_sock.c | 52 +++++++++++++++----------------------- 1 file changed, 20 insertions(+), 32 deletions(-)
net/bluetooth/l2cap_sock.c +20 -32
@@ -726,7 +726,7 @@ static int l2cap_sock_setsockopt_old(struct socket *sock, int optname,struct sock *sk = sock->sk;struct l2cap_chan *chan =l2cap_pi (sk)->chan;struct l2cap_options opts; -int len, err =0 ; +int err =0 ;u32 opt;BT_DBG ("sk %p" , sk);
@@ -753,11 +753,9 @@ static int l2cap_sock_setsockopt_old(struct socket *sock, int optname, opts.max_tx = chan->max_tx; opts.txwin_size = chan->tx_win; - len = min_t(unsigned int, sizeof(opts), optlen); -if (copy_from_sockptr (&opts, optval, len)) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opts,sizeof (opts), optval, optlen); +if (err)break ; - }if (opts.txwin_size >L2CAP_DEFAULT_EXT_WINDOW ) { err = -EINVAL ;
@@ -800,10 +798,9 @@ static int l2cap_sock_setsockopt_old(struct socket *sock, int optname,break ;case L2CAP_LM : -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - }if (opt &L2CAP_LM_FIPS ) { err = -EINVAL ;
@@ -884,7 +881,7 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,struct bt_security sec;struct bt_power pwr;struct l2cap_conn *conn; -int len, err =0 ; +int err =0 ;u32 opt;u16 mtu;u8 mode;
@@ -910,11 +907,9 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname, sec.level =BT_SECURITY_LOW ; - len = min_t(unsigned int, sizeof(sec), optlen); -if (copy_from_sockptr (&sec, optval, len)) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&sec,sizeof (sec), optval, optlen); +if (err)break ; - }if (sec.level <BT_SECURITY_LOW || sec.level >BT_SECURITY_FIPS ) {
@@ -959,10 +954,9 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,break ; } -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - }if (opt) {set_bit (BT_SK_DEFER_SETUP , &bt_sk (sk)->flags);
@@ -974,10 +968,9 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,break ;case BT_FLUSHABLE : -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - }if (opt >BT_FLUSHABLE_ON ) { err = -EINVAL ;
@@ -1009,11 +1002,9 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname, pwr.force_active =BT_POWER_FORCE_ACTIVE_ON ; - len = min_t(unsigned int, sizeof(pwr), optlen); -if (copy_from_sockptr (&pwr, optval, len)) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&pwr,sizeof (pwr), optval, optlen); +if (err)break ; - }if (pwr.force_active)set_bit (FLAG_FORCE_ACTIVE , &chan->flags);
@@ -1022,10 +1013,9 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,break ;case BT_CHANNEL_POLICY : -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - } err = -EOPNOTSUPP ;break ;
@@ -1054,10 +1044,9 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,break ; } -if (copy_from_sockptr (&mtu, optval,sizeof (u16))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&mtu,sizeof (mtu), optval, optlen); +if (err)break ; - }if (chan->mode ==L2CAP_MODE_EXT_FLOWCTL && sk->sk_state ==BT_CONNECTED )
@@ -1085,10 +1074,9 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,break ; } -if (copy_from_sockptr (&mode, optval,sizeof (u8))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&mode,sizeof (mode), optval, optlen); +if (err)break ; - }BT_DBG ("mode %u" , mode);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 78/172 Bluetooth: ISO: Fix not validating setsockopt user input
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit 9e8742cdfc4b0e65266bb4a901a19462bda9285e ]
Check user input length before copying data.
--- net/bluetooth/iso.c | 36 ++++++++++++------------------------ 1 file changed, 12 insertions(+), 24 deletions(-)
net/bluetooth/iso.c +12 -24
@@ -1479,7 +1479,7 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval, unsigned int optlen) {struct sock *sk = sock->sk; -int len, err =0 ; +int err =0 ;struct bt_iso_qos qos = default_qos;u32 opt;
@@ -1494,10 +1494,9 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname,break ; } -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - }if (opt)set_bit (BT_SK_DEFER_SETUP , &bt_sk (sk)->flags);
@@ -1506,10 +1505,9 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname,break ;case BT_PKT_STATUS : -if (copy_from_sockptr (&opt, optval,sizeof (u32))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, optlen); +if (err)break ; - }if (opt)set_bit (BT_SK_PKT_STATUS , &bt_sk (sk)->flags);
@@ -1524,17 +1522,9 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname,break ; } - len = min_t(unsigned int, sizeof(qos), optlen); - -if (copy_from_sockptr (&qos, optval, len)) { - err = -EFAULT ; -break ; - } - -if (len ==sizeof (qos.ucast) && !check_ucast_qos (&qos)) { - err = -EINVAL ; + err =bt_copy_from_sockptr (&qos,sizeof (qos), optval, optlen); +if (err)break ; - }iso_pi (sk)->qos = qos;iso_pi (sk)->qos_user_set = true;
@@ -1549,18 +1539,16 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname, }if (optlen >sizeof (iso_pi (sk)->base)) { - err = -EOVERFLOW ; + err = -EINVAL ;break ; } - len = min_t(unsigned int, sizeof(iso_pi(sk)->base), optlen); - -if (copy_from_sockptr (iso_pi (sk)->base, optval, len)) { - err = -EFAULT ; + err =bt_copy_from_sockptr (iso_pi (sk)->base, optlen, optval, + optlen); +if (err)break ; - } -iso_pi (sk)->base_len = len; +iso_pi (sk)->base_len = optlen;break ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 79/172 Bluetooth: hci_sock: Fix not validating setsockopt user input
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luiz Augusto von Dentz <[email protected]>
[ Upstream commit b2186061d6043d6345a97100460363e990af0d46 ]
Check user input length before copying data.
--- net/bluetooth/hci_sock.c | 21 ++++++++------------- 1 file changed, 8 insertions(+), 13 deletions(-)
net/bluetooth/hci_sock.c +8 -13
@@ -1946,10 +1946,9 @@ static int hci_sock_setsockopt_old(struct socket *sock, int level, int optname,switch (optname) {case HCI_DATA_DIR : -if (copy_from_sockptr (&opt, optval,sizeof (opt))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, len); +if (err)break ; - }if (opt)hci_pi (sk)->cmsg_mask |=HCI_CMSG_DIR ;
@@ -1958,10 +1957,9 @@ static int hci_sock_setsockopt_old(struct socket *sock, int level, int optname,break ;case HCI_TIME_STAMP : -if (copy_from_sockptr (&opt, optval,sizeof (opt))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, len); +if (err)break ; - }if (opt)hci_pi (sk)->cmsg_mask |=HCI_CMSG_TSTAMP ;
@@ -1979,11 +1977,9 @@ static int hci_sock_setsockopt_old(struct socket *sock, int level, int optname, uf.event_mask[1 ] = *((u32 *) f->event_mask +1 ); } - len = min_t(unsigned int, len, sizeof(uf)); -if (copy_from_sockptr (&uf, optval, len)) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&uf,sizeof (uf), optval, len); +if (err)break ; - }if (!capable (CAP_NET_RAW )) { uf.type_mask &= hci_sec_filter.type_mask;
@@ -2042,10 +2038,9 @@ static int hci_sock_setsockopt(struct socket *sock, int level, int optname, goto done; } -if (copy_from_sockptr (&opt, optval,sizeof (opt))) { - err = -EFAULT ; + err =bt_copy_from_sockptr (&opt,sizeof (opt), optval, len); +if (err)break ; - }hci_pi (sk)->mtu = opt;break ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 80/172 Bluetooth: l2cap: Dont double set the HCI_CONN_MGMT_CONNECTED bit
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Archie Pusaka <[email protected]>
[ Upstream commit 600b0bbe73d3a9a264694da0e4c2c0800309141e ]
The bit is set and tested inside mgmt_device_connected(), therefore we must not set it just outside the function.
--- net/bluetooth/l2cap_core.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
net/bluetooth/l2cap_core.c +1 -2
@@ -4054,8 +4054,7 @@ static int l2cap_connect_req(struct l2cap_conn *conn,return -EPROTO ;hci_dev_lock (hdev); -if (hci_dev_test_flag (hdev,HCI_MGMT ) && - !test_and_set_bit (HCI_CONN_MGMT_CONNECTED , &hcon->flags)) +if (hci_dev_test_flag (hdev,HCI_MGMT ))mgmt_device_connected (hdev, hcon,NULL ,0 );hci_dev_unlock (hdev);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 81/172 netfilter: complete validation of user input
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit 65acf6e0501ac8880a4f73980d01b5d27648b956 ]
In my recent commit, I missed that do_replace() handlers use copy_from_sockptr() (which I fixed), followed by unsafe copy_from_sockptr_offset() calls.
In all functions, we can perform the @optlen validation before even calling xt_alloc_table_info() with the following check:
if ((u64)optlen < (u64)tmp.size + sizeof(tmp))
return -EINVAL;
--- net/ipv4/netfilter/arp_tables.c | 4 ++++ net/ipv4/netfilter/ip_tables.c | 4 ++++ net/ipv6/netfilter/ip6_tables.c | 4 ++++ 3 files changed, 12 insertions(+)
net/ipv4/netfilter/arp_tables.c +4 -0
@@ -966,6 +966,8 @@ static int do_replace(struct net *net, sockptr_t arg, unsigned int len)return -ENOMEM ;if (tmp.num_counters ==0 )return -EINVAL ; +if ((u64 )len < (u64 )tmp.size +sizeof (tmp)) +return -EINVAL ; tmp.name[sizeof (tmp.name)-1 ] =0 ;
@@ -1266,6 +1268,8 @@ static int compat_do_replace(struct net *net, sockptr_t arg, unsigned int len)return -ENOMEM ;if (tmp.num_counters ==0 )return -EINVAL ; +if ((u64 )len < (u64 )tmp.size +sizeof (tmp)) +return -EINVAL ; tmp.name[sizeof (tmp.name)-1 ] =0 ;
net/ipv4/netfilter/ip_tables.c +4 -0
@@ -1118,6 +1118,8 @@ do_replace(struct net *net, sockptr_t arg, unsigned int len)return -ENOMEM ;if (tmp.num_counters ==0 )return -EINVAL ; +if ((u64 )len < (u64 )tmp.size +sizeof (tmp)) +return -EINVAL ; tmp.name[sizeof (tmp.name)-1 ] =0 ;
@@ -1504,6 +1506,8 @@ compat_do_replace(struct net *net, sockptr_t arg, unsigned int len)return -ENOMEM ;if (tmp.num_counters ==0 )return -EINVAL ; +if ((u64 )len < (u64 )tmp.size +sizeof (tmp)) +return -EINVAL ; tmp.name[sizeof (tmp.name)-1 ] =0 ;
net/ipv6/netfilter/ip6_tables.c +4 -0
@@ -1135,6 +1135,8 @@ do_replace(struct net *net, sockptr_t arg, unsigned int len)return -ENOMEM ;if (tmp.num_counters ==0 )return -EINVAL ; +if ((u64 )len < (u64 )tmp.size +sizeof (tmp)) +return -EINVAL ; tmp.name[sizeof (tmp.name)-1 ] =0 ;
@@ -1513,6 +1515,8 @@ compat_do_replace(struct net *net, sockptr_t arg, unsigned int len)return -ENOMEM ;if (tmp.num_counters ==0 )return -EINVAL ; +if ((u64 )len < (u64 )tmp.size +sizeof (tmp)) +return -EINVAL ; tmp.name[sizeof (tmp.name)-1 ] =0 ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 82/172 net/mlx5: SF, Stop waiting for FW as teardown was called
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Moshe Shemesh <[email protected]>
[ Upstream commit 137cef6d55564fb687d12fbc5f85be43ff7b53a7 ]
When PF/VF teardown is called the driver sets the flag MLX5_BREAK_FW_WAIT to stop waiting for FW loading and initializing. Same should be applied to SF driver teardown to cut waiting time. On mlx5_sf_dev_remove() set the flag before draining health WQ as recovery flow may also wait for FW reloading while it is not relevant anymore.
Stable-dep-of: c6e77aa9dd82 ("net/mlx5: Register devlink first under devlink lock")
--- .../mellanox/mlx5/core/sf/dev/driver.c | 21 ++++++++++++------- 1 file changed, 13 insertions(+), 8 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c +13 -8
@@ -95,24 +95,29 @@ static int mlx5_sf_dev_probe(struct auxiliary_device *adev, const struct auxiliastatic void mlx5_sf_dev_remove (struct auxiliary_device *adev) { struct mlx5_sf_dev *sf_dev = container_of(adev, struct mlx5_sf_dev, adev); -struct devlink *devlink =priv_to_devlink (sf_dev->mdev); +struct mlx5_core_dev *mdev = sf_dev->mdev; +struct devlink *devlink; -mlx5_drain_health_wq (sf_dev->mdev); + devlink =priv_to_devlink (mdev); +set_bit (MLX5_BREAK_FW_WAIT , &mdev->intf_state); +mlx5_drain_health_wq (mdev);devlink_unregister (devlink); -if (mlx5_dev_is_lightweight (sf_dev->mdev)) -mlx5_uninit_one_light (sf_dev->mdev); +if (mlx5_dev_is_lightweight (mdev)) +mlx5_uninit_one_light (mdev);else -mlx5_uninit_one (sf_dev->mdev); -iounmap (sf_dev->mdev->iseg); -mlx5_mdev_uninit (sf_dev->mdev); +mlx5_uninit_one (mdev); +iounmap (mdev->iseg); +mlx5_mdev_uninit (mdev);mlx5_devlink_free (devlink); }static void mlx5_sf_dev_shutdown (struct auxiliary_device *adev) { struct mlx5_sf_dev *sf_dev = container_of(adev, struct mlx5_sf_dev, adev); +struct mlx5_core_dev *mdev = sf_dev->mdev; -mlx5_unload_one (sf_dev->mdev, false); +set_bit (MLX5_BREAK_FW_WAIT , &mdev->intf_state); +mlx5_unload_one (mdev, false); }static const struct auxiliary_device_id mlx5_sf_dev_id_table[] = {
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 83/172 net/mlx5: Register devlink first under devlink lock
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Shay Drory <[email protected]>
[ Upstream commit c6e77aa9dd82bc18a89bf49418f8f7e961cfccc8 ]
In case device is having a non fatal FW error during probe, the driver will report the error to user via devlink. This will trigger a WARN_ON, since mlx5 is calling devlink_register() last.
In order to avoid the WARN_ON[1], change mlx5 to invoke devl_register() first under devlink lock.
[1]
WARNING: CPU: 5 PID: 227 at net/devlink/health.c:483 devlink_recover_notify.constprop.0+0xb8/0xc0 CPU: 5 PID: 227 Comm: kworker/u16:3 Not tainted 6.4.0-rc5_for_upstream_min_debug_2023_06_12_12_38 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Workqueue: mlx5_health0000:08:00.0 mlx5_fw_reporter_err_work [mlx5_core] RIP: 0010:devlink_recover_notify.constprop.0+0xb8/0xc0 Call Trace: <TASK> ? __warn+0x79/0x120 ? devlink_recover_notify.constprop.0+0xb8/0xc0 ? report_bug+0x17c/0x190 ? handle_bug+0x3c/0x60 ? exc_invalid_op+0x14/0x70 ? asm_exc_invalid_op+0x16/0x20 ? devlink_recover_notify.constprop.0+0xb8/0xc0 devlink_health_report+0x4a/0x1c0 mlx5_fw_reporter_err_work+0xa4/0xd0 [mlx5_core] process_one_work+0x1bb/0x3c0 ? process_one_work+0x3c0/0x3c0 worker_thread+0x4d/0x3c0 ? process_one_work+0x3c0/0x3c0 kthread+0xc6/0xf0 ? kthread_complete_and_exit+0x20/0x20 ret_from_fork+0x1f/0x30 </TASK>
--- .../net/ethernet/mellanox/mlx5/core/main.c | 37 ++++++++++--------- .../mellanox/mlx5/core/sf/dev/driver.c | 1 - 2 files changed, 20 insertions(+), 18 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/main.c +20 -17
@@ -1480,6 +1480,14 @@ int mlx5_init_one_devl_locked(struct mlx5_core_dev *dev)if (err) goto err_register; + err =mlx5_crdump_enable (dev); +if (err) +mlx5_core_err (dev,"mlx5_crdump_enable failed with error code %d\n" , err); + + err =mlx5_hwmon_dev_register (dev); +if (err) +mlx5_core_err (dev,"mlx5_hwmon_dev_register failed with error code %d\n" , err); +mutex_unlock (&dev->intf_state_mutex);return 0 ;
@@ -1505,7 +1513,10 @@ int mlx5_init_one(struct mlx5_core_dev *dev)int err;devl_lock (devlink); +devl_register (devlink); err =mlx5_init_one_devl_locked (dev); +if (err) +devl_unregister (devlink);devl_unlock (devlink);return err; }
@@ -1517,6 +1528,8 @@ void mlx5_uninit_one(struct mlx5_core_dev *dev)devl_lock (devlink);mutex_lock (&dev->intf_state_mutex); +mlx5_hwmon_dev_unregister (dev); +mlx5_crdump_disable (dev);mlx5_unregister_device (dev);if (!test_bit (MLX5_INTERFACE_STATE_UP , &dev->intf_state)) {
@@ -1534,6 +1547,7 @@ void mlx5_uninit_one(struct mlx5_core_dev *dev)mlx5_function_teardown (dev, true); out:mutex_unlock (&dev->intf_state_mutex); +devl_unregister (devlink);devl_unlock (devlink); }
@@ -1680,16 +1694,20 @@ int mlx5_init_one_light(struct mlx5_core_dev *dev) }devl_lock (devlink); +devl_register (devlink); + err =mlx5_devlink_params_register (priv_to_devlink (dev)); -devl_unlock (devlink);if (err) {mlx5_core_warn (dev,"mlx5_devlink_param_reg err = %d\n" , err); goto query_hca_caps_err; } +devl_unlock (devlink);return 0 ; query_hca_caps_err: +devl_unregister (devlink); +devl_unlock (devlink);mlx5_function_disable (dev, true); out: dev->state =MLX5_DEVICE_STATE_INTERNAL_ERROR ;
@@ -1702,6 +1720,7 @@ void mlx5_uninit_one_light(struct mlx5_core_dev *dev)devl_lock (devlink);mlx5_devlink_params_unregister (priv_to_devlink (dev)); +devl_unregister (devlink);devl_unlock (devlink);if (dev->state !=MLX5_DEVICE_STATE_UP )return ;
@@ -1943,16 +1962,7 @@ static int probe_one(struct pci_dev *pdev, const struct pci_device_id *id) goto err_init_one; } - err =mlx5_crdump_enable (dev); -if (err) -dev_err (&pdev->dev,"mlx5_crdump_enable failed with error code %d\n" , err); - - err =mlx5_hwmon_dev_register (dev); -if (err) -mlx5_core_err (dev,"mlx5_hwmon_dev_register failed with error code %d\n" , err); -pci_save_state (pdev); -devlink_register (devlink);return 0 ; err_init_one:
@@ -1973,16 +1983,9 @@ static void remove_one(struct pci_dev *pdev)struct devlink *devlink =priv_to_devlink (dev);set_bit (MLX5_BREAK_FW_WAIT , &dev->intf_state); -/* mlx5_drain_fw_reset() and mlx5_drain_health_wq() are using -* devlink notify APIs. -* Hence, we must drain them before unregistering the devlink. -*/ mlx5_drain_fw_reset (dev);mlx5_drain_health_wq (dev); -devlink_unregister (devlink);mlx5_sriov_disable (pdev, false); -mlx5_hwmon_dev_unregister (dev); -mlx5_crdump_disable (dev);mlx5_uninit_one (dev);mlx5_pci_close (dev);mlx5_mdev_uninit (dev);
drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c +0 -1
@@ -101,7 +101,6 @@ static void mlx5_sf_dev_remove(struct auxiliary_device *adev) devlink =priv_to_devlink (mdev);set_bit (MLX5_BREAK_FW_WAIT , &mdev->intf_state);mlx5_drain_health_wq (mdev); -devlink_unregister (devlink);if (mlx5_dev_is_lightweight (mdev))mlx5_uninit_one_light (mdev);else
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 84/172 net/mlx5: offset comp irq index in name by one
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michael Liang <[email protected]>
[ Upstream commit 9f7e8fbb91f8fa29548e2f6ab50c03b628c67ede ]
The mlx5 comp irq name scheme is changed a little bit between commit 3663ad34bc70 ("net/mlx5: Shift control IRQ to the last index") and commit 3354822cde5a ("net/mlx5: Use dynamic msix vectors allocation"). The index in the comp irq name used to start from 0 but now it starts from 1. There is nothing critical here, but it's harmless to change back to the old behavior, a.k.a starting from 0.
--- drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c +3 -1
@@ -19,6 +19,7 @@#define MLX5_IRQ_CTRL_SF_MAX 8/* min num of vectors for SFs to be enabled */ #define MLX5_IRQ_VEC_COMP_BASE_SF 2 +#define MLX5_IRQ_VEC_COMP_BASE 1#define MLX5_EQ_SHARE_IRQ_MAX_COMP (8)#define MLX5_EQ_SHARE_IRQ_MAX_CTRL (UINT_MAX)
@@ -246,6 +247,7 @@ static void irq_set_name(struct mlx5_irq_pool *pool, char *name, int vecidx)return ; } + vecidx -=MLX5_IRQ_VEC_COMP_BASE ;snprintf (name,MLX5_MAX_IRQ_NAME ,"mlx5_comp%d" , vecidx); }
@@ -585,7 +587,7 @@ struct mlx5_irq *mlx5_irq_request_vector(struct mlx5_core_dev *dev, u16 cpu,struct mlx5_irq_table *table =mlx5_irq_table_get (dev);struct mlx5_irq_pool *pool = table->pcif_pool;struct irq_affinity_desc af_desc; -int offset =1 ; +int offset =MLX5_IRQ_VEC_COMP_BASE ;if (!pool->xa_num_irqs.max) offset =0 ;
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 85/172 net/mlx5: Properly link new fs rules into the tree
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Cosmin Ratiu <[email protected]>
[ Upstream commit 7c6782ad4911cbee874e85630226ed389ff2e453 ]
Previously, add_rule_fg would only add newly created rules from the handle into the tree when they had a refcount of 1. On the other hand, create_flow_handle tries hard to find and reference already existing identical rules instead of creating new ones.
These two behaviors can result in a situation where create_flow_handle
1) creates a new rule and references it, then
2) in a subsequent step during the same handle creation references it again,
resulting in a rule with a refcount of 2 that is not linked into the tree, will have a NULL parent and root and will result in a crash when the flow group is deleted because del_sw_hw_rule, invoked on rule deletion, assumes node->parent is != NULL.
This happened in the wild, due to another bug related to incorrect handling of duplicate pkt_reformat ids, which lead to the code in create_flow_handle incorrectly referencing a just-added rule in the same flow handle, resulting in the problem described above. Full details are at [1].
This patch changes add_rule_fg to add new rules without parents into the tree, properly initializing them and avoiding the crash. This makes it more consistent with how rules are added to an FTE in create_flow_handle.
--- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +2 -1
@@ -1808,8 +1808,9 @@ static struct mlx5_flow_handle *add_rule_fg(struct mlx5_flow_group *fg, }trace_mlx5_fs_set_fte (fte, false); +/* Link newly added rules into the tree. */ for (i =0 ; i < handle->num_rules; i++) { -if (refcount_read (&handle->rule[i]->node.refcount) ==1 ) { +if (!handle->rule[i]->node.parent) {tree_add_node (&handle->rule[i]->node, &fte->node);trace_mlx5_fs_add_rule (handle->rule[i]); }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 86/172 net/mlx5: Correctly compare pkt reformat ids
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Cosmin Ratiu <[email protected]>
[ Upstream commit 9eca93f4d5ab03905516a68683674d9c50ff95bd ]
struct mlx5_pkt_reformat contains a naked union of a u32 id and a dr_action pointer which is used when the action is SW-managed (when pkt_reformat.owner is set to MLX5_FLOW_RESOURCE_OWNER_SW). Using id directly in that case is incorrect, as it maps to the least significant 32 bits of the 64-bit pointer in mlx5_fs_dr_action and not to the pkt reformat id allocated in firmware.
For the purpose of comparing whether two rules are identical, interpreting the least significant 32 bits of the mlx5_fs_dr_action pointer as an id mostly works... until it breaks horribly and produces the outcome described in [1].
This patch fixes mlx5_flow_dests_cmp to correctly compare ids using mlx5_fs_dr_action_get_pkt_reformat_id for the SW-managed rules.
--- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +12 -2
@@ -1664,6 +1664,16 @@ static int create_auto_flow_group(struct mlx5_flow_table *ft,return err; } +static bool mlx5_pkt_reformat_cmp (struct mlx5_pkt_reformat *p1, +struct mlx5_pkt_reformat *p2) +{ +return p1->owner == p2->owner && + (p1->owner ==MLX5_FLOW_RESOURCE_OWNER_FW ? + p1->id == p2->id : +mlx5_fs_dr_action_get_pkt_reformat_id (p1) == +mlx5_fs_dr_action_get_pkt_reformat_id (p2)); +} +static bool mlx5_flow_dests_cmp (struct mlx5_flow_destination *d1,struct mlx5_flow_destination *d2) {
@@ -1675,8 +1685,8 @@ static bool mlx5_flow_dests_cmp(struct mlx5_flow_destination *d1, ((d1->vport.flags &MLX5_FLOW_DEST_VPORT_VHCA_ID ) ? (d1->vport.vhca_id == d2->vport.vhca_id) : true) && ((d1->vport.flags &MLX5_FLOW_DEST_VPORT_REFORMAT_ID ) ? - (d1->vport.pkt_reformat->id == - d2->vport.pkt_reformat->id) : true)) || +mlx5_pkt_reformat_cmp (d1->vport.pkt_reformat, + d2->vport.pkt_reformat) : true)) || (d1->type ==MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE && d1->ft == d2->ft) || (d1->type == MLX5_FLOW_DESTINATION_TYPE_TIR &&
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 87/172 net/mlx5e: RSS, Block changing channels number when RXFH is configured
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Carolina Jubran <[email protected]>
[ Upstream commit ee3572409f74a838154af74ce1e56e62c17786a8 ]
Changing the channels number after configuring the receive flow hash indirection table may affect the RSS table size. The previous configuration may no longer be compatible with the new receive flow hash indirection table.
Block changing the channels number when RXFH is configured and changing the channels number requires resizing the RSS table size.
--- .../ethernet/mellanox/mlx5/core/en_ethtool.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)
drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +17 -0
@@ -451,6 +451,23 @@ int mlx5e_ethtool_set_channels(struct mlx5e_priv *priv,mutex_lock (&priv->state_lock); +/* If RXFH is configured, changing the channels number is allowed only if +* it does not require resizing the RSS table. This is because the previous +* configuration may no longer be compatible with the new RSS table. +*/ +if (netif_is_rxfh_configured (priv->netdev)) { +int cur_rqt_size =mlx5e_rqt_size (priv->mdev, cur_params->num_channels); +int new_rqt_size =mlx5e_rqt_size (priv->mdev, count); + +if (new_rqt_size != cur_rqt_size) { + err = -EINVAL ; +netdev_err (priv->netdev, +"%s: RXFH is configured, block changing channels number that affects RSS table size (new: %d, current: %d)\n" , + __func__, new_rqt_size, cur_rqt_size); + goto out; + } + } +/* Don't allow changing the number of channels if HTB offload is active, * because the numeration of the QoS SQs will change, while per-queue * qdiscs are attached.
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 88/172 net/mlx5e: Fix mlx5e_priv_init() cleanup flow
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Carolina Jubran <[email protected]>
[ Upstream commit ecb829459a841198e142f72fadab56424ae96519 ]
When mlx5e_priv_init() fails, the cleanup flow calls mlx5e_selq_cleanup which calls mlx5e_selq_apply() that assures that the priv->state_lock is held using lockdep_is_held().
Acquire the state_lock in mlx5e_selq_cleanup().
Kernel log:
=============================
WARNING: suspicious RCU usage
6.8.0-rc3_net_next_841a9b5 #1 Not tainted
-----------------------------drivers/net/ethernet/mellanox/mlx5/core/en/selq.c:124 suspicious rcu_dereference_protected() usage!
other info that might help us debug this:
rcu_scheduler_active = 2, debug_locks = 1
2 locks held by systemd-modules/293:
#0: ffffffffa05067b0 (devices_rwsem){++++}-{3:3}, at: ib_register_client+0x109/0x1b0 [ib_core] #1: ffff8881096c65c0 (&device->client_data_rwsem){++++}-{3:3}, at: add_client_context+0x104/0x1c0 [ib_core]
stack backtrace: CPU: 4 PID: 293 Comm: systemd-modules Not tainted 6.8.0-rc3_net_next_841a9b5 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x8a/0xa0 lockdep_rcu_suspicious+0x154/0x1a0 mlx5e_selq_apply+0x94/0xa0 [mlx5_core] mlx5e_selq_cleanup+0x3a/0x60 [mlx5_core] mlx5e_priv_init+0x2be/0x2f0 [mlx5_core] mlx5_rdma_setup_rn+0x7c/0x1a0 [mlx5_core] rdma_init_netdev+0x4e/0x80 [ib_core] ? mlx5_rdma_netdev_free+0x70/0x70 [mlx5_core] ipoib_intf_init+0x64/0x550 [ib_ipoib] ipoib_intf_alloc+0x4e/0xc0 [ib_ipoib] ipoib_add_one+0xb0/0x360 [ib_ipoib] add_client_context+0x112/0x1c0 [ib_core] ib_register_client+0x166/0x1b0 [ib_core] ? 0xffffffffa0573000 ipoib_init_module+0xeb/0x1a0 [ib_ipoib] do_one_initcall+0x61/0x250 do_init_module+0x8a/0x270 init_module_from_file+0x8b/0xd0 idempotent_init_module+0x17d/0x230 __x64_sys_finit_module+0x61/0xb0 do_syscall_64+0x71/0x140 entry_SYSCALL_64_after_hwframe+0x46/0x4e </TASK>
--- drivers/net/ethernet/mellanox/mlx5/core/en/selq.c | 2 ++ drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 -- 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en/selq.c +2 -0
@@ -57,6 +57,7 @@ int mlx5e_selq_init(struct mlx5e_selq *selq, struct mutex *state_lock)void mlx5e_selq_cleanup (struct mlx5e_selq *selq) { +mutex_lock (selq->state_lock);WARN_ON_ONCE (selq->is_prepared);kvfree (selq->standby);
@@ -67,6 +68,7 @@ void mlx5e_selq_cleanup(struct mlx5e_selq *selq)kvfree (selq->standby); selq->standby =NULL ; +mutex_unlock (selq->state_lock); }void mlx5e_selq_prepare_params (struct mlx5e_selq *selq,struct mlx5e_params *params)
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +0 -2
@@ -5695,9 +5695,7 @@ void mlx5e_priv_cleanup(struct mlx5e_priv *priv)kfree (priv->tx_rates);kfree (priv->txq2sq);destroy_workqueue (priv->wq); -mutex_lock (&priv->state_lock);mlx5e_selq_cleanup (&priv->selq); -mutex_unlock (&priv->state_lock);free_cpumask_var (priv->scratchpad.cpumask);for (i =0 ; i < priv->htb_max_qos_sqs; i++)
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 89/172 net/mlx5e: HTB, Fix inconsistencies with QoS SQs number
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Carolina Jubran <[email protected]>
[ Upstream commit 2f436f1869771d46e1a9f85738d5a1a7c5653a4e ]
When creating a new HTB class while the interface is down, the variable that follows the number of QoS SQs (htb_max_qos_sqs) may not be consistent with the number of HTB classes.
Previously, we compared these two values to ensure that the node_qid is lower than the number of QoS SQs, and we allocated stats for that SQ when they are equal.
Change the check to compare the node_qid with the current number of leaf nodes and fix the checking conditions to ensure allocation of stats_list and stats for each node.
--- .../net/ethernet/mellanox/mlx5/core/en/qos.c | 33 ++++++++++--------- 1 file changed, 17 insertions(+), 16 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en/qos.c +17 -16
@@ -83,24 +83,25 @@ int mlx5e_open_qos_sq(struct mlx5e_priv *priv, struct mlx5e_channels *chs, txq_ix =mlx5e_qid_from_qos (chs, node_qid); -WARN_ON (node_qid > priv->htb_max_qos_sqs); -if (node_qid == priv->htb_max_qos_sqs) { -struct mlx5e_sq_stats *stats, **stats_list =NULL ; - -if (priv->htb_max_qos_sqs ==0 ) { - stats_list =kvcalloc (mlx5e_qos_max_leaf_nodes (priv->mdev), -sizeof (*stats_list), -GFP_KERNEL ); -if (!stats_list) -return -ENOMEM ; - } +WARN_ON (node_qid >=mlx5e_htb_cur_leaf_nodes (priv->htb)); +if (!priv->htb_qos_sq_stats) { +struct mlx5e_sq_stats **stats_list; + + stats_list =kvcalloc (mlx5e_qos_max_leaf_nodes (priv->mdev), +sizeof (*stats_list),GFP_KERNEL ); +if (!stats_list) +return -ENOMEM ; + +WRITE_ONCE (priv->htb_qos_sq_stats, stats_list); + } + +if (!priv->htb_qos_sq_stats[node_qid]) { +struct mlx5e_sq_stats *stats; + stats =kzalloc (sizeof (*stats),GFP_KERNEL ); -if (!stats) { -kvfree (stats_list); +if (!stats)return -ENOMEM ; - } -if (stats_list) -WRITE_ONCE (priv->htb_qos_sq_stats, stats_list); +WRITE_ONCE (priv->htb_qos_sq_stats[node_qid], stats);/* Order htb_max_qos_sqs increment after writing the array pointer. * Pairs with smp_load_acquire in en_stats.c.
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 90/172 net/mlx5e: Do not produce metadata freelist entries in Tx port ts WQE xmit
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Rahul Rameshbabu <[email protected]>
[ Upstream commit 86b0ca5b118d3a0bae5e5645a13e66f8a4f6c525 ]
Free Tx port timestamping metadata entries in the NAPI poll context and consume metadata enties in the WQE xmit path. Do not free a Tx port timestamping metadata entry in the WQE xmit path even in the error path to avoid a race between two metadata entry producers.
--- drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h | 8 +++++++- drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 7 +++---- 2 files changed, 10 insertions(+), 5 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h +7 -1
@@ -95,9 +95,15 @@ static inline void mlx5e_ptp_metadata_fifo_push(struct mlx5e_ptp_metadata_fifo * }static inline u8 +mlx5e_ptp_metadata_fifo_peek (struct mlx5e_ptp_metadata_fifo *fifo) +{ +return fifo->data[fifo->mask & fifo->cc]; +} + +static inline void mlx5e_ptp_metadata_fifo_pop (struct mlx5e_ptp_metadata_fifo *fifo) { -return fifo->data[fifo->mask & fifo->cc++]; + fifo->cc++; } static inline void
drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +3 -4
@@ -398,6 +398,8 @@ mlx5e_txwqe_complete(struct mlx5e_txqsq *sq, struct sk_buff *skb, (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP))) {u8 metadata_index =be32_to_cpu (eseg->flow_table_metadata); +mlx5e_ptp_metadata_fifo_pop (&sq->ptpsq->metadata_freelist); +mlx5e_skb_cb_hwtstamp_init (skb);mlx5e_ptp_metadata_map_put (&sq->ptpsq->metadata_map, skb, metadata_index);
@@ -496,9 +498,6 @@ mlx5e_sq_xmit_wqe(struct mlx5e_txqsq *sq, struct sk_buff *skb, err_drop: stats->dropped++; -if (unlikely (sq->ptpsq && (skb_shinfo (skb)->tx_flags &SKBTX_HW_TSTAMP ))) -mlx5e_ptp_metadata_fifo_push (&sq->ptpsq->metadata_freelist, -be32_to_cpu (eseg->flow_table_metadata));dev_kfree_skb_any (skb);mlx5e_tx_flush (sq); }
@@ -657,7 +656,7 @@ static void mlx5e_cqe_ts_id_eseg(struct mlx5e_ptpsq *ptpsq, struct sk_buff *skb, {if (unlikely (skb_shinfo (skb)->tx_flags &SKBTX_HW_TSTAMP )) eseg->flow_table_metadata = -cpu_to_be32 (mlx5e_ptp_metadata_fifo_pop (&ptpsq->metadata_freelist)); +cpu_to_be32 (mlx5e_ptp_metadata_fifo_peek (&ptpsq->metadata_freelist)); } static void mlx5e_txwqe_build_eseg(struct mlx5e_priv *priv, struct mlx5e_txqsq *sq,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 91/172 net: sparx5: fix wrong config being used when reconfiguring PCS
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Machon <[email protected]>
[ Upstream commit 33623113a48ea906f1955cbf71094f6aa4462e8f ]
The wrong port config is being used if the PCS is reconfigured. Fix this by correctly using the new config instead of the old one.
--- drivers/net/ethernet/microchip/sparx5/sparx5_port.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/microchip/sparx5/sparx5_port.c +2 -2
@@ -731,7 +731,7 @@ static int sparx5_port_pcs_low_set(struct sparx5 *sparx5,bool sgmii = false, inband_aneg = false;int err; -if (port->conf.inband) { +if (conf->inband) {if (conf->portmode ==PHY_INTERFACE_MODE_SGMII || conf->portmode ==PHY_INTERFACE_MODE_QSGMII ) inband_aneg = true;/* Cisco-SGMII in-band-aneg */
@@ -948,7 +948,7 @@ int sparx5_port_pcs_set(struct sparx5 *sparx5,if (err)return -EINVAL ; -if (port->conf.inband) { +if (conf->inband) {/* Enable/disable 1G counters in ASM */ spx5_rmw(ASM_PORT_CFG_CSC_STAT_DIS_SET(high_speed_dev), ASM_PORT_CFG_CSC_STAT_DIS,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 92/172 Revert "s390/ism: fix receive message buffer allocation"
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gerd Bayer <[email protected]>
[ Upstream commit d51dc8dd6ab6f93a894ff8b38d3b8d02c98eb9fb ]
This reverts commit 58effa3476536215530c9ec4910ffc981613b413. Review was not finished on this patch. So it's not ready for upstreaming.
--- drivers/s390/net/ism_drv.c | 38 +++++++++----------------------------- 1 file changed, 9 insertions(+), 29 deletions(-)
drivers/s390/net/ism_drv.c +9 -29
@@ -14,8 +14,6 @@#include <linux/err.h> #include <linux/ctype.h> #include <linux/processor.h> -#include <linux/dma-mapping.h> -#include <linux/mm.h> #include "ism.h"
@@ -294,15 +292,13 @@ static int ism_read_local_gid(struct ism_dev *ism)static void ism_free_dmb (struct ism_dev *ism,struct ism_dmb *dmb) {clear_bit (dmb->sba_idx, ism->sba_bitmap); -dma_unmap_page (&ism->pdev->dev, dmb->dma_addr, dmb->dmb_len, -DMA_FROM_DEVICE ); -folio_put (virt_to_folio (dmb->cpu_addr)); +dma_free_coherent (&ism->pdev->dev, dmb->dmb_len, + dmb->cpu_addr, dmb->dma_addr); }static int ism_alloc_dmb (struct ism_dev *ism,struct ism_dmb *dmb) {unsigned long bit; -int rc;if (PAGE_ALIGN (dmb->dmb_len) >dma_get_max_seg_size (&ism->pdev->dev))return -EINVAL ;
@@ -319,30 +315,14 @@ static int ism_alloc_dmb(struct ism_dev *ism, struct ism_dmb *dmb) test_and_set_bit(dmb->sba_idx, ism->sba_bitmap)) return -EINVAL; - dmb->cpu_addr = -folio_address (folio_alloc (GFP_KERNEL | __GFP_NOWARN | - __GFP_NOMEMALLOC | __GFP_NORETRY, -get_order (dmb->dmb_len))); + dmb->cpu_addr =dma_alloc_coherent (&ism->pdev->dev, dmb->dmb_len, + &dmb->dma_addr, +GFP_KERNEL | __GFP_NOWARN | + __GFP_NOMEMALLOC | __GFP_NORETRY); +if (!dmb->cpu_addr) +clear_bit (dmb->sba_idx, ism->sba_bitmap); -if (!dmb->cpu_addr) { - rc = -ENOMEM ; - goto out_bit; - } - dmb->dma_addr =dma_map_page (&ism->pdev->dev, -virt_to_page (dmb->cpu_addr),0 , - dmb->dmb_len,DMA_FROM_DEVICE ); -if (dma_mapping_error (&ism->pdev->dev, dmb->dma_addr)) { - rc = -ENOMEM ; - goto out_free; - } - -return 0 ; - -out_free: -kfree (dmb->cpu_addr); -out_bit: -clear_bit (dmb->sba_idx, ism->sba_bitmap); -return rc; +return dmb->cpu_addr ?0 : -ENOMEM ; } int ism_register_dmb(struct ism_dev *ism, struct ism_dmb *dmb,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 93/172 net: dsa: mt7530: trap link-local frames regardless of ST Port State
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arınç ÜNAL <[email protected]>
[ Upstream commit 17c560113231ddc20088553c7b499b289b664311 ]
In Clause 5 of IEEE Std 802-2014, two sublayers of the data link layer (DLL) of the Open Systems Interconnection basic reference model (OSI/RM) are described; the medium access control (MAC) and logical link control (LLC) sublayers. The MAC sublayer is the one facing the physical layer.
In 8.2 of IEEE Std 802.1Q-2022, the Bridge architecture is described. A Bridge component comprises a MAC Relay Entity for interconnecting the Ports of the Bridge, at least two Ports, and higher layer entities with at least a Spanning Tree Protocol Entity included.
Each Bridge Port also functions as an end station and shall provide the MAC Service to an LLC Entity. Each instance of the MAC Service is provided to a distinct LLC Entity that supports protocol identification, multiplexing, and demultiplexing, for protocol data unit (PDU) transmission and reception by one or more higher layer entities.
It is described in 8.13.9 of IEEE Std 802.1Q-2022 that in a Bridge, the LLC Entity associated with each Bridge Port is modeled as being directly connected to the attached Local Area Network (LAN).
On the switch with CPU port architecture, CPU port functions as Management Port, and the Management Port functionality is provided by software which functions as an end station. Software is connected to an IEEE 802 LAN that is wholly contained within the system that incorporates the Bridge. Software provides access to the LLC Entity associated with each Bridge Port by the value of the source port field on the special tag on the frame received by software.
We call frames that carry control information to determine the active topology and current extent of each Virtual Local Area Network (VLAN), i.e., spanning tree or Shortest Path Bridging (SPB) and Multiple VLAN Registration Protocol Data Units (MVRPDUs), and frames from other link constrained protocols, such as Extensible Authentication Protocol over LAN (EAPOL) and Link Layer Discovery Protocol (LLDP), link-local frames. They are not forwarded by a Bridge. Permanently configured entries in the filtering database (FDB) ensure that such frames are discarded by the Forwarding Process. In 8.6.3 of IEEE Std 802.1Q-2022, this is described in detail:
Each of the reserved MAC addresses specified in Table 8-1 (01-80-C2-00-00-[00,01,02,03,04,05,06,07,08,09,0A,0B,0C,0D,0E,0F]) shall be permanently configured in the FDB in C-VLAN components and ERs.
Each of the reserved MAC addresses specified in Table 8-2 (01-80-C2-00-00-[01,02,03,04,05,06,07,08,09,0A,0E]) shall be permanently configured in the FDB in S-VLAN components.
Each of the reserved MAC addresses specified in Table 8-3 (01-80-C2-00-00-[01,02,04,0E]) shall be permanently configured in the FDB in TPMR components.
The FDB entries for reserved MAC addresses shall specify filtering for all Bridge Ports and all VIDs. Management shall not provide the capability to modify or remove entries for reserved MAC addresses.
The addresses in Table 8-1, Table 8-2, and Table 8-3 determine the scope of propagation of PDUs within a Bridged Network, as follows:
The Nearest Bridge group address (01-80-C2-00-00-0E) is an address that no conformant Two-Port MAC Relay (TPMR) component, Service VLAN (S-VLAN) component, Customer VLAN (C-VLAN) component, or MAC Bridge can forward. PDUs transmitted using this destination address, or any other addresses that appear in Table 8-1, Table 8-2, and Table 8-3 (01-80-C2-00-00-[00,01,02,03,04,05,06,07,08,09,0A,0B,0C,0D,0E,0F]), can therefore travel no further than those stations that can be reached via a single individual LAN from the originating station.
The Nearest non-TPMR Bridge group address (01-80-C2-00-00-03), is an address that no conformant S-VLAN component, C-VLAN component, or MAC Bridge can forward; however, this address is relayed by a TPMR component. PDUs using this destination address, or any of the other addresses that appear in both Table 8-1 and Table 8-2 but not in Table 8-3 (01-80-C2-00-00-[00,03,05,06,07,08,09,0A,0B,0C,0D,0F]), will be relayed by any TPMRs but will propagate no further than the nearest S-VLAN component, C-VLAN component, or MAC Bridge.
The Nearest Customer Bridge group address (01-80-C2-00-00-00) is an address that no conformant C-VLAN component, MAC Bridge can forward; however, it is relayed by TPMR components and S-VLAN components. PDUs using this destination address, or any of the other addresses that appear in Table 8-1 but not in either Table 8-2 or Table 8-3 (01-80-C2-00-00-[00,0B,0C,0D,0F]), will be relayed by TPMR components and S-VLAN components but will propagate no further than the nearest C-VLAN component or MAC Bridge.
Because the LLC Entity associated with each Bridge Port is provided via CPU port, we must not filter these frames but forward them to CPU port.
In a Bridge, the transmission Port is majorly decided by ingress and egress rules, FDB, and spanning tree Port State functions of the Forwarding Process. For link-local frames, only CPU port should be designated as destination port in the FDB, and the other functions of the Forwarding Process must not interfere with the decision of the transmission Port. We call this process trapping frames to CPU port.
Therefore, on the switch with CPU port architecture, link-local frames must be trapped to CPU port, and certain link-local frames received by a Port of a Bridge comprising a TPMR component or an S-VLAN component must be excluded from it.
A Bridge of the switch with CPU port architecture cannot comprise a Two-Port MAC Relay (TPMR) component as a TPMR component supports only a subset of the functionality of a MAC Bridge. A Bridge comprising two Ports (Management Port doesn't count) of this architecture will either function as a standard MAC Bridge or a standard VLAN Bridge.
Therefore, a Bridge of this architecture can only comprise S-VLAN components, C-VLAN components, or MAC Bridge components. Since there's no TPMR component, we don't need to relay PDUs using the destination addresses specified on the Nearest non-TPMR section, and the proportion of the Nearest Customer Bridge section where they must be relayed by TPMR components.
One option to trap link-local frames to CPU port is to add static FDB entries with CPU port designated as destination port. However, because that Independent VLAN Learning (IVL) is being used on every VID, each entry only applies to a single VLAN Identifier (VID). For a Bridge comprising a MAC Bridge component or a C-VLAN component, there would have to be 16 times 4096 entries. This switch intellectual property can only hold a maximum of 2048 entries. Using this option, there also isn't a mechanism to prevent link-local frames from being discarded when the spanning tree Port State of the reception Port is discarding.
The remaining option is to utilise the BPC, RGAC1, RGAC2, RGAC3, and RGAC4 registers. Whilst this applies to every VID, it doesn't contain all of the reserved MAC addresses without affecting the remaining Standard Group MAC Addresses. The REV_UN frame tag utilised using the RGAC4 register covers the remaining 01-80-C2-00-00-[04,05,06,07,08,09,0A,0B,0C,0D,0F] destination addresses. It also includes the 01-80-C2-00-00-22 to 01-80-C2-00-00-FF destination addresses which may be relayed by MAC Bridges or VLAN Bridges. The latter option provides better but not complete conformance.
This switch intellectual property also does not provide a mechanism to trap link-local frames with specific destination addresses to CPU port by Bridge, to conform to the filtering rules for the distinct Bridge components.
Therefore, regardless of the type of the Bridge component, link-local frames with these destination addresses will be trapped to CPU port:
01-80-C2-00-00-[00,01,02,03,0E]
In a Bridge comprising a MAC Bridge component or a C-VLAN component:
Link-local frames with these destination addresses won't be trapped to CPU port which won't conform to IEEE Std 802.1Q-2022:
01-80-C2-00-00-[04,05,06,07,08,09,0A,0B,0C,0D,0F]
In a Bridge comprising an S-VLAN component:
Link-local frames with these destination addresses will be trapped to CPU port which won't conform to IEEE Std 802.1Q-2022:
01-80-C2-00-00-00
Link-local frames with these destination addresses won't be trapped to CPU port which won't conform to IEEE Std 802.1Q-2022:
01-80-C2-00-00-[04,05,06,07,08,09,0A]
Currently on this switch intellectual property, if the spanning tree Port State of the reception Port is discarding, link-local frames will be discarded.
To trap link-local frames regardless of the spanning tree Port State, make the switch regard them as Bridge Protocol Data Units (BPDUs). This switch intellectual property only lets the frames regarded as BPDUs bypass the spanning tree Port State function of the Forwarding Process.
With this change, the only remaining interference is the ingress rules. When the reception Port has no PVID assigned on software, VLAN-untagged frames won't be allowed in. There doesn't seem to be a mechanism on the switch intellectual property to have link-local frames bypass this function of the Forwarding Process.
--- drivers/net/dsa/mt7530.c | 229 +++++++++++++++++++++++++++++++++------ drivers/net/dsa/mt7530.h | 5 + 2 files changed, 200 insertions(+), 34 deletions(-)
drivers/net/dsa/mt7530.c +195 -34
@@ -998,20 +998,173 @@ static void mt7530_setup_port5(struct dsa_switch *ds, phy_interface_t interface)mutex_unlock (&priv->reg_mutex); } -/* On page 205, section "8.6.3 Frame filtering" of the active standard, IEEE Std -* 802.1Q™-2022, it is stated that frames with 01:80:C2:00:00:00-0F as MAC DA -* must only be propagated to C-VLAN and MAC Bridge components. That means -* VLAN-aware and VLAN-unaware bridges. On the switch designs with CPU ports, -* these frames are supposed to be processed by the CPU (software). So we make -* the switch only forward them to the CPU port. And if received from a CPU -* port, forward to a single port. The software is responsible of making the -* switch conform to the latter by setting a single port as destination port on -* the special tag. +/* In Clause 5 of IEEE Std 802-2014, two sublayers of the data link layer (DLL) +* of the Open Systems Interconnection basic reference model (OSI/RM) are +* described; the medium access control (MAC) and logical link control (LLC) +* sublayers. The MAC sublayer is the one facing the physical layer. * -* This switch intellectual property cannot conform to this part of the standard -* fully. Whilst the REV_UN frame tag covers the remaining :04-0D and :0F MAC -* DAs, it also includes :22-FF which the scope of propagation is not supposed -* to be restricted for these MAC DAs. +* In 8.2 of IEEE Std 802.1Q-2022, the Bridge architecture is described. A +* Bridge component comprises a MAC Relay Entity for interconnecting the Ports +* of the Bridge, at least two Ports, and higher layer entities with at least a +* Spanning Tree Protocol Entity included. +* +* Each Bridge Port also functions as an end station and shall provide the MAC +* Service to an LLC Entity. Each instance of the MAC Service is provided to a +* distinct LLC Entity that supports protocol identification, multiplexing, and +* demultiplexing, for protocol data unit (PDU) transmission and reception by +* one or more higher layer entities. +* +* It is described in 8.13.9 of IEEE Std 802.1Q-2022 that in a Bridge, the LLC +* Entity associated with each Bridge Port is modeled as being directly +* connected to the attached Local Area Network (LAN). +* +* On the switch with CPU port architecture, CPU port functions as Management +* Port, and the Management Port functionality is provided by software which +* functions as an end station. Software is connected to an IEEE 802 LAN that is +* wholly contained within the system that incorporates the Bridge. Software +* provides access to the LLC Entity associated with each Bridge Port by the +* value of the source port field on the special tag on the frame received by +* software. +* +* We call frames that carry control information to determine the active +* topology and current extent of each Virtual Local Area Network (VLAN), i.e., +* spanning tree or Shortest Path Bridging (SPB) and Multiple VLAN Registration +* Protocol Data Units (MVRPDUs), and frames from other link constrained +* protocols, such as Extensible Authentication Protocol over LAN (EAPOL) and +* Link Layer Discovery Protocol (LLDP), link-local frames. They are not +* forwarded by a Bridge. Permanently configured entries in the filtering +* database (FDB) ensure that such frames are discarded by the Forwarding +* Process. In 8.6.3 of IEEE Std 802.1Q-2022, this is described in detail: +* +* Each of the reserved MAC addresses specified in Table 8-1 +* (01-80-C2-00-00-[00,01,02,03,04,05,06,07,08,09,0A,0B,0C,0D,0E,0F]) shall be +* permanently configured in the FDB in C-VLAN components and ERs. +* +* Each of the reserved MAC addresses specified in Table 8-2 +* (01-80-C2-00-00-[01,02,03,04,05,06,07,08,09,0A,0E]) shall be permanently +* configured in the FDB in S-VLAN components. +* +* Each of the reserved MAC addresses specified in Table 8-3 +* (01-80-C2-00-00-[01,02,04,0E]) shall be permanently configured in the FDB in +* TPMR components. +* +* The FDB entries for reserved MAC addresses shall specify filtering for all +* Bridge Ports and all VIDs. Management shall not provide the capability to +* modify or remove entries for reserved MAC addresses. +* +* The addresses in Table 8-1, Table 8-2, and Table 8-3 determine the scope of +* propagation of PDUs within a Bridged Network, as follows: +* +* The Nearest Bridge group address (01-80-C2-00-00-0E) is an address that no +* conformant Two-Port MAC Relay (TPMR) component, Service VLAN (S-VLAN) +* component, Customer VLAN (C-VLAN) component, or MAC Bridge can forward. +* PDUs transmitted using this destination address, or any other addresses +* that appear in Table 8-1, Table 8-2, and Table 8-3 +* (01-80-C2-00-00-[00,01,02,03,04,05,06,07,08,09,0A,0B,0C,0D,0E,0F]), can +* therefore travel no further than those stations that can be reached via a +* single individual LAN from the originating station. +* +* The Nearest non-TPMR Bridge group address (01-80-C2-00-00-03), is an +* address that no conformant S-VLAN component, C-VLAN component, or MAC +* Bridge can forward; however, this address is relayed by a TPMR component. +* PDUs using this destination address, or any of the other addresses that +* appear in both Table 8-1 and Table 8-2 but not in Table 8-3 +* (01-80-C2-00-00-[00,03,05,06,07,08,09,0A,0B,0C,0D,0F]), will be relayed by +* any TPMRs but will propagate no further than the nearest S-VLAN component, +* C-VLAN component, or MAC Bridge. +* +* The Nearest Customer Bridge group address (01-80-C2-00-00-00) is an address +* that no conformant C-VLAN component, MAC Bridge can forward; however, it is +* relayed by TPMR components and S-VLAN components. PDUs using this +* destination address, or any of the other addresses that appear in Table 8-1 +* but not in either Table 8-2 or Table 8-3 (01-80-C2-00-00-[00,0B,0C,0D,0F]), +* will be relayed by TPMR components and S-VLAN components but will propagate +* no further than the nearest C-VLAN component or MAC Bridge. +* +* Because the LLC Entity associated with each Bridge Port is provided via CPU +* port, we must not filter these frames but forward them to CPU port. +* +* In a Bridge, the transmission Port is majorly decided by ingress and egress +* rules, FDB, and spanning tree Port State functions of the Forwarding Process. +* For link-local frames, only CPU port should be designated as destination port +* in the FDB, and the other functions of the Forwarding Process must not +* interfere with the decision of the transmission Port. We call this process +* trapping frames to CPU port. +* +* Therefore, on the switch with CPU port architecture, link-local frames must +* be trapped to CPU port, and certain link-local frames received by a Port of a +* Bridge comprising a TPMR component or an S-VLAN component must be excluded +* from it. +* +* A Bridge of the switch with CPU port architecture cannot comprise a Two-Port +* MAC Relay (TPMR) component as a TPMR component supports only a subset of the +* functionality of a MAC Bridge. A Bridge comprising two Ports (Management Port +* doesn't count) of this architecture will either function as a standard MAC +* Bridge or a standard VLAN Bridge. +* +* Therefore, a Bridge of this architecture can only comprise S-VLAN components, +* C-VLAN components, or MAC Bridge components. Since there's no TPMR component, +* we don't need to relay PDUs using the destination addresses specified on the +* Nearest non-TPMR section, and the proportion of the Nearest Customer Bridge +* section where they must be relayed by TPMR components. +* +* One option to trap link-local frames to CPU port is to add static FDB entries +* with CPU port designated as destination port. However, because that +* Independent VLAN Learning (IVL) is being used on every VID, each entry only +* applies to a single VLAN Identifier (VID). For a Bridge comprising a MAC +* Bridge component or a C-VLAN component, there would have to be 16 times 4096 +* entries. This switch intellectual property can only hold a maximum of 2048 +* entries. Using this option, there also isn't a mechanism to prevent +* link-local frames from being discarded when the spanning tree Port State of +* the reception Port is discarding. +* +* The remaining option is to utilise the BPC, RGAC1, RGAC2, RGAC3, and RGAC4 +* registers. Whilst this applies to every VID, it doesn't contain all of the +* reserved MAC addresses without affecting the remaining Standard Group MAC +* Addresses. The REV_UN frame tag utilised using the RGAC4 register covers the +* remaining 01-80-C2-00-00-[04,05,06,07,08,09,0A,0B,0C,0D,0F] destination +* addresses. It also includes the 01-80-C2-00-00-22 to 01-80-C2-00-00-FF +* destination addresses which may be relayed by MAC Bridges or VLAN Bridges. +* The latter option provides better but not complete conformance. +* +* This switch intellectual property also does not provide a mechanism to trap +* link-local frames with specific destination addresses to CPU port by Bridge, +* to conform to the filtering rules for the distinct Bridge components. +* +* Therefore, regardless of the type of the Bridge component, link-local frames +* with these destination addresses will be trapped to CPU port: +* +* 01-80-C2-00-00-[00,01,02,03,0E] +* +* In a Bridge comprising a MAC Bridge component or a C-VLAN component: +* +* Link-local frames with these destination addresses won't be trapped to CPU +* port which won't conform to IEEE Std 802.1Q-2022: +* +* 01-80-C2-00-00-[04,05,06,07,08,09,0A,0B,0C,0D,0F] +* +* In a Bridge comprising an S-VLAN component: +* +* Link-local frames with these destination addresses will be trapped to CPU +* port which won't conform to IEEE Std 802.1Q-2022: +* +* 01-80-C2-00-00-00 +* +* Link-local frames with these destination addresses won't be trapped to CPU +* port which won't conform to IEEE Std 802.1Q-2022: +* +* 01-80-C2-00-00-[04,05,06,07,08,09,0A] +* +* To trap link-local frames to CPU port as conformant as this switch +* intellectual property can allow, link-local frames are made to be regarded as +* Bridge Protocol Data Units (BPDUs). This is because this switch intellectual +* property only lets the frames regarded as BPDUs bypass the spanning tree Port +* State function of the Forwarding Process. +* +* The only remaining interference is the ingress rules. When the reception Port +* has no PVID assigned on software, VLAN-untagged frames won't be allowed in. +* There doesn't seem to be a mechanism on the switch intellectual property to +* have link-local frames bypass this function of the Forwarding Process. */ static void mt753x_trap_frames (struct mt7530_priv *priv)
@@ -1019,35 +1172,43 @@ mt753x_trap_frames(struct mt7530_priv *priv)/* Trap 802.1X PAE frames and BPDUs to the CPU port(s) and egress them * VLAN-untagged. */ -mt7530_rmw (priv,MT753X_BPC ,MT753X_PAE_EG_TAG_MASK | -MT753X_PAE_PORT_FW_MASK |MT753X_BPDU_EG_TAG_MASK | -MT753X_BPDU_PORT_FW_MASK , -MT753X_PAE_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | -MT753X_PAE_PORT_FW (MT753X_BPDU_CPU_ONLY ) | -MT753X_BPDU_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | -MT753X_BPDU_CPU_ONLY ); +mt7530_rmw (priv,MT753X_BPC , +MT753X_PAE_BPDU_FR |MT753X_PAE_EG_TAG_MASK | +MT753X_PAE_PORT_FW_MASK |MT753X_BPDU_EG_TAG_MASK | +MT753X_BPDU_PORT_FW_MASK , +MT753X_PAE_BPDU_FR | +MT753X_PAE_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | +MT753X_PAE_PORT_FW (MT753X_BPDU_CPU_ONLY ) | +MT753X_BPDU_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | +MT753X_BPDU_CPU_ONLY );/* Trap frames with :01 and :02 MAC DAs to the CPU port(s) and egress * them VLAN-untagged. */ -mt7530_rmw (priv,MT753X_RGAC1 ,MT753X_R02_EG_TAG_MASK | -MT753X_R02_PORT_FW_MASK |MT753X_R01_EG_TAG_MASK | -MT753X_R01_PORT_FW_MASK , -MT753X_R02_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | -MT753X_R02_PORT_FW (MT753X_BPDU_CPU_ONLY ) | -MT753X_R01_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | -MT753X_BPDU_CPU_ONLY ); +mt7530_rmw (priv,MT753X_RGAC1 , +MT753X_R02_BPDU_FR |MT753X_R02_EG_TAG_MASK | +MT753X_R02_PORT_FW_MASK |MT753X_R01_BPDU_FR | +MT753X_R01_EG_TAG_MASK |MT753X_R01_PORT_FW_MASK , +MT753X_R02_BPDU_FR | +MT753X_R02_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | +MT753X_R02_PORT_FW (MT753X_BPDU_CPU_ONLY ) | +MT753X_R01_BPDU_FR | +MT753X_R01_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | +MT753X_BPDU_CPU_ONLY );/* Trap frames with :03 and :0E MAC DAs to the CPU port(s) and egress * them VLAN-untagged. */ -mt7530_rmw (priv,MT753X_RGAC2 ,MT753X_R0E_EG_TAG_MASK | -MT753X_R0E_PORT_FW_MASK |MT753X_R03_EG_TAG_MASK | -MT753X_R03_PORT_FW_MASK , -MT753X_R0E_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | -MT753X_R0E_PORT_FW (MT753X_BPDU_CPU_ONLY ) | -MT753X_R03_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | -MT753X_BPDU_CPU_ONLY ); +mt7530_rmw (priv,MT753X_RGAC2 , +MT753X_R0E_BPDU_FR |MT753X_R0E_EG_TAG_MASK | +MT753X_R0E_PORT_FW_MASK |MT753X_R03_BPDU_FR | +MT753X_R03_EG_TAG_MASK |MT753X_R03_PORT_FW_MASK , +MT753X_R0E_BPDU_FR | +MT753X_R0E_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | +MT753X_R0E_PORT_FW (MT753X_BPDU_CPU_ONLY ) | +MT753X_R03_BPDU_FR | +MT753X_R03_EG_TAG (MT7530_VLAN_EG_UNTAGGED ) | +MT753X_BPDU_CPU_ONLY ); } static int
drivers/net/dsa/mt7530.h +5 -0
@@ -65,6 +65,7 @@ enum mt753x_id {/* Registers for BPDU and PAE frame control*/ #define MT753X_BPC 0x24 +#define MT753X_PAE_BPDU_FR BIT(25)#define MT753X_PAE_EG_TAG_MASK GENMASK(24, 22)#define MT753X_PAE_EG_TAG (x) FIELD_PREP(MT753X_PAE_EG_TAG_MASK, x)#define MT753X_PAE_PORT_FW_MASK GENMASK(18, 16)
@@ -75,20 +76,24 @@ enum mt753x_id {/* Register for :01 and :02 MAC DA frame control */ #define MT753X_RGAC1 0x28 +#define MT753X_R02_BPDU_FR BIT(25)#define MT753X_R02_EG_TAG_MASK GENMASK(24, 22)#define MT753X_R02_EG_TAG (x) FIELD_PREP(MT753X_R02_EG_TAG_MASK, x)#define MT753X_R02_PORT_FW_MASK GENMASK(18, 16)#define MT753X_R02_PORT_FW (x) FIELD_PREP(MT753X_R02_PORT_FW_MASK, x) +#define MT753X_R01_BPDU_FR BIT(9)#define MT753X_R01_EG_TAG_MASK GENMASK(8, 6)#define MT753X_R01_EG_TAG (x) FIELD_PREP(MT753X_R01_EG_TAG_MASK, x)#define MT753X_R01_PORT_FW_MASK GENMASK(2, 0)/* Register for :03 and :0E MAC DA frame control */ #define MT753X_RGAC2 0x2c +#define MT753X_R0E_BPDU_FR BIT(25)#define MT753X_R0E_EG_TAG_MASK GENMASK(24, 22)#define MT753X_R0E_EG_TAG (x) FIELD_PREP(MT753X_R0E_EG_TAG_MASK, x)#define MT753X_R0E_PORT_FW_MASK GENMASK(18, 16)#define MT753X_R0E_PORT_FW (x) FIELD_PREP(MT753X_R0E_PORT_FW_MASK, x) +#define MT753X_R03_BPDU_FR BIT(9)#define MT753X_R03_EG_TAG_MASK GENMASK(8, 6)#define MT753X_R03_EG_TAG (x) FIELD_PREP(MT753X_R03_EG_TAG_MASK, x)#define MT753X_R03_PORT_FW_MASK GENMASK(2, 0)
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 94/172 af_unix: Do not use atomic ops for unix_sk(sk)->inflight.
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 97af84a6bba2ab2b9c704c08e67de3b5ea551bb2 ]
When touching unix_sk(sk)->inflight, we are always under spin_lock(&unix_gc_lock).
Let's convert unix_sk(sk)->inflight to the normal unsigned long.
Stable-dep-of: 47d8ac011fe1 ("af_unix: Fix garbage collector racing against connect()")
--- include/net/af_unix.h | 2 +- net/unix/af_unix.c | 4 ++-- net/unix/garbage.c | 17 ++++++++--------- net/unix/scm.c | 8 +++++--- 4 files changed, 16 insertions(+), 15 deletions(-)
include/net/af_unix.h +1 -1
@@ -55,7 +55,7 @@ struct unix_sock {struct mutex iolock, bindlock;struct sock *peer;struct list_head link; -atomic_long_t inflight; +unsigned long inflight;spinlock_t lock;unsigned long gc_flags;#define UNIX_GC_CANDIDATE 0
net/unix/af_unix.c +2 -2
@@ -980,11 +980,11 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern, sk->sk_write_space = unix_write_space; sk->sk_max_ack_backlog = net->unx.sysctl_max_dgram_qlen; sk->sk_destruct = unix_sock_destructor; - u =unix_sk (sk); + u =unix_sk (sk); + u->inflight =0 ; u->path.dentry =NULL ; u->path.mnt =NULL ;spin_lock_init (&u->lock); -atomic_long_set (&u->inflight,0 );INIT_LIST_HEAD (&u->link);mutex_init (&u->iolock);/* single task reading lock */ mutex_init (&u->bindlock);/* single task binding lock */
net/unix/garbage.c +8 -9
@@ -166,17 +166,18 @@ static void scan_children(struct sock *x, void (*func)(struct unix_sock *),static void dec_inflight (struct unix_sock *usk) { -atomic_long_dec (&usk->inflight); + usk->inflight--; }static void inc_inflight (struct unix_sock *usk) { -atomic_long_inc (&usk->inflight); + usk->inflight++; }static void inc_inflight_move_tail (struct unix_sock *u) { -atomic_long_inc (&u->inflight); + u->inflight++; +/* If this still might be part of a cycle, move it to the end * of the list, so that it's checked even if it was already * passed over
@@ -237,14 +238,12 @@ void unix_gc(void)*/ list_for_each_entry_safe (u, next, &gc_inflight_list, link) {long total_refs; -long inflight_refs; total_refs =file_count (u->sk.sk_socket->file); - inflight_refs =atomic_long_read (&u->inflight); -BUG_ON (inflight_refs <1 ); -BUG_ON (total_refs < inflight_refs); -if (total_refs == inflight_refs) { +BUG_ON (!u->inflight); +BUG_ON (total_refs < u->inflight); +if (total_refs == u->inflight) {list_move_tail (&u->link, &gc_candidates);__set_bit (UNIX_GC_CANDIDATE , &u->gc_flags);__set_bit (UNIX_GC_MAYBE_CYCLE , &u->gc_flags);
@@ -271,7 +270,7 @@ void unix_gc(void)/* Move cursor to after the current position. */ list_move (&cursor, &u->link); -if (atomic_long_read (&u->inflight) >0 ) { +if (u->inflight) {list_move_tail (&u->link, ¬_cycle_list);__clear_bit (UNIX_GC_MAYBE_CYCLE , &u->gc_flags);scan_children (&u->sk, inc_inflight_move_tail,NULL );
net/unix/scm.c +5 -3
@@ -53,12 +53,13 @@ void unix_inflight(struct user_struct *user, struct file *fp)if (s) {struct unix_sock *u =unix_sk (s); -if (atomic_long_inc_return (&u->inflight) ==1 ) { +if (!u->inflight) {BUG_ON (!list_empty (&u->link));list_add_tail (&u->link, &gc_inflight_list); }else {BUG_ON (list_empty (&u->link)); } + u->inflight++;/* Paired with READ_ONCE() in wait_for_unix_gc() */ WRITE_ONCE (unix_tot_inflight, unix_tot_inflight +1 ); }
@@ -75,10 +76,11 @@ void unix_notinflight(struct user_struct *user, struct file *fp)if (s) {struct unix_sock *u =unix_sk (s); -BUG_ON (!atomic_long_read (&u->inflight)); +BUG_ON (!u->inflight);BUG_ON (list_empty (&u->link)); -if (atomic_long_dec_and_test (&u->inflight)) + u->inflight--; +if (!u->inflight)list_del_init (&u->link);/* Paired with READ_ONCE() in wait_for_unix_gc() */ WRITE_ONCE (unix_tot_inflight, unix_tot_inflight -1 );
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 95/172 af_unix: Fix garbage collector racing against connect()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michal Luczaj <[email protected]>
[ Upstream commit 47d8ac011fe1c9251070e1bd64cb10b48193ec51 ]
Garbage collector does not take into account the risk of embryo getting enqueued during the garbage collection. If such embryo has a peer that carries SCM_RIGHTS, two consecutive passes of scan_children() may see a different set of children. Leading to an incorrectly elevated inflight count, and then a dangling pointer within the gc_inflight_list.
sockets are AF_UNIX/SOCK_STREAM
S is an unconnected socket
L is a listening in-flight socket bound to addr, not in fdtable V's fd will be passed via sendmsg(), gets inflight count bumped
connect(S, addr) sendmsg(S, [V]); close(V) __unix_gc()
---------------- ------------------------- -----------
NS = unix_create1()
skb1 = sock_wmalloc(NS)
L = unix_find_other(addr)unix_state_lock(L)unix_peer(S) = NS
// V count=1 inflight=0
NS = unix_peer(S)
skb2 = sock_alloc()
skb_queue_tail(NS, skb2[V])
// V became in-flight
// V count=2 inflight=1
close(V)
// V count=1 inflight=1
// GC candidate condition met
for u in gc_inflight_list:
if (total_refs == inflight_refs)
add u to gc_candidates
// gc_candidates={L, V}
for u in gc_candidates:
scan_children(u, dec_inflight)
// embryo (skb1) was not
// reachable from L yet, so V's
// inflight remains unchanged
__skb_queue_tail(L, skb1)
unix_state_unlock(L)
for u in gc_candidates:
if (u.inflight)
scan_children(u, inc_inflight_move_tail)
// V count=1 inflight=2 (!)If there is a GC-candidate listening socket, lock/unlock its state. This makes GC wait until the end of any ongoing connect() to that socket. After flipping the lock, a possibly SCM-laden embryo is already enqueued. And if there is another embryo coming, it can not possibly carry SCM_RIGHTS. At this point, unix_inflight() can not happen because unix_gc_lock is already taken. Inflight graph remains unaffected.
--- net/unix/garbage.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-)
net/unix/garbage.c +17 -1
@@ -235,11 +235,22 @@ void unix_gc(void)* receive queues. Other, non candidate sockets _can_ be * added to queue, so we must make sure only to touch * candidates. +* +* Embryos, though never candidates themselves, affect which +* candidates are reachable by the garbage collector. Before +* being added to a listener's queue, an embryo may already +* receive data carrying SCM_RIGHTS, potentially making the +* passed socket a candidate that is not yet reachable by the +* collector. It becomes reachable once the embryo is +* enqueued. Therefore, we must ensure that no SCM-laden +* embryo appears in a (candidate) listener's queue between +* consecutive scan_children() calls. */ list_for_each_entry_safe (u, next, &gc_inflight_list, link) { +struct sock *sk = &u->sk;long total_refs; - total_refs =file_count (u->sk.sk_socket->file); + total_refs =file_count (sk->sk_socket->file);BUG_ON (!u->inflight);BUG_ON (total_refs < u->inflight);
@@ -247,6 +258,11 @@ void unix_gc(void)list_move_tail (&u->link, &gc_candidates);__set_bit (UNIX_GC_CANDIDATE , &u->gc_flags);__set_bit (UNIX_GC_MAYBE_CYCLE , &u->gc_flags); + +if (sk->sk_state ==TCP_LISTEN ) { +unix_state_lock (sk); +unix_state_unlock (sk); + } } }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 96/172 net: ena: Fix potential sign extension issue
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Arinzon <[email protected]>
[ Upstream commit 713a85195aad25d8a26786a37b674e3e5ec09e3c ]
Small unsigned types are promoted to larger signed types in the case of multiplication, the result of which may overflow. In case the result of such a multiplication has its MSB turned on, it will be sign extended with '1's.
This changes the multiplication result.
Code example of the phenomenon:
-------------------------------
u16 x, y;size_t z1, z2;
x = y = 0xffff;printk("x=%x y=%x\n",x,y);
z1 = x*y;
z2 = (size_t)x*y;
printk("z1=%lx z2=%lx\n", z1, z2);
Output:
-------
x=ffff y=ffff
z1=fffffffffffe0001 z2=fffe0001
The expected result of ffff*ffff is fffe0001, and without the explicit casting to avoid the unwanted sign extension we got fffffffffffe0001.
This commit adds an explicit casting to avoid the sign extension issue.
--- drivers/net/ethernet/amazon/ena/ena_com.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/amazon/ena/ena_com.c +1 -1
@@ -362,7 +362,7 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev,ENA_COM_BOUNCE_BUFFER_CNTRL_CNT ; io_sq->bounce_buf_ctrl.next_to_use =0 ; - size = io_sq->bounce_buf_ctrl.buffer_size * + size = (size_t )io_sq->bounce_buf_ctrl.buffer_size * io_sq->bounce_buf_ctrl.buffers_num; dev_node =dev_to_node (ena_dev->dmadev);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 97/172 net: ena: Wrong missing IO completions check order
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Arinzon <[email protected]>
[ Upstream commit f7e417180665234fdb7af2ebe33d89aaa434d16f ]
Missing IO completions check is called every second (HZ jiffies). This commit fixes several issues with this check:
1. Duplicate queues check:
Max of 4 queues are scanned on each check due to monitor budget. Once reaching the budget, this check exits under the assumption that the next check will continue to scan the remainder of the queues, but in practice, next check will first scan the last already scanned queue which is not necessary and may cause the full queue scan to last a couple of seconds longer.
The fix is to start every check with the next queue to scan. For example, on 8 IO queues:
Bug: [0,1,2,3], [3,4,5,6], [6,7]
Fix: [0,1,2,3], [4,5,6,7]
2. Unbalanced queues check:
In case the number of active IO queues is not a multiple of budget, there will be checks which don't utilize the full budget because the full scan exits when reaching the last queue id. The fix is to run every TX completion check with exact queue budget regardless of the queue id.
For example, on 7 IO queues:
Bug: [0,1,2,3], [4,5,6], [0,1,2,3]
Fix: [0,1,2,3], [4,5,6,0], [1,2,3,4]
The budget may be lowered in case the number of IO queues is less than the budget (4) to make sure there are no duplicate queues on the same check.
For example, on 3 IO queues:
Bug: [0,1,2,0], [1,2,0,1]
Fix: [0,1,2], [0,1,2]
--- drivers/net/ethernet/amazon/ena/ena_netdev.c | 21 +++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-)
drivers/net/ethernet/amazon/ena/ena_netdev.c +12 -9
@@ -3421,10 +3421,11 @@ static void check_for_missing_completions(struct ena_adapter *adapter) {struct ena_ring *tx_ring;struct ena_ring *rx_ring; -int i, budget, rc; +int qid, budget, rc;int io_queue_count; io_queue_count = adapter->xdp_num_queues + adapter->num_io_queues; +/* Make sure the driver doesn't turn the device in other process */ smp_rmb ();
@@ -3437,27 +3438,29 @@ static void check_for_missing_completions(struct ena_adapter *adapter)if (adapter->missing_tx_completion_to ==ENA_HW_HINTS_NO_TIMEOUT )return ; - budget =ENA_MONITORED_TX_QUEUES ; + budget =min_t (u32, io_queue_count,ENA_MONITORED_TX_QUEUES ); -for (i = adapter->last_monitored_tx_qid; i < io_queue_count; i++) { - tx_ring = &adapter->tx_ring[i]; - rx_ring = &adapter->rx_ring[i]; + qid = adapter->last_monitored_tx_qid; + +while (budget) { + qid = (qid +1 ) % io_queue_count; + + tx_ring = &adapter->tx_ring[qid]; + rx_ring = &adapter->rx_ring[qid]; rc =check_missing_comp_in_tx_queue (adapter, tx_ring);if (unlikely (rc))return ; - rc = !ENA_IS_XDP_INDEX (adapter, i) ? + rc = !ENA_IS_XDP_INDEX (adapter, qid) ?check_for_rx_interrupt_queue (adapter, rx_ring) :0 ;if (unlikely (rc))return ; budget--; -if (!budget) -break ; } - adapter->last_monitored_tx_qid = i % io_queue_count; + adapter->last_monitored_tx_qid = qid; }/* trigger napi schedule after 2 consecutive detections */
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 98/172 net: ena: Fix incorrect descriptor free behavior
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Arinzon <[email protected]>
[ Upstream commit bf02d9fe00632d22fa91d34749c7aacf397b6cde ]
ENA has two types of TX queues:
- queues which only process TX packets arriving from the network stack
- queues which only process TX packets forwarded to it by XDP_REDIRECT or XDP_TX instructions
The ena_free_tx_bufs() cycles through all descriptors in a TX queue and unmaps + frees every descriptor that hasn't been acknowledged yet by the device (uncompleted TX transactions).
The function assumes that the processed TX queue is necessarily from the first category listed above and ends up using napi_consume_skb() for descriptors belonging to an XDP specific queue.
This patch solves a bug in which, in case of a VF reset, the descriptors aren't freed correctly, leading to crashes.
--- drivers/net/ethernet/amazon/ena/ena_netdev.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)
drivers/net/ethernet/amazon/ena/ena_netdev.c +11 -3
@@ -696,8 +696,11 @@ void ena_unmap_tx_buff(struct ena_ring *tx_ring,static void ena_free_tx_bufs (struct ena_ring *tx_ring) {bool print_once = true; +bool is_xdp_ring;u32 i; + is_xdp_ring =ENA_IS_XDP_INDEX (tx_ring->adapter, tx_ring->qid); +for (i =0 ; i < tx_ring->ring_size; i++) {struct ena_tx_buffer *tx_info = &tx_ring->tx_buffer_info[i];
@@ -717,10 +720,15 @@ static void ena_free_tx_bufs(struct ena_ring *tx_ring)ena_unmap_tx_buff (tx_ring, tx_info); -dev_kfree_skb_any (tx_info->skb); +if (is_xdp_ring) +xdp_return_frame (tx_info->xdpf); +else +dev_kfree_skb_any (tx_info->skb); } -netdev_tx_reset_queue (netdev_get_tx_queue (tx_ring->netdev, - tx_ring->qid)); + +if (!is_xdp_ring) +netdev_tx_reset_queue (netdev_get_tx_queue (tx_ring->netdev, + tx_ring->qid)); }static void ena_free_all_tx_bufs (struct ena_adapter *adapter)
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 99/172 net: ena: Set tx_info->xdpf value to NULL
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Arinzon <[email protected]>
[ Upstream commit 36a1ca01f0452f2549420e7279c2588729bd94df ]
The patch mentioned in the Fixes tag removed the explicit assignment of tx_info->xdpf to NULL with the justification that there's no need to set tx_info->xdpf to NULL and tx_info->num_of_bufs to 0 in case of a mapping error. Both values won't be used once the mapping function returns an error, and their values would be overridden by the next transmitted packet.
While both values do indeed get overridden in the next transmission call, the value of tx_info->xdpf is also used to check whether a TX descriptor's transmission has been completed (i.e. a completion for it was polled).
An example scenario:
1. Mapping failed, tx_info->xdpf wasn't set to NULL
2. A VF reset occurred leading to IO resource destruction and a call to ena_free_tx_bufs() function
3. Although the descriptor whose mapping failed was freed by the transmission function, it still passes the check
if (!tx_info->skb)
(skb and xdp_frame are in a union)
4. The xdp_frame associated with the descriptor is freed twice
This patch returns the assignment of NULL to tx_info->xdpf to make the cleaning function knows that the descriptor is already freed.
--- drivers/net/ethernet/amazon/ena/ena_xdp.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/amazon/ena/ena_xdp.c +3 -1
@@ -89,7 +89,7 @@ int ena_xdp_xmit_frame(struct ena_ring *tx_ring, rc =ena_xdp_tx_map_frame (tx_ring, tx_info, xdpf, &ena_tx_ctx);if (unlikely (rc)) -return rc; + goto err; ena_tx_ctx.req_id = req_id;
@@ -112,7 +112,9 @@ int ena_xdp_xmit_frame(struct ena_ring *tx_ring, error_unmap_dma:ena_unmap_tx_buff (tx_ring, tx_info); +err: tx_info->xdpf =NULL ; +return rc; }
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 100/172 drm/xe/display: Fix double mutex initialization
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lucas De Marchi <[email protected]>
[ Upstream commit 50a9b7fc151e67b9e642232d32e8c5a5ac13e64a ]
All of these mutexes are already initialized by the display side since commit 3fef3e6ff86a ("drm/i915: move display mutex inits to display code"), so the xe shouldn´t initialize them.
--- drivers/gpu/drm/xe/xe_display.c | 5 ----- 1 file changed, 5 deletions(-)
drivers/gpu/drm/xe/xe_display.c +0 -5
@@ -108,11 +108,6 @@ int xe_display_create(struct xe_device *xe) xe->display.hotplug.dp_wq =alloc_ordered_workqueue ("xe-dp" ,0 );drmm_mutex_init (&xe->drm, &xe->sb_lock); -drmm_mutex_init (&xe->drm, &xe->display.backlight.lock); -drmm_mutex_init (&xe->drm, &xe->display.audio.mutex); -drmm_mutex_init (&xe->drm, &xe->display.wm.wm_mutex); -drmm_mutex_init (&xe->drm, &xe->display.pps.mutex); -drmm_mutex_init (&xe->drm, &xe->display.hdcp.hdcp_mutex); xe->enabled_irq_mask = ~0 ; err =drmm_add_action_or_reset (&xe->drm, display_destroy,NULL );
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 101/172 drm/xe/hwmon: Cast result to output precision on left shift of operand
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Karthik Poosa <[email protected]>
[ Upstream commit a8ad8715472bb8f6a2ea8b4072a28151eb9f4f24 ]
Address potential overflow in result of left shift of a lower precision (u32) operand before assignment to higher precision (u64) variable.
v2:
- Update commit message. (Himal)
--- drivers/gpu/drm/xe/xe_hwmon.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/xe/xe_hwmon.c +2 -2
@@ -288,7 +288,7 @@ xe_hwmon_power1_max_interval_show(struct device *dev, struct device_attribute *a* As y can be < 2, we compute tau4 = (4 | x) << y * and then add 2 when doing the final right shift to account for units */ - tau4 = ((1 << x_w) | x) << y; + tau4 = (u64)((1 << x_w) | x) << y;/* val in hwmon interface units (millisec) */ out =mul_u64_u32_shr (tau4,SF_TIME , hwmon->scl_shift_time + x_w);
@@ -328,7 +328,7 @@ xe_hwmon_power1_max_interval_store(struct device *dev, struct device_attribute * r =FIELD_PREP (PKG_MAX_WIN ,PKG_MAX_WIN_DEFAULT ); x =REG_FIELD_GET (PKG_MAX_WIN_X , r); y =REG_FIELD_GET (PKG_MAX_WIN_Y , r); - tau4 = ((1 << x_w) | x) << y; + tau4 = (u64)((1 << x_w) | x) << y; max_win =mul_u64_u32_shr (tau4,SF_TIME , hwmon->scl_shift_time + x_w);if (val > max_win)
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 102/172 tracing: hide unused ftrace_event_id_fops
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit 5281ec83454d70d98b71f1836fb16512566c01cd ]
When CONFIG_PERF_EVENTS, a 'make W=1' build produces a warning about the unused ftrace_event_id_fops variable:
kernel/trace/trace_events.c:2155:37: error: 'ftrace_event_id_fops' defined but not used [-Werror=unused-const-variable=]
2155 | static const struct file_operations ftrace_event_id_fops = {
Hide this in the same #ifdef as the reference to it.
--- kernel/trace/trace_events.c | 4 ++++ 1 file changed, 4 insertions(+)
kernel/trace/trace_events.c +4 -0
@@ -1670,6 +1670,7 @@ static int trace_format_open(struct inode *inode, struct file *file)return 0 ; } +#ifdef CONFIG_PERF_EVENTS static ssize_t event_id_read (struct file *filp,char __user *ubuf,size_t cnt,loff_t *ppos) {
@@ -1684,6 +1685,7 @@ event_id_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos)return simple_read_from_buffer (ubuf, cnt, ppos, buf, len); } +#endif static ssize_t event_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
@@ -2152,10 +2154,12 @@ static const struct file_operations ftrace_event_format_fops = { .release = seq_release, }; +#ifdef CONFIG_PERF_EVENTS static const struct file_operations ftrace_event_id_fops = { .read = event_id_read, .llseek = default_llseek, }; +#endif static const struct file_operations ftrace_event_filter_fops = { .open = tracing_open_file_tr,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 103/172 iommu/vt-d: Fix wrong use of pasid config
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xuchun Shang <[email protected]>
[ Upstream commit 5b3625a4f6422e8982f90f0c11b5546149c962b8 ]
The commit "iommu/vt-d: Add IOMMU perfmon support" introduce IOMMU PMU feature, but use the wrong config when set pasid filter.
--- drivers/iommu/intel/perfmon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iommu/intel/perfmon.c +1 -1
@@ -438,7 +438,7 @@ static int iommu_pmu_assign_event(struct iommu_pmu *iommu_pmu,iommu_pmu_set_filter (domain, event->attr.config1,IOMMU_PMU_FILTER_DOMAIN , idx, event->attr.config1); -iommu_pmu_set_filter (pasid, event->attr.config1, +iommu_pmu_set_filter (pasid, event->attr.config2,IOMMU_PMU_FILTER_PASID , idx, event->attr.config1); iommu_pmu_set_filter(ats, event->attr.config2,
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 104/172 iommu/vt-d: Allocate local memory for page request queue
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jacob Pan <[email protected]>
[ Upstream commit a34f3e20ddff02c4f12df2c0635367394e64c63d ]
The page request queue is per IOMMU, its allocation should be made NUMA-aware for performance reasons.
--- drivers/iommu/intel/svm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iommu/intel/svm.c +1 -1
@@ -67,7 +67,7 @@ int intel_svm_enable_prq(struct intel_iommu *iommu)struct page *pages;int irq, ret; - pages =alloc_pages (GFP_KERNEL | __GFP_ZERO,PRQ_ORDER ); + pages =alloc_pages_node (iommu->node,GFP_KERNEL | __GFP_ZERO,PRQ_ORDER );if (!pages) {pr_warn ("IOMMU: %s: Failed to allocate page request queue\n" , iommu->name);
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 105/172 iommu/vt-d: Fix WARN_ON in iommu probe path
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lu Baolu <[email protected]>
[ Upstream commit 89436f4f54125b1297aec1f466efd8acb4ec613d ]
Commit 1a75cc710b95 ("iommu/vt-d: Use rbtree to track iommu probed devices") adds all devices probed by the iommu driver in a rbtree indexed by the source ID of each device. It assumes that each device has a unique source ID. This assumption is incorrect and the VT-d spec doesn't state this requirement either.
The reason for using a rbtree to track devices is to look up the device with PCI bus and devfunc in the paths of handling ATS invalidation time out error and the PRI I/O page faults. Both are PCI ATS feature related.
Only track the devices that have PCI ATS capabilities in the rbtree to avoid unnecessary WARN_ON in the iommu probe path. Otherwise, on some platforms below kernel splat will be displayed and the iommu probe results in failure.
WARNING: CPU: 3 PID: 166 at drivers/iommu/intel/iommu.c:158 intel_iommu_probe_device+0x319/0xd90 Call Trace: <TASK> ? __warn+0x7e/0x180 ? intel_iommu_probe_device+0x319/0xd90 ? report_bug+0x1f8/0x200 ? handle_bug+0x3c/0x70 ? exc_invalid_op+0x18/0x70 ? asm_exc_invalid_op+0x1a/0x20 ? intel_iommu_probe_device+0x319/0xd90 ? debug_mutex_init+0x37/0x50 __iommu_probe_device+0xf2/0x4f0 iommu_probe_device+0x22/0x70 iommu_bus_notifier+0x1e/0x40 notifier_call_chain+0x46/0x150 blocking_notifier_call_chain+0x42/0x60 bus_notify+0x2f/0x50 device_add+0x5ed/0x7e0 platform_device_add+0xf5/0x240 mfd_add_devices+0x3f9/0x500 ? preempt_count_add+0x4c/0xa0 ? up_write+0xa2/0x1b0 ? __debugfs_create_file+0xe3/0x150 intel_lpss_probe+0x49f/0x5b0 ? pci_conf1_write+0xa3/0xf0 intel_lpss_pci_probe+0xcf/0x110 [intel_lpss_pci] pci_device_probe+0x95/0x120 really_probe+0xd9/0x370 ? __pfx___driver_attach+0x10/0x10 __driver_probe_device+0x73/0x150 driver_probe_device+0x19/0xa0 __driver_attach+0xb6/0x180 ? __pfx___driver_attach+0x10/0x10 bus_for_each_dev+0x77/0xd0 bus_add_driver+0x114/0x210 driver_register+0x5b/0x110 ? __pfx_intel_lpss_pci_driver_init+0x10/0x10 [intel_lpss_pci] do_one_initcall+0x57/0x2b0 ? kmalloc_trace+0x21e/0x280 ? do_init_module+0x1e/0x210 do_init_module+0x5f/0x210 load_module+0x1d37/0x1fc0 ? init_module_from_file+0x86/0xd0 init_module_from_file+0x86/0xd0 idempotent_init_module+0x17c/0x230 __x64_sys_finit_module+0x56/0xb0 do_syscall_64+0x6e/0x140 entry_SYSCALL_64_after_hwframe+0x71/0x79
--- drivers/iommu/intel/iommu.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)
drivers/iommu/intel/iommu.c +7 -4
@@ -4381,9 +4381,11 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev) }dev_iommu_priv_set (dev, info); - ret =device_rbtree_insert (iommu, info); -if (ret) - goto free; +if (pdev &&pci_ats_supported (pdev)) { + ret =device_rbtree_insert (iommu, info); +if (ret) + goto free; + }if (sm_supported (iommu) && !dev_is_real_dma_subdevice (dev)) { ret =intel_pasid_alloc_table (dev);
@@ -4410,7 +4412,8 @@ static void intel_iommu_release_device(struct device *dev)struct intel_iommu *iommu = info->iommu;mutex_lock (&iommu->iopf_lock); -device_rbtree_remove (info); +if (dev_is_pci (dev) &&pci_ats_supported (to_pci_dev (dev))) +device_rbtree_remove (info);mutex_unlock (&iommu->iopf_lock); if (sm_supported(iommu) && !dev_is_real_dma_subdevice(dev) &&
Signature · 1 line
2.43.0
Greg Kroah-Hartman
maintainerPATCH 106/172 io_uring: refactor DEFER_TASKRUN multishot checks
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Pavel Begunkov <[email protected]>
Commit e0e4ab52d17096d96c21a6805ccd424b283c3c6d upstream.
We disallow DEFER_TASKRUN multishots from running by io-wq, which is checked by individual opcodes in the issue path. We can consolidate all it in io_wq_submit_work() at the same time moving the checks out of the hot path.
--- io_uring/io_uring.c | 20 ++++++++++++++++++++ io_uring/net.c | 21 --------------------- io_uring/rw.c | 2 -- 3 files changed, 20 insertions(+), 23 deletions(-)
io_uring/io_uring.c +20 -0
@@ -949,6 +949,8 @@ bool io_fill_cqe_req_aux(struct io_kiocbu64 user_data = req->cqe.user_data;struct io_uring_cqe *cqe; +lockdep_assert (!io_wq_current_is_worker ()); +if (!defer)return __io_post_aux_cqe (ctx, user_data, res, cflags, false);
@@ -1950,6 +1952,24 @@ fail: goto fail; } +/* +* If DEFER_TASKRUN is set, it's only allowed to post CQEs from the +* submitter task context. Final request completions are handed to the +* right context, however this is not the case of auxiliary CQEs, +* which is the main mean of operation for multishot requests. +* Don't allow any multishot execution from io-wq. It's more restrictive +* than necessary and also cleaner. +*/ +if (req->flags &REQ_F_APOLL_MULTISHOT ) { + err = -EBADFD ; +if (!file_can_poll (req->file)) + goto fail; + err = -ECANCELED ; +if (io_arm_poll_handler (req, issue_flags) !=IO_APOLL_OK ) + goto fail; +return ; + } +if (req->flags &REQ_F_FORCE_ASYNC ) {bool opcode_poll = def->pollin || def->pollout;
io_uring/net.c +0 -21
@@ -78,19 +78,6 @@ struct io_sr_msg {*/ #define MULTISHOT_MAX_RETRY 32 -static inline bool io_check_multishot (struct io_kiocb *req, -unsigned int issue_flags) -{ -/* -* When ->locked_cq is set we only allow to post CQEs from the original -* task context. Usual request completions will be handled in other -* generic paths but multipoll may decide to post extra cqes. -*/ -return !(issue_flags &IO_URING_F_IOWQ ) || - !(req->flags &REQ_F_APOLL_MULTISHOT ) || - !req->ctx->task_complete; -} -int io_shutdown_prep (struct io_kiocb *req,const struct io_uring_sqe *sqe) { struct io_shutdown *shutdown = io_kiocb_to_cmd(req, struct io_shutdown);
@@ -837,9 +824,6 @@ int io_recvmsg(struct io_kiocb *req, uns (sr->flags & IORING_RECVSEND_POLL_FIRST))return io_setup_async_msg (req, kmsg, issue_flags); -if (!io_check_multishot (req, issue_flags)) -return io_setup_async_msg (req, kmsg, issue_flags); - retry_multishot:if (io_do_buffer_select (req)) {void __user *buf;
@@ -935,9 +919,6 @@ int io_recv(struct io_kiocb *req, unsign (sr->flags & IORING_RECVSEND_POLL_FIRST))return -EAGAIN ; -if (!io_check_multishot (req, issue_flags)) -return -EAGAIN ; - sock =sock_from_file (req->file);if (unlikely (!sock))return -ENOTSOCK ;
@@ -1386,8 +1367,6 @@ int io_accept(struct io_kiocb *req, unsistruct file *file;int ret, fd; -if (!io_check_multishot (req, issue_flags)) -return -EAGAIN ; retry:if (!fixed) { fd =__get_unused_fd_flags (accept->flags, accept->nofile);
io_uring/rw.c +0 -2
@@ -932,8 +932,6 @@ int io_read_mshot(struct io_kiocb *req,*/ if (!file_can_poll (req->file))return -EBADFD ; -if (issue_flags &IO_URING_F_IOWQ ) -return -EAGAIN ; ret =__io_read (req, issue_flags);
Greg Kroah-Hartman
maintainerPATCH 107/172 io_uring: disable io-wq execution of multishot NOWAIT requests
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <[email protected]>
Commit bee1d5becdf5bf23d4ca0cd9c6b60bdf3c61d72b upstream.
Do the same check for direct io-wq execution for multishot requests that commit 2a975d426c82 did for the inline execution, and disable multishot mode (and revert to single shot) if the file type doesn't support NOWAIT, and isn't opened in O_NONBLOCK mode. For multishot to work properly, it's a requirement that nonblocking read attempts can be done.
--- io_uring/io_uring.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
io_uring/io_uring.c +9 -4
@@ -1964,10 +1964,15 @@ fail: err = -EBADFD ;if (!file_can_poll (req->file)) goto fail; - err = -ECANCELED ; -if (io_arm_poll_handler (req, issue_flags) !=IO_APOLL_OK ) - goto fail; -return ; +if (req->file->f_flags &O_NONBLOCK || + req->file->f_mode &FMODE_NOWAIT ) { + err = -ECANCELED ; +if (io_arm_poll_handler (req, issue_flags) !=IO_APOLL_OK ) + goto fail; +return ; + }else { + req->flags &= ~REQ_F_APOLL_MULTISHOT ; + } }if (req->flags &REQ_F_FORCE_ASYNC ) {
Greg Kroah-Hartman
maintainerPATCH 108/172 btrfs: qgroup: correctly model root qgroup rsv in convert
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Boris Burkov <[email protected]>
commit 141fb8cd206ace23c02cd2791c6da52c1d77d42a upstream.
We use add_root_meta_rsv and sub_root_meta_rsv to track prealloc and pertrans reservations for subvolumes when quotas are enabled. The convert function does not properly increment pertrans after decrementing prealloc, so the count is not accurate.
Note: we check that the fs is not read-only to mirror the logic in qgroup_convert_meta, which checks that before adding to the pertrans rsv.
--- fs/btrfs/qgroup.c | 2 ++ 1 file changed, 2 insertions(+)
fs/btrfs/qgroup.c +2 -0
@@ -4432,6 +4432,8 @@ void btrfs_qgroup_convert_reserved_meta( BTRFS_QGROUP_RSV_META_PREALLOC);trace_qgroup_meta_convert (root, num_bytes);qgroup_convert_meta (fs_info, root->root_key.objectid, num_bytes); +if (!sb_rdonly (fs_info->sb)) +add_root_meta_rsv (root, num_bytes,BTRFS_QGROUP_RSV_META_PERTRANS ); }/*
Greg Kroah-Hartman
maintainerPATCH 109/172 btrfs: qgroup: fix qgroup prealloc rsv leak in subvolume operations
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Boris Burkov <[email protected]>
commit 74e97958121aa1f5854da6effba70143f051b0cd upstream.
Create subvolume, create snapshot and delete subvolume all use btrfs_subvolume_reserve_metadata() to reserve metadata for the changes done to the parent subvolume's fs tree, which cannot be mediated in the normal way via start_transaction. When quota groups (squota or qgroups) are enabled, this reserves qgroup metadata of type PREALLOC. Once the operation is associated to a transaction, we convert PREALLOC to PERTRANS, which gets cleared in bulk at the end of the transaction.
However, the error paths of these three operations were not implementing this lifecycle correctly. They unconditionally converted the PREALLOC to PERTRANS in a generic cleanup step regardless of errors or whether the operation was fully associated to a transaction or not. This resulted in error paths occasionally converting this rsv to PERTRANS without calling record_root_in_trans successfully, which meant that unless that root got recorded in the transaction by some other thread, the end of the transaction would not free that root's PERTRANS, leaking it. Ultimately, this resulted in hitting a WARN in CONFIG_BTRFS_DEBUG builds at unmount for the leaked reservation.
The fix is to ensure that every qgroup PREALLOC reservation observes the following properties:
1. any failure before record_root_in_trans is called successfully results in freeing the PREALLOC reservation.
2. after record_root_in_trans, we convert to PERTRANS, and now the transaction owns freeing the reservation.
This patch enforces those properties on the three operations. Without it, generic/269 with squotas enabled at mkfs time would fail in ~5-10 runs on my system. With this patch, it ran successfully 1000 times in a row.
--- fs/btrfs/inode.c | 13 ++++++++++++- fs/btrfs/ioctl.c | 37 ++++++++++++++++++++++++++++--------- fs/btrfs/root-tree.c | 10 ---------- fs/btrfs/root-tree.h | 2 -- 4 files changed, 40 insertions(+), 22 deletions(-)
fs/btrfs/inode.c +12 -1
@@ -4476,6 +4476,7 @@ int btrfs_delete_subvolume(struct btrfs_struct btrfs_trans_handle *trans;struct btrfs_block_rsv block_rsv;u64 root_flags; +u64 qgroup_reserved =0 ;int ret;down_write (&fs_info->subvol_sem);
@@ -4520,12 +4521,20 @@ int btrfs_delete_subvolume(struct btrfs_ ret =btrfs_subvolume_reserve_metadata (root, &block_rsv,5 , true);if (ret) goto out_undead; + qgroup_reserved = block_rsv.qgroup_rsv_reserved; trans =btrfs_start_transaction (root,0 );if (IS_ERR (trans)) { ret =PTR_ERR (trans); goto out_release; } + ret =btrfs_record_root_in_trans (trans, root); +if (ret) { +btrfs_abort_transaction (trans, ret); + goto out_end_trans; + } +btrfs_qgroup_convert_reserved_meta (root, qgroup_reserved); + qgroup_reserved =0 ; trans->block_rsv = &block_rsv; trans->bytes_reserved = block_rsv.size;
@@ -4584,7 +4593,9 @@ out_end_trans: ret =btrfs_end_transaction (trans); inode->i_flags |=S_DEAD ; out_release: -btrfs_subvolume_release_metadata (root, &block_rsv); +btrfs_block_rsv_release (fs_info, &block_rsv, (u64)-1 ,NULL ); +if (qgroup_reserved) +btrfs_qgroup_free_meta_prealloc (root, qgroup_reserved); out_undead:if (ret) {spin_lock (&dest->root_item_lock);
fs/btrfs/ioctl.c +28 -9
@@ -603,6 +603,7 @@ static noinline int create_subvol(structint ret;dev_t anon_dev;u64 objectid; +u64 qgroup_reserved =0 ; root_item =kzalloc (sizeof (*root_item),GFP_KERNEL );if (!root_item)
@@ -640,13 +641,18 @@ static noinline int create_subvol(struct trans_num_items, false);if (ret) goto out_new_inode_args; + qgroup_reserved = block_rsv.qgroup_rsv_reserved; trans =btrfs_start_transaction (root,0 );if (IS_ERR (trans)) { ret =PTR_ERR (trans); -btrfs_subvolume_release_metadata (root, &block_rsv); - goto out_new_inode_args; + goto out_release_rsv; } + ret =btrfs_record_root_in_trans (trans,BTRFS_I (dir)->root); +if (ret) + goto out; +btrfs_qgroup_convert_reserved_meta (root, qgroup_reserved); + qgroup_reserved =0 ; trans->block_rsv = &block_rsv; trans->bytes_reserved = block_rsv.size;/* Tree log can't currently deal with an inode which is a new root. */
@@ -757,9 +763,11 @@ static noinline int create_subvol(struct out: trans->block_rsv =NULL ; trans->bytes_reserved =0 ; -btrfs_subvolume_release_metadata (root, &block_rsv); -btrfs_end_transaction (trans); +out_release_rsv: +btrfs_block_rsv_release (fs_info, &block_rsv, (u64)-1 ,NULL ); +if (qgroup_reserved) +btrfs_qgroup_free_meta_prealloc (root, qgroup_reserved); out_new_inode_args:btrfs_new_inode_args_destroy (&new_inode_args); out_inode:
@@ -781,6 +789,8 @@ static int create_snapshot(struct btrfs_struct btrfs_pending_snapshot *pending_snapshot;unsigned int trans_num_items;struct btrfs_trans_handle *trans; +struct btrfs_block_rsv *block_rsv; +u64 qgroup_reserved =0 ;int ret;/* We do not support snapshotting right now. */
@@ -817,19 +827,19 @@ static int create_snapshot(struct btrfs_ goto free_pending; } -btrfs_init_block_rsv (&pending_snapshot->block_rsv, -BTRFS_BLOCK_RSV_TEMP ); + block_rsv = &pending_snapshot->block_rsv; +btrfs_init_block_rsv (block_rsv,BTRFS_BLOCK_RSV_TEMP );/* * 1 to add dir item * 1 to add dir index * 1 to update parent inode item */ trans_num_items =create_subvol_num_items (inherit) +3 ; - ret =btrfs_subvolume_reserve_metadata (BTRFS_I (dir)->root, - &pending_snapshot->block_rsv, + ret =btrfs_subvolume_reserve_metadata (BTRFS_I (dir)->root, block_rsv, trans_num_items, false);if (ret) goto free_pending; + qgroup_reserved = block_rsv->qgroup_rsv_reserved; pending_snapshot->dentry = dentry; pending_snapshot->root = root;
@@ -842,6 +852,13 @@ static int create_snapshot(struct btrfs_ ret =PTR_ERR (trans); goto fail; } + ret =btrfs_record_root_in_trans (trans,BTRFS_I (dir)->root); +if (ret) { +btrfs_end_transaction (trans); + goto fail; + } +btrfs_qgroup_convert_reserved_meta (root, qgroup_reserved); + qgroup_reserved =0 ; trans->pending_snapshot = pending_snapshot;
@@ -871,7 +888,9 @@ fail:if (ret && pending_snapshot->snap) pending_snapshot->snap->anon_dev =0 ;btrfs_put_root (pending_snapshot->snap); -btrfs_subvolume_release_metadata (root, &pending_snapshot->block_rsv); +btrfs_block_rsv_release (fs_info, block_rsv, (u64)-1 ,NULL ); +if (qgroup_reserved) +btrfs_qgroup_free_meta_prealloc (root, qgroup_reserved); free_pending:if (pending_snapshot->anon_dev)free_anon_bdev (pending_snapshot->anon_dev);
fs/btrfs/root-tree.c +0 -10
@@ -539,13 +539,3 @@ int btrfs_subvolume_reserve_metadata(str }return ret; } - -void btrfs_subvolume_release_metadata (struct btrfs_root *root, -struct btrfs_block_rsv *rsv) -{ -struct btrfs_fs_info *fs_info = root->fs_info; -u64 qgroup_to_release; - -btrfs_block_rsv_release (fs_info, rsv, (u64)-1 , &qgroup_to_release); -btrfs_qgroup_convert_reserved_meta (root, qgroup_to_release); -}
fs/btrfs/root-tree.h +0 -2
@@ -8,8 +8,6 @@ struct fscrypt_str;int btrfs_subvolume_reserve_metadata (struct btrfs_root *root,struct btrfs_block_rsv *rsv,int nitems,bool use_global_rsv); -void btrfs_subvolume_release_metadata (struct btrfs_root *root, -struct btrfs_block_rsv *rsv);int btrfs_add_root_ref (struct btrfs_trans_handle *trans,u64 root_id,u64 ref_id,u64 dirid,u64 sequence,const struct fscrypt_str *name);
Greg Kroah-Hartman
maintainerPATCH 110/172 btrfs: record delayed inode root in transaction
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Boris Burkov <[email protected]>
commit 71537e35c324ea6fbd68377a4f26bb93a831ae35 upstream.
When running delayed inode updates, we do not record the inode's root in the transaction, but we do allocate PREALLOC and thus converted PERTRANS space for it. To be sure we free that PERTRANS meta rsv, we must ensure that we record the root in the transaction.
--- fs/btrfs/delayed-inode.c | 3 +++ 1 file changed, 3 insertions(+)
fs/btrfs/delayed-inode.c +3 -0
@@ -1128,6 +1128,9 @@ __btrfs_commit_inode_delayed_items(strucif (ret)return ret; + ret =btrfs_record_root_in_trans (trans, node->root); +if (ret) +return ret; ret =btrfs_update_delayed_inode (trans, node->root, path, node);return ret; }
Greg Kroah-Hartman
maintainerPATCH 111/172 btrfs: qgroup: convert PREALLOC to PERTRANS after record_root_in_trans
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Boris Burkov <[email protected]>
commit 211de93367304ab395357f8cb12568a4d1e20701 upstream.
The transaction is only able to free PERTRANS reservations for a root once that root has been recorded with the TRANS tag on the roots radix tree. Therefore, until we are sure that this root will get tagged, it isn't safe to convert. Generally, this is not an issue as *some* transaction will likely tag the root before long and this reservation will get freed in that transaction, but technically it could stick around until unmount and result in a warning about leaked metadata reservation space.
This path is most exercised by running the generic/269 fstest with CONFIG_BTRFS_DEBUG.
--- fs/btrfs/transaction.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-)
fs/btrfs/transaction.c +8 -9
@@ -747,14 +747,6 @@ again: h->reloc_reserved = reloc_reserved; } -/* -* Now that we have found a transaction to be a part of, convert the -* qgroup reservation from prealloc to pertrans. A different transaction -* can't race in and free our pertrans out from under us. -*/ -if (qgroup_reserved) -btrfs_qgroup_convert_reserved_meta (root, qgroup_reserved); - got_it:if (!current->journal_info) current->journal_info = h;
@@ -788,8 +780,15 @@ got_it:* not just freed. */ btrfs_end_transaction (h); -return ERR_PTR (ret); + goto reserve_fail; } +/* +* Now that we have found a transaction to be a part of, convert the +* qgroup reservation from prealloc to pertrans. A different transaction +* can't race in and free our pertrans out from under us. +*/ +if (qgroup_reserved) +btrfs_qgroup_convert_reserved_meta (root, qgroup_reserved);return h;
Greg Kroah-Hartman
maintainerPATCH 112/172 io_uring/net: restore msg_control on sendzc retry
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Pavel Begunkov <[email protected]>
commit 4fe82aedeb8a8cb09bfa60f55ab57b5c10a74ac4 upstream.
cac9e4418f4cb ("io_uring/net: save msghdr->msg_control for retries") reinstatiates msg_control before every __sys_sendmsg_sock(), since the function can overwrite the value in msghdr. We need to do same for zerocopy sendmsg.
--- io_uring/net.c | 1 + 1 file changed, 1 insertion(+)
io_uring/net.c +1 -0
@@ -1255,6 +1255,7 @@ int io_sendmsg_zc(struct io_kiocb *req,if (req_has_async_data (req)) { kmsg = req->async_data; + kmsg->msg.msg_control_user = sr->msg_control; }else { ret =io_sendmsg_copy_hdr (req, &iomsg);if (ret)
Greg Kroah-Hartman
maintainerPATCH 113/172 kprobes: Fix possible use-after-free issue on kprobe registration
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zheng Yejian <[email protected]>
commit 325f3fb551f8cd672dbbfc4cf58b14f9ee3fc9e8 upstream.
When unloading a module, its state is changing MODULE_STATE_LIVE ->
MODULE_STATE_GOING -> MODULE_STATE_UNFORMED. Each change will take
a time. is_module_text_address() and __module_text_address() works with MODULE_STATE_LIVE and MODULE_STATE_GOING.
If we use is_module_text_address() and __module_text_address() separately, there is a chance that the first one is succeeded but the next one is failed because module->state becomes MODULE_STATE_UNFORMED between those operations.
In check_kprobe_address_safe(), if the second __module_text_address() is failed, that is ignored because it expected a kernel_text address. But it may have failed simply because module->state has been changed to MODULE_STATE_UNFORMED. In this case, arm_kprobe() will try to modify non-exist module text address (use-after-free).
To fix this problem, we should not use separated is_module_text_address() and __module_text_address(), but use only __module_text_address() once and do try_module_get(module) which is only available with MODULE_STATE_LIVE.
--- kernel/kprobes.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-)
kernel/kprobes.c +12 -6
@@ -1567,10 +1567,17 @@ static int check_kprobe_address_safe(strjump_label_lock ();preempt_disable (); -/* Ensure it is not in reserved area nor out of text */ - if (!(core_kernel_text((unsigned long) p->addr) || - is_module_text_address((unsigned long) p->addr)) || - in_gate_area_no_mm((unsigned long) p->addr) || +/* Ensure the address is in a text area, and find a module if exists. */ + *probed_mod =NULL ; +if (!core_kernel_text ((unsigned long ) p->addr)) { + *probed_mod =__module_text_address ((unsigned long ) p->addr); +if (!(*probed_mod)) { + ret = -EINVAL ; + goto out; + } + } +/* Ensure it is not in reserved area. */ + if (in_gate_area_no_mm((unsigned long) p->addr) || within_kprobe_blacklist((unsigned long) p->addr) || jump_label_text_reserved(p->addr, p->addr) || static_call_text_reserved(p->addr, p->addr) ||
@@ -1580,8 +1587,7 @@ static int check_kprobe_address_safe(str goto out; } -/* Check if 'p' is probing a module. */ - *probed_mod =__module_text_address ((unsigned long ) p->addr); +/* Get module refcount and reject __init functions for loaded modules. */ if (*probed_mod) {/* * We must hold a refcount of the probed module while updating
Greg Kroah-Hartman
maintainerPATCH 114/172 fs/proc: remove redundant comments from /proc/bootconfig
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zhenhua Huang <[email protected]>
commit fbbdc255fbee59b4207a5398fdb4f04590681a79 upstream.
commit 717c7c894d4b ("fs/proc: Add boot loader arguments as comment to /proc/bootconfig") adds bootloader argument comments into /proc/bootconfig.
/proc/bootconfig shows boot_command_line[] multiple times following every xbc key value pair, that's duplicated and not necessary. Remove redundant ones.
Output before and after the fix is like:
key1 = value1
*bootloader argument comments*
key2 = value2
*bootloader argument comments*
key3 = value3
*bootloader argument comments*
...
key1 = value1
key2 = value2
key3 = value3
*bootloader argument comments*
...
--- fs/proc/bootconfig.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
fs/proc/bootconfig.c +6 -6
@@ -62,12 +62,12 @@ static int __init copy_xbc_key_value_list(char *dst, size_t size)break ; dst += ret; } -if (ret >=0 && boot_command_line[0 ]) { - ret =snprintf (dst,rest (dst, end),"# Parameters from bootloader:\n# %s\n" , - boot_command_line); -if (ret >0 ) - dst += ret; - } + } +if (ret >=0 && boot_command_line[0 ]) { + ret =snprintf (dst,rest (dst, end),"# Parameters from bootloader:\n# %s\n" , + boot_command_line); +if (ret >0 ) + dst += ret; } out:kfree (key);
Signature · 1 line
2.44.0
Greg Kroah-Hartman
maintainerPATCH 115/172 fs/proc: Skip bootloader comment if no embedded kernel parameters
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Masami Hiramatsu <[email protected]>
commit c722cea208789d9e2660992bcd05fb9fac3adb56 upstream.
If the "bootconfig" kernel command-line argument was specified or if the kernel was built with CONFIG_BOOT_CONFIG_FORCE, but if there are no embedded kernel parameter, omit the "# Parameters from bootloader:" comment from the /proc/bootconfig file. This will cause automation to fall back to the /proc/cmdline file, which will be identical to the comment in this no-embedded-kernel-parameters case.
--- fs/proc/bootconfig.c | 2 +- include/linux/bootconfig.h | 1 + init/main.c | 5 +++++ 3 files changed, 7 insertions(+), 1 deletion(-)
fs/proc/bootconfig.c +1 -1
@@ -63,7 +63,7 @@ static int __init copy_xbc_key_value_lis dst += ret; } } -if (ret >=0 && boot_command_line[0 ]) { +if (cmdline_has_extra_options () && ret >=0 && boot_command_line[0 ]) { ret =snprintf (dst,rest (dst, end),"# Parameters from bootloader:\n# %s\n" , boot_command_line);if (ret >0 )
include/linux/bootconfig.h +1 -0
@@ -10,6 +10,7 @@#ifdef __KERNEL__#include <linux/kernel.h> #include <linux/types.h> +bool __init cmdline_has_extra_options (void );#else /* !__KERNEL__ */ /* * NOTE: This is only for tools/bootconfig, because tools/bootconfig will
init/main.c +5 -0
@@ -485,6 +485,11 @@ static int __init warn_bootconfig(char *early_param ("bootconfig" , warn_bootconfig); +bool __init cmdline_has_extra_options (void ) +{ +return extra_command_line || extra_init_args; +} +/* Change NUL term back to "=", to make "param" the whole string. */ static void __init repair_env_string (char *param,char *val) {
Greg Kroah-Hartman
maintainerPATCH 116/172 scsi: sg: Avoid sg device teardown race
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alexander Wetzel <[email protected]>
commit 27f58c04a8f438078583041468ec60597841284d upstream.
sg_remove_sfp_usercontext() must not use sg_device_destroy() after calling scsi_device_put().
sg_device_destroy() is accessing the parent scsi_device request_queue which will already be set to NULL when the preceding call to scsi_device_put() removed the last reference to the parent scsi_device.
The resulting NULL pointer exception will then crash the kernel.
--- drivers/scsi/sg.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/scsi/sg.c +3 -1
@@ -2207,6 +2207,7 @@ sg_remove_sfp_usercontext(struct work_st { struct sg_fd *sfp = container_of(work, struct sg_fd, ew.work);struct sg_device *sdp = sfp->parentdp; +struct scsi_device *device = sdp->device;Sg_request *srp;unsigned long iflags;
@@ -2232,8 +2233,9 @@ sg_remove_sfp_usercontext(struct work_st "sg_remove_sfp: sfp=0x%p\n", sfp));kfree (sfp); -scsi_device_put (sdp->device); +WARN_ON_ONCE (kref_read (&sdp->d_ref) !=1 );kref_put (&sdp->d_ref, sg_device_destroy); +scsi_device_put (device);module_put (THIS_MODULE ); }
Greg Kroah-Hartman
maintainerPATCH 117/172 scsi: sg: Avoid race in error handling & drop bogus warn
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alexander Wetzel <[email protected]>
commit d4e655c49f474deffaf5ed7e65034b8167ee39c8 upstream.
Commit 27f58c04a8f4 ("scsi: sg: Avoid sg device teardown race") introduced an incorrect WARN_ON_ONCE() and missed a sequence where sg_device_destroy() was used after scsi_device_put().
sg_device_destroy() is accessing the parent scsi_device request_queue which will already be set to NULL when the preceding call to scsi_device_put() removed the last reference to the parent scsi_device.
Drop the incorrect WARN_ON_ONCE() - allowing more than one concurrent access to the sg device - and make sure sg_device_destroy() is not used after scsi_device_put() in the error handling.
--- drivers/scsi/sg.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)
drivers/scsi/sg.c +10 -8
@@ -285,6 +285,7 @@ sg_open(struct inode *inode, struct fileint dev =iminor (inode);int flags = filp->f_flags;struct request_queue *q; +struct scsi_device *device;Sg_device *sdp;Sg_fd *sfp;int retval;
@@ -301,11 +302,12 @@ sg_open(struct inode *inode, struct file/* This driver's module count bumped by fops_get in <linux/fs.h> */ /* Prevent the device driver from vanishing while we sleep */ - retval =scsi_device_get (sdp->device); + device = sdp->device; + retval =scsi_device_get (device);if (retval) goto sg_put; - retval =scsi_autopm_get_device (sdp->device); + retval =scsi_autopm_get_device (device);if (retval) goto sdp_put;
@@ -313,7 +315,7 @@ sg_open(struct inode *inode, struct file* check if O_NONBLOCK. Permits SCSI commands to be issued * during error recovery. Tread carefully. */ if (!((flags &O_NONBLOCK ) || -scsi_block_when_processing_errors (sdp->device))) { +scsi_block_when_processing_errors (device))) { retval = -ENXIO ;/* we are in error recovery for this device */ goto error_out;
@@ -344,7 +346,7 @@ sg_open(struct inode *inode, struct fileif (sdp->open_cnt <1 ) {/* no existing opens */ sdp->sgdebug =0 ; - q = sdp->device->request_queue; + q = device->request_queue; sdp->sg_tablesize =queue_max_segments (q); } sfp =sg_add_sfp (sdp);
@@ -370,10 +372,11 @@ out_undo: error_mutex_locked:mutex_unlock (&sdp->open_rel_lock); error_out: -scsi_autopm_put_device (sdp->device); +scsi_autopm_put_device (device); sdp_put: -scsi_device_put (sdp->device); - goto sg_put; +kref_put (&sdp->d_ref, sg_device_destroy); +scsi_device_put (device); +return retval; }/* Release resources associated with a successful sg_open()
@@ -2233,7 +2236,6 @@ sg_remove_sfp_usercontext(struct work_st "sg_remove_sfp: sfp=0x%p\n", sfp));kfree (sfp); -WARN_ON_ONCE (kref_read (&sdp->d_ref) !=1 );kref_put (&sdp->d_ref, sg_device_destroy);scsi_device_put (device);module_put (THIS_MODULE );
Greg Kroah-Hartman
maintainerPATCH 118/172 accel/ivpu: Check return code of ipc->lock init
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Wachowski, Karol <[email protected]>
commit f0cf7ffcd02953c72fed5995378805883d16203e upstream.
Return value of drmm_mutex_init(ipc->lock) was unchecked.
--- drivers/accel/ivpu/ivpu_ipc.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/accel/ivpu/ivpu_ipc.c +6 -2
@@ -1,6 +1,6 @@// SPDX-License-Identifier: GPL-2.0-only /* -* Copyright (C) 2020-2023 Intel Corporation +* Copyright (C) 2020-2024 Intel Corporation */ #include <linux/genalloc.h>
@@ -501,7 +501,11 @@ int ivpu_ipc_init(struct ivpu_device *vdspin_lock_init (&ipc->cons_lock);INIT_LIST_HEAD (&ipc->cons_list);INIT_LIST_HEAD (&ipc->cb_msg_list); -drmm_mutex_init (&vdev->drm, &ipc->lock); + ret =drmm_mutex_init (&vdev->drm, &ipc->lock); +if (ret) { +ivpu_err (vdev,"Failed to initialize ipc->lock, ret %d\n" , ret); + goto err_free_rx; + }ivpu_ipc_reset (vdev);return 0 ;
Greg Kroah-Hartman
maintainerPATCH 119/172 accel/ivpu: Fix PCI D0 state entry in resume
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Wachowski, Karol <[email protected]>
commit 3534eacbf101f6e66105f03d869a03893407c384 upstream.
In case of failed power up we end up left in PCI D3hot state making it impossible to access NPU registers on retry. Enter D0 state on retry before proceeding with power up sequence.
--- drivers/accel/ivpu/ivpu_pm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/accel/ivpu/ivpu_pm.c +2 -2
@@ -74,10 +74,10 @@ static int ivpu_resume(struct ivpu_devic {int ret; -pci_set_power_state (to_pci_dev (vdev->drm.dev),PCI_D0 ); +retry:pci_restore_state (to_pci_dev (vdev->drm.dev)); +pci_set_power_state (to_pci_dev (vdev->drm.dev),PCI_D0 ); -retry: ret =ivpu_hw_power_up (vdev);if (ret) {ivpu_err (vdev,"Failed to power up HW: %d\n" , ret);
Greg Kroah-Hartman
maintainerPATCH 120/172 accel/ivpu: Put NPU back to D3hot after failed resume
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jacek Lawrynowicz <[email protected]>
commit 875bc9cd1b33eb027a5663f5e6878a43d98e9a16 upstream.
Put NPU in D3hot after ivpu_resume() fails to power up the device. This will assure that D3->D0 power cycle will be performed before the next resume and also will minimize power usage in this corner case.
--- drivers/accel/ivpu/ivpu_pm.c | 1 + 1 file changed, 1 insertion(+)
drivers/accel/ivpu/ivpu_pm.c +1 -0
@@ -100,6 +100,7 @@ err_mmu_disable:ivpu_mmu_disable (vdev); err_power_down:ivpu_hw_power_down (vdev); +pci_set_power_state (to_pci_dev (vdev->drm.dev), PCI_D3hot);if (!ivpu_fw_is_cold_boot (vdev)) {ivpu_pm_prepare_cold_boot (vdev);
Greg Kroah-Hartman
maintainerPATCH 121/172 accel/ivpu: Return max freq for DRM_IVPU_PARAM_CORE_CLOCK_RATE
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jacek Lawrynowicz <[email protected]>
commit c52c35e5b404b95a5bcff39af9be1b9293be3434 upstream.
DRM_IVPU_PARAM_CORE_CLOCK_RATE returns current NPU frequency which could be 0 if device was sleeping. This value isn't really useful to the user space, so return max freq instead which can be used to estimate NPU performance.
--- drivers/accel/ivpu/ivpu_drv.c | 18 +----------------- drivers/accel/ivpu/ivpu_hw.h | 6 ++++++ drivers/accel/ivpu/ivpu_hw_37xx.c | 7 ++++--- drivers/accel/ivpu/ivpu_hw_40xx.c | 6 ++++++ 4 files changed, 17 insertions(+), 20 deletions(-)
drivers/accel/ivpu/ivpu_drv.c +1 -17
@@ -131,22 +131,6 @@ static int ivpu_get_capabilities(structreturn 0 ; } -static int ivpu_get_core_clock_rate (struct ivpu_device *vdev,u64 *clk_rate) -{ -int ret; - - ret =ivpu_rpm_get_if_active (vdev); -if (ret <0 ) -return ret; - - *clk_rate = ret ?ivpu_hw_reg_pll_freq_get (vdev) :0 ; - -if (ret) -ivpu_rpm_put (vdev); - -return 0 ; -} -static int ivpu_get_param_ioctl (struct drm_device *dev,void *data,struct drm_file *file) {struct ivpu_file_priv *file_priv = file->driver_priv;
@@ -170,7 +154,7 @@ static int ivpu_get_param_ioctl(struct d args->value = vdev->platform;break ;case DRM_IVPU_PARAM_CORE_CLOCK_RATE : - ret =ivpu_get_core_clock_rate (vdev, &args->value); + args->value =ivpu_hw_ratio_to_freq (vdev, vdev->hw->pll.max_ratio);break ;case DRM_IVPU_PARAM_NUM_CONTEXTS : args->value =ivpu_get_context_count (vdev);
drivers/accel/ivpu/ivpu_hw.h +6 -0
@@ -21,6 +21,7 @@ struct ivpu_hw_ops {u32 (*profiling_freq_get)(struct ivpu_device *vdev);void (*profiling_freq_drive)(struct ivpu_device *vdev,bool enable);u32 (*reg_pll_freq_get)(struct ivpu_device *vdev); +u32 (*ratio_to_freq)(struct ivpu_device *vdev,u32 ratio);u32 (*reg_telemetry_offset_get)(struct ivpu_device *vdev);u32 (*reg_telemetry_size_get)(struct ivpu_device *vdev);u32 (*reg_telemetry_enable_get)(struct ivpu_device *vdev);
@@ -130,6 +131,11 @@ static inline u32 ivpu_hw_reg_pll_freq_greturn vdev->hw->ops->reg_pll_freq_get (vdev); }; +static inline u32 ivpu_hw_ratio_to_freq (struct ivpu_device *vdev,u32 ratio) +{ +return vdev->hw->ops->ratio_to_freq (vdev, ratio); +} +static inline u32 ivpu_hw_reg_telemetry_offset_get (struct ivpu_device *vdev) {return vdev->hw->ops->reg_telemetry_offset_get (vdev);
drivers/accel/ivpu/ivpu_hw_37xx.c +4 -3
@@ -805,12 +805,12 @@ static void ivpu_hw_37xx_profiling_freq_/* Profiling freq - is a debug feature. Unavailable on VPU 37XX. */ } -static u32 ivpu_hw_37xx_pll_to_freq (u32 ratio,u32 config) +static u32 ivpu_hw_37xx_ratio_to_freq (struct ivpu_device *vdev,u32 ratio) {u32 pll_clock =PLL_REF_CLK_FREQ * ratio;u32 cpu_clock; -if ((config &0xff ) ==PLL_RATIO_4_3 ) +if ((vdev->hw->config &0xff ) ==PLL_RATIO_4_3 ) cpu_clock = pll_clock *2 /4 ;else cpu_clock = pll_clock *2 /5 ;
@@ -829,7 +829,7 @@ static u32 ivpu_hw_37xx_reg_pll_freq_getif (!ivpu_is_silicon (vdev))return PLL_SIMULATION_FREQ ; -return ivpu_hw_37xx_pll_to_freq (pll_curr_ratio, vdev->hw->config); +return ivpu_hw_37xx_ratio_to_freq (vdev, pll_curr_ratio); }static u32 ivpu_hw_37xx_reg_telemetry_offset_get (struct ivpu_device *vdev)
@@ -1052,6 +1052,7 @@ const struct ivpu_hw_ops ivpu_hw_37xx_op .profiling_freq_get = ivpu_hw_37xx_profiling_freq_get, .profiling_freq_drive = ivpu_hw_37xx_profiling_freq_drive, .reg_pll_freq_get = ivpu_hw_37xx_reg_pll_freq_get, + .ratio_to_freq = ivpu_hw_37xx_ratio_to_freq, .reg_telemetry_offset_get = ivpu_hw_37xx_reg_telemetry_offset_get, .reg_telemetry_size_get = ivpu_hw_37xx_reg_telemetry_size_get, .reg_telemetry_enable_get = ivpu_hw_37xx_reg_telemetry_enable_get,
drivers/accel/ivpu/ivpu_hw_40xx.c +6 -0
@@ -980,6 +980,11 @@ static u32 ivpu_hw_40xx_reg_pll_freq_getreturn PLL_RATIO_TO_FREQ (pll_curr_ratio); } +static u32 ivpu_hw_40xx_ratio_to_freq (struct ivpu_device *vdev,u32 ratio) +{ +return PLL_RATIO_TO_FREQ (ratio); +} +static u32 ivpu_hw_40xx_reg_telemetry_offset_get (struct ivpu_device *vdev) {return REGB_RD32 (VPU_40XX_BUTTRESS_VPU_TELEMETRY_OFFSET );
@@ -1230,6 +1235,7 @@ const struct ivpu_hw_ops ivpu_hw_40xx_op .profiling_freq_get = ivpu_hw_40xx_profiling_freq_get, .profiling_freq_drive = ivpu_hw_40xx_profiling_freq_drive, .reg_pll_freq_get = ivpu_hw_40xx_reg_pll_freq_get, + .ratio_to_freq = ivpu_hw_40xx_ratio_to_freq, .reg_telemetry_offset_get = ivpu_hw_40xx_reg_telemetry_offset_get, .reg_telemetry_size_get = ivpu_hw_40xx_reg_telemetry_size_get, .reg_telemetry_enable_get = ivpu_hw_40xx_reg_telemetry_enable_get,
Greg Kroah-Hartman
maintainerPATCH 122/172 accel/ivpu: Fix deadlock in context_xa
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jacek Lawrynowicz <[email protected]>
commit fd7726e75968b27fe98534ccbf47ccd6fef686f3 upstream.
ivpu_device->context_xa is locked both in kernel thread and IRQ context. It requires XA_FLAGS_LOCK_IRQ flag to be passed during initialization otherwise the lock could be acquired from a thread and interrupted by an IRQ that locks it for the second time causing the deadlock.
This deadlock was reported by lockdep and observed in internal tests.
--- drivers/accel/ivpu/ivpu_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/accel/ivpu/ivpu_drv.c +1 -1
@@ -514,7 +514,7 @@ static int ivpu_dev_init(struct ivpu_dev vdev->context_xa_limit.min =IVPU_USER_CONTEXT_MIN_SSID ; vdev->context_xa_limit.max =IVPU_USER_CONTEXT_MAX_SSID ;atomic64_set (&vdev->unique_id_counter,0 ); -xa_init_flags (&vdev->context_xa,XA_FLAGS_ALLOC ); +xa_init_flags (&vdev->context_xa,XA_FLAGS_ALLOC |XA_FLAGS_LOCK_IRQ );xa_init_flags (&vdev->submitted_jobs_xa,XA_FLAGS_ALLOC1 );lockdep_set_class (&vdev->submitted_jobs_xa.xa_lock, &submitted_jobs_xa_lock_class_key);INIT_LIST_HEAD (&vdev->bo_list);
Greg Kroah-Hartman
maintainerPATCH 123/172 drm/vmwgfx: Enable DMA mappings with SEV
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zack Rusin <[email protected]>
commit 4c08f01934ab67d1d283d5cbaa52b923abcfe4cd upstream.
Enable DMA mappings in vmwgfx after TTM has been fixed in commit 3bf3710e3718 ("drm/ttm: Add a generic TTM memcpy move for page-based iomem")
This enables full guest-backed memory support and in particular allows usage of screen targets as the presentation mechanism.
--- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +6 -5
@@ -666,11 +666,12 @@ static int vmw_dma_select_mode(struct vm [vmw_dma_map_populate] = "Caching DMA mappings.", [vmw_dma_map_bind] = "Giving up DMA mappings early."}; -/* TTM currently doesn't fully support SEV encryption. */ -if (cc_platform_has (CC_ATTR_MEM_ENCRYPT )) -return -EINVAL ; - -if (vmw_force_coherent) +/* +* When running with SEV we always want dma mappings, because +* otherwise ttm tt pool pages will bounce through swiotlb running +* out of available space. +*/ +if (vmw_force_coherent ||cc_platform_has (CC_ATTR_MEM_ENCRYPT )) dev_priv->map_mode = vmw_dma_alloc_coherent;else if (vmw_restrict_iommu) dev_priv->map_mode = vmw_dma_map_bind;
Greg Kroah-Hartman
maintainerPATCH 124/172 drm/i915/vrr: Disable VRR when using bigjoiner
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ville Syrjälä <[email protected]>
commit dcd8992e47f13afb5c11a61e8d9c141c35e23751 upstream.
All joined pipes share the same transcoder/timing generator. Currently we just do the commits per-pipe, which doesn't really work if we need to change switch between non-VRR and VRR timings generators on the fly, or even when sending the push to the transcoder. For now just disable VRR when bigjoiner is needed.
--- drivers/gpu/drm/i915/display/intel_vrr.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/gpu/drm/i915/display/intel_vrr.c +7 -0
@@ -117,6 +117,13 @@ intel_vrr_compute_config(struct intel_crconst struct drm_display_info *info = &connector->base.display_info;int vmin, vmax; +/* +* FIXME all joined pipes share the same transcoder. +* Need to account for that during VRR toggle/push/etc. +*/ +if (crtc_state->bigjoiner_pipes) +return ; +if (adjusted_mode->flags &DRM_MODE_FLAG_INTERLACE )return ;
Greg Kroah-Hartman
maintainerPATCH 125/172 drm/amdkfd: Reset GPU on queue preemption failure
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Harish Kasiviswanathan <[email protected]>
commit 8bdfb4ea95ca738d33ef71376c21eba20130f2eb upstream.
Currently, with F32 HWS GPU reset is only when unmap queue fails.
However, if compute queue doesn't repond to preemption request in time unmap will return without any error. In this case, only preemption error is logged and Reset is not triggered. Call GPU reset in this case also.
--- drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +1 -0
@@ -1997,6 +1997,7 @@ static int unmap_queues_cpsch(struct devdev_err (dev,"HIQ MQD's queue_doorbell_id0 is not 0, Queue preemption time out\n" );while (halt_if_hws_hang)schedule (); +kfd_hws_hang (dqm);return -ETIME ; }
Greg Kroah-Hartman
maintainerPATCH 126/172 drm/ast: Fix soft lockup
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jammy Huang <[email protected]>
commit bc004f5038220b1891ef4107134ccae44be55109 upstream.
There is a while-loop in ast_dp_set_on_off() that could lead to infinite-loop. This is because the register, VGACRI-Dx, checked in this API is a scratch register actually controlled by a MCU, named DPMCU, in BMC.
These scratch registers are protected by scu-lock. If suc-lock is not off, DPMCU can not update these registers and then host will have soft lockup due to never updated status.
DPMCU is used to control DP and relative registers to handshake with host's VGA driver. Even the most time-consuming task, DP's link training, is less than 100ms. 200ms should be enough.
--- drivers/gpu/drm/ast/ast_dp.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/gpu/drm/ast/ast_dp.c +3 -0
@@ -180,6 +180,7 @@ void ast_dp_set_on_off(struct drm_device {struct ast_device *ast =to_ast_device (dev);u8 video_on_off = on; +u32 i =0 ;// Video On/Off ast_set_index_reg_mask (ast,AST_IO_VGACRI ,0xE3 , (u8 ) ~AST_DP_VIDEO_ENABLE , on);
@@ -192,6 +193,8 @@ void ast_dp_set_on_off(struct drm_device ASTDP_MIRROR_VIDEO_ENABLE) != video_on_off) {// wait 1 ms mdelay (1 ); +if (++i >200 ) +break ; } } }
Greg Kroah-Hartman
maintainerPATCH 127/172 drm/panfrost: Fix the error path in panfrost_mmu_map_fault_addr()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Boris Brezillon <[email protected]>
commit 1fc9af813b25e146d3607669247d0f970f5a87c3 upstream.
Subject: [PATCH 6.8 127/172] drm/panfrost: Fix the error path in panfrost_mmu_map_fault_addr()
If some the pages or sgt allocation failed, we shouldn't release the pages ref we got earlier, otherwise we will end up with unbalanced get/put_pages() calls. We should instead leave everything in place and let the BO release function deal with extra cleanup when the object is destroyed, or let the fault handler try again next time it's called.
--- drivers/gpu/drm/panfrost/panfrost_mmu.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
drivers/gpu/drm/panfrost/panfrost_mmu.c +9 -4
@@ -502,11 +502,18 @@ static int panfrost_mmu_map_fault_addr(smapping_set_unevictable (mapping);for (i = page_offset; i < page_offset +NUM_FAULT_PAGES ; i++) { +/* Can happen if the last fault only partially filled this +* section of the pages array before failing. In that case +* we skip already filled pages. +*/ +if (pages[i]) +continue ; + pages[i] =shmem_read_mapping_page (mapping, i);if (IS_ERR (pages[i])) { ret =PTR_ERR (pages[i]); pages[i] =NULL ; - goto err_pages; + goto err_unlock; } }
@@ -514,7 +521,7 @@ static int panfrost_mmu_map_fault_addr(s ret =sg_alloc_table_from_pages (sgt, pages + page_offset,NUM_FAULT_PAGES ,0 ,SZ_2M ,GFP_KERNEL );if (ret) - goto err_pages; + goto err_unlock; ret =dma_map_sgtable (pfdev->dev, sgt,DMA_BIDIRECTIONAL ,0 );if (ret)
@@ -537,8 +544,6 @@ out: err_map:sg_free_table (sgt); -err_pages: -drm_gem_shmem_put_pages (&bo->base); err_unlock:dma_resv_unlock (obj->resv); err_bo:
Greg Kroah-Hartman
maintainerPATCH 128/172 drm/client: Fully protect modes[] with dev->mode_config.mutex
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ville Syrjälä <[email protected]>
commit 3eadd887dbac1df8f25f701e5d404d1b90fd0fea upstream.
The modes[] array contains pointers to modes on the connectors' mode lists, which are protected by dev->mode_config.mutex. Thus we need to extend modes[] the same protection or by the time we use it the elements may already be pointing to freed/reused memory.
--- drivers/gpu/drm/drm_client_modeset.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/drm_client_modeset.c +2 -1
@@ -777,6 +777,7 @@ int drm_client_modeset_probe(struct drm_unsigned int total_modes_count =0 ;struct drm_client_offset *offsets;unsigned int connector_count =0 ; +/* points to modes protected by mode_config.mutex */ struct drm_display_mode **modes;struct drm_crtc **crtcs;int i, ret =0 ;
@@ -845,7 +846,6 @@ int drm_client_modeset_probe(struct drm_drm_client_pick_crtcs (client, connectors, connector_count, crtcs, modes,0 , width, height); } -mutex_unlock (&dev->mode_config.mutex);drm_client_modeset_release (client);
@@ -875,6 +875,7 @@ int drm_client_modeset_probe(struct drm_ modeset->y = offset->y; } } +mutex_unlock (&dev->mode_config.mutex);mutex_unlock (&client->modeset_mutex); out:
Greg Kroah-Hartman
maintainerPATCH 129/172 drm/msm/dp: fix runtime PM leak on disconnect
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hovold <[email protected]>
commit 0640f47b742667fca6aac174f7cd62b6c2c7532c upstream.
Make sure to put the runtime PM usage count (and suspend) also when receiving a disconnect event while in the ST_MAINLINK_READY state.
This specifically avoids leaking a runtime PM usage count on every disconnect with display servers that do not automatically enable external displays when receiving a hotplug notification.
Patchwork: https://patchwork.freedesktop.org/patch/582744/
--- drivers/gpu/drm/msm/dp/dp_display.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/msm/dp/dp_display.c +1 -0
@@ -655,6 +655,7 @@ static int dp_hpd_unplug_handle(struct ddp_display_host_phy_exit (dp); dp->hpd_state =ST_DISCONNECTED ;dp_display_notify_disconnect (&dp->dp_display.pdev->dev); +pm_runtime_put_sync (&pdev->dev);mutex_unlock (&dp->event_mutex);return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 130/172 drm/msm/dp: fix runtime PM leak on connect failure
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hovold <[email protected]>
commit e86750b01a1560f198e4b3e21bb3f78bfd5bb2c3 upstream.
Make sure to balance the runtime PM usage counter (and suspend) before returning on connect failures (e.g. DPCD read failures after a spurious connect event or if link training fails).
Patchwork: https://patchwork.freedesktop.org/patch/582746/
--- drivers/gpu/drm/msm/dp/dp_display.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/msm/dp/dp_display.c +1 -0
@@ -598,6 +598,7 @@ static int dp_hpd_plug_handle(struct dp_ ret =dp_display_usbpd_configure_cb (&pdev->dev);if (ret) {/* link train failed */ dp->hpd_state =ST_DISCONNECTED ; +pm_runtime_put_sync (&pdev->dev); }else { dp->hpd_state =ST_MAINLINK_READY ; }
Greg Kroah-Hartman
maintainerPATCH 131/172 drm/amdgpu/umsch: reinitialize write pointer in hw init
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lang Yu <[email protected]>
commit 0f1bbcc2bab25d5fb2dfb1ee3e08131437690d3d upstream.
Otherwise the old one will be used during GPU reset. That's not expected.
--- drivers/gpu/drm/amd/amdgpu/umsch_mm_v4_0.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/gpu/drm/amd/amdgpu/umsch_mm_v4_0.c +2 -0
@@ -225,6 +225,8 @@ static int umsch_mm_v4_0_ring_start(struWREG32_SOC15 (VCN ,0 , regVCN_UMSCH_RB_SIZE, ring->ring_size); + ring->wptr =0 ; + data =RREG32_SOC15 (VCN ,0 , regVCN_RB_ENABLE); data &= ~(VCN_RB_ENABLE__AUDIO_RB_EN_MASK );WREG32_SOC15 (VCN ,0 , regVCN_RB_ENABLE, data);
Greg Kroah-Hartman
maintainerPATCH 132/172 arm64: dts: imx8qm-ss-dma: fix can lpcg indices
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Frank Li <[email protected]>
commit 00b436182138310bb8d362b912b12a9df8f72ca3 upstream.
can1_lpcg: clock-controller@5ace0000 {
... Col1 Col2
clocks = <&clk IMX_SC_R_CAN_1 IMX_SC_PM_CLK_PER>,// 0 0
<&dma_ipg_clk>, // 1 4
<&dma_ipg_clk>; // 2 5
clock-indices = <IMX_LPCG_CLK_0>,
<IMX_LPCG_CLK_4>,
<IMX_LPCG_CLK_5>;
};Col1: index, which existing dts try to get.
Col2: actual index in lpcg driver
&flexcan2 {
clocks = <&can1_lpcg 1>, <&can1_lpcg 0>;
^^ ^^
Should be:
clocks = <&can1_lpcg IMX_LPCG_CLK_4>, <&can1_lpcg IMX_LPCG_CLK_0>;
};Arg0 is divided by 4 in lpcg driver. So flexcan get IMX_SC_PM_CLK_PER by <&can1_lpcg 1> and <&can1_lpcg 0>. Although function work, code logic is wrong. Fix it by using correct clock indices.
--- arch/arm64/boot/dts/freescale/imx8qm-ss-dma.dtsi | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
arch/arm64/boot/dts/freescale/imx8qm-ss-dma.dtsi +4 -4
@@ -127,15 +127,15 @@ }; &flexcan2 { - clocks = <&can1_lpcg 1>, - <&can1_lpcg 0>; + clocks = <&can1_lpcg IMX_LPCG_CLK_4>, + <&can1_lpcg IMX_LPCG_CLK_0>; assigned-clocks = <&clk IMX_SC_R_CAN_1 IMX_SC_PM_CLK_PER>; fsl,clk-source = /bits/ 8 <1>; }; &flexcan3 { - clocks = <&can2_lpcg 1>, - <&can2_lpcg 0>; + clocks = <&can2_lpcg IMX_LPCG_CLK_4>, + <&can2_lpcg IMX_LPCG_CLK_0>; assigned-clocks = <&clk IMX_SC_R_CAN_2 IMX_SC_PM_CLK_PER>; fsl,clk-source = /bits/ 8 <1>; };
Greg Kroah-Hartman
maintainerPATCH 133/172 arm64: dts: imx8-ss-dma: fix can lpcg indices
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Frank Li <[email protected]>
commit 0893392334b5dffdf616a53679c6a2942c46391b upstream.
can0_lpcg: clock-controller@5acd0000 {
... Col1 Col2
clocks = <&clk IMX_SC_R_CAN_0 IMX_SC_PM_CLK_PER>, // 0 0
<&dma_ipg_clk>, // 1 4
<&dma_ipg_clk>; // 2 5
clock-indices = <IMX_LPCG_CLK_0>,
<IMX_LPCG_CLK_4>,
<IMX_LPCG_CLK_5>;
}Col1: index, which existing dts try to get.
Col2: actual index in lpcg driver.
flexcan1: can@5a8d0000 {
clocks = <&can0_lpcg 1>, <&can0_lpcg 0>;
^^ ^^
Should be:
clocks = <&can0_lpcg IMX_LPCG_CLK_4>, <&can0_lpcg IMX_LPCG_CLK_0>;
};Arg0 is divided by 4 in lpcg driver. flexcan driver get IMX_SC_PM_CLK_PER by <&can0_lpcg 1> and <&can0_lpcg 0>. Although function can work, code logic is wrong. Fix it by using correct clock indices.
--- arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi +6 -6
@@ -406,8 +406,8 @@ dma_subsys: bus@5a000000 { reg = <0x5a8d0000 0x10000>; interrupts = <GIC_SPI 235 IRQ_TYPE_LEVEL_HIGH>; interrupt-parent = <&gic>; - clocks = <&can0_lpcg 1>, - <&can0_lpcg 0>; + clocks = <&can0_lpcg IMX_LPCG_CLK_4>, + <&can0_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "per"; assigned-clocks = <&clk IMX_SC_R_CAN_0 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <40000000>;
@@ -427,8 +427,8 @@ dma_subsys: bus@5a000000 { * CAN1 shares CAN0's clock and to enable CAN0's clock it * has to be powered on. */ - clocks = <&can0_lpcg 1>, - <&can0_lpcg 0>; + clocks = <&can0_lpcg IMX_LPCG_CLK_4>, + <&can0_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "per"; assigned-clocks = <&clk IMX_SC_R_CAN_0 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <40000000>;
@@ -448,8 +448,8 @@ dma_subsys: bus@5a000000 { * CAN2 shares CAN0's clock and to enable CAN0's clock it * has to be powered on. */ - clocks = <&can0_lpcg 1>, - <&can0_lpcg 0>; + clocks = <&can0_lpcg IMX_LPCG_CLK_4>, + <&can0_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "per"; assigned-clocks = <&clk IMX_SC_R_CAN_0 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <40000000>;
Greg Kroah-Hartman
maintainerPATCH 134/172 arm64: dts: imx8-ss-dma: fix adc lpcg indices
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Frank Li <[email protected]>
commit 81975080f14167610976e968e8016e92d836266f upstream.
adc0_lpcg: clock-controller@5ac80000 {
... Col1 Col2
clocks = <&clk IMX_SC_R_ADC_0 IMX_SC_PM_CLK_PER>, // 0 0
<&dma_ipg_clk>; // 1 4
clock-indices = <IMX_LPCG_CLK_0>, <IMX_LPCG_CLK_4>;
};Col1: index, which existing dts try to get.
Col2: actual index in lpcg driver.
adc0: adc@5a880000 {
clocks = <&adc0_lpcg 0>, <&adc0_lpcg 1>;
^^ ^^
clocks = <&adc0_lpcg IMX_LPCG_CLK_0>, <&adc0_lpcg IMX_LPCG_CLK_4>;Arg0 is divided by 4 in lpcg driver. So adc get IMX_SC_PM_CLK_PER by <&adc0_lpcg 0>, <&adc0_lpcg 1>. Although function can work, code logic is wrong. Fix it by using correct indices.
--- arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi +4 -4
@@ -377,8 +377,8 @@ dma_subsys: bus@5a000000 { reg = <0x5a880000 0x10000>; interrupts = <GIC_SPI 240 IRQ_TYPE_LEVEL_HIGH>; interrupt-parent = <&gic>; - clocks = <&adc0_lpcg 0>, - <&adc0_lpcg 1>; + clocks = <&adc0_lpcg IMX_LPCG_CLK_0>, + <&adc0_lpcg IMX_LPCG_CLK_4>; clock-names = "per", "ipg"; assigned-clocks = <&clk IMX_SC_R_ADC_0 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <24000000>;
@@ -392,8 +392,8 @@ dma_subsys: bus@5a000000 { reg = <0x5a890000 0x10000>; interrupts = <GIC_SPI 241 IRQ_TYPE_LEVEL_HIGH>; interrupt-parent = <&gic>; - clocks = <&adc1_lpcg 0>, - <&adc1_lpcg 1>; + clocks = <&adc1_lpcg IMX_LPCG_CLK_0>, + <&adc1_lpcg IMX_LPCG_CLK_4>; clock-names = "per", "ipg"; assigned-clocks = <&clk IMX_SC_R_ADC_1 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <24000000>;
Greg Kroah-Hartman
maintainerPATCH 135/172 arm64: dts: imx8-ss-conn: fix usb lpcg indices
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Frank Li <[email protected]>
commit 808e7716edcdb39d3498b9f567ef6017858b49aa upstream.
usb2_lpcg: clock-controller@5b270000 {
... Col1 Col2
clocks = <&conn_ahb_clk>, <&conn_ipg_clk>; // 0 6
clock-indices = <IMX_LPCG_CLK_6>, <IMX_LPCG_CLK_7>; // 0 7
...
};Col1: index, which existing dts try to get.
Col2: actual index in lpcg driver.
usbotg1: usb@5b0d0000 {
...
clocks = <&usb2_lpcg 0>;
^^
Should be:
clocks = <&usb2_lpcg IMX_LPCG_CLK_6>;
};usbphy1: usbphy@5b100000 {
clocks = <&usb2_lpcg 1>;
^^
SHould be:
clocks = <&usb2_lpcg IMX_LPCG_CLK_7>;
};Arg0 is divided by 4 in lpcg driver. So lpcg will do dummy enable. Fix it by use correct clock indices.
--- arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi +2 -2
@@ -41,7 +41,7 @@ conn_subsys: bus@5b000000 { interrupts = <GIC_SPI 267 IRQ_TYPE_LEVEL_HIGH>; fsl,usbphy = <&usbphy1>; fsl,usbmisc = <&usbmisc1 0>; - clocks = <&usb2_lpcg 0>; + clocks = <&usb2_lpcg IMX_LPCG_CLK_6>; ahb-burst-config = <0x0>; tx-burst-size-dword = <0x10>; rx-burst-size-dword = <0x10>;
@@ -58,7 +58,7 @@ conn_subsys: bus@5b000000 { usbphy1: usbphy@5b100000 { compatible = "fsl,imx7ulp-usbphy"; reg = <0x5b100000 0x1000>; - clocks = <&usb2_lpcg 1>; + clocks = <&usb2_lpcg IMX_LPCG_CLK_7>; power-domains = <&pd IMX_SC_R_USB_0_PHY>; status = "disabled"; };
Greg Kroah-Hartman
maintainerPATCH 136/172 arm64: dts: imx8-ss-dma: fix pwm lpcg indices
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Frank Li <[email protected]>
commit 9055d87bce7276234173fa90e9702af31b3f5353 upstream.
adma_pwm_lpcg: clock-controller@5a590000 {
... col1 col2
clocks = <&clk IMX_SC_R_LCD_0_PWM_0 IMX_SC_PM_CLK_PER>,// 0 0
<&dma_ipg_clk>; // 1 4
clock-indices = <IMX_LPCG_CLK_0>, <IMX_LPCG_CLK_4>;
...
};Col1: index, which existing dts try to get.
Col2: actual index in lpcg driver.
adma_pwm: pwm@5a190000 {
...
clocks = <&adma_pwm_lpcg 1>, <&adma_pwm_lpcg 0>;
^^ ^^
Should be
clocks = <&adma_pwm_lpcg IMX_LPCG_CLK_4>,
<&adma_pwm_lpcg IMX_LPCG_CLK_0>;
};Arg0 will be divided by 4 in lcpg driver, so pwm will get IMX_SC_PM_CLK_PER by <&adma_pwm_lpcg 1>, <&adma_pwm_lpcg 0>. Although function can work, code logic is wrong. Fix it by use correct indices.
--- arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi +2 -2
@@ -144,8 +144,8 @@ dma_subsys: bus@5a000000 { compatible = "fsl,imx8qxp-pwm", "fsl,imx27-pwm"; reg = <0x5a190000 0x1000>; interrupts = <GIC_SPI 127 IRQ_TYPE_LEVEL_HIGH>; - clocks = <&adma_pwm_lpcg 1>, - <&adma_pwm_lpcg 0>; + clocks = <&adma_pwm_lpcg IMX_LPCG_CLK_4>, + <&adma_pwm_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "per"; assigned-clocks = <&clk IMX_SC_R_LCD_0_PWM_0 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <24000000>;
Greg Kroah-Hartman
maintainerPATCH 137/172 arm64: dts: imx8-ss-lsio: fix pwm lpcg indices
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Frank Li <[email protected]>
commit 1d86c2b3946e69d6b0b93568d312aae6247847c0 upstream.
lpcg's arg0 should use clock indices instead of index.
pwm0_lpcg: clock-controller@5d400000 {
... // Col1 Col2
clocks = <&clk IMX_SC_R_PWM_0 IMX_SC_PM_CLK_PER>, // 0 0
<&clk IMX_SC_R_PWM_0 IMX_SC_PM_CLK_PER>, // 1 1
<&clk IMX_SC_R_PWM_0 IMX_SC_PM_CLK_PER>, // 2 4
<&lsio_bus_clk>, // 3 5
<&clk IMX_SC_R_PWM_0 IMX_SC_PM_CLK_PER>; // 4 6
clock-indices = <IMX_LPCG_CLK_0>, <IMX_LPCG_CLK_1>,
<IMX_LPCG_CLK_4>, <IMX_LPCG_CLK_5>,
<IMX_LPCG_CLK_6>;
};Col1: index, which existing dts try to get.
Col2: actual index in lpcg driver.
pwm1 {
....
clocks = <&pwm1_lpcg 4>, <&pwm1_lpcg 1>;
^^ ^^should be:
clocks = <&pwm1_lpcg IMX_LPCG_CLK_6>, <&pwm1_lpcg IMX_LPCG_CLK_1>; };
Arg0 is divided by 4 in lpcg driver, so index 0 and 1 will be get by pwm driver, which are same as IMX_LPCG_CLK_6 and IMX_LPCG_CLK_1. Even it can work, but code logic is wrong. Fixed it by use correct indices.
--- arch/arm64/boot/dts/freescale/imx8-ss-lsio.dtsi | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
arch/arm64/boot/dts/freescale/imx8-ss-lsio.dtsi +8 -8
@@ -25,8 +25,8 @@ lsio_subsys: bus@5d000000 { compatible = "fsl,imx27-pwm"; reg = <0x5d000000 0x10000>; clock-names = "ipg", "per"; - clocks = <&pwm0_lpcg 4>, - <&pwm0_lpcg 1>; + clocks = <&pwm0_lpcg IMX_LPCG_CLK_6>, + <&pwm0_lpcg IMX_LPCG_CLK_1>; assigned-clocks = <&clk IMX_SC_R_PWM_0 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <24000000>; #pwm-cells = <3>;
@@ -38,8 +38,8 @@ lsio_subsys: bus@5d000000 { compatible = "fsl,imx27-pwm"; reg = <0x5d010000 0x10000>; clock-names = "ipg", "per"; - clocks = <&pwm1_lpcg 4>, - <&pwm1_lpcg 1>; + clocks = <&pwm1_lpcg IMX_LPCG_CLK_6>, + <&pwm1_lpcg IMX_LPCG_CLK_1>; assigned-clocks = <&clk IMX_SC_R_PWM_1 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <24000000>; #pwm-cells = <3>;
@@ -51,8 +51,8 @@ lsio_subsys: bus@5d000000 { compatible = "fsl,imx27-pwm"; reg = <0x5d020000 0x10000>; clock-names = "ipg", "per"; - clocks = <&pwm2_lpcg 4>, - <&pwm2_lpcg 1>; + clocks = <&pwm2_lpcg IMX_LPCG_CLK_6>, + <&pwm2_lpcg IMX_LPCG_CLK_1>; assigned-clocks = <&clk IMX_SC_R_PWM_2 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <24000000>; #pwm-cells = <3>;
@@ -64,8 +64,8 @@ lsio_subsys: bus@5d000000 { compatible = "fsl,imx27-pwm"; reg = <0x5d030000 0x10000>; clock-names = "ipg", "per"; - clocks = <&pwm3_lpcg 4>, - <&pwm3_lpcg 1>; + clocks = <&pwm3_lpcg IMX_LPCG_CLK_6>, + <&pwm3_lpcg IMX_LPCG_CLK_1>; assigned-clocks = <&clk IMX_SC_R_PWM_3 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <24000000>; #pwm-cells = <3>;
Greg Kroah-Hartman
maintainerPATCH 138/172 arm64: dts: imx8-ss-dma: fix spi lpcg indices
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Frank Li <[email protected]>
commit f72b544a514c07d34a0d9d5380f5905b3731e647 upstream.
spi0_lpcg: clock-controller@5a400000 {
... Col0 Col1
clocks = <&clk IMX_SC_R_SPI_0 IMX_SC_PM_CLK_PER>,// 0 1
<&dma_ipg_clk>; // 1 4
clock-indices = <IMX_LPCG_CLK_0>, <IMX_LPCG_CLK_4>;
};Col1: index, which existing dts try to get.
Col2: actual index in lpcg driver.
lpspi0: spi@5a000000 {
...
clocks = <&spi0_lpcg 0>, <&spi0_lpcg 1>;
^ ^
Should be:
clocks = <&spi0_lpcg IMX_LPCG_CLK_0>, <&spi0_lpcg IMX_LPCG_CLK_4>;
};Arg0 is divided by 4 in lpcg driver. <&spi0_lpcg 0> and <&spi0_lpcg 1> are IMX_SC_PM_CLK_PER. Although code can work, code logic is wrong. It should use IMX_LPCG_CLK_0 and IMX_LPCG_CLK_4 for lpcg arg0.
--- arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi +8 -8
@@ -27,8 +27,8 @@ dma_subsys: bus@5a000000 { #size-cells = <0>; interrupts = <GIC_SPI 336 IRQ_TYPE_LEVEL_HIGH>; interrupt-parent = <&gic>; - clocks = <&spi0_lpcg 0>, - <&spi0_lpcg 1>; + clocks = <&spi0_lpcg IMX_LPCG_CLK_0>, + <&spi0_lpcg IMX_LPCG_CLK_4>; clock-names = "per", "ipg"; assigned-clocks = <&clk IMX_SC_R_SPI_0 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <60000000>;
@@ -43,8 +43,8 @@ dma_subsys: bus@5a000000 { #size-cells = <0>; interrupts = <GIC_SPI 337 IRQ_TYPE_LEVEL_HIGH>; interrupt-parent = <&gic>; - clocks = <&spi1_lpcg 0>, - <&spi1_lpcg 1>; + clocks = <&spi1_lpcg IMX_LPCG_CLK_0>, + <&spi1_lpcg IMX_LPCG_CLK_4>; clock-names = "per", "ipg"; assigned-clocks = <&clk IMX_SC_R_SPI_1 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <60000000>;
@@ -59,8 +59,8 @@ dma_subsys: bus@5a000000 { #size-cells = <0>; interrupts = <GIC_SPI 338 IRQ_TYPE_LEVEL_HIGH>; interrupt-parent = <&gic>; - clocks = <&spi2_lpcg 0>, - <&spi2_lpcg 1>; + clocks = <&spi2_lpcg IMX_LPCG_CLK_0>, + <&spi2_lpcg IMX_LPCG_CLK_4>; clock-names = "per", "ipg"; assigned-clocks = <&clk IMX_SC_R_SPI_2 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <60000000>;
@@ -75,8 +75,8 @@ dma_subsys: bus@5a000000 { #size-cells = <0>; interrupts = <GIC_SPI 339 IRQ_TYPE_LEVEL_HIGH>; interrupt-parent = <&gic>; - clocks = <&spi3_lpcg 0>, - <&spi3_lpcg 1>; + clocks = <&spi3_lpcg IMX_LPCG_CLK_0>, + <&spi3_lpcg IMX_LPCG_CLK_4>; clock-names = "per", "ipg"; assigned-clocks = <&clk IMX_SC_R_SPI_3 IMX_SC_PM_CLK_PER>; assigned-clock-rates = <60000000>;
Greg Kroah-Hartman
maintainerPATCH 139/172 vhost: Add smp_rmb() in vhost_vq_avail_empty()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gavin Shan <[email protected]>
commit 22e1992cf7b034db5325660e98c41ca5afa5f519 upstream.
A smp_rmb() has been missed in vhost_vq_avail_empty(), spotted by Will. Otherwise, it's not ensured the available ring entries pushed by guest can be observed by vhost in time, leading to stale available ring entries fetched by vhost in vhost_get_vq_desc(), as reported by Yihuang Yu on NVidia's grace-hopper (ARM64) platform.
/home/gavin/sandbox/qemu.main/build/qemu-system-aarch64 \ -accel kvm -machine virt,gic-version=host -cpu host \ -smp maxcpus=1,cpus=1,sockets=1,clusters=1,cores=1,threads=1 \ -m 4096M,slots=16,maxmem=64G \ -object memory-backend-ram,id=mem0,size=4096M \ : \ -netdev tap,id=vnet0,vhost=true \ -device virtio-net-pci,bus=pcie.8,netdev=vnet0,mac=52:54:00:f1:26:b0 : guest# netperf -H 10.26.1.81 -l 60 -C -c -t UDP_STREAM virtio_net virtio0: output.0:id 100 is not a head!
Add the missed smp_rmb() in vhost_vq_avail_empty(). When tx_can_batch() returns true, it means there's still pending tx buffers. Since it might read indices, so it still can bypass the smp_rmb() in vhost_get_vq_desc(). Note that it should be safe until vq->avail_idx is changed by commit 275bf960ac697 ("vhost: better detection of available buffers").
Message-Id: <[email protected]>
--- drivers/vhost/vhost.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/vhost/vhost.c +11 -1
@@ -2799,9 +2799,19 @@ bool vhost_vq_avail_empty(struct vhost_d r =vhost_get_avail_idx (vq, &avail_idx);if (unlikely (r))return false; + vq->avail_idx =vhost16_to_cpu (vq, avail_idx); +if (vq->avail_idx != vq->last_avail_idx) { +/* Since we have updated avail_idx, the following +* call to vhost_get_vq_desc() will read available +* ring entries. Make sure that read happens after +* the avail_idx read. +*/ +smp_rmb (); +return false; + } -return vq->avail_idx == vq->last_avail_idx; +return true; }EXPORT_SYMBOL_GPL (vhost_vq_avail_empty);
Greg Kroah-Hartman
maintainerPATCH 140/172 vhost: Add smp_rmb() in vhost_enable_notify()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gavin Shan <[email protected]>
commit df9ace7647d4123209395bb9967e998d5758c645 upstream.
A smp_rmb() has been missed in vhost_enable_notify(), inspired by Will. Otherwise, it's not ensured the available ring entries pushed by guest can be observed by vhost in time, leading to stale available ring entries fetched by vhost in vhost_get_vq_desc(), as reported by Yihuang Yu on NVidia's grace-hopper (ARM64) platform.
/home/gavin/sandbox/qemu.main/build/qemu-system-aarch64 \ -accel kvm -machine virt,gic-version=host -cpu host \ -smp maxcpus=1,cpus=1,sockets=1,clusters=1,cores=1,threads=1 \ -m 4096M,slots=16,maxmem=64G \ -object memory-backend-ram,id=mem0,size=4096M \ : \ -netdev tap,id=vnet0,vhost=true \ -device virtio-net-pci,bus=pcie.8,netdev=vnet0,mac=52:54:00:f1:26:b0 : guest# netperf -H 10.26.1.81 -l 60 -C -c -t UDP_STREAM virtio_net virtio0: output.0:id 100 is not a head!
Add the missed smp_rmb() in vhost_enable_notify(). When it returns true, it means there's still pending tx buffers. Since it might read indices, so it still can bypass the smp_rmb() in vhost_get_vq_desc(). Note that it should be safe until vq->avail_idx is changed by commit d3bb267bbdcb ("vhost: cache avail index in vhost_enable_notify()").
Message-Id: <[email protected]>
--- drivers/vhost/vhost.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/vhost/vhost.c +11 -1
@@ -2848,9 +2848,19 @@ bool vhost_enable_notify(struct vhost_de &vq->avail->idx, r);return false; } + vq->avail_idx =vhost16_to_cpu (vq, avail_idx); +if (vq->avail_idx != vq->last_avail_idx) { +/* Since we have updated avail_idx, the following +* call to vhost_get_vq_desc() will read available +* ring entries. Make sure that read happens after +* the avail_idx read. +*/ +smp_rmb (); +return true; + } -return vq->avail_idx != vq->last_avail_idx; +return false; }EXPORT_SYMBOL_GPL (vhost_enable_notify);
Greg Kroah-Hartman
maintainerPATCH 141/172 perf/x86: Fix out of range data
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Namhyung Kim <[email protected]>
commit dec8ced871e17eea46f097542dd074d022be4bd1 upstream.
On x86 each struct cpu_hw_events maintains a table for counter assignment but it missed to update one for the deleted event in x86_pmu_del(). This can make perf_clear_dirty_counters() reset used counter if it's called before event scheduling or enabling. Then it would return out of range data which doesn't make sense.
The following code can reproduce the problem.
$ cat repro.c #include <pthread.h> #include <stdio.h> #include <stdlib.h> #include <unistd.h> #include <linux/perf_event.h> #include <sys/ioctl.h> #include <sys/mman.h> #include <sys/syscall.h>
struct perf_event_attr attr = {
.type = PERF_TYPE_HARDWARE,
.config = PERF_COUNT_HW_CPU_CYCLES,
.disabled = 1,
};
void *worker(void *arg)
{
int cpu = (long)arg;
int fd1 = syscall(SYS_perf_event_open, &attr, -1, cpu, -1, 0);
int fd2 = syscall(SYS_perf_event_open, &attr, -1, cpu, -1, 0);
void *p;
do {
ioctl(fd1, PERF_EVENT_IOC_ENABLE, 0);
p = mmap(NULL, 4096, PROT_READ, MAP_SHARED, fd1, 0);
ioctl(fd2, PERF_EVENT_IOC_ENABLE, 0);
ioctl(fd2, PERF_EVENT_IOC_DISABLE, 0);
munmap(p, 4096);
ioctl(fd1, PERF_EVENT_IOC_DISABLE, 0);
} while (1);
return NULL;
}
int main(void)
{
int i;
int n = sysconf(_SC_NPROCESSORS_ONLN);
pthread_t *th = calloc(n, sizeof(*th));
for (i = 0; i < n; i++)
pthread_create(&th[i], NULL, worker, (void *)(long)i);
for (i = 0; i < n; i++)
pthread_join(th[i], NULL);
free(th);
return 0;
}And you can see the out of range data using perf stat like this. Probably it'd be easier to see on a large machine.
$ gcc -o repro repro.c -pthread
$ ./repro &
$ sudo perf stat -A -I 1000 2>&1 | awk '{ if (length($3) > 15) print }'
1.001028462 CPU6 196,719,295,683,763 cycles # 194290.996 GHz (71.54%)
1.001028462 CPU3 396,077,485,787,730 branch-misses # 15804359784.80% of all branches (71.07%)
1.001028462 CPU17 197,608,350,727,877 branch-misses # 14594186554.56% of all branches (71.22%)
2.020064073 CPU4 198,372,472,612,140 cycles # 194681.113 GHz (70.95%)
2.020064073 CPU6 199,419,277,896,696 cycles # 195720.007 GHz (70.57%)
2.020064073 CPU20 198,147,174,025,639 cycles # 194474.654 GHz (71.03%)
2.020064073 CPU20 198,421,240,580,145 stalled-cycles-frontend # 100.14% frontend cycles idle (70.93%)
3.037443155 CPU4 197,382,689,923,416 cycles # 194043.065 GHz (71.30%)
3.037443155 CPU20 196,324,797,879,414 cycles # 193003.773 GHz (71.69%)
3.037443155 CPU5 197,679,956,608,205 stalled-cycles-backend # 1315606428.66% backend cycles idle (71.19%)
3.037443155 CPU5 198,571,860,474,851 instructions # 13215422.58 insn per cycleIt should move the contents in the cpuc->assign as well.
--- arch/x86/events/core.c | 1 + 1 file changed, 1 insertion(+)
arch/x86/events/core.c +1 -0
@@ -1644,6 +1644,7 @@ static void x86_pmu_del(struct perf_evenwhile (++i < cpuc->n_events) { cpuc->event_list[i-1 ] = cpuc->event_list[i]; cpuc->event_constraint[i-1 ] = cpuc->event_constraint[i]; + cpuc->assign[i-1 ] = cpuc->assign[i]; } cpuc->event_constraint[i-1 ] =NULL ; --cpuc->n_events;
Greg Kroah-Hartman
maintainerPATCH 142/172 x86/cpu: Actually turn off mitigations by default for SPECULATION_MITIGATIONS=n
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sean Christopherson <[email protected]>
commit f337a6a21e2fd67eadea471e93d05dd37baaa9be upstream.
Initialize cpu_mitigations to CPU_MITIGATIONS_OFF if the kernel is built with CONFIG_SPECULATION_MITIGATIONS=n, as the help text quite clearly states that disabling SPECULATION_MITIGATIONS is supposed to turn off all mitigations by default.
│ If you say N, all mitigations will be disabled. You really │ should know what you are doing to say so.
As is, the kernel still defaults to CPU_MITIGATIONS_AUTO, which results in some mitigations being enabled in spite of SPECULATION_MITIGATIONS=n.
--- kernel/cpu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
kernel/cpu.c +2 -1
@@ -3207,7 +3207,8 @@ enum cpu_mitigations { };static enum cpu_mitigations cpu_mitigations__ro_after_init = -CPU_MITIGATIONS_AUTO ; +IS_ENABLED (CONFIG_SPECULATION_MITIGATIONS ) ?CPU_MITIGATIONS_AUTO : +CPU_MITIGATIONS_OFF ;static int __init mitigations_parse_cmdline (char *arg) {
Greg Kroah-Hartman
maintainerPATCH 143/172 selftests/timers/posix_timers: Reimplement check_timer_distribution()
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Oleg Nesterov <[email protected]>
commit 6d029c25b71f2de2838a6f093ce0fa0e69336154 upstream.
check_timer_distribution() runs ten threads in a busy loop and tries to test that the kernel distributes a process posix CPU timer signal to every thread over time.
There is not guarantee that this is true even after commit bcb7ee79029d ("posix-timers: Prefer delivery of signals to the current thread") because that commit only avoids waking up the sleeping process leader thread, but that has nothing to do with the actual signal delivery.
As the signal is process wide the first thread which observes sigpending and wins the race to lock sighand will deliver the signal. Testing shows that this hangs on a regular base because some threads never win the race.
The comment "This primarily tests that the kernel does not favour any one." is wrong. The kernel does favour a thread which hits the timer interrupt when CLOCK_PROCESS_CPUTIME_ID expires.
Rewrite the test so it only checks that the group leader sleeping in join() never receives SIGALRM and the thread which burns CPU cycles receives all signals.
In older kernels which do not have commit bcb7ee79029d ("posix-timers:
Prefer delivery of signals to the current thread") the test-case fails immediately, the very 1st tick wakes the leader up. Otherwise it quickly succeeds after 100 ticks.
CI testing wants to use newer selftest versions on stable kernels. In this case the test is guaranteed to fail.
So check in the failure case whether the kernel version is less than v6.3 and skip the test result in that case.
[ tglx: Massaged change log, renamed the version check helper ]
--- tools/testing/selftests/kselftest.h | 13 +++ tools/testing/selftests/timers/posix_timers.c | 111 +++++++++++--------------- 2 files changed, 64 insertions(+), 60 deletions(-)
tools/testing/selftests/kselftest.h +13 -0
@@ -50,6 +50,7 @@#include <stdarg.h> #include <string.h> #include <stdio.h> +#include <sys/utsname.h> #endif #ifndef ARRAY_SIZE
@@ -343,4 +344,16 @@ static inline __printf(1, 2) int ksft_ex exit(KSFT_SKIP); } +static inline int ksft_min_kernel_version (unsigned int min_major, +unsigned int min_minor) +{ +unsigned int major, minor; +struct utsname info; + +if (uname (&info) ||sscanf (info.release,"%u.%u." , &major, &minor) !=2 ) +ksft_exit_fail_msg ("Can't parse kernel version\n" ); + +return major > min_major || (major == min_major && minor >= min_minor); +} +#endif /* __KSELFTEST_H */
tools/testing/selftests/timers/posix_timers.c +51 -60
131 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 144/172 selftests: timers: Fix posix_timers ksft_print_msg() warning
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: John Stultz <[email protected]>
commit e4a6bceac98eba3c00e874892736b34ea5fdaca3 upstream.
After commit 6d029c25b71f ("selftests/timers/posix_timers: Reimplement check_timer_distribution()") the following warning occurs when building with an older gcc:
posix_timers.c:250:2: warning: format not a string literal and no format arguments [-Wformat-security]
250 | ksft_print_msg(errmsg);
| ^~~~~~~~~~~~~~Fix this up by changing it to ksft_print_msg("%s", errmsg)
--- tools/testing/selftests/timers/posix_timers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/timers/posix_timers.c +1 -1
@@ -247,7 +247,7 @@ static int check_timer_distribution(void)ksft_test_result_skip ("check signal distribution (old kernel)\n" );return 0 ; err: -ksft_print_msg (errmsg); +ksft_print_msg ("%s" , errmsg);return -1 ; }
Signature · 1 line
2.44.0
Greg Kroah-Hartman
maintainerPATCH 145/172 selftests: timers: Fix abs() warning in posix_timers test
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: John Stultz <[email protected]>
commit ed366de8ec89d4f960d66c85fc37d9de22f7bf6d upstream.
Building with clang results in the following warning:
posix_timers.c:69:6: warning: absolute value function 'abs' given an
argument of type 'long long' but has parameter of type 'int' which may cause truncation of value [-Wabsolute-value]
if (abs(diff - DELAY * USECS_PER_SEC) > USECS_PER_SEC / 2) {
^
So switch to using llabs() instead.
--- tools/testing/selftests/timers/posix_timers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/timers/posix_timers.c +1 -1
@@ -66,7 +66,7 @@ static int check_diff(struct timeval sta diff = end.tv_usec - start.tv_usec; diff += (end.tv_sec - start.tv_sec) *USECS_PER_SEC ; -if (abs (diff -DELAY *USECS_PER_SEC ) >USECS_PER_SEC /2 ) { +if (llabs (diff -DELAY *USECS_PER_SEC ) >USECS_PER_SEC /2 ) {printf ("Diff too high: %lld.." , diff);return -1 ; }
Greg Kroah-Hartman
maintainerPATCH 146/172 selftests: kselftest: Mark functions that unconditionally call exit() as __noreturn
In reply to Greg Kroah-Hartman6.8-stable review patch. If anyone has any objections, please let me know.
------------------
From: Nathan Chancellor <[email protected]>
commit f7d5bcd35d427daac7e206b1073ca14f5db85c27 upstream.
After commit 6d029c25b71f ("selftests/timers/posix_timers: Reimplement check_timer_distribution()"), clang warns:
tools/testing/selftests/timers/../kselftest.h:398:6: warning: variable 'major' is used uninitialized whenever '||' condition is true [-Wsometimes-uninitialized]
398 | if (uname(&info) || sscanf(info.release, "%u.%u.", &major, &minor) != 2)
| ^~~~~~~~~~~~
tools/testing/selftests/timers/../kselftest.h:401:9: note: uninitialized use occurs here
401 | return major > min_major || (major == min_major && minor >= min_minor);
| ^~~~~
tools/testing/selftests/timers/../kselftest.h:398:6: note: remove the '||' if its condition is always false
398 | if (uname(&info) || sscanf(info.release, "%u.%u.", &major, &minor) != 2)
| ^~~~~~~~~~~~~~~
tools/testing/selftests/timers/../kselftest.h:395:20: note: initialize the variable 'major' to silence this warning
395 | unsigned int major, minor;
| ^
| = 0This is a false positive because if uname() fails, ksft_exit_fail_msg() will be called, which unconditionally calls exit(), a noreturn function. However, clang does not know that ksft_exit_fail_msg() will call exit() at the point in the pipeline that the warning is emitted because inlining has not occurred, so it assumes control flow will resume normally after ksft_exit_fail_msg() is called.
Make it clear to clang that all of the functions that call exit() unconditionally in kselftest.h are noreturn transitively by marking them explicitly with '__attribute__((__noreturn__))', which clears up the warning above and any future warnings that may appear for the same reason.
--- tools/testing/selftests/kselftest.h | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-)
tools/testing/selftests/kselftest.h +9 -6
@@ -79,6 +79,9 @@#define KSFT_XPASS 3#define KSFT_SKIP 4 +#ifndef __noreturn +#define __noreturn __attribute__((__noreturn__)) +#endif #define __printf (a, b) __attribute__((format(printf, a, b)))/* counters */
@@ -255,13 +258,13 @@ static inline __printf(1, 2) void ksft_tva_end (args); } -static inline int ksft_exit_pass (void ) +static inline __noreturn int ksft_exit_pass (void ) {ksft_print_cnts ();exit (KSFT_PASS ); } -static inline int ksft_exit_fail (void ) +static inline __noreturn int ksft_exit_fail (void ) {ksft_print_cnts ();exit (KSFT_FAIL );
@@ -288,7 +291,7 @@ static inline int ksft_exit_fail(void) ksft_cnt.ksft_xfail + \ ksft_cnt.ksft_xskip) -static inline __printf(1, 2) int ksft_exit_fail_msg(const char *msg, ...) +static inline __noreturn __printf(1, 2) int ksft_exit_fail_msg(const char *msg, ...) {int saved_errno = errno;va_list args;
@@ -303,19 +306,19 @@ static inline __printf(1, 2) int ksft_exexit (KSFT_FAIL ); } -static inline int ksft_exit_xfail (void ) +static inline __noreturn int ksft_exit_xfail (void ) {ksft_print_cnts ();exit (KSFT_XFAIL ); } -static inline int ksft_exit_xpass (void ) +static inline __noreturn int ksft_exit_xpass (void ) {ksft_print_cnts ();exit (KSFT_XPASS ); } -static inline __printf(1, 2) int ksft_exit_skip(const char *msg, ...) +static inline __noreturn __printf(1, 2) int ksft_exit_skip(const char *msg, ...) { int saved_errno = errno;va_list args;