6.12.95-rc1 review
Subsystems (5)
Greg Kroah-Hartman
maintainercover 6.12.95-rc1 review
This is the start of the stable review cycle for the 6.12.95 release. There are 204 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Sat, 04 Jul 2026 15:50:58 +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-6.12.95-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-6.12.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 6.12.95-rc1
Hangbin Liu <[email protected]>
bonding: do not set usable_slaves for broadcast mode
Eric Dumazet <[email protected]>
bonding: annotate data-races arcound churn variables
Tonghao Zhang <[email protected]>
net: bonding: update the slave array for broadcast mode
John Stultz <[email protected]>
locking: rtmutex: Fix wake_q logic in task_blocks_on_rt_mutex
HanQuan <[email protected]>
net/tcp-ao: fix use-after-free of key in del_async path
Giovanni Cabiddu <[email protected]>
crypto: qat - remove unused character device and IOCTLs
Herbert Xu <[email protected]>
crypto: qat - Return pointer directly in adf_ctl_alloc_resources
Thorsten Blum <[email protected]>
crypto: qat - Replace kzalloc() + copy_from_user() with memdup_user()
Bagas Sanjaya <[email protected]>
Documentation: ioctl-number: Extend "Include File" column width
Bagas Sanjaya <[email protected]>
Documentation: ioctl-number: Fix linuxppc-dev mailto link
Georgi Djakov <[email protected]>
drivers/base/memory: set mem->altmap after successful device registration
Stepan Ionichev <[email protected]>
serial: 8250_dw: unregister 8250 port if clk_notifier_register() fails
Hem Parekh <[email protected]>
ksmbd: fix out-of-bounds read in smb_check_perm_dacl()
Markus Elfring <[email protected]>
NFS: Prevent resource leak in nfs_alloc_server()
Michael Bommarito <[email protected]>
NFSv4/pNFS: reject zero-length r_addr in nfs4_decode_mp_ds_addr
Jeff Layton <[email protected]>
nfsd: reset write verifier on deferred writeback errors
Jeff Layton <[email protected]>
nfsd: avoid leaking pre-allocated openowner on unconfirmed retry race
Dominik Woźniak <[email protected]>
nfsd: check get_user() return when reading princhashlen
Jeff Layton <[email protected]>
nfsd: fix posix_acl leak on SETACL decode failure
Guannan Wang <[email protected]>
NFSD: Fix SECINFO_NO_NAME decode error cleanup
Johan Hovold <[email protected]>
i2c: core: fix adapter registration race
Steffen Persvold <[email protected]>
fbdev: modedb: Fix misaligned fields in the 1920x1080-60 mode
Tuo Li <[email protected]>
fbdev: modedb: fix a possible UAF in fb_find_mode()
Ian Bridges <[email protected]>
fbdev: Fix fb_new_modelist to prevent null-ptr-deref in fb_videomode_to_var
Vivian Wang <[email protected]>
riscv: kfence: Call mark_new_valid_map() for kfence_unprotect()
Wentao Liang <[email protected]>
power: reset: linkstation-poweroff: fix use-after-free in the linkstation_poweroff_init()
Ashutosh Desai <[email protected]>
KVM: SVM: Fix page overflow in sev_dbg_crypt() for ENCRYPT path
Hyunwoo Kim <[email protected]>
KVM: x86: hyper-v: Bound the bank index when querying sparse banks
Jonas Jelonek <[email protected]>
MIPS: smp: report dying CPU to RCU in stop_this_cpu()
Yizhou Zhao <[email protected]>
9p: avoid putting oldfid in p9_client_walk() error path
Zhang Cen <[email protected]>
ocfs2: reject oversized group bitmap descriptors
Yuho Choi <[email protected]>
rpmsg: char: Fix use-after-free on probe error path
Wentao Liang <[email protected]>
fpga: region: fix use-after-free in child_regions_with_firmware()
Qingshuang Fu <[email protected]>
irqchip/imgpdc: Fix resource leak, add missing chained handler cleanup on remove
Wentao Liang <[email protected]>
pNFS: Fix use-after-free in pnfs_update_layout()
Huacai Chen <[email protected]>
LoongArch: Report dying CPU to RCU in stop_this_cpu()
Doruk Tan Ozturk <[email protected]>
tipc: fix slab-use-after-free Read in tipc_aead_decrypt_done
Michal Koutný <[email protected]>
blk-cgroup: fix UAF in __blkcg_rstat_flush()
Fan Wu <[email protected]>
hdlc_ppp: sync per-proto timers before freeing hdlc state
Wentao Liang <[email protected]>
pwrseq: core: fix use-after-free in pwrseq_debugfs_seq_next()
Tristan Madani <[email protected]>
gfs2: fix use-after-free in gfs2_qd_dealloc
Sean Christopherson <[email protected]>
KVM: Replace guest-triggerable BUG_ON() in ioeventfd datamatch with get_unaligned()
Michael Bommarito <[email protected]>
exfat: fix potential use-after-free in exfat_find_dir_entry()
Maciej W. Rozycki <[email protected]>
MIPS: DEC: Prevent initial console buffer from landing in XKPHYS
Dawei Feng <[email protected]>
bpf: use kvfree() for replaced sysctl write buffer
Denis Arefev <[email protected]>
block: Avoid mounting the bdev pseudo-filesystem in userspace
Wenjie Qi <[email protected]>
f2fs: keep atomic write retry from zeroing original data
Yongpeng Yang <[email protected]>
f2fs: fix incorrect FI_NO_EXTENT handling in __destroy_extent_node()
Zhang Cen <[email protected]>
f2fs: validate ACL entry sizes in f2fs_acl_from_disk()
Sunmin Jeong <[email protected]>
f2fs: fix to round down start offset of fallocate for pin file
Wenjie Qi <[email protected]>
f2fs: validate compress cache inode only when enabled
Junjie Cao <[email protected]>
wifi: iwlwifi: mvm: fix race condition in PTP removal
Luka Gejak <[email protected]>
wifi: rtw88: usb: fix memory leaks on USB write failures
Luka Gejak <[email protected]>
wifi: rtw88: increase TX report timeout to fix race condition
Bitterblue Smith <[email protected]>
wifi: rtlwifi: rtl8821ae: Fix C2H bit location in RX descriptor
Jose Ignacio Tornos Martinez <[email protected]>
wifi: ath11k: fix warning when unbinding
ElXreno <[email protected]>
wifi: mt76: mt7925: don't disable AP BSS when removing TDLS peer
Zenm Chen <[email protected]>
wifi: mt76: mt76x2u: Add support for ELECOM WDC-867SU3S
Shaomin Chen <[email protected]>
keys: Pin request_key_auth payload in instantiate paths
Jarkko Sakkinen <[email protected]>
KEYS: fix overflow in keyctl_pkey_params_get_2()
Arnd Bergmann <[email protected]>
err.h: use __always_inline on all error pointer helpers
Usama Arif <[email protected]>
block: invalidate cached plug timestamp after task switch
Ian Bridges <[email protected]>
fbdev: fix use-after-free in store_modes()
Koichiro Den <[email protected]>
NTB: epf: Avoid pci_iounmap() with offset when PEER_SPAD and CONFIG share BAR
Ruslan Valiyev <[email protected]>
apparmor: fix use-after-free in rawdata dedup loop
Bryam Vargas <[email protected]>
apparmor: mediate the implicit connect of TCP fast open sendmsg
Maoyi Xie <[email protected]>
net: ip_gre: require CAP_NET_ADMIN in the device netns for changelink
Yiming Qian <[email protected]>
net: skmsg: preserve sg.copy across SG transforms
Doruk Tan Ozturk <[email protected]>
mac802154: llsec: add skb_cow_data() before in-place crypto
Cheng Ming Lin <[email protected]>
mtd: spi-nor: macronix: add support for mx66{l2, u1}g45g
Cheng Ming Lin <[email protected]>
mtd: spi-nor: macronix: Add post_sfdp fixups for Quad Input Page Program
Kuniyuki Iwashima <[email protected]>
af_unix: Set gc_in_progress to true in unix_gc().
Sean Christopherson <[email protected]>
KVM: SEV: Unmap and unpin the GHCB as needed on vCPU free
Sean Christopherson <[email protected]>
KVM: SEV: Move sev_free_vcpu() down below sev_es_unmap_ghcb()
Konstantin Komarov <[email protected]>
ntfs3: reject direct userspace writes to reserved $LX* xattrs
Wongi Lee <[email protected]>
ipv4: account for fraggap on the paged allocation path
Eric Dumazet <[email protected]>
inet: add indirect call wrapper for getfrag() calls
Paul Moore <[email protected]>
selinux: fix overlayfs mmap() and mprotect() access checks
Paul Moore <[email protected]>
lsm: add backing_file LSM hooks
Amir Goldstein <[email protected]>
fs: constify file ptr in backing_file accessor helpers
Wongi Lee <[email protected]>
ipv6: account for fraggap on the paged allocation path
Sven Eckelmann <[email protected]>
batman-adv: tvlv: avoid race of cifsnotfound handler state
Sven Eckelmann <[email protected]>
batman-adv: tvlv: enforce 2-byte alignment
Sven Eckelmann <[email protected]>
batman-adv: dat: prevent false sharing between VLANs
Sven Eckelmann <[email protected]>
batman-adv: tt: track roam count per VID
Sven Eckelmann <[email protected]>
batman-adv: tt: don't merge change entries with different VIDs
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: handle overlapping packets
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: prevent parallel modifications of last_recv
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: annotate last_recv_time access with READ/WRITE_ONCE
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: restrict number of unacked list entries
Sven Eckelmann <[email protected]>
batman-adv: v: prevent OGM aggregation on disabled hardif
Sven Eckelmann <[email protected]>
batman-adv: frag: avoid underflow of TTL
Sven Eckelmann <[email protected]>
batman-adv: frag: ensure fragment is writable before modifying TTL
Sven Eckelmann <[email protected]>
batman-adv: fix (m|b)cast csum after decrementing TTL
Sven Eckelmann <[email protected]>
batman-adv: ensure bcast is writable before modifying TTL
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: initialize last_recv_time during init
Sven Eckelmann <[email protected]>
batman-adv: prevent ELP transmission interval underflow
Sven Eckelmann <[email protected]>
batman-adv: bla: annotate lasttime access with READ/WRITE_ONCE
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: add only finished tp_vars to lists
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: handle seqno wrap-around for fast recovery detection
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: fix fast recovery precondition
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: avoid divide-by-zero for dec_cwnd
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: avoid window underflow
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: initialize dec_cwnd explicitly
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: initialize dup_acks explicitly
Sven Eckelmann <[email protected]>
batman-adv: tp_meter: keep unacked list in ascending ordered
Sean Christopherson <[email protected]>
KVM: SEV: Ignore Port I/O requests of length '0'
Sean Christopherson <[email protected]>
KVM: SEV: Reject MMIO requests larger than 8 bytes with GHCB v2+
Sean Christopherson <[email protected]>
KVM: SEV: Ignore MMIO requests of length '0'
Sasha Levin <[email protected]>
Revert "PCI: qcom: Advertise Hotplug Slot Capability with no Command Completion support"
Sean Christopherson <[email protected]>
KVM: x86/mmu: Ensure hugepage is in by slot before checking max mapping level
Paolo Bonzini <[email protected]>
KVM: x86: Fix shadow paging use-after-free due to unexpected role
Miklos Szeredi <[email protected]>
virtiofs: fix UAF on submount umount
Ruslan Valiyev <[email protected]>
media: vidtv: fix NULL pointer dereference in vidtv_mux_push_si
Gil Portnoy <[email protected]>
ksmbd: reject non-VALID session in compound request branch
Viken Dadhaniya <[email protected]>
serial: qcom_geni: Fix RX DMA stall when SE_DMA_RX_LEN_IN is zero
Yi Yang <[email protected]>
vc_screen: fix null-ptr-deref in vcs_notifier() during concurrent vcs_write
Sam Daly <[email protected]>
iio: adc: ti-ads1298: add bounds check to pga_settings index
Sam Daly <[email protected]>
iio: light: veml6075: add bounds check to veml6075_it_ms index
Xin Long <[email protected]>
sctp: disable BH before calling udp_tunnel_xmit_skb()
Petr Machata <[email protected]>
net: ipv6: Make udp_tunnel6_xmit_skb() void
Vasily Gorbik <[email protected]>
scripts/sorttable: Fix endianness handling in build-time mcount sort
Steven Rostedt <[email protected]>
scripts/sorttable: Allow matches to functions before function entry
Steven Rostedt <[email protected]>
scripts/sorttable: Use normal sort if theres no relocs in the mcount section
Steven Rostedt <[email protected]>
ftrace: Check against is_kernel_text() instead of kaslr_offset()
Steven Rostedt <[email protected]>
ftrace: Test mcount_loc addr before calling ftrace_call_addr()
Guenter Roeck <[email protected]>
ftrace: Do not over-allocate ftrace memory
Steven Rostedt <[email protected]>
ftrace: Have ftrace pages output reflect freed pages
Steven Rostedt <[email protected]>
ftrace: Update the mcount_loc check of skipped entries
Steven Rostedt <[email protected]>
scripts/sorttable: Zero out weak functions in mcount_loc table
Steven Rostedt <[email protected]>
scripts/sorttable: Always use an array for the mcount_loc sorting
Steven Rostedt <[email protected]>
scripts/sorttable: Have mcount rela sort use direct values
Steven Rostedt <[email protected]>
arm64: scripts/sorttable: Implement sorting mcount_loc at boot for arm64
Steven Rostedt <[email protected]>
scripts/sorttable: Use a structure of function pointers for elf helpers
Steven Rostedt <[email protected]>
scripts/sorttable: Get start/stop_mcount_loc from ELF file directly
Steven Rostedt <[email protected]>
scripts/sorttable: Move code from sorttable.h into sorttable.c
Steven Rostedt <[email protected]>
scripts/sorttable: Use uint64_t for mcount sorting
Steven Rostedt <[email protected]>
scripts/sorttable: Add helper functions for Elf_Sym
Steven Rostedt <[email protected]>
scripts/sorttable: Add helper functions for Elf_Shdr
Steven Rostedt <[email protected]>
scripts/sorttable: Add helper functions for Elf_Ehdr
Steven Rostedt <[email protected]>
scripts/sorttable: Convert Elf_Sym MACRO over to a union
Steven Rostedt <[email protected]>
scripts/sorttable: Replace Elf_Shdr Macro with a union
Steven Rostedt <[email protected]>
scripts/sorttable: Convert Elf_Ehdr to union
Steven Rostedt <[email protected]>
scripts/sorttable: Make compare_extable() into two functions
Steven Rostedt <[email protected]>
scripts/sorttable: Have the ORC code use the _r() functions to read
Steven Rostedt <[email protected]>
scripts/sorttable: Remove unneeded Elf_Rel
Steven Rostedt <[email protected]>
scripts/sorttable: Remove unused write functions
Steven Rostedt <[email protected]>
scripts/sorttable: Remove unused macro defines
Joanne Koong <[email protected]>
fuse: re-lock request before replacing page cache folio
David Howells <[email protected]>
rxrpc: Fix the ACK parser to extract the SACK table for parsing
Santosh Kalluri <[email protected]>
net: phonet: free phonet_device after RCU grace period
Kuniyuki Iwashima <[email protected]>
phonet: Pass net and ifindex to phonet_address_notify().
Kuniyuki Iwashima <[email protected]>
phonet: Pass ifindex to fill_addr().
Davidlohr Bueso <[email protected]>
locking/rtmutex: Skip remove_waiter() when waiter is not enqueued
Peter Zijlstra <[email protected]>
locking/mutex: Remove wakeups from under mutex::wait_lock
Ji'an Zhou <[email protected]>
futex/requeue: Prevent NULL pointer dereference in remove_waiter() on self-deadlock
Thorsten Blum <[email protected]>
hv: utils: handle and propagate errors in kvp_register
Dexuan Cui <[email protected]>
Drivers: hv: vmbus: Improve the logic of reserving fb_mmio on Gen2 VMs
Jann Horn <[email protected]>
fhandle: fix UAF due to unlocked ->mnt_ns read in may_decode_fh()
Salman Alghamdi <[email protected]>
staging: rtl8723bs: fix buffer over-read in rtw_update_protection
Hangbin Liu <[email protected]>
bonding: fix NULL pointer dereference in actor_port_prio setting
Xiang Mei <[email protected]>
net: bonding: fix use-after-free in bond_xmit_broadcast()
Eric Dumazet <[email protected]>
bonding: 3ad: implement proper RCU rules for port->aggregator
Hangbin Liu <[email protected]>
bonding: print churn state via netlink
Hangbin Liu <[email protected]>
bonding: add support for per-port LACP actor priority
Tonghao Zhang <[email protected]>
net: bonding: add broadcast_neighbor option for 802.3ad
Kevin Berry <[email protected]>
Revert "net: bonding: fix use-after-free in bond_xmit_broadcast()"
Yingjie Gao <[email protected]>
xfs: fix error returns in CoW fork repair
Christoph Hellwig <[email protected]>
xfs: remove the expr argument to XFS_TEST_ERROR
Thadeu Lima de Souza Cascardo <[email protected]>
dlm: prevent NPD when writing a positive value to event_done
André Draszik <[email protected]>
regulator: core: fix locking in regulator_resolve_supply() error path
Yicong Yang <[email protected]>
ACPI: scan: Use async schedule function in acpi_scan_clear_dep_fn()
Varun R Mallya <[email protected]>
selftests/bpf: Add test to ensure kprobe_multi is not sleepable
Varun R Mallya <[email protected]>
bpf: Reject sleepable kprobe_multi programs at attach time
Mingyu Wang <[email protected]>
agp/amd64: Fix broken error propagation in agp_amd64_probe()
Weiming Shi <[email protected]>
net: qualcomm: rmnet: fix endpoint use-after-free in rmnet_dellink()
Weiming Shi <[email protected]>
i2c: stub: Reject I2C block transfers with invalid length
Lord Ulf Henrik Holmberg <[email protected]>
RDMA/bnxt_re: zero shared page before exposing to userspace
Waiman Long <[email protected]>
debugobjects: Dont call fill_pool() in early boot hardirq context
Helen Koike <[email protected]>
debugobjects: Do not fill_pool() if pi_blocked_on
Sebastian Andrzej Siewior <[email protected]>
debugobjects: Use LD_WAIT_CONFIG instead of LD_WAIT_SLEEP
Sebastian Andrzej Siewior <[email protected]>
debugobjects: Allow to refill the pool before SYSTEM_SCHEDULING
Petr Machata <[email protected]>
Reapply "selftest/ptp: update ptp selftest to exercise the gettimex options"
Eric Dumazet <[email protected]>
ip6_vti: set netns_immutable on the fallback device.
Sebastian Andrzej Siewior <[email protected]>
net: Drop the lock in skb_may_tx_timestamp()
Antoniu Miclaus <[email protected]>
iio: light: bh1780: fix PM runtime leak on error path
Christian Brauner <[email protected]>
eventpoll: fix ep_remove struct eventpoll / struct file UAF
Christian Brauner <[email protected]>
eventpoll: move epi_fget() up
Christian Brauner <[email protected]>
eventpoll: rename ep_remove_safe() back to ep_remove()
Christian Brauner <[email protected]>
eventpoll: drop vestigial __ prefix from ep_remove_{file,epi}()
Christian Brauner <[email protected]>
eventpoll: kill __ep_remove()
Christian Brauner <[email protected]>
eventpoll: split __ep_remove()
Christian Brauner <[email protected]>
eventpoll: use hlist_is_singular_node() in __ep_remove()
Maíra Canal <[email protected]>
drm/v3d: Skip CSD when it has zeroed workgroups
Maíra Canal <[email protected]>
drm/v3d: Store the active job inside the queue's state
Jani Nikula <[email protected]>
drm/xe/display: fix oops in suspend/shutdown without display
Gabriel Krisman Bertazi <[email protected]>
io_uring/net: Avoid msghdr on op_connect/op_bind async data
Tzung-Bi Shih <[email protected]>
gpio: Fix resource leaks on errors in gpiochip_add_data_with_key()
Andy Shevchenko <[email protected]>
gpiolib: Remove redundant assignment of return variable
Andy Shevchenko <[email protected]>
gpiolib: Extract gpiochip_choose_fwnode() for wider use
Jann Horn <[email protected]>
fuse: limit FUSE_NOTIFY_RETRIEVE to uptodate folios
Sean Wang <[email protected]>
wifi: mt76: mt7921: fix potential deadlock in mt7921_roc_abort_sync
Quan Zhou <[email protected]>
wifi: mt76: mt7921: fix a potential scan no APs
Leon Yen <[email protected]>
wifi: mt76: mt7921: avoid undesired changes of the preset regulatory domain
-------------
Diffstat:
Documentation/networking/bonding.rst | 15 + Documentation/userspace-api/ioctl/ioctl-number.rst | 485 +++++---- Makefile | 4 +- arch/arm64/Kconfig | 1 + arch/loongarch/kernel/smp.c | 1 + arch/mips/dec/prom/console.c | 7 +- arch/mips/kernel/smp.c | 2 + arch/riscv/include/asm/kfence.h | 7 +- arch/riscv/kernel/entry.S | 6 +- arch/x86/kvm/hyperv.c | 5 + arch/x86/kvm/mmu/mmu.c | 28 +- arch/x86/kvm/svm/sev.c | 113 +- block/bdev.c | 5 - block/blk-cgroup.c | 21 +- drivers/acpi/scan.c | 41 +- drivers/base/memory.c | 3 +- drivers/char/agp/amd64-agp.c | 2 +- drivers/crypto/intel/qat/qat_common/adf_cfg.c | 10 - drivers/crypto/intel/qat/qat_common/adf_cfg.h | 1 - .../crypto/intel/qat/qat_common/adf_cfg_common.h | 32 - drivers/crypto/intel/qat/qat_common/adf_cfg_user.h | 38 - .../crypto/intel/qat/qat_common/adf_common_drv.h | 3 - drivers/crypto/intel/qat/qat_common/adf_ctl_drv.c | 416 +------- drivers/crypto/intel/qat/qat_common/adf_dev_mgr.c | 70 -- drivers/fpga/of-fpga-region.c | 3 +- drivers/gpio/gpiolib.c | 156 +-- drivers/gpu/drm/v3d/v3d_drv.h | 7 +- drivers/gpu/drm/v3d/v3d_gem.c | 7 +- drivers/gpu/drm/v3d/v3d_irq.c | 62 +- drivers/gpu/drm/v3d/v3d_sched.c | 42 +- drivers/gpu/drm/xe/display/xe_display.c | 11 +- drivers/hv/hv_kvp.c | 25 +- drivers/hv/vmbus_drv.c | 29 +- drivers/i2c/i2c-core-base.c | 8 +- drivers/i2c/i2c-stub.c | 5 + drivers/iio/adc/ti-ads1298.c | 7 +- drivers/iio/light/bh1780.c | 4 +- drivers/iio/light/veml6075.c | 8 +- drivers/infiniband/hw/bnxt_re/ib_verbs.c | 2 +- drivers/irqchip/irq-imgpdc.c | 6 + drivers/media/test-drivers/vidtv/vidtv_mux.c | 8 +- drivers/mtd/spi-nor/macronix.c | 31 + drivers/net/bonding/bond_3ad.c | 131 ++- drivers/net/bonding/bond_main.c | 86 +- drivers/net/bonding/bond_netlink.c | 37 +- drivers/net/bonding/bond_options.c | 71 ++ drivers/net/bonding/bond_procfs.c | 11 +- drivers/net/bonding/bond_sysfs_slave.c | 17 +- drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c | 8 +- drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h | 1 + drivers/net/wan/hdlc_ppp.c | 15 +- drivers/net/wireless/ath/ath11k/dp.c | 1 + drivers/net/wireless/intel/iwlwifi/mvm/ptp.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/usb.c | 1 + drivers/net/wireless/mediatek/mt76/mt7921/init.c | 7 + drivers/net/wireless/mediatek/mt76/mt7921/main.c | 14 +- drivers/net/wireless/mediatek/mt76/mt7925/main.c | 3 + .../net/wireless/realtek/rtlwifi/rtl8821ae/trx.h | 2 +- drivers/net/wireless/realtek/rtw88/tx.c | 7 +- drivers/net/wireless/realtek/rtw88/usb.c | 13 +- drivers/ntb/hw/epf/ntb_hw_epf.c | 3 +- drivers/pci/controller/dwc/pcie-qcom.c | 17 +- drivers/power/reset/linkstation-poweroff.c | 2 +- drivers/power/sequencing/core.c | 14 +- drivers/regulator/core.c | 10 +- drivers/rpmsg/rpmsg_char.c | 15 +- drivers/staging/rtl8723bs/core/rtw_mlme.c | 10 +- drivers/tty/serial/8250/8250_dw.c | 4 +- drivers/tty/serial/qcom_geni_serial.c | 9 +- drivers/tty/vt/vc_screen.c | 2 +- drivers/video/fbdev/core/fbmem.c | 12 + drivers/video/fbdev/core/fbsysfs.c | 10 +- drivers/video/fbdev/core/modedb.c | 5 +- fs/backing-file.c | 22 +- fs/dlm/lockspace.c | 2 +- fs/eventpoll.c | 142 +-- fs/exfat/dir.c | 4 +- fs/f2fs/acl.c | 18 +- fs/f2fs/data.c | 16 +- fs/f2fs/extent_cache.c | 19 +- fs/f2fs/file.c | 9 +- fs/f2fs/inode.c | 9 +- fs/fhandle.c | 16 +- fs/file_table.c | 40 +- fs/fuse/dev.c | 23 +- fs/fuse/file.c | 8 +- fs/fuse/passthrough.c | 2 +- fs/gfs2/super.c | 1 + fs/internal.h | 4 +- fs/mount.h | 10 +- fs/namespace.c | 6 +- fs/nfs/client.c | 1 + fs/nfs/pnfs.c | 2 +- fs/nfs/pnfs_nfs.c | 4 +- fs/nfsd/nfs2acl.c | 17 +- fs/nfsd/nfs3acl.c | 17 +- fs/nfsd/nfs4recover.c | 3 +- fs/nfsd/nfs4state.c | 1 + fs/nfsd/nfs4xdr.c | 3 +- fs/nfsd/vfs.c | 6 +- fs/ntfs3/xattr.c | 12 + fs/ocfs2/suballoc.c | 22 + fs/overlayfs/dir.c | 2 +- fs/overlayfs/file.c | 3 +- fs/smb/server/smb2pdu.c | 5 + fs/smb/server/smbacl.c | 4 +- fs/xfs/libxfs/xfs_ag_resv.c | 8 +- fs/xfs/libxfs/xfs_alloc.c | 5 +- fs/xfs/libxfs/xfs_attr_leaf.c | 2 +- fs/xfs/libxfs/xfs_bmap.c | 17 +- fs/xfs/libxfs/xfs_btree.c | 2 +- fs/xfs/libxfs/xfs_da_btree.c | 2 +- fs/xfs/libxfs/xfs_dir2.c | 2 +- fs/xfs/libxfs/xfs_exchmaps.c | 4 +- fs/xfs/libxfs/xfs_ialloc.c | 2 +- fs/xfs/libxfs/xfs_inode_buf.c | 4 +- fs/xfs/libxfs/xfs_inode_fork.c | 3 +- fs/xfs/libxfs/xfs_refcount.c | 5 +- fs/xfs/libxfs/xfs_rmap.c | 2 +- fs/xfs/scrub/cow_repair.c | 7 +- fs/xfs/scrub/repair.c | 2 +- fs/xfs/xfs_attr_item.c | 2 +- fs/xfs/xfs_buf.c | 4 +- fs/xfs/xfs_error.c | 5 +- fs/xfs/xfs_error.h | 10 +- fs/xfs/xfs_inode.c | 28 +- fs/xfs/xfs_iomap.c | 2 +- fs/xfs/xfs_log.c | 8 +- fs/xfs/xfs_trans_ail.c | 2 +- include/keys/request_key_auth-type.h | 2 + include/linux/backing-file.h | 4 +- include/linux/blkdev.h | 16 +- include/linux/err.h | 12 +- include/linux/fs.h | 19 +- include/linux/kvm_host.h | 7 +- include/linux/lsm_audit.h | 2 +- include/linux/lsm_hook_defs.h | 5 + include/linux/lsm_hooks.h | 1 + include/linux/security.h | 22 + include/linux/skmsg.h | 15 +- include/net/bond_3ad.h | 3 +- include/net/bond_options.h | 2 + include/net/bonding.h | 3 + include/net/phonet/pn_dev.h | 2 +- include/net/rtnetlink.h | 2 + include/net/sock.h | 2 +- include/net/udp_tunnel.h | 14 +- include/uapi/linux/if_link.h | 3 + io_uring/net.c | 36 +- io_uring/opdef.c | 4 +- kernel/bpf/cgroup.c | 2 +- kernel/futex/pi.c | 6 +- kernel/futex/requeue.c | 6 + kernel/locking/mutex.c | 16 +- kernel/locking/rtmutex.c | 51 +- kernel/locking/rtmutex_api.c | 14 +- kernel/locking/rtmutex_common.h | 3 +- kernel/locking/rwbase_rt.c | 8 +- kernel/locking/rwsem.c | 4 +- kernel/locking/spinlock_rt.c | 5 +- kernel/locking/ww_mutex.h | 30 +- kernel/sched/core.c | 12 +- kernel/trace/bpf_trace.c | 4 + kernel/trace/ftrace.c | 68 +- lib/debugobjects.c | 56 +- net/9p/client.c | 3 +- net/batman-adv/bat_iv_ogm.c | 11 +- net/batman-adv/bat_v.c | 1 + net/batman-adv/bat_v_ogm.c | 23 +- net/batman-adv/bridge_loop_avoidance.c | 28 +- net/batman-adv/distributed-arp-table.c | 12 +- net/batman-adv/fragmentation.c | 22 +- net/batman-adv/fragmentation.h | 3 +- net/batman-adv/netlink.c | 10 +- net/batman-adv/routing.c | 73 +- net/batman-adv/tp_meter.c | 115 +- net/batman-adv/translation-table.c | 12 +- net/batman-adv/tvlv.c | 69 +- net/batman-adv/types.h | 21 +- net/core/filter.c | 27 + net/core/rtnetlink.c | 8 + net/core/skbuff.c | 23 +- net/core/skmsg.c | 2 + net/ipv4/ip_gre.c | 6 + net/ipv4/ip_output.c | 20 +- net/ipv4/tcp_ao.c | 4 + net/ipv6/ip6_output.c | 22 +- net/ipv6/ip6_udp_tunnel.c | 15 +- net/ipv6/ip6_vti.c | 1 + net/mac802154/llsec.c | 14 + net/phonet/pn_dev.c | 12 +- net/phonet/pn_netlink.c | 23 +- net/rxrpc/input.c | 21 +- net/sctp/ipv6.c | 9 +- net/sctp/protocol.c | 2 + net/socket.c | 2 +- net/tipc/crypto.c | 9 + net/tipc/udp_media.c | 10 +- net/tls/tls_sw.c | 4 + net/unix/garbage.c | 2 + scripts/link-vmlinux.sh | 4 +- scripts/sorttable.c | 1119 +++++++++++++++++++- scripts/sorttable.h | 500 --------- security/apparmor/include/policy_unpack.h | 19 + security/apparmor/lsm.c | 16 +- security/apparmor/policy.c | 8 +- security/keys/internal.h | 2 + security/keys/keyctl.c | 24 +- security/keys/keyctl_pkey.c | 9 +- security/keys/request_key_auth.c | 33 +- security/security.c | 109 ++ security/selinux/hooks.c | 256 +++-- security/selinux/include/objsec.h | 11 + .../selftests/bpf/prog_tests/kprobe_multi_test.c | 35 +- .../selftests/bpf/progs/kprobe_multi_sleepable.c | 25 + tools/testing/selftests/ptp/testptp.c | 62 +- virt/kvm/eventfd.c | 12 +- 217 files changed, 3867 insertions(+), 2335 deletions(-)
Greg Kroah-Hartman
maintainerPATCH 1/204 wifi: mt76: mt7921: avoid undesired changes of the preset regulatory domain
6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Leon Yen <[email protected]>
commit 2425dc7beaadc39c2636f97f8bdc22dc3cf88149 upstream.
Some countries have strict RF restrictions where changing the regulatory domain dynamically based on the connected AP is not acceptable. This patch disables Beacon country IE hinting when a valid country code is set from usersland (e.g., by system using iw or CRDA).
--- drivers/net/wireless/mediatek/mt76/mt7921/init.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/net/wireless/mediatek/mt76/mt7921/init.c +7 -0
@@ -137,6 +137,13 @@ mt7921_regd_notifier(struct wiphy *wiphy, dev->mt76.region = request->dfs_region; dev->country_ie_env = request->country_ie_env; +if (request->initiator ==NL80211_REGDOM_SET_BY_USER ) { +if (dev->mt76.alpha2[0 ] =='0' && dev->mt76.alpha2[1 ] =='0' ) + wiphy->regulatory_flags &= ~REGULATORY_COUNTRY_IE_IGNORE ; +else + wiphy->regulatory_flags |=REGULATORY_COUNTRY_IE_IGNORE ; + } +if (pm->suspended)return ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 2/204 wifi: mt76: mt7921: fix a potential scan no APs
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Quan Zhou <[email protected]>
commit 5ed54896b6bd444223092cab361b0785932119ab upstream.
In multi-channel scenarios, the granted channel must be aborted before station remove. Otherwise, the firmware will be put into a wrong state, resulting in have chance to make subsequence scan no APs. With this patch, the granted channel will be always aborted before station remove.
--- drivers/net/wireless/mediatek/mt76/mt7921/main.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7921/main.c +4 -3
@@ -368,9 +368,9 @@ void mt7921_roc_abort_sync(struct mt792x_dev *dev)del_timer_sync (&phy->roc_timer);cancel_work_sync (&phy->roc_work);if (test_and_clear_bit (MT76_STATE_ROC , &phy->mt76->state)) -ieee80211_iterate_active_interfaces (mt76_hw (dev), -IEEE80211_IFACE_ITER_RESUME_ALL , - mt7921_roc_iter, (void *)phy); +ieee80211_iterate_interfaces (mt76_hw (dev), +IEEE80211_IFACE_ITER_RESUME_ALL , + mt7921_roc_iter, (void *)phy); }EXPORT_SYMBOL_GPL (mt7921_roc_abort_sync);
@@ -881,6 +881,7 @@ void mt7921_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif, struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv; +mt7921_roc_abort_sync (dev);mt76_connac_free_pending_tx_skbs (&dev->pm, &msta->deflink.wcid);mt76_connac_pm_wake (&dev->mphy, &dev->pm);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 3/204 wifi: mt76: mt7921: fix potential deadlock in mt7921_roc_abort_sync
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sean Wang <[email protected]>
commit d5059e52fd8bc624ec4255c9fa01a266513d126b upstream.
roc_abort_sync() can deadlock with roc_work(). roc_work() holds dev->mt76.mutex, while cancel_work_sync() waits for roc_work() to finish. If the caller already owns the same mutex, both sides block and no progress is possible.
This deadlock can occur during station removal when mt76_sta_state() -> mt76_sta_remove() -> mt7921_mac_sta_remove() -> mt7921_roc_abort_sync() invokes cancel_work_sync() while roc_work() is still running and holding dev->mt76.mutex.
This avoids the mutex deadlock and preserves exactly-once work ownership.
[Ajrat: keep del_timer_sync() instead of timer_delete_sync() -- the
timer API rename is not present in 6.12.y. ]
--- drivers/net/wireless/mediatek/mt76/mt7921/main.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7921/main.c +8 -5
@@ -365,12 +365,15 @@ void mt7921_roc_abort_sync(struct mt792x_dev *dev) {struct mt792x_phy *phy = &dev->phy; +if (!test_and_clear_bit (MT76_STATE_ROC , &phy->mt76->state)) +return ; +del_timer_sync (&phy->roc_timer); -cancel_work_sync (&phy->roc_work); -if (test_and_clear_bit (MT76_STATE_ROC , &phy->mt76->state)) -ieee80211_iterate_interfaces (mt76_hw (dev), -IEEE80211_IFACE_ITER_RESUME_ALL , - mt7921_roc_iter, (void *)phy); +cancel_work (&phy->roc_work); + +ieee80211_iterate_interfaces (mt76_hw (dev), +IEEE80211_IFACE_ITER_RESUME_ALL , + mt7921_roc_iter, (void *)phy); }EXPORT_SYMBOL_GPL (mt7921_roc_abort_sync);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 4/204 fuse: limit FUSE_NOTIFY_RETRIEVE to uptodate folios
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jann Horn <[email protected]>
[ Upstream commit 4e3d1b2c48ca6c55f1e9ca7f8dccc76f120f276c ]
FUSE_NOTIFY_RETRIEVE must be limited to uptodate folios; !uptodate folios can contain uninitialized data.
Since FUSE_NOTIFY_RETRIEVE is intended to only return data that is already in the page cache and not wait for data from the FUSE daemon, treat !uptodate folios as if they weren't present.
This only has security impact on systems that don't enable automatic zero-initialization of all page allocations via CONFIG_INIT_ON_ALLOC_DEFAULT_ON or init_on_alloc=1.
--- fs/fuse/dev.c | 4 ++++ 1 file changed, 4 insertions(+)
fs/fuse/dev.c +4 -0
@@ -1769,6 +1769,10 @@ static int fuse_retrieve(struct fuse_mount *fm, struct inode *inode, page =find_get_page (mapping, index);if (!page)break ; +if (!PageUptodate (page)) { +put_page (page); +break ; + } this_num =min_t (unsigned, num,PAGE_SIZE - offset); ap->pages[ap->num_pages] = page;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 5/204 gpiolib: Extract gpiochip_choose_fwnode() for wider use
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Andy Shevchenko <[email protected]>
[ Upstream commit 375790f18396b2ba706e031b150c58cd37b45a11 ]
Extract gpiochip_choose_fwnode() for the future use in another function.
Stable-dep-of: 16fdabe143fc ("gpio: Fix resource leaks on errors in gpiochip_add_data_with_key()")
--- drivers/gpio/gpiolib.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-)
drivers/gpio/gpiolib.c +16 -8
@@ -883,6 +883,21 @@ void *gpiochip_get_data(struct gpio_chip *gc) }EXPORT_SYMBOL_GPL (gpiochip_get_data); +/* +* If the calling driver provides the specific firmware node, +* use it. Otherwise use the one from the parent device, if any. +*/ +static struct fwnode_handle *gpiochip_choose_fwnode (struct gpio_chip *gc) +{ +if (gc->fwnode) +return gc->fwnode; + +if (gc->parent) +return dev_fwnode (gc->parent); + +return NULL ; +} +int gpiochip_get_ngpios (struct gpio_chip *gc,struct device *dev) {u32 ngpios = gc->ngpio;
@@ -942,14 +957,7 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data, gc->gpiodev = gdev;gpiochip_set_data (gc, data); -/* -* If the calling driver did not initialize firmware node, -* do it here using the parent device, if any. -*/ -if (gc->fwnode) -device_set_node (&gdev->dev, gc->fwnode); -else if (gc->parent) -device_set_node (&gdev->dev,dev_fwnode (gc->parent)); +device_set_node (&gdev->dev,gpiochip_choose_fwnode (gc)); gdev->id =ida_alloc (&gpio_ida,GFP_KERNEL );if (gdev->id <0 ) {
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 6/204 gpiolib: Remove redundant assignment of return variable
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Andy Shevchenko <[email protected]>
[ Upstream commit 550300b9a295a591e0721a31f8c964a4bc08d51c ]
In some functions the returned variable is assigned to 0 and then reassigned to the actual value. Remove redundant assignments.
In one case make it more clear that the assignment is not needed.
Stable-dep-of: 16fdabe143fc ("gpio: Fix resource leaks on errors in gpiochip_add_data_with_key()")
--- drivers/gpio/gpiolib.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-)
drivers/gpio/gpiolib.c +6 -7
@@ -939,7 +939,7 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,struct gpio_device *gdev;unsigned int desc_index;int base =0 ; -int ret =0 ; +int ret;/* * First: allocate and populate the internal stat container, and
@@ -959,11 +959,10 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,device_set_node (&gdev->dev,gpiochip_choose_fwnode (gc)); - gdev->id =ida_alloc (&gpio_ida,GFP_KERNEL ); -if (gdev->id <0 ) { - ret = gdev->id; + ret =ida_alloc (&gpio_ida,GFP_KERNEL ); +if (ret <0 ) goto err_free_gdev; - } + gdev->id = ret; ret =dev_set_name (&gdev->dev,GPIOCHIP_NAME "%d" , gdev->id);if (ret)
@@ -2882,7 +2881,7 @@ EXPORT_SYMBOL_GPL(gpiod_direction_output);*/ int gpiod_enable_hw_timestamp_ns (struct gpio_desc *desc,unsigned long flags) { -int ret =0 ; +int ret;VALIDATE_DESC (desc);
@@ -2915,7 +2914,7 @@ EXPORT_SYMBOL_GPL(gpiod_enable_hw_timestamp_ns);*/ int gpiod_disable_hw_timestamp_ns (struct gpio_desc *desc,unsigned long flags) { -int ret =0 ; +int ret;VALIDATE_DESC (desc);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 7/204 gpio: Fix resource leaks on errors in gpiochip_add_data_with_key()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tzung-Bi Shih <[email protected]>
[ Upstream commit 16fdabe143fce2cbf89139677728e17e21b46c28 ]
Since commit aab5c6f20023 ("gpio: set device type for GPIO chips"), gdev->dev.release is unset. As a result, the reference count to gdev->dev isn't dropped on the error handling paths.
Drop the reference on errors.
Also reorder the instructions to make the error handling simpler. Now gpiochip_add_data_with_key() roughly looks like:
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Some memory allocation. Go to ERR ZONE 1 on errors.
device_initialize().
gpiodev_release() takes over the responsibility for freeing the resources of gdev->dev. The subsequent error handling paths shouldn't go through ERR ZONE 1 again which leads to double free.
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Some initialization mainly on
gdev.
The rest of initialization. Go to ERR ZONE 2 on errors. Chip registration success and exit.ERR ZONE 2.
gpio_device_put()and exit.
ERR ZONE 1.
[missing commit fcc8b637c542 ("gpiolib: switch the line state notifier
to atomic"), commit dcb73cbaaeb3 ("gpio: cdev: use raw notifier for line state events") and commit d4f335b410dd ("gpiolib: rename GPIO chip printk macros") in 6.12.y.
s/gpiochip_err/chip_err/ as well as replaced rwlock_init+RAW_INIT_NOTIFIER_HEAD with BLOCKING_INIT_NOTIFIER_HEAD based on missing commits, following same logic as in 16fdabe143fc.]
--- drivers/gpio/gpiolib.c | 101 ++++++++++++++++++++--------------------- 1 file changed, 48 insertions(+), 53 deletions(-)
drivers/gpio/gpiolib.c +48 -53
@@ -785,13 +785,15 @@ static const struct device_type gpio_dev_type = {#define gcdev_unregister (gdev) device_del(&(gdev)->dev)#endif +/* +* An initial reference count has been held in gpiochip_add_data_with_key(). +* The caller should drop the reference via gpio_device_put() on errors. +*/ static int gpiochip_setup_dev (struct gpio_device *gdev) {struct fwnode_handle *fwnode =dev_fwnode (&gdev->dev);int ret; -device_initialize (&gdev->dev); -/* * If fwnode doesn't belong to another device, it's safe to clear its * initialized flag.
@@ -859,9 +861,11 @@ static void gpiochip_setup_devs(void)list_for_each_entry_srcu (gdev, &gpio_devices, list,srcu_read_lock_held (&gpio_devices_srcu)) { ret =gpiochip_setup_dev (gdev); -if (ret) +if (ret) { +gpio_device_put (gdev);dev_err (&gdev->dev,"Failed to initialize gpio device (%d)\n" , ret); + } } }
@@ -941,70 +945,71 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,int base =0 ;int ret; -/* -* First: allocate and populate the internal stat container, and -* set up the struct device. -*/ gdev =kzalloc (sizeof (*gdev),GFP_KERNEL );if (!gdev)return -ENOMEM ; - - gdev->dev.type = &gpio_dev_type; - gdev->dev.bus = &gpio_bus_type; - gdev->dev.parent = gc->parent; -rcu_assign_pointer (gdev->chip, gc); - gc->gpiodev = gdev;gpiochip_set_data (gc, data); -device_set_node (&gdev->dev,gpiochip_choose_fwnode (gc)); - ret =ida_alloc (&gpio_ida,GFP_KERNEL );if (ret <0 ) goto err_free_gdev; gdev->id = ret; - ret =dev_set_name (&gdev->dev,GPIOCHIP_NAME "%d" , gdev->id); + ret =init_srcu_struct (&gdev->srcu);if (ret) goto err_free_ida; +rcu_assign_pointer (gdev->chip, gc); -if (gc->parent && gc->parent->driver) - gdev->owner = gc->parent->driver->owner; -else if (gc->owner) -/* TODO: remove chip->owner */ - gdev->owner = gc->owner; -else - gdev->owner =THIS_MODULE ; + ret =init_srcu_struct (&gdev->desc_srcu); +if (ret) + goto err_cleanup_gdev_srcu; + + ret =dev_set_name (&gdev->dev,GPIOCHIP_NAME "%d" , gdev->id); +if (ret) + goto err_cleanup_desc_srcu; + +device_initialize (&gdev->dev); +/* +* After this point any allocated resources to `gdev` will be +* free():ed by gpiodev_release(). If you add new resources +* then make sure they get free():ed there. +*/ + gdev->dev.type = &gpio_dev_type; + gdev->dev.bus = &gpio_bus_type; + gdev->dev.parent = gc->parent; +device_set_node (&gdev->dev,gpiochip_choose_fwnode (gc)); ret =gpiochip_get_ngpios (gc, &gdev->dev);if (ret) - goto err_free_dev_name; + goto err_put_device; + gdev->ngpio = gc->ngpio; gdev->descs =kcalloc (gc->ngpio,sizeof (*gdev->descs),GFP_KERNEL );if (!gdev->descs) { ret = -ENOMEM ; - goto err_free_dev_name; + goto err_put_device; } gdev->label =kstrdup_const (gc->label ?:"unknown" ,GFP_KERNEL );if (!gdev->label) { ret = -ENOMEM ; - goto err_free_descs; + goto err_put_device; } - gdev->ngpio = gc->ngpio; gdev->can_sleep = gc->can_sleep; -BLOCKING_INIT_NOTIFIER_HEAD (&gdev->line_state_notifier);BLOCKING_INIT_NOTIFIER_HEAD (&gdev->device_notifier); - - ret =init_srcu_struct (&gdev->srcu); -if (ret) - goto err_free_label; - - ret =init_srcu_struct (&gdev->desc_srcu); -if (ret) - goto err_cleanup_gdev_srcu; +#ifdef CONFIG_PINCTRL +INIT_LIST_HEAD (&gdev->pin_ranges); +#endif +if (gc->parent && gc->parent->driver) + gdev->owner = gc->parent->driver->owner; +else if (gc->owner) +/* TODO: remove chip->owner */ + gdev->owner = gc->owner; +else + gdev->owner =THIS_MODULE ; scoped_guard(mutex, &gpio_devices_lock) {/*
@@ -1020,7 +1025,7 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,if (base <0 ) { ret = base; base =0 ; - goto err_cleanup_desc_srcu; + goto err_put_device; }/*
@@ -1040,14 +1045,10 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data, ret =gpiodev_add_to_list_unlocked (gdev);if (ret) {chip_err (gc,"GPIO integer space overlap, cannot add chip\n" ); - goto err_cleanup_desc_srcu; + goto err_put_device; } } -#ifdef CONFIG_PINCTRL -INIT_LIST_HEAD (&gdev->pin_ranges); -#endif -if (gc->names)gpiochip_set_desc_names (gc);
@@ -1128,25 +1129,19 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data, scoped_guard(mutex, &gpio_devices_lock)list_del_rcu (&gdev->list);synchronize_srcu (&gpio_devices_srcu); -if (gdev->dev.release) { -/* release() has been registered by gpiochip_setup_dev() */ -gpio_device_put (gdev); - goto err_print_message; - } +err_put_device: +gpio_device_put (gdev); + goto err_print_message; + err_cleanup_desc_srcu:cleanup_srcu_struct (&gdev->desc_srcu); err_cleanup_gdev_srcu:cleanup_srcu_struct (&gdev->srcu); -err_free_label: -kfree_const (gdev->label); -err_free_descs: -kfree (gdev->descs); -err_free_dev_name: -kfree (dev_name (&gdev->dev)); err_free_ida:ida_free (&gpio_ida, gdev->id); err_free_gdev:kfree (gdev); + err_print_message:/* failures here can mean systems won't boot... */ if (ret != -EPROBE_DEFER ) {
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 8/204 io_uring/net: Avoid msghdr on op_connect/op_bind async data
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gabriel Krisman Bertazi <[email protected]>
[ Upstream commit 3979840cd858f30f43ea9f4e7f7f1f56de82d698 ] This fixes a memory leak due to the lack of the cleanup hook for the iovec. The stable backport differs from upstream by dropping the io_connect_bpf_populate hunk, which didn't exist at the time and by fixing the merge conflict due to the introduction of io_bind_file_create and by using the older async_data allocation API.
Both IORING_OP_CONNECT and IORING_OP_BIND reuse the msghdr object just to store the sockaddr. Beyond allocating a much larger object than needed, msghdr can also wrap an iovec, which will be recycled unnecessarily. This uses the sockaddr directly.
--- io_uring/net.c | 36 ++++++++++++++++++------------------ io_uring/opdef.c | 4 ++-- 2 files changed, 20 insertions(+), 20 deletions(-)
io_uring/net.c +18 -18
@@ -1719,7 +1719,7 @@ int io_socket(struct io_kiocb *req, unsigned int issue_flags)int io_connect_prep (struct io_kiocb *req,const struct io_uring_sqe *sqe) { struct io_connect *conn = io_kiocb_to_cmd(req, struct io_connect); -struct io_async_msghdr *io; +struct sockaddr_storage *addr;if (sqe->len || sqe->buf_index || sqe->rw_flags || sqe->splice_fd_in)return -EINVAL ;
@@ -1728,17 +1728,17 @@ int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) conn->addr_len =READ_ONCE (sqe->addr2); conn->in_progress = conn->seen_econnaborted = false; - io =io_msg_alloc_async (req); -if (unlikely (!io)) +if (io_alloc_async_data (req))return -ENOMEM ; + addr = req->async_data; -return move_addr_to_kernel (conn->addr, conn->addr_len, &io->addr); +return move_addr_to_kernel (conn->addr, conn->addr_len, addr); }int io_connect (struct io_kiocb *req,unsigned int issue_flags) { struct io_connect *connect = io_kiocb_to_cmd(req, struct io_connect); -struct io_async_msghdr *io = req->async_data; +struct sockaddr_storage *addr = req->async_data;unsigned file_flags;int ret;bool force_nonblock = issue_flags &IO_URING_F_NONBLOCK ;
@@ -1752,8 +1752,7 @@ int io_connect(struct io_kiocb *req, unsigned int issue_flags) file_flags = force_nonblock ?O_NONBLOCK :0 ; - ret =__sys_connect_file (req->file, &io->addr, connect->addr_len, - file_flags); + ret =__sys_connect_file (req->file, addr, connect->addr_len, file_flags);if ((ret == -EAGAIN || ret == -EINPROGRESS || ret == -ECONNABORTED ) && force_nonblock) {if (ret == -EINPROGRESS ) {
@@ -1782,7 +1781,6 @@ int io_connect(struct io_kiocb *req, unsigned int issue_flags) out:if (ret <0 )req_set_fail (req); -io_req_msg_cleanup (req, issue_flags);io_req_set_res (req, ret,0 );return IOU_OK ; }
@@ -1792,15 +1790,15 @@ int io_connect(struct io_kiocb *req, unsigned int issue_flags)* which in turn end up in mnt_want_write() which will grab the fs * percpu start write sem. This can trigger a lockdep warning. */ -static int io_bind_file_create (const struct io_async_msghdr *io,int addr_len) +static int io_bind_file_create (const struct sockaddr_storage *addr,int addr_len) {const struct sockaddr_un *sun; -if (io->addr.ss_family !=AF_UNIX ) +if (addr->ss_family !=AF_UNIX )return 0 ;if (addr_len <= offsetof(struct sockaddr_un , sun_path))return 0 ; - sun = (const struct sockaddr_un *) &io->addr; + sun = (const struct sockaddr_un *) addr;return sun->sun_path[0 ] !='\0' ; }
@@ -1808,7 +1806,7 @@ int io_bind_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_bind *bind = io_kiocb_to_cmd(req, struct io_bind);struct sockaddr __user *uaddr; -struct io_async_msghdr *io; +struct sockaddr_storage *addr;int ret;if (sqe->len || sqe->buf_index || sqe->rw_flags || sqe->splice_fd_in)
@@ -1817,21 +1815,23 @@ int io_bind_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) uaddr =u64_to_user_ptr (READ_ONCE (sqe->addr)); bind->addr_len =READ_ONCE (sqe->addr2); - io =io_msg_alloc_async (req); -if (unlikely (!io)) +if (io_alloc_async_data (req))return -ENOMEM ; - ret =move_addr_to_kernel (uaddr, bind->addr_len, &io->addr); + addr = req->async_data; + + ret =move_addr_to_kernel (uaddr, bind->addr_len, addr);if (unlikely (ret))return ret; -if (io_bind_file_create (io, bind->addr_len)) +if (io_bind_file_create (addr, bind->addr_len)) req->flags |=REQ_F_FORCE_ASYNC ;return 0 ; } +int io_bind (struct io_kiocb *req,unsigned int issue_flags) { struct io_bind *bind = io_kiocb_to_cmd(req, struct io_bind); -struct io_async_msghdr *io = req->async_data; +struct sockaddr_storage *addr = req->async_data;struct socket *sock;int ret;
@@ -1839,7 +1839,7 @@ int io_bind(struct io_kiocb *req, unsigned int issue_flags)if (unlikely (!sock))return -ENOTSOCK ; - ret =__sys_bind_socket (sock, &io->addr, bind->addr_len); + ret =__sys_bind_socket (sock, addr, bind->addr_len);if (ret <0 )req_set_fail (req);io_req_set_res (req, ret,0 );
io_uring/opdef.c +2 -2
@@ -205,7 +205,7 @@ const struct io_issue_def io_issue_defs[] = { .unbound_nonreg_file = 1, .pollout = 1,#if defined(CONFIG_NET ) - .async_size = sizeof(struct io_async_msghdr), + .async_size = sizeof(struct sockaddr_storage), .prep = io_connect_prep, .issue = io_connect,#else
@@ -501,7 +501,7 @@ const struct io_issue_def io_issue_defs[] = { .needs_file = 1, .prep = io_bind_prep, .issue = io_bind, - .async_size = sizeof(struct io_async_msghdr), + .async_size = sizeof(struct sockaddr_storage), #else .prep = io_eopnotsupp_prep,#endif
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 9/204 drm/xe/display: fix oops in suspend/shutdown without display
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jani Nikula <[email protected]>
[ Upstream commit 68938cc08e23a94fd881e845837ff918de005ce7 ]
The xe driver keeps track of whether to probe display, and whether display hardware is there, using xe->info.probe_display. It gets set to false if there's no display after intel_display_device_probe(). However, the display may also be disabled via fuses, detected at a later time in intel_display_device_info_runtime_init().
In this case, the xe driver does for_each_intel_crtc() on uninitialized mode config in xe_display_flush_cleanup_work(), leading to a NULL pointer dereference, and generally calls display code with display info cleared.
Check for intel_display_device_present() after intel_display_device_info_runtime_init(), and reset xe->info.probe_display as necessary. Also do unset_display_features() for completeness, although display runtime init has already done that. This will need to be unified across all cases later.
Move intel_display_device_info_runtime_init() call slightly earlier, similar to i915, to avoid a bunch of unnecessary setup for no display cases.
Note #1: The xe driver has no business doing low level display plumbing like for_each_intel_crtc() to begin with. It all needs to happen in display code.
Note #2: The actual bug is present already in commit 44e694958b95 ("drm/xe/display: Implement display support"), but the oops was likely introduced later at commit ddf6492e0e50 ("drm/xe/display: Make display suspend/resume work on discrete").
--- drivers/gpu/drm/xe/display/xe_display.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)
drivers/gpu/drm/xe/display/xe_display.c +9 -2
@@ -148,6 +148,15 @@ int xe_display_init_noirq(struct xe_device *xe)intel_display_driver_early_probe (xe); +intel_display_device_info_runtime_init (xe); + +/* Display may have been disabled at runtime init */ +if (!has_display (xe)) { + xe->info.probe_display = false; +unset_display_features (xe); +return 0 ; + } +/* Early display init.. */ intel_opregion_setup (display);
@@ -159,8 +168,6 @@ int xe_display_init_noirq(struct xe_device *xe)intel_bw_init_hw (xe); -intel_display_device_info_runtime_init (xe); - err =intel_display_driver_probe_noirq (xe);if (err) {intel_opregion_cleanup (display);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 10/204 drm/v3d: Store the active job inside the queues state
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Maíra Canal <[email protected]>
[ Upstream commit 0d3768826d38c0ac740f8b45cd13346630535f2b ]
Instead of storing the queue's active job in four different variables, store the active job inside the queue's state. This way, it's possible to access all active jobs using an index based in enum v3d_queue.
Stable-dep-of: 7f93fad5ea0a ("drm/v3d: Skip CSD when it has zeroed workgroups")
--- drivers/gpu/drm/v3d/v3d_drv.h | 7 ++-- drivers/gpu/drm/v3d/v3d_gem.c | 7 ++-- drivers/gpu/drm/v3d/v3d_irq.c | 62 +++++++++++++-------------------- drivers/gpu/drm/v3d/v3d_sched.c | 26 +++++++++----- 4 files changed, 48 insertions(+), 54 deletions(-)
drivers/gpu/drm/v3d/v3d_drv.h +3 -4
@@ -59,6 +59,9 @@ struct v3d_queue_state {/* Stores the GPU stats for this queue in the global context. */ struct v3d_stats stats; + +/* Currently active job for this queue */ +struct v3d_job *active_job; };/* Performance monitor object. The perform lifetime is controlled by userspace
@@ -147,10 +150,6 @@ struct v3d_dev {struct work_struct overflow_mem_work; -struct v3d_bin_job *bin_job; -struct v3d_render_job *render_job; -struct v3d_tfu_job *tfu_job; -struct v3d_csd_job *csd_job;struct v3d_cpu_job *cpu_job;struct v3d_queue_state queue[V3D_MAX_QUEUES ];
drivers/gpu/drm/v3d/v3d_gem.c +3 -4
@@ -304,16 +304,15 @@ void v3d_gem_destroy(struct drm_device *dev) {struct v3d_dev *v3d =to_v3d_dev (dev); +enum v3d_queue q;v3d_sched_fini (v3d);/* Waiting for jobs to finish would need to be done before * unregistering V3D. */ -WARN_ON (v3d->bin_job); -WARN_ON (v3d->render_job); -WARN_ON (v3d->tfu_job); -WARN_ON (v3d->csd_job); +for (q =0 ; q <V3D_MAX_QUEUES ; q++) +WARN_ON (v3d->queue[q].active_job);drm_mm_takedown (&v3d->mm);
drivers/gpu/drm/v3d/v3d_irq.c +24 -38
105 lines not on this page · raw message
drivers/gpu/drm/v3d/v3d_sched.c +18 -8
@@ -208,14 +208,18 @@ static struct dma_fence *v3d_bin_job_run(struct drm_sched_job *sched_job)struct dma_fence *fence;unsigned long irqflags; -if (unlikely (job->base.base.s_fence->finished.error)) +if (unlikely (job->base.base.s_fence->finished.error)) { +spin_lock_irqsave (&v3d->job_lock, irqflags); + v3d->queue[V3D_BIN ].active_job =NULL ; +spin_unlock_irqrestore (&v3d->job_lock, irqflags);return NULL ; + }/* Lock required around bin_job update vs * v3d_overflow_mem_work(). */ spin_lock_irqsave (&v3d->job_lock, irqflags); - v3d->bin_job = job; + v3d->queue[V3D_BIN ].active_job = &job->base;/* Clear out the overflow allocation, so we don't * reuse the overflow attached to a previous job. */
@@ -263,10 +267,12 @@ static struct dma_fence *v3d_render_job_run(struct drm_sched_job *sched_job)struct drm_device *dev = &v3d->drm;struct dma_fence *fence; -if (unlikely (job->base.base.s_fence->finished.error)) +if (unlikely (job->base.base.s_fence->finished.error)) { + v3d->queue[V3D_RENDER ].active_job =NULL ;return NULL ; + } - v3d->render_job = job; + v3d->queue[V3D_RENDER ].active_job = &job->base;/* Can we avoid this flush? We need to be careful of * scheduling, though -- imagine job0 rendering to texture and
@@ -309,10 +315,12 @@ v3d_tfu_job_run(struct drm_sched_job *sched_job)struct drm_device *dev = &v3d->drm;struct dma_fence *fence; -if (unlikely (job->base.base.s_fence->finished.error)) +if (unlikely (job->base.base.s_fence->finished.error)) { + v3d->queue[V3D_TFU ].active_job =NULL ;return NULL ; + } - v3d->tfu_job = job; + v3d->queue[V3D_TFU ].active_job = &job->base; fence =v3d_fence_create (v3d,V3D_TFU );if (IS_ERR (fence))
@@ -355,10 +363,12 @@ v3d_csd_job_run(struct drm_sched_job *sched_job)struct dma_fence *fence;int i, csd_cfg0_reg; -if (unlikely (job->base.base.s_fence->finished.error)) +if (unlikely (job->base.base.s_fence->finished.error)) { + v3d->queue[V3D_CSD ].active_job =NULL ;return NULL ; + } - v3d->csd_job = job; + v3d->queue[V3D_CSD ].active_job = &job->base;v3d_invalidate_caches (v3d);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 11/204 drm/v3d: Skip CSD when it has zeroed workgroups
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Maíra Canal <[email protected]>
[ Upstream commit 7f93fad5ea0affc9e1505dd0f7596c0fdb496213 ]
A compute shader dispatch encodes its workgroup counts in the CFG0..CFG2 registers. Kicking off a dispatch with a zero count in any of the three dimensions is invalid. First, the hardware will process 0 as 65536, while the user-space driver exposes a maximum of 65535. Over that, a submission with a zeroed workgroup dimension should be a no-op.
These zeroed counts can reach the dispatch path through an indirect CSD job, whose workgroup counts are only known once the indirect buffer is read and may legitimately be zero, but such scenario should only result in a no-op.
Overwrite the indirect CSD job workgroup counts with the indirect BO ones, even if they are zeroed, and don't submit the job to the hardware when any of the workgroup counts is zero, so the job completes immediately instead of running the shader.
--- drivers/gpu/drm/v3d/v3d_sched.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/gpu/drm/v3d/v3d_sched.c +13 -3
@@ -368,6 +368,16 @@ v3d_csd_job_run(struct drm_sched_job *sched_job)return NULL ; } +/* The HW interprets a workgroup size of 0 as 65536; however, the +* user-space driver exposes a maximum of 65535. Therefore, a 0 in +* any dimension means that we have no workgroups and the compute +* shader should not be dispatched. +*/ +if (!V3D_GET_FIELD (job->args.cfg[0 ],V3D_CSD_QUEUED_CFG0_NUM_WGS_X ) || + !V3D_GET_FIELD (job->args.cfg[1 ],V3D_CSD_QUEUED_CFG1_NUM_WGS_Y ) || + !V3D_GET_FIELD (job->args.cfg[2 ],V3D_CSD_QUEUED_CFG2_NUM_WGS_Z )) +return NULL ; + v3d->queue[V3D_CSD ].active_job = &job->base;v3d_invalidate_caches (v3d);
@@ -418,13 +428,13 @@ v3d_rewrite_csd_job_wg_counts_from_indirect(struct v3d_cpu_job *job) wg_counts = (uint32_t *)(bo->vaddr + indirect_csd->offset); -if (wg_counts[0 ] ==0 || wg_counts[1 ] ==0 || wg_counts[2 ] ==0 ) - goto unmap_bo; - args->cfg[0 ] = wg_counts[0 ] <<V3D_CSD_CFG012_WG_COUNT_SHIFT ; args->cfg[1 ] = wg_counts[1 ] <<V3D_CSD_CFG012_WG_COUNT_SHIFT ; args->cfg[2 ] = wg_counts[2 ] <<V3D_CSD_CFG012_WG_COUNT_SHIFT ; +if (wg_counts[0 ] ==0 || wg_counts[1 ] ==0 || wg_counts[2 ] ==0 ) + goto unmap_bo; + num_batches =DIV_ROUND_UP (indirect_csd->wg_size,16 ) * (wg_counts[0 ] * wg_counts[1 ] * wg_counts[2 ]);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 12/204 eventpoll: use hlist_is_singular_node() in __ep_remove()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <[email protected]>
[ Upstream commit 3d9fd0abc94d8cd430cc7cd7d37ce5e5aae2cd2b ]
Replace the open-coded "epi is the only entry in file->f_ep" check with hlist_is_singular_node(). Same semantics, and the helper avoids the head-cacheline access in the common false case.
Stable-dep-of: a6dc643c6931 ("eventpoll: fix ep_remove struct eventpoll / struct file UAF")
--- fs/eventpoll.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/eventpoll.c +1 -1
@@ -827,7 +827,7 @@ static bool __ep_remove(struct eventpoll *ep, struct epitem *epi, bool force) to_free =NULL ; head = file->f_ep; -if (head->first == &epi->fllink && !epi->fllink.next) { +if (hlist_is_singular_node (&epi->fllink, head)) {/* See eventpoll_release() for details. */ WRITE_ONCE (file->f_ep,NULL );if (!is_file_epoll (file)) {
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 13/204 eventpoll: split __ep_remove()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <[email protected]>
[ Upstream commit 0f7bdfd413000985de09fc39eb9efa1e091a3ce0 ]
Split __ep_remove() to delineate file removal from epoll item removal.
Stable-dep-of: a6dc643c6931 ("eventpoll: fix ep_remove struct eventpoll / struct file UAF")
--- fs/eventpoll.c | 27 +++++++++++++++++++++++---- 1 file changed, 23 insertions(+), 4 deletions(-)
fs/eventpoll.c +23 -4
@@ -797,6 +797,9 @@ static void ep_free(struct eventpoll *ep)kfree_rcu (ep, rcu); } +static void __ep_remove_file (struct eventpoll *ep,struct epitem *epi,struct file *file); +static bool __ep_remove_epi (struct eventpoll *ep,struct epitem *epi); +/* * Removes a "struct epitem" from the eventpoll RB tree and deallocates * all the associated resources. Must be called with "mtx" held.
@@ -808,8 +811,6 @@ static void ep_free(struct eventpoll *ep)static bool __ep_remove (struct eventpoll *ep,struct epitem *epi,bool force) {struct file *file = epi->ffd.file; -struct epitems_head *to_free; -struct hlist_head *head;lockdep_assert_irqs_enabled ();
@@ -825,8 +826,21 @@ static bool __ep_remove(struct eventpoll *ep, struct epitem *epi, bool force)return false; } - to_free =NULL ; - head = file->f_ep; +__ep_remove_file (ep, epi, file); +return __ep_remove_epi (ep, epi); +} + +/* +* Called with &file->f_lock held, +* returns with it released +*/ +static void __ep_remove_file (struct eventpoll *ep,struct epitem *epi,struct file *file) +{ +struct epitems_head *to_free =NULL ; +struct hlist_head *head = file->f_ep; + +lockdep_assert_held (&ep->mtx); +if (hlist_is_singular_node (&epi->fllink, head)) {/* See eventpoll_release() for details. */ WRITE_ONCE (file->f_ep,NULL );
@@ -840,6 +854,11 @@ static bool __ep_remove(struct eventpoll *ep, struct epitem *epi, bool force)hlist_del_rcu (&epi->fllink);spin_unlock (&file->f_lock);free_ephead (to_free); +} + +static bool __ep_remove_epi (struct eventpoll *ep,struct epitem *epi) +{ +lockdep_assert_held (&ep->mtx);rb_erase_cached (&epi->rbn, &ep->rbr);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 14/204 eventpoll: kill __ep_remove()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <[email protected]>
[ Upstream commit e9e5cd40d7c403e19f21d0f7b8b8ba3a76b58330 ]
Remove the boolean conditional in __ep_remove() and restructure the code so the check for racing with eventpoll_release_file() are only done in the ep_remove_safe() path where they belong.
Stable-dep-of: a6dc643c6931 ("eventpoll: fix ep_remove struct eventpoll / struct file UAF")
--- fs/eventpoll.c | 67 ++++++++++++++++++++++---------------------------- 1 file changed, 30 insertions(+), 37 deletions(-)
fs/eventpoll.c +30 -37
@@ -797,49 +797,18 @@ static void ep_free(struct eventpoll *ep)kfree_rcu (ep, rcu); } -static void __ep_remove_file (struct eventpoll *ep,struct epitem *epi,struct file *file); -static bool __ep_remove_epi (struct eventpoll *ep,struct epitem *epi); - -/* -* Removes a "struct epitem" from the eventpoll RB tree and deallocates -* all the associated resources. Must be called with "mtx" held. -* If the dying flag is set, do the removal only if force is true. -* This prevents ep_clear_and_put() from dropping all the ep references -* while running concurrently with eventpoll_release_file(). -* Returns true if the eventpoll can be disposed. -*/ -static bool __ep_remove (struct eventpoll *ep,struct epitem *epi,bool force) -{ -struct file *file = epi->ffd.file; - -lockdep_assert_irqs_enabled (); - -/* -* Removes poll wait queue hooks. -*/ -ep_unregister_pollwait (ep, epi); - -/* Remove the current item from the list of epoll hooks */ -spin_lock (&file->f_lock); -if (epi->dying && !force) { -spin_unlock (&file->f_lock); -return false; - } - -__ep_remove_file (ep, epi, file); -return __ep_remove_epi (ep, epi); -} -/* * Called with &file->f_lock held, * returns with it released */ -static void __ep_remove_file (struct eventpoll *ep,struct epitem *epi,struct file *file) +static void __ep_remove_file (struct eventpoll *ep,struct epitem *epi, +struct file *file) {struct epitems_head *to_free =NULL ;struct hlist_head *head = file->f_ep;lockdep_assert_held (&ep->mtx); +lockdep_assert_held (&file->f_lock);if (hlist_is_singular_node (&epi->fllink, head)) {/* See eventpoll_release() for details. */
@@ -886,7 +855,25 @@ static bool __ep_remove_epi(struct eventpoll *ep, struct epitem *epi)*/ static void ep_remove_safe (struct eventpoll *ep,struct epitem *epi) { -if (__ep_remove (ep, epi, false)) +struct file *file = epi->ffd.file; + +lockdep_assert_irqs_enabled (); +lockdep_assert_held (&ep->mtx); + +ep_unregister_pollwait (ep, epi); + +/* sync with eventpoll_release_file() */ +if (unlikely (READ_ONCE (epi->dying))) +return ; + +spin_lock (&file->f_lock); +if (epi->dying) { +spin_unlock (&file->f_lock); +return ; + } +__ep_remove_file (ep, epi, file); + +if (__ep_remove_epi (ep, epi))WARN_ON_ONCE (ep_refcount_dec_and_test (ep)); }
@@ -1118,7 +1105,7 @@ void eventpoll_release_file(struct file *file)spin_lock (&file->f_lock);if (file->f_ep && file->f_ep->first) { epi = hlist_entry(file->f_ep->first, struct epitem, fllink); - epi->dying = true; +WRITE_ONCE (epi->dying, true);spin_unlock (&file->f_lock);/*
@@ -1127,7 +1114,13 @@ void eventpoll_release_file(struct file *file)*/ ep = epi->ep;mutex_lock (&ep->mtx); - dispose =__ep_remove (ep, epi, true); + +ep_unregister_pollwait (ep, epi); + +spin_lock (&file->f_lock); +__ep_remove_file (ep, epi, file); + dispose =__ep_remove_epi (ep, epi); +mutex_unlock (&ep->mtx);if (dispose &&ep_refcount_dec_and_test (ep))
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 15/204 eventpoll: drop vestigial __ prefix from ep_remove_{file,epi}()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <[email protected]>
[ Upstream commit 0feaf644f7180c4a91b6b405a881afbfd958f1cf ]
With __ep_remove() gone, the double-underscore on __ep_remove_file() and __ep_remove_epi() no longer contrasts with a __-less parent and just reads as noise. Rename both to ep_remove_file() and ep_remove_epi(). No functional change.
Stable-dep-of: a6dc643c6931 ("eventpoll: fix ep_remove struct eventpoll / struct file UAF")
--- fs/eventpoll.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
fs/eventpoll.c +6 -6
@@ -801,7 +801,7 @@ static void ep_free(struct eventpoll *ep)* Called with &file->f_lock held, * returns with it released */ -static void __ep_remove_file (struct eventpoll *ep,struct epitem *epi, +static void ep_remove_file (struct eventpoll *ep,struct epitem *epi,struct file *file) {struct epitems_head *to_free =NULL ;
@@ -825,7 +825,7 @@ static void __ep_remove_file(struct eventpoll *ep, struct epitem *epi,free_ephead (to_free); } -static bool __ep_remove_epi (struct eventpoll *ep,struct epitem *epi) +static bool ep_remove_epi (struct eventpoll *ep,struct epitem *epi) {lockdep_assert_held (&ep->mtx);
@@ -871,9 +871,9 @@ static void ep_remove_safe(struct eventpoll *ep, struct epitem *epi)spin_unlock (&file->f_lock);return ; } -__ep_remove_file (ep, epi, file); +ep_remove_file (ep, epi, file); -if (__ep_remove_epi (ep, epi)) +if (ep_remove_epi (ep, epi))WARN_ON_ONCE (ep_refcount_dec_and_test (ep)); }
@@ -1118,8 +1118,8 @@ void eventpoll_release_file(struct file *file)ep_unregister_pollwait (ep, epi);spin_lock (&file->f_lock); -__ep_remove_file (ep, epi, file); - dispose =__ep_remove_epi (ep, epi); +ep_remove_file (ep, epi, file); + dispose =ep_remove_epi (ep, epi);mutex_unlock (&ep->mtx);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 16/204 eventpoll: rename ep_remove_safe() back to ep_remove()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <[email protected]>
[ Upstream commit 0bade234723e40e4937be912e105785d6a51464e ]
The current name is just confusing and doesn't clarify anything.
Stable-dep-of: a6dc643c6931 ("eventpoll: fix ep_remove struct eventpoll / struct file UAF")
--- fs/eventpoll.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
fs/eventpoll.c +8 -8
@@ -853,7 +853,7 @@ static bool ep_remove_epi(struct eventpoll *ep, struct epitem *epi)/* * ep_remove variant for callers owing an additional reference to the ep */ -static void ep_remove_safe (struct eventpoll *ep,struct epitem *epi) +static void ep_remove (struct eventpoll *ep,struct epitem *epi) {struct file *file = epi->ffd.file;
@@ -900,7 +900,7 @@ static void ep_clear_and_put(struct eventpoll *ep)/* * Walks through the whole tree and try to free each "struct epitem". -* Note that ep_remove_safe() will not remove the epitem in case of a +* Note that ep_remove() will not remove the epitem in case of a * racing eventpoll_release_file(); the latter will do the removal. * At this point we are sure no poll callbacks will be lingering around. * Since we still own a reference to the eventpoll struct, the loop can't
@@ -909,7 +909,7 @@ static void ep_clear_and_put(struct eventpoll *ep)for (rbp =rb_first_cached (&ep->rbr); rbp; rbp = next) { next =rb_next (rbp); epi = rb_entry(rbp, struct epitem, rbn); -ep_remove_safe (ep, epi); +ep_remove (ep, epi);cond_resched (); }
@@ -1602,21 +1602,21 @@ static int ep_insert(struct eventpoll *ep, const struct epoll_event *event,mutex_unlock (&tep->mtx);/* -* ep_remove_safe() calls in the later error paths can't lead to +* ep_remove() calls in the later error paths can't lead to * ep_free() as the ep file itself still holds an ep reference. */ ep_get (ep);/* now check if we've created too many backpaths */ if (unlikely (full_check &&reverse_path_check ())) { -ep_remove_safe (ep, epi); +ep_remove (ep, epi);return -EINVAL ; }if (epi->event.events &EPOLLWAKEUP ) { error =ep_create_wakeup_source (epi);if (error) { -ep_remove_safe (ep, epi); +ep_remove (ep, epi);return error; } }
@@ -1640,7 +1640,7 @@ static int ep_insert(struct eventpoll *ep, const struct epoll_event *event,* high memory pressure. */ if (unlikely (!epq.epi)) { -ep_remove_safe (ep, epi); +ep_remove (ep, epi);return -ENOMEM ; }
@@ -2329,7 +2329,7 @@ int do_epoll_ctl(int epfd, int op, int fd, struct epoll_event *epds,* The eventpoll itself is still alive: the refcount * can't go to zero here. */ -ep_remove_safe (ep, epi); +ep_remove (ep, epi); error =0 ; }else { error = -ENOENT ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 17/204 eventpoll: move epi_fget() up
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <[email protected]>
[ Upstream commit 86e87059e6d1fd5115a31949726450ed03c1073b ]
We'll need it when removing files so move it up. No functional change.
Stable-dep-of: a6dc643c6931 ("eventpoll: fix ep_remove struct eventpoll / struct file UAF") [file_ref_get(&file->f_ref) from original commit left as
atomic_long_inc_not_zero(&file->f_count) due to v6.12.y missing commit 90ee6ed776c0 ("fs: port files to file_ref") and its dependent commit 08ef26ea9ab3 ("fs: add file_ref")]
--- fs/eventpoll.c | 56 +++++++++++++++++++++++++------------------------- 1 file changed, 28 insertions(+), 28 deletions(-)
fs/eventpoll.c +28 -28
@@ -797,6 +797,34 @@ static void ep_free(struct eventpoll *ep)kfree_rcu (ep, rcu); } +/* +* The ffd.file pointer may be in the process of being torn down due to +* being closed, but we may not have finished eventpoll_release() yet. +* +* Normally, even with the atomic_long_inc_not_zero, the file may have +* been free'd and then gotten re-allocated to something else (since +* files are not RCU-delayed, they are SLAB_TYPESAFE_BY_RCU). +* +* But for epoll, users hold the ep->mtx mutex, and as such any file in +* the process of being free'd will block in eventpoll_release_file() +* and thus the underlying file allocation will not be free'd, and the +* file re-use cannot happen. +* +* For the same reason we can avoid a rcu_read_lock() around the +* operation - 'ffd.file' cannot go away even if the refcount has +* reached zero (but we must still not call out to ->poll() functions +* etc). +*/ +static struct file *epi_fget (const struct epitem *epi) +{ +struct file *file; + + file = epi->ffd.file; +if (!atomic_long_inc_not_zero (&file->f_count)) + file =NULL ; +return file; +} +/* * Called with &file->f_lock held, * returns with it released
@@ -989,34 +1017,6 @@ static __poll_t __ep_eventpoll_poll(struct file *file, poll_table *wait, int depreturn res; } -/* -* The ffd.file pointer may be in the process of being torn down due to -* being closed, but we may not have finished eventpoll_release() yet. -* -* Normally, even with the atomic_long_inc_not_zero, the file may have -* been free'd and then gotten re-allocated to something else (since -* files are not RCU-delayed, they are SLAB_TYPESAFE_BY_RCU). -* -* But for epoll, users hold the ep->mtx mutex, and as such any file in -* the process of being free'd will block in eventpoll_release_file() -* and thus the underlying file allocation will not be free'd, and the -* file re-use cannot happen. -* -* For the same reason we can avoid a rcu_read_lock() around the -* operation - 'ffd.file' cannot go away even if the refcount has -* reached zero (but we must still not call out to ->poll() functions -* etc). -*/ -static struct file *epi_fget (const struct epitem *epi) -{ -struct file *file; - - file = epi->ffd.file; -if (!atomic_long_inc_not_zero (&file->f_count)) - file =NULL ; -return file; -} -/* * Differs from ep_eventpoll_poll() in that internal callers already have * the ep->mtx so we need to start from depth=1, such that mutex_lock_nested()
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 18/204 eventpoll: fix ep_remove struct eventpoll / struct file UAF
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <[email protected]>
[ Upstream commit a6dc643c69311677c574a0f17a3f4d66a5f3744b ]
ep_remove() (via ep_remove_file()) cleared file->f_ep under file->f_lock but then kept using @file inside the critical section (is_file_epoll(), hlist_del_rcu() through the head, spin_unlock). A concurrent __fput() taking the eventpoll_release() fastpath in that window observed the transient NULL, skipped eventpoll_release_file() and ran to f_op->release / file_free().
For the epoll-watches-epoll case, f_op->release is ep_eventpoll_release() -> ep_clear_and_put() -> ep_free(), which kfree()s the watched struct eventpoll. Its embedded ->refs hlist_head is exactly where epi->fllink.pprev points, so the subsequent hlist_del_rcu()'s "*pprev = next" scribbles into freed kmalloc-192 memory.
In addition, struct file is SLAB_TYPESAFE_BY_RCU, so the slot backing @file could be recycled by alloc_empty_file() -- reinitializing f_lock and f_ep -- while ep_remove() is still nominally inside that lock. The upshot is an attacker-controllable kmem_cache_free() against the wrong slab cache.
Pin @file via epi_fget() at the top of ep_remove() and gate the critical section on the pin succeeding. With the pin held @file cannot reach refcount zero, which holds __fput() off and transitively keeps the watched struct eventpoll alive across the hlist_del_rcu() and the f_lock use, closing both UAFs.
If the pin fails @file has already reached refcount zero and its __fput() is in flight. Because we bailed before clearing f_ep, that path takes the eventpoll_release() slow path into eventpoll_release_file() and blocks on ep->mtx until the waiter side's ep_clear_and_put() drops it. The bailed epi's share of ep->refcount stays intact, so the trailing ep_refcount_dec_and_test() in ep_clear_and_put() cannot free the eventpoll out from under eventpoll_release_file(); the orphaned epi is then cleaned up there.
A successful pin also proves we are not racing eventpoll_release_file() on this epi, so drop the now-redundant re-check of epi->dying under f_lock. The cheap lockless READ_ONCE(epi->dying) fast-path bailout stays.
--- fs/eventpoll.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-)
fs/eventpoll.c +10 -6
@@ -883,22 +883,26 @@ static bool ep_remove_epi(struct eventpoll *ep, struct epitem *epi)*/ static void ep_remove (struct eventpoll *ep,struct epitem *epi) { -struct file *file = epi->ffd.file; + struct file *file __free(fput) = NULL;lockdep_assert_irqs_enabled ();lockdep_assert_held (&ep->mtx);ep_unregister_pollwait (ep, epi); -/* sync with eventpoll_release_file() */ +/* cheap sync with eventpoll_release_file() */ if (unlikely (READ_ONCE (epi->dying)))return ; -spin_lock (&file->f_lock); -if (epi->dying) { -spin_unlock (&file->f_lock); +/* +* If we manage to grab a reference it means we're not in +* eventpoll_release_file() and aren't going to be. +*/ + file =epi_fget (epi); +if (!file)return ; - } + +spin_lock (&file->f_lock);ep_remove_file (ep, epi, file);if (ep_remove_epi (ep, epi))
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 19/204 iio: light: bh1780: fix PM runtime leak on error path
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Antoniu Miclaus <[email protected]>
commit dd72e6c3cdea05cad24e99710939086f7a113fb5 upstream.
Move pm_runtime_put_autosuspend() before the error check to ensure the PM runtime reference count is always decremented after pm_runtime_get_sync(), regardless of whether the read operation succeeds or fails.
--- drivers/iio/light/bh1780.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/iio/light/bh1780.c +2 -2
@@ -109,10 +109,10 @@ static int bh1780_read_raw(struct iio_dev *indio_dev,case IIO_LIGHT :pm_runtime_get_sync (&bh1780->client->dev); value =bh1780_read_word (bh1780,BH1780_REG_DLOW ); -if (value <0 ) -return value;pm_runtime_mark_last_busy (&bh1780->client->dev);pm_runtime_put_autosuspend (&bh1780->client->dev); +if (value <0 ) +return value; *val = value;return IIO_VAL_INT ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 20/204 net: Drop the lock in skb_may_tx_timestamp()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sebastian Andrzej Siewior <[email protected]>
[ Upstream commit 983512f3a87fd8dc4c94dfa6b596b6e57df5aad7 ]
skb_may_tx_timestamp() may acquire sock::sk_callback_lock. The lock must not be taken in IRQ context, only softirq is okay. A few drivers receive the timestamp via a dedicated interrupt and complete the TX timestamp from that handler. This will lead to a deadlock if the lock is already write-locked on the same CPU.
Taking the lock can be avoided. The socket (pointed by the skb) will remain valid until the skb is released. The ->sk_socket and ->file member will be set to NULL once the user closes the socket which may happen before the timestamp arrives.
If we happen to observe the pointer while the socket is closing but before the pointer is set to NULL then we may use it because both pointer (and the file's cred member) are RCU freed.
Drop the lock. Use READ_ONCE() to obtain the individual pointer. Add a matching WRITE_ONCE() where the pointer are cleared.
[adapted sk_set_socket() in include/net/sock.h to fix the conflict from
not having commit 5d6b58c932ec ("net: lockless sock_i_ino()") and the additional previous changes required by it.
It comes down to just now having the lines of
if (sock) {
WRITE_ONCE(sk->sk_uid, SOCK_INODE(sock)->i_uid);
WRITE_ONCE(sk->sk_ino, SOCK_INODE(sock)->i_ino);
} below the changed line.
I've tested this on a device running an nfs-root and did some additional network stress-testing.]
--- include/net/sock.h | 2 +- net/core/skbuff.c | 23 ++++++++++++++++++----- net/socket.c | 2 +- 3 files changed, 20 insertions(+), 7 deletions(-)
include/net/sock.h +1 -1
@@ -2040,7 +2040,7 @@ static inline int sk_rx_queue_get(const struct sock *sk)static inline void sk_set_socket (struct sock *sk,struct socket *sock) { - sk->sk_socket = sock; +WRITE_ONCE (sk->sk_socket, sock); }static inline wait_queue_head_t *sk_sleep (struct sock *sk)
net/core/skbuff.c +18 -5
@@ -5525,15 +5525,28 @@ static void __skb_complete_tx_timestamp(struct sk_buff *skb,static bool skb_may_tx_timestamp (struct sock *sk,bool tsonly) { -bool ret; +struct socket *sock; +struct file *file; +bool ret = false;if (likely (READ_ONCE (sysctl_tstamp_allow_data) || tsonly))return true; -read_lock_bh (&sk->sk_callback_lock); - ret = sk->sk_socket && sk->sk_socket->file && -file_ns_capable (sk->sk_socket->file, &init_user_ns,CAP_NET_RAW ); -read_unlock_bh (&sk->sk_callback_lock); +/* The sk pointer remains valid as long as the skb is. The sk_socket and +* file pointer may become NULL if the socket is closed. Both structures +* (including file->cred) are RCU freed which means they can be accessed +* within a RCU read section. +*/ +rcu_read_lock (); + sock =READ_ONCE (sk->sk_socket); +if (!sock) + goto out; + file =READ_ONCE (sock->file); +if (!file) + goto out; + ret =file_ns_capable (file, &init_user_ns,CAP_NET_RAW ); +out: +rcu_read_unlock ();return ret; }
net/socket.c +1 -1
@@ -652,7 +652,7 @@ static void __sock_release(struct socket *sock, struct inode *inode)iput (SOCK_INODE (sock));return ; } - sock->file =NULL ; +WRITE_ONCE (sock->file,NULL ); }/**
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 21/204 ip6_vti: set netns_immutable on the fallback device.
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit d289d5307762d1838aaece22c6b6fcad9e8865f9 ]
john1988 and Noam Rathaus reported that vti6_init_net() does not set the netns_immutable flag on the per-netns fallback tunnel device (ip6_vti0).
Other similar tunnel drivers (like ip6_tunnel, sit, ip6_gre, and ip_tunnel) correctly set this flag during their fallback device initialization to prevent them from being moved to another network namespace.
[Salvatore Bonaccorso: Backport for version without 0c493da86374 ("net:
rename netns_local to netns_immutable") in v6.15-rc1 and use netns_local.]
--- net/ipv6/ip6_vti.c | 1 + 1 file changed, 1 insertion(+)
net/ipv6/ip6_vti.c +1 -0
@@ -1159,6 +1159,7 @@ static int __net_init vti6_init_net(struct net *net) goto err_alloc_dev;dev_net_set (ip6n->fb_tnl_dev, net); ip6n->fb_tnl_dev->rtnl_link_ops = &vti6_link_ops; + ip6n->fb_tnl_dev->netns_local = true; err =vti6_fb_tnl_dev_init (ip6n->fb_tnl_dev);if (err <0 )
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 22/204 Reapply "selftest/ptp: update ptp selftest to exercise the gettimex options"
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Petr Machata <[email protected]>
This reverts commit 6b2176a5c99b33f3c4acc04faadaa9c75da7b163, which in turn reverts commit fa361565a7275cc43c6ca1abec9ec4fcc9ec51f1, which is commit 3d07b691ee707c00afaf365440975e81bb96cd9b upstream.
The reason for the original revert was that struct ptp_sys_offset_extended does not contain the field clock_id in 6.12.y. However the claim was false:
6.12.y does in fact contain the field. Reapply therefore the original patch.
--- tools/testing/selftests/ptp/testptp.c | 62 +++++++++++++++++++++++++++++++--- 1 file changed, 57 insertions(+), 5 deletions(-)
tools/testing/selftests/ptp/testptp.c +57 -5
@@ -147,6 +147,7 @@ static void usage(char *progname)" -T val set the ptp clock time to 'val' seconds\n" " -x val get an extended ptp clock time with the desired number of samples (up to %d)\n" " -X get a ptp clock cross timestamp\n" +" -y val pre/post tstamp timebase to use {realtime|monotonic|monotonic-raw}\n" " -z test combinations of rising/falling external time stamp flags\n" , progname, PTP_MAX_SAMPLES); }
@@ -191,6 +192,7 @@ int main(int argc, char *argv[])int readonly =0 ;int settime =0 ;int channel =-1 ; +clockid_t ext_clockid =CLOCK_REALTIME ;int64_t t1, t2, tp;int64_t interval, offset;
@@ -200,7 +202,7 @@ int main(int argc, char *argv[]) progname =strrchr (argv[0 ],'/' ); progname = progname ?1 +progname : argv[0 ]; -while (EOF != (c =getopt (argc, argv,"cd:e:f:F:ghH:i:k:lL:n:o:p:P:rsSt:T:w:x:Xz" ))) { +while (EOF != (c =getopt (argc, argv,"cd:e:f:F:ghH:i:k:lL:n:o:p:P:rsSt:T:w:x:Xy:z" ))) {switch (c) {case 'c' : capabilities =1 ;
@@ -283,6 +285,21 @@ int main(int argc, char *argv[])case 'X' : getcross =1 ;break ; +case 'y' : +if (!strcasecmp (optarg,"realtime" )) + ext_clockid =CLOCK_REALTIME ; +else if (!strcasecmp (optarg,"monotonic" )) + ext_clockid =CLOCK_MONOTONIC ; +else if (!strcasecmp (optarg,"monotonic-raw" )) + ext_clockid =CLOCK_MONOTONIC_RAW ; +else { +fprintf (stderr, +"type needs to be realtime, monotonic or monotonic-raw; was given %s\n" , + optarg); +return -1 ; + } +break ; +case 'z' : flagtest =1 ;break ;
@@ -575,6 +592,7 @@ int main(int argc, char *argv[]) } soe->n_samples = getextended; + soe->clockid = ext_clockid;if (ioctl (fd,PTP_SYS_OFFSET_EXTENDED , soe)) {perror ("PTP_SYS_OFFSET_EXTENDED" );
@@ -583,12 +601,46 @@ int main(int argc, char *argv[]) getextended);for (i =0 ; i < getextended; i++) { -printf ("sample #%2d: system time before: %lld.%09u\n" , - i, soe->ts[i][0 ].sec, soe->ts[i][0 ].nsec); +switch (ext_clockid) { +case CLOCK_REALTIME : +printf ("sample #%2d: real time before: %lld.%09u\n" , + i, soe->ts[i][0 ].sec, + soe->ts[i][0 ].nsec); +break ; +case CLOCK_MONOTONIC : +printf ("sample #%2d: monotonic time before: %lld.%09u\n" , + i, soe->ts[i][0 ].sec, + soe->ts[i][0 ].nsec); +break ; +case CLOCK_MONOTONIC_RAW : +printf ("sample #%2d: monotonic-raw time before: %lld.%09u\n" , + i, soe->ts[i][0 ].sec, + soe->ts[i][0 ].nsec); +break ; +default : +break ; + }printf (" phc time: %lld.%09u\n" , soe->ts[i][1 ].sec, soe->ts[i][1 ].nsec); -printf (" system time after: %lld.%09u\n" , - soe->ts[i][2 ].sec, soe->ts[i][2 ].nsec); +switch (ext_clockid) { +case CLOCK_REALTIME : +printf (" real time after: %lld.%09u\n" , + soe->ts[i][2 ].sec, + soe->ts[i][2 ].nsec); +break ; +case CLOCK_MONOTONIC : +printf (" monotonic time after: %lld.%09u\n" , + soe->ts[i][2 ].sec, + soe->ts[i][2 ].nsec); +break ; +case CLOCK_MONOTONIC_RAW : +printf (" monotonic-raw time after: %lld.%09u\n" , + soe->ts[i][2 ].sec, + soe->ts[i][2 ].nsec); +break ; +default : +break ; + } } }
Greg Kroah-Hartman
maintainerPATCH 23/204 debugobjects: Allow to refill the pool before SYSTEM_SCHEDULING
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sebastian Andrzej Siewior <[email protected]>
commit 06e0ae988f6e3499785c407429953ade19c1096b upstream.
The pool of free objects is refilled on several occasions such as object initialisation. On PREEMPT_RT refilling is limited to preemptible sections due to sleeping locks used by the memory allocator. The system boots with disabled interrupts so the pool can not be refilled.
If too many objects are initialized and the pool gets empty then debugobjects disables itself.
Refiling can also happen early in the boot with disabled interrupts as long as the scheduler is not operational. If the scheduler can not preempt a task then a sleeping lock can not be contended.
Allow to additionally refill the pool if the scheduler is not operational.
--- lib/debugobjects.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
lib/debugobjects.c +1 -1
@@ -604,7 +604,7 @@ static void debug_objects_fill_pool(void)* raw_spinlock_t are basically the same type and this lock-type * inversion works just fine. */ -if (!IS_ENABLED (CONFIG_PREEMPT_RT ) ||preemptible ()) { +if (!IS_ENABLED (CONFIG_PREEMPT_RT ) ||preemptible () || system_state <SYSTEM_SCHEDULING ) {/* * Annotate away the spinlock_t inside raw_spinlock_t warning * by temporarily raising the wait-type to WAIT_SLEEP, matching
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 24/204 debugobjects: Use LD_WAIT_CONFIG instead of LD_WAIT_SLEEP
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sebastian Andrzej Siewior <[email protected]>
commit 37de2dbc318ee10577c1c2704de5a803e75e55a2 upstream.
fill_pool_map is used to suppress nesting violations caused by acquiring a spinlock_t (from within the memory allocator) while holding a raw_spinlock_t. The used annotation is wrong.
LD_WAIT_SLEEP is for always sleeping lock types such as mutex_t. LD_WAIT_CONFIG is for lock type which are sleeping while spinning on PREEMPT_RT such as spinlock_t.
Use LD_WAIT_CONFIG as override.
--- lib/debugobjects.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
lib/debugobjects.c +2 -2
@@ -607,10 +607,10 @@ static void debug_objects_fill_pool(void)if (!IS_ENABLED (CONFIG_PREEMPT_RT ) ||preemptible () || system_state <SYSTEM_SCHEDULING ) {/* * Annotate away the spinlock_t inside raw_spinlock_t warning -* by temporarily raising the wait-type to WAIT_SLEEP, matching +* by temporarily raising the wait-type to LD_WAIT_CONFIG, matching * the preemptible() condition above. */ - static DEFINE_WAIT_OVERRIDE_MAP(fill_pool_map, LD_WAIT_SLEEP); + static DEFINE_WAIT_OVERRIDE_MAP(fill_pool_map, LD_WAIT_CONFIG);lock_map_acquire_try (&fill_pool_map);fill_pool ();lock_map_release (&fill_pool_map);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 25/204 debugobjects: Do not fill_pool() if pi_blocked_on
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Helen Koike <[email protected]>
commit 5f41161059fd0f1bbf18c90f3180e38cc45a14eb upstream.
On RT enabled kernels, fill_pool() ends up calling rtlock_lock(), which asserts if current::pi_blocked_on is set, because a task can obviously only block on one lock as otherwise the priority inheritenace chain gets corrupted.
Prevent this by expanding the conditional to take current::pi_blocked_on into account.
--- lib/debugobjects.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-)
lib/debugobjects.c +14 -4
@@ -596,15 +596,25 @@ static struct debug_obj *lookup_object_or_alloc(void *addr, struct debug_bucketreturn NULL ; } +static inline bool debug_objects_is_pi_blocked_on (void ) +{ +#ifdef CONFIG_RT_MUTEXES +return current->pi_blocked_on !=NULL ; +#else +return false; +#endif +} +static void debug_objects_fill_pool (void ) {/* * On RT enabled kernels the pool refill must happen in preemptible -* context -- for !RT kernels we rely on the fact that spinlock_t and -* raw_spinlock_t are basically the same type and this lock-type -* inversion works just fine. +* context and not enqueued on an rt_mutex -- for !RT kernels we rely +* on the fact that spinlock_t and raw_spinlock_t are basically the +* same type and this lock-type inversion works just fine. */ -if (!IS_ENABLED (CONFIG_PREEMPT_RT ) ||preemptible () || system_state <SYSTEM_SCHEDULING ) { +if (!IS_ENABLED (CONFIG_PREEMPT_RT ) || system_state <SYSTEM_SCHEDULING || + (preemptible () && !debug_objects_is_pi_blocked_on ())) {/* * Annotate away the spinlock_t inside raw_spinlock_t warning * by temporarily raising the wait-type to LD_WAIT_CONFIG, matching
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 26/204 debugobjects: Dont call fill_pool() in early boot hardirq context
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Waiman Long <[email protected]>
commit 0d046ae106255cba5eb83b23f78ee93f3620247d upstream.
When booting a debug PREEMPT_RT kernel on an ARM64 system, a "inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-W} usage" lockdep warning message was reported to the console.
During early boot, interrupts are enabled before the scheduler is enabled. In this window (before SYSTEM_SCHEDULING is set) interrupts can fire and in the hard interrupt context handler attempt to fill the pool
This can lead to a deadlock when the interrupt occurred when the interrupt hits a region which holds a lock that is required to be taken in the allocation path.
Add a new can_fill_pool() helper and reorder the exception rule and forbid this scenario by excluding allocations from hard interrupt context.
--- lib/debugobjects.c | 44 ++++++++++++++++++++++++++++++++++++-------- 1 file changed, 36 insertions(+), 8 deletions(-)
lib/debugobjects.c +36 -8
@@ -605,20 +605,48 @@ static inline bool debug_objects_is_pi_blocked_on(void)#endif } -static void debug_objects_fill_pool (void ) +static inline bool can_fill_pool (void ) {/* -* On RT enabled kernels the pool refill must happen in preemptible -* context and not enqueued on an rt_mutex -- for !RT kernels we rely -* on the fact that spinlock_t and raw_spinlock_t are basically the -* same type and this lock-type inversion works just fine. +* On !RT enabled kernels there are no restrictions and spinlock_t and +* raw_spinlock_t are the same types. +*/ +if (!IS_ENABLED (CONFIG_PREEMPT_RT )) +return true; + +/* +* On RT enabled kernels, the task must not be blocked on a lock as +* that could corrupt the PI state when blocking on a lock in the +* allocation path. +*/ +if (debug_objects_is_pi_blocked_on ()) +return false; + +/* +* On RT enabled kernels the pool refill should happen in preemptible +* context. */ -if (!IS_ENABLED (CONFIG_PREEMPT_RT ) || system_state <SYSTEM_SCHEDULING || - (preemptible () && !debug_objects_is_pi_blocked_on ())) { +if (preemptible ()) +return true; + +/* +* Though during system boot before scheduling is set up, preemption is +* disabled and the pool can get exhausted. Before scheduling is active +* a task cannot be blocked on a sleeping lock, but it might hold a lock +* and if interrupted then hard interrupt context might run into a lock +* inversion. So exclude hard interrupt context from allocations before +* scheduling is active. +*/ +return system_state <SYSTEM_SCHEDULING && !in_hardirq (); +} + +static void debug_objects_fill_pool (void ) +{ +if (can_fill_pool ()) {/* * Annotate away the spinlock_t inside raw_spinlock_t warning * by temporarily raising the wait-type to LD_WAIT_CONFIG, matching -* the preemptible() condition above. +* the preemptible() condition in can_fill_pool(). */ static DEFINE_WAIT_OVERRIDE_MAP(fill_pool_map, LD_WAIT_CONFIG);lock_map_acquire_try (&fill_pool_map);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 27/204 RDMA/bnxt_re: zero shared page before exposing to userspace
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lord Ulf Henrik Holmberg <[email protected]>
commit f6b079629becfa977f9c51fe53ad2e6dcc55ef44 upstream.
bnxt_re_alloc_ucontext() allocates uctx->shpg via __get_free_page(GFP_KERNEL). The buddy allocator does not zero pages without __GFP_ZERO, so the page contains stale kernel data from whatever object most recently freed it.
The page is then mapped into userspace via vm_insert_page() under BNXT_RE_MMAP_SH_PAGE in bnxt_re_mmap(). The driver only ever writes 4 bytes (a u32 AVID) at offset BNXT_RE_AVID_OFFT (0x10) inside bnxt_re_create_ah(); the remaining 4092 bytes of the page are exposed to userspace unsanitised, leaking kernel memory contents.
Any user with access to /dev/infiniband/uverbsX on a host with a bnxt_re device (typically rdma group membership) can read this data via a single mmap() at pgoff 0 after IB_USER_VERBS_CMD_GET_CONTEXT.
Other shared pages in the same file already use get_zeroed_page() correctly:
drivers/infiniband/hw/bnxt_re/ib_verbs.c
srq->uctx_srq_page = (void *)get_zeroed_page(GFP_KERNEL);
cq->uctx_cq_page = (void *)get_zeroed_page(GFP_KERNEL);uctx->shpg is the only outlier. Bring it in line with the existing convention by switching to get_zeroed_page().
--- drivers/infiniband/hw/bnxt_re/ib_verbs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/hw/bnxt_re/ib_verbs.c +1 -1
@@ -4231,7 +4231,7 @@ int bnxt_re_alloc_ucontext(struct ib_uco uctx->rdev = rdev; - uctx->shpg = (void *)__get_free_page (GFP_KERNEL ); + uctx->shpg = (void *)get_zeroed_page (GFP_KERNEL );if (!uctx->shpg) { rc = -ENOMEM ; goto fail;
Greg Kroah-Hartman
maintainerPATCH 28/204 i2c: stub: Reject I2C block transfers with invalid length
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Weiming Shi <[email protected]>
commit 6036b5067a8199ba7a2dc7b377d4b9dd276d5f9e upstream.
The I2C_SMBUS_I2C_BLOCK_DATA case in stub_xfer() uses data->block[0] as the transfer length. The existing check only clamps it to avoid overrunning the chip->words[256] register array, but does not validate it against I2C_SMBUS_BLOCK_MAX (32), which is the limit of the union i2c_smbus_data.block buffer (34 bytes total). The driver is a development/test tool (CONFIG_I2C_STUB=m, not built by default) that must be loaded with a chip_addr= parameter.
A local user with access to /dev/i2c-* can issue an I2C_SMBUS ioctl with I2C_SMBUS_I2C_BLOCK_DATA and data->block[0] > 32, causing stub_xfer() to read or write past the end of the union i2c_smbus_data.block buffer:
BUG: KASAN: stack-out-of-bounds in stub_xfer (drivers/i2c/i2c-stub.c:223) Read of size 1 at addr ffff88800abcfd92 by task exploit/81 Call Trace: <TASK> stub_xfer (drivers/i2c/i2c-stub.c:223) __i2c_smbus_xfer (drivers/i2c/i2c-core-smbus.c:593) i2c_smbus_xfer (drivers/i2c/i2c-core-smbus.c:536) i2cdev_ioctl_smbus (drivers/i2c/i2c-dev.c:391) i2cdev_ioctl (drivers/i2c/i2c-dev.c:478) __x64_sys_ioctl (fs/ioctl.c:583) do_syscall_64 (arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) </TASK>
The bug exists because i2c-stub implements .smbus_xfer directly, bypassing the I2C_SMBUS_BLOCK_MAX validation in i2c_smbus_xfer_emulated(). The I2C_SMBUS_BLOCK_DATA case in the same function correctly validates against I2C_SMBUS_BLOCK_MAX, but the I2C_SMBUS_I2C_BLOCK_DATA case does not.
Fix by rejecting transfers with data->block[0] == 0 or data->block[0] > I2C_SMBUS_BLOCK_MAX with -EINVAL, consistent with both the I2C_SMBUS_BLOCK_DATA case in the same function and the I2C_SMBUS_I2C_BLOCK_DATA validation in i2c_smbus_xfer_emulated().
--- drivers/i2c/i2c-stub.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/i2c/i2c-stub.c +5 -0
@@ -214,6 +214,11 @@ static s32 stub_xfer(struct i2c_adapter* We ignore banks here, because banked chips don't use I2C * block transfers */ +if (data->block[0 ] ==0 || + data->block[0 ] >I2C_SMBUS_BLOCK_MAX ) { + ret = -EINVAL ; +break ; + }if (data->block[0 ] >256 - command)/* Avoid overrun */ data->block[0 ] =256 - command; len = data->block[0 ];
Greg Kroah-Hartman
maintainerPATCH 29/204 net: qualcomm: rmnet: fix endpoint use-after-free in rmnet_dellink()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Weiming Shi <[email protected]>
commit d00c953a8f69921f484b629801766da68f27f658 upstream.
rmnet_dellink() removes the endpoint from the hash table with hlist_del_init_rcu() and then immediately frees it with kfree(). However, RCU readers on the receive path (rmnet_rx_handler -> __rmnet_map_ingress_handler) may still hold a reference to the endpoint and dereference ep->egress_dev after the memory has been freed. The endpoint is a kmalloc-32 object, and the stale read at offset 8 corresponds to the egress_dev pointer.
BUG: unable to handle page fault for address: ffffffffde942eef Oops: 0002 [#1] SMP NOPTI CPU: 1 UID: 0 PID: 137 Comm: poc_write Not tainted 7.0.0+ #4 PREEMPTLAZY RIP: 0010:rmnet_vnd_rx_fixup (rmnet_vnd.c:27) Call Trace: <TASK> __rmnet_map_ingress_handler (rmnet_handlers.c:48 rmnet_handlers.c:101) rmnet_rx_handler (rmnet_handlers.c:129 rmnet_handlers.c:235) __netif_receive_skb_core.constprop.0 (net/core/dev.c:6096) __netif_receive_skb_one_core (net/core/dev.c:6208) netif_receive_skb (net/core/dev.c:6467) tun_get_user (drivers/net/tun.c:1955) tun_chr_write_iter (drivers/net/tun.c:2003) vfs_write (fs/read_write.c:688) ksys_write (fs/read_write.c:740) </TASK>
Add an rcu_head field to struct rmnet_endpoint and replace kfree() with kfree_rcu() so the endpoint memory remains valid through the RCU grace period. Also remove the rmnet_vnd_dellink() call and inline only the nr_rmnet_devs decrement, since rmnet_vnd_dellink() would set ep->egress_dev to NULL during the grace period, creating a data race with lockless readers.
--- drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c | 8 ++++---- drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h | 1 + 2 files changed, 5 insertions(+), 4 deletions(-)
drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c +4 -4
@@ -210,8 +210,8 @@ static void rmnet_dellink(struct net_dev ep =rmnet_get_endpoint (real_port, mux_id);if (ep) {hlist_del_init_rcu (&ep->hlnode); -rmnet_vnd_dellink (mux_id, real_port, ep); -kfree (ep); + real_port->nr_rmnet_devs--; +kfree_rcu (ep, rcu); }netdev_upper_dev_unlink (real_dev, dev);
@@ -235,9 +235,9 @@ static void rmnet_force_unassociate_devihash_for_each_safe (port->muxed_ep, bkt_ep, tmp_ep, ep, hlnode) {unregister_netdevice_queue (ep->egress_dev, &list);netdev_upper_dev_unlink (real_dev, ep->egress_dev); -rmnet_vnd_dellink (ep->mux_id, port, ep);hlist_del_init_rcu (&ep->hlnode); -kfree (ep); + port->nr_rmnet_devs--; +kfree_rcu (ep, rcu); }rmnet_unregister_real_device (real_dev);unregister_netdevice_many (&list);
drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h +1 -0
@@ -18,6 +18,7 @@ struct rmnet_endpoint {u8 mux_id;struct net_device *egress_dev;struct hlist_node hlnode; +struct rcu_head rcu; };struct rmnet_egress_agg_params {
Greg Kroah-Hartman
maintainerPATCH 30/204 agp/amd64: Fix broken error propagation in agp_amd64_probe()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mingyu Wang <[email protected]>
commit b08472db93b1ccff84a7adec5779d47f0e9d3a30 upstream.
A NULL pointer dereference was observed in the AMD64 AGP driver when running in a virtualized environment (e.g. qemu/kvm) without a physical AMD northbridge. The crash occurs in amd64_fetch_size() when attempting to dereference the pointer returned by node_to_amd_nb(0).
The root cause of this crash is broken error propagation in agp_amd64_probe(): When no AMD northbridges are found, cache_nbs() correctly returns -ENODEV. However, the probe function erroneously checks the return value against exactly -1, rather than < 0.
As a result, the hardware absence error is masked, allowing the driver to improperly proceed with initialization. It eventually calls agp_add_bridge(), which invokes amd64_fetch_size(). Since the hardware does not exist, node_to_amd_nb(0) returns NULL, leading to a General Protection Fault (GPF) when accessing its ->misc member.
Fix the issue by correcting the error check in agp_amd64_probe() to abort properly when cache_nbs() returns any negative error code. This prevents the driver from erroneously proceeding without hardware, thereby avoiding the subsequent NULL pointer dereference at its source.
--- drivers/char/agp/amd64-agp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/char/agp/amd64-agp.c +1 -1
@@ -546,7 +546,7 @@ static int agp_amd64_probe(struct pci_de/* Fill in the mode register */ pci_read_config_dword (pdev, bridge->capndx+PCI_AGP_STATUS , &bridge->mode); -if (cache_nbs (pdev, cap_ptr) ==-1 ) { +if (cache_nbs (pdev, cap_ptr) <0 ) {agp_put_bridge (bridge);return -ENODEV ; }
Greg Kroah-Hartman
maintainerPATCH 31/204 bpf: Reject sleepable kprobe_multi programs at attach time
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Varun R Mallya <[email protected]>
commit eb7024bfcc5f68ed11ed9dd4891a3073c15f04a8 upstream.
kprobe.multi programs run in atomic/RCU context and cannot sleep. However, bpf_kprobe_multi_link_attach() did not validate whether the program being attached had the sleepable flag set, allowing sleepable helpers such as bpf_copy_from_user() to be invoked from a non-sleepable context.
This causes a "sleeping function called from invalid context" splat:
BUG: sleeping function called from invalid context at ./include/linux/uaccess.h:169 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 1787, name: sudo preempt_count: 1, expected: 0 RCU nest depth: 2, expected: 0
Fix this by rejecting sleepable programs early in bpf_kprobe_multi_link_attach(), before any further processing.
--- kernel/trace/bpf_trace.c | 4 ++++ 1 file changed, 4 insertions(+)
kernel/trace/bpf_trace.c +4 -0
@@ -2943,6 +2943,10 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *prif (!is_kprobe_multi (prog))return -EINVAL ; +/* kprobe_multi is not allowed to be sleepable. */ +if (prog->sleepable) +return -EINVAL ; + flags = attr->link_create.kprobe_multi.flags;if (flags & ~BPF_F_KPROBE_MULTI_RETURN )return -EINVAL ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 32/204 selftests/bpf: Add test to ensure kprobe_multi is not sleepable
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Varun R Mallya <[email protected]>
commit c7cab53f9d5273f0cf2a26bdf178c4e074bdfb50 upstream.
Add a selftest to ensure that kprobe_multi programs cannot be attached using the BPF_F_SLEEPABLE flag. This test succeeds when the kernel rejects attachment of kprobe_multi when the BPF_F_SLEEPABLE flag is set.
--- .../bpf/prog_tests/kprobe_multi_test.c | 35 ++++++++++++++++++- .../bpf/progs/kprobe_multi_sleepable.c | 25 +++++++++++++ 2 files changed, 59 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/bpf/progs/kprobe_multi_sleepable.c
tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c +34 -1
@@ -6,6 +6,7 @@#include "kprobe_multi_override.skel.h" #include "kprobe_multi_session.skel.h" #include "kprobe_multi_session_cookie.skel.h" +#include "kprobe_multi_sleepable.skel.h" #include "bpf/libbpf_internal.h" #include "bpf/hashmap.h"
@@ -216,7 +217,9 @@ static void test_attach_api_syms(void)static void test_attach_api_fails (void ) {LIBBPF_OPTS (bpf_kprobe_multi_opts, opts); +LIBBPF_OPTS (bpf_test_run_opts, topts);struct kprobe_multi *skel =NULL ; +struct kprobe_multi_sleepable *sl_skel =NULL ;struct bpf_link *link =NULL ;unsigned long long addrs[2 ];const char *syms[2 ] = {
@@ -224,7 +227,7 @@ static void test_attach_api_fails(void) "bpf_fentry_test2", };__u64 cookies[2 ]; -int saved_error; +int saved_error, err; addrs[0 ] =ksym_get_addr ("bpf_fentry_test1" ); addrs[1 ] =ksym_get_addr ("bpf_fentry_test2" );
@@ -323,9 +326,39 @@ static void test_attach_api_fails(void)if (!ASSERT_EQ (saved_error, -E2BIG ,"fail_6_error" )) goto cleanup; +/* fail_9 - sleepable kprobe multi should not attach */ + sl_skel =kprobe_multi_sleepable__open (); +if (!ASSERT_OK_PTR (sl_skel,"sleep_skel_open" )) + goto cleanup; + + sl_skel->bss->user_ptr = sl_skel; + + err =bpf_program__set_flags (sl_skel->progs.handle_kprobe_multi_sleepable, +BPF_F_SLEEPABLE ); +if (!ASSERT_OK (err,"sleep_skel_set_flags" )) + goto cleanup; + + err =kprobe_multi_sleepable__load (sl_skel); +if (!ASSERT_OK (err,"sleep_skel_load" )) + goto cleanup; + + link =bpf_program__attach_kprobe_multi_opts (sl_skel->progs.handle_kprobe_multi_sleepable, +"bpf_fentry_test1" ,NULL ); + saved_error = -errno; + +if (!ASSERT_ERR_PTR (link,"fail_9" )) + goto cleanup; + +if (!ASSERT_EQ (saved_error, -EINVAL ,"fail_9_error" )) + goto cleanup; + + err =bpf_prog_test_run_opts (bpf_program__fd (sl_skel->progs.fentry), &topts); +ASSERT_OK (err,"bpf_prog_test_run_opts" ); + cleanup:bpf_link__destroy (link);kprobe_multi__destroy (skel); +kprobe_multi_sleepable__destroy (sl_skel); }static void test_session_skel_api (void )
tools/testing/selftests/bpf/progs/kprobe_multi_sleepable.c new file +25 -0
@@ -0,0 +1,25 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "vmlinux.h" +#include <bpf/bpf_helpers.h> +#include <bpf/bpf_tracing.h> + +void *user_ptr =0 ; + +SEC("kprobe.multi") +int handle_kprobe_multi_sleepable (struct pt_regs *ctx) +{ +int a, err; + + err =bpf_copy_from_user (&a,sizeof (a), user_ptr); +barrier_var (a); +return err; +} + +SEC("fentry/bpf_fentry_test1") +int BPF_PROG (fentry ) +{ +return 0 ; +} + +char _license[] SEC("license") = "GPL";
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 33/204 ACPI: scan: Use async schedule function in acpi_scan_clear_dep_fn()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yicong Yang <[email protected]>
[ Upstream commit 7cf28b3797a81b616bb7eb3e90cf131afc452919 ]
The device object rescan in acpi_scan_clear_dep_fn() is scheduled on a system workqueue which is not guaranteed to be finished before entering userspace. This may cause some key devices to be missing when userspace init task tries to find them. Two issues observed on RISCV platforms:
- Kernel panic due to userspace init cannot have an opened console.
The console device scanning is queued by acpi_scan_clear_dep_queue() and not finished by the time userspace init process running, thus by the time userspace init runs, no console is present.
- Entering rescue shell due to the lack of root devices (PCIe nvme in our case).
Same reason as above, the PCIe host bridge scanning is queued on a system workqueue and finished after init process runs.
The reason is because both devices (console, PCIe host bridge) depend on riscv-aplic irqchip to serve their interrupts (console's wired interrupt and PCI's INTx interrupts). In order to keep the dependency, these devices are scanned and created after initializing riscv-aplic. The riscv-aplic is initialized in device_initcall() and a device scan work is queued via acpi_scan_clear_dep_queue(), which is close to the time userspace init process is run. Since system_dfl_wq is used in acpi_scan_clear_dep_queue() with no synchronization, the issues will happen if userspace init runs before these devices are ready.
The solution is to wait for the queued work to complete before entering userspace init. One possible way would be to use a dedicated workqueue instead of system_dfl_wq, and explicitly flush it somewhere in the initcall stage before entering userspace. Another way is to use async_schedule_dev_nocall() for scanning these devices. It's designed for asynchronous initialization and will work in the same way as before because it's using a dedicated unbound workqueue as well, but the kernel init code calls async_synchronize_full() right before entering userspace init which will wait for the work to complete.
Compared to a dedicated workqueue, the second approach is simpler because the async schedule framework takes care of all of the details. The ACPI code only needs to focus on its job. A dedicated workqueue for this could also be redundant because some platforms don't need acpi_scan_clear_dep_queue() for their device scanning.
--- drivers/acpi/scan.c | 41 +++++++++++++++-------------------------- 1 file changed, 15 insertions(+), 26 deletions(-)
drivers/acpi/scan.c +15 -26
@@ -5,6 +5,7 @@#define pr_fmt (fmt) "ACPI: " fmt +#include <linux/async.h> #include <linux/module.h> #include <linux/init.h> #include <linux/slab.h>
@@ -2364,46 +2365,34 @@ static int acpi_dev_get_next_consumer_dev_cb(struct acpi_dep_data *dep, void *dareturn 0 ; } -struct acpi_scan_clear_dep_work { -struct work_struct work; -struct acpi_device *adev; -}; - -static void acpi_scan_clear_dep_fn (struct work_struct *work) +static void acpi_scan_clear_dep_fn (void *dev,async_cookie_t cookie) { -struct acpi_scan_clear_dep_work *cdw; - - cdw = container_of(work, struct acpi_scan_clear_dep_work, work); +struct acpi_device *adev =to_acpi_device (dev);acpi_scan_lock_acquire (); -acpi_bus_attach (cdw->adev, (void *)true); +acpi_bus_attach (adev, (void *)true);acpi_scan_lock_release (); -acpi_dev_put (cdw->adev); -kfree (cdw); +acpi_dev_put (adev); }static bool acpi_scan_clear_dep_queue (struct acpi_device *adev) { -struct acpi_scan_clear_dep_work *cdw; -if (adev->dep_unmet)return false; - cdw =kmalloc (sizeof (*cdw),GFP_KERNEL ); -if (!cdw) -return false; - - cdw->adev = adev; -INIT_WORK (&cdw->work, acpi_scan_clear_dep_fn);/* -* Since the work function may block on the lock until the entire -* initial enumeration of devices is complete, put it into the unbound -* workqueue. +* Async schedule the deferred acpi_scan_clear_dep_fn() since: +* - acpi_bus_attach() needs to hold acpi_scan_lock which cannot +* be acquired under acpi_dep_list_lock (held here) +* - the deferred work at boot stage is ensured to be finished +* before userspace init task by the async_synchronize_full() +* barrier +* +* Use _nocall variant since it'll return on failure instead of +* run the function synchronously. */ -queue_work (system_unbound_wq, &cdw->work); - -return true; +return async_schedule_dev_nocall (acpi_scan_clear_dep_fn, &adev->dev); }static void acpi_scan_delete_dep_data (struct acpi_dep_data *dep)
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 34/204 regulator: core: fix locking in regulator_resolve_supply() error path
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: André Draszik <[email protected]>
commit 497330b203d2c59c5ff3fa4c34d14494d7203bc3 upstream.
If late enabling of a supply regulator fails in regulator_resolve_supply(), the code currently triggers a lockdep warning:
WARNING: drivers/regulator/core.c:2649 at _regulator_put+0x80/0xa0, CPU#6: kworker/u32:4/596
...
Call trace:
_regulator_put+0x80/0xa0 (P)
regulator_resolve_supply+0x7cc/0xbe0
regulator_register_resolve_supply+0x28/0xb8as the regulator_list_mutex must be held when calling _regulator_put().
To solve this, simply switch to using regulator_put().
While at it, we should also make sure that no concurrent access happens to our rdev while we clear out the supply pointer. Add appropriate locking to ensure that.
While the code in question will be removed altogether in a follow-up commit, I believe it is still beneficial to have this corrected before removal for future reference.
--- drivers/regulator/core.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/regulator/core.c +9 -1
@@ -2116,8 +2116,16 @@ static int regulator_resolve_supply(struif (rdev->use_count) { ret =regulator_enable (rdev->supply);if (ret <0 ) { -_regulator_put (rdev->supply); +struct regulator *supply; + +regulator_lock_two (rdev, rdev->supply->rdev, &ww_ctx); + + supply = rdev->supply; rdev->supply =NULL ; + +regulator_unlock_two (rdev, supply->rdev, &ww_ctx); + +regulator_put (supply); goto out; } }
Greg Kroah-Hartman
maintainerPATCH 35/204 dlm: prevent NPD when writing a positive value to event_done
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thadeu Lima de Souza Cascardo <[email protected]>
commit 8e2bad543eca5c25cd02cbc63d72557934d45f13 upstream.
do_uevent returns the value written to event_done. In case it is a positive value, new_lockspace would undo all the work, and lockspace would not be set. __dlm_new_lockspace, however, would treat that positive value as a success due to commit 8511a2728ab8 ("dlm: fix use count with multiple joins").
Down the line, device_create_lockspace would pass that NULL lockspace to dlm_find_lockspace_local, leading to a NULL pointer dereference.
Treating such positive values as successes prevents the problem. Given this has been broken for so long, this is unlikely to break userspace expectations.
--- fs/dlm/lockspace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/dlm/lockspace.c +1 -1
@@ -576,7 +576,7 @@ static int new_lockspace(const char *namlockspace to start running (via sysfs) in dlm_ls_start(). */ error =do_uevent (ls,1 ); -if (error) +if (error <0 ) goto out_recoverd;/* wait until recovery is successful or failed */
Greg Kroah-Hartman
maintainerPATCH 36/204 xfs: remove the expr argument to XFS_TEST_ERROR
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christoph Hellwig <[email protected]>
[ Upstream commit 807df3227d7674d7957c576551d552acf15bb96f ]
Don't pass expr to XFS_TEST_ERROR. Most calls pass a constant false, and the places that do pass an expression become cleaner by moving it out.
Stable-dep-of: fcf4faba9f98 ("xfs: fix error returns in CoW fork repair")
--- fs/xfs/libxfs/xfs_ag_resv.c | 8 ++++---- fs/xfs/libxfs/xfs_alloc.c | 5 ++--- fs/xfs/libxfs/xfs_attr_leaf.c | 2 +- fs/xfs/libxfs/xfs_bmap.c | 17 ++++++++--------- fs/xfs/libxfs/xfs_btree.c | 2 +- fs/xfs/libxfs/xfs_da_btree.c | 2 +- fs/xfs/libxfs/xfs_dir2.c | 2 +- fs/xfs/libxfs/xfs_exchmaps.c | 4 ++-- fs/xfs/libxfs/xfs_ialloc.c | 2 +- fs/xfs/libxfs/xfs_inode_buf.c | 4 ++-- fs/xfs/libxfs/xfs_inode_fork.c | 3 +-- fs/xfs/libxfs/xfs_refcount.c | 5 ++--- fs/xfs/libxfs/xfs_rmap.c | 2 +- fs/xfs/scrub/cow_repair.c | 2 +- fs/xfs/scrub/repair.c | 2 +- fs/xfs/xfs_attr_item.c | 2 +- fs/xfs/xfs_buf.c | 4 ++-- fs/xfs/xfs_error.c | 5 ++--- fs/xfs/xfs_error.h | 10 +++++----- fs/xfs/xfs_inode.c | 28 +++++++++++++--------------- fs/xfs/xfs_iomap.c | 2 +- fs/xfs/xfs_log.c | 8 ++++---- fs/xfs/xfs_trans_ail.c | 2 +- 23 files changed, 58 insertions(+), 65 deletions(-)
fs/xfs/libxfs/xfs_ag_resv.c +4 -4
@@ -91,9 +91,9 @@ xfs_ag_resv_critical(trace_xfs_ag_resv_critical (pag, type, avail);/* Critically low if less than 10% or max btree height remains. */ -return XFS_TEST_ERROR (avail < orig /10 || - avail < pag->pag_mount->m_agbtree_maxlevels, - pag->pag_mount,XFS_ERRTAG_AG_RESV_CRITICAL ); +return avail < orig /10 || + avail < pag->pag_mount->m_agbtree_maxlevels || +XFS_TEST_ERROR (pag->pag_mount,XFS_ERRTAG_AG_RESV_CRITICAL ); }/*
@@ -201,7 +201,7 @@ __xfs_ag_resv_init(return -EINVAL ; } -if (XFS_TEST_ERROR (false, mp,XFS_ERRTAG_AG_RESV_FAIL )) +if (XFS_TEST_ERROR (mp,XFS_ERRTAG_AG_RESV_FAIL )) error = -ENOSPC ;else error =xfs_dec_fdblocks (mp, hidden_space, true);
fs/xfs/libxfs/xfs_alloc.c +2 -3
@@ -3312,7 +3312,7 @@ xfs_agf_read_verify(xfs_verifier_error (bp, -EFSBADCRC , __this_address); else { fa =xfs_agf_verify (bp); -if (XFS_TEST_ERROR (fa, mp,XFS_ERRTAG_ALLOC_READ_AGF )) +if (fa ||XFS_TEST_ERROR (mp,XFS_ERRTAG_ALLOC_READ_AGF ))xfs_verifier_error (bp, -EFSCORRUPTED , fa); } }
@@ -3986,8 +3986,7 @@ __xfs_free_extent(ASSERT (len !=0 );ASSERT (type !=XFS_AG_RESV_AGFL ); -if (XFS_TEST_ERROR (false, mp, -XFS_ERRTAG_FREE_EXTENT )) +if (XFS_TEST_ERROR (mp,XFS_ERRTAG_FREE_EXTENT ))return -EIO ; error =xfs_free_extent_fix_freelist (tp, pag, &agbp);
fs/xfs/libxfs/xfs_attr_leaf.c +1 -1
@@ -1225,7 +1225,7 @@ xfs_attr3_leaf_to_node(trace_xfs_attr_leaf_to_node (args); -if (XFS_TEST_ERROR (false, mp,XFS_ERRTAG_ATTR_LEAF_TO_NODE )) { +if (XFS_TEST_ERROR (mp,XFS_ERRTAG_ATTR_LEAF_TO_NODE )) { error = -EIO ; goto out; }
fs/xfs/libxfs/xfs_bmap.c +8 -9
@@ -3766,8 +3766,7 @@ xfs_bmap_btalloc(/* Trim the allocation back to the maximum an AG can fit. */ args.maxlen =min (ap->length, mp->m_ag_max_usable); -if (unlikely (XFS_TEST_ERROR (false, mp, -XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT ))) +if (unlikely (XFS_TEST_ERROR (mp,XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT ))) error =xfs_bmap_exact_minlen_extent_alloc (ap, &args);else if ((ap->datatype &XFS_ALLOC_USERDATA ) &&xfs_inode_is_filestream (ap->ip))
@@ -3953,7 +3952,7 @@ xfs_bmapi_read( }if (XFS_IS_CORRUPT (mp, !xfs_ifork_has_extents (ifp)) || -XFS_TEST_ERROR (false, mp,XFS_ERRTAG_BMAPIFORMAT )) { +XFS_TEST_ERROR (mp,XFS_ERRTAG_BMAPIFORMAT )) {xfs_bmap_mark_sick (ip, whichfork);return -EFSCORRUPTED ; }
@@ -4442,7 +4441,7 @@ xfs_bmapi_write( (XFS_BMAPI_PREALLOC | XFS_BMAPI_ZERO));if (XFS_IS_CORRUPT (mp, !xfs_ifork_has_extents (ifp)) || -XFS_TEST_ERROR (false, mp,XFS_ERRTAG_BMAPIFORMAT )) { +XFS_TEST_ERROR (mp,XFS_ERRTAG_BMAPIFORMAT )) {xfs_bmap_mark_sick (ip, whichfork);return -EFSCORRUPTED ; }
@@ -4785,7 +4784,7 @@ xfs_bmapi_remap( (XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC));if (XFS_IS_CORRUPT (mp, !xfs_ifork_has_extents (ifp)) || -XFS_TEST_ERROR (false, mp,XFS_ERRTAG_BMAPIFORMAT )) { +XFS_TEST_ERROR (mp,XFS_ERRTAG_BMAPIFORMAT )) {xfs_bmap_mark_sick (ip, whichfork);return -EFSCORRUPTED ; }
@@ -5873,7 +5872,7 @@ xfs_bmap_collapse_extents(int logflags =0 ;if (XFS_IS_CORRUPT (mp, !xfs_ifork_has_extents (ifp)) || -XFS_TEST_ERROR (false, mp,XFS_ERRTAG_BMAPIFORMAT )) { +XFS_TEST_ERROR (mp,XFS_ERRTAG_BMAPIFORMAT )) {xfs_bmap_mark_sick (ip, whichfork);return -EFSCORRUPTED ; }
@@ -5988,7 +5987,7 @@ xfs_bmap_insert_extents(int logflags =0 ;if (XFS_IS_CORRUPT (mp, !xfs_ifork_has_extents (ifp)) || -XFS_TEST_ERROR (false, mp,XFS_ERRTAG_BMAPIFORMAT )) { +XFS_TEST_ERROR (mp,XFS_ERRTAG_BMAPIFORMAT )) {xfs_bmap_mark_sick (ip, whichfork);return -EFSCORRUPTED ; }
@@ -6092,7 +6091,7 @@ xfs_bmap_split_extent(int i =0 ;if (XFS_IS_CORRUPT (mp, !xfs_ifork_has_extents (ifp)) || -XFS_TEST_ERROR (false, mp,XFS_ERRTAG_BMAPIFORMAT )) { +XFS_TEST_ERROR (mp,XFS_ERRTAG_BMAPIFORMAT )) {xfs_bmap_mark_sick (ip, whichfork);return -EFSCORRUPTED ; }
@@ -6257,7 +6256,7 @@ xfs_bmap_finish_one(trace_xfs_bmap_deferred (bi); -if (XFS_TEST_ERROR (false, tp->t_mountp,XFS_ERRTAG_BMAP_FINISH_ONE )) +if (XFS_TEST_ERROR (tp->t_mountp,XFS_ERRTAG_BMAP_FINISH_ONE ))return -EIO ;switch (bi->bi_type) {
fs/xfs/libxfs/xfs_btree.c +1 -1
@@ -300,7 +300,7 @@ xfs_btree_check_block( fa =__xfs_btree_check_block (cur, block, level, bp);if (XFS_IS_CORRUPT (mp, fa !=NULL ) || -XFS_TEST_ERROR (false, mp,xfs_btree_block_errtag (cur))) { +XFS_TEST_ERROR (mp,xfs_btree_block_errtag (cur))) {if (bp)trace_xfs_btree_corrupt (bp, _RET_IP_);xfs_btree_mark_sick (cur);
fs/xfs/libxfs/xfs_da_btree.c +1 -1
@@ -565,7 +565,7 @@ xfs_da3_split(trace_xfs_da_split (state->args); -if (XFS_TEST_ERROR (false, state->mp,XFS_ERRTAG_DA_LEAF_SPLIT )) +if (XFS_TEST_ERROR (state->mp,XFS_ERRTAG_DA_LEAF_SPLIT ))return -EIO ;/*
fs/xfs/libxfs/xfs_dir2.c +1 -1
@@ -223,7 +223,7 @@ xfs_dir_ino_validate(bool ino_ok =xfs_verify_dir_ino (mp, ino);if (XFS_IS_CORRUPT (mp, !ino_ok) || -XFS_TEST_ERROR (false, mp,XFS_ERRTAG_DIR_INO_VALIDATE )) { +XFS_TEST_ERROR (mp,XFS_ERRTAG_DIR_INO_VALIDATE )) {xfs_warn (mp,"Invalid inode number 0x%Lx" , (unsigned long long ) ino);return -EFSCORRUPTED ;
fs/xfs/libxfs/xfs_exchmaps.c +2 -2
@@ -616,7 +616,7 @@ xfs_exchmaps_finish_one(return error; } -if (XFS_TEST_ERROR (false, tp->t_mountp,XFS_ERRTAG_EXCHMAPS_FINISH_ONE )) +if (XFS_TEST_ERROR (tp->t_mountp,XFS_ERRTAG_EXCHMAPS_FINISH_ONE ))return -EIO ;/* If we still have work to do, ask for a new transaction. */
@@ -880,7 +880,7 @@ xmi_ensure_delta_nextents( &new_nextents))return -EFBIG ; -if (XFS_TEST_ERROR (false, mp,XFS_ERRTAG_REDUCE_MAX_IEXTENTS ) && +if (XFS_TEST_ERROR (mp,XFS_ERRTAG_REDUCE_MAX_IEXTENTS ) && new_nextents >10 )return -EFBIG ;
fs/xfs/libxfs/xfs_ialloc.c +1 -1
@@ -2690,7 +2690,7 @@ xfs_agi_read_verify(xfs_verifier_error (bp, -EFSBADCRC , __this_address); else { fa =xfs_agi_verify (bp); -if (XFS_TEST_ERROR (fa, mp,XFS_ERRTAG_IALLOC_READ_AGI )) +if (fa ||XFS_TEST_ERROR (mp,XFS_ERRTAG_IALLOC_READ_AGI ))xfs_verifier_error (bp, -EFSCORRUPTED , fa); } }
fs/xfs/libxfs/xfs_inode_buf.c +2 -2
@@ -60,8 +60,8 @@ xfs_inode_buf_verify( di_ok =xfs_verify_magic16 (bp, dip->di_magic) &&xfs_dinode_good_version (mp, dip->di_version) &&xfs_verify_agino_or_null (bp->b_pag, unlinked_ino); -if (unlikely (XFS_TEST_ERROR (!di_ok, mp, -XFS_ERRTAG_ITOBP_INOTOBP ))) { +if (unlikely (!di_ok || +XFS_TEST_ERROR (mp,XFS_ERRTAG_ITOBP_INOTOBP ))) {if (readahead) { bp->b_flags &= ~XBF_DONE ;xfs_buf_ioerror (bp, -EIO );
fs/xfs/libxfs/xfs_inode_fork.c +1 -2
@@ -795,8 +795,7 @@ xfs_iext_count_extend(if (nr_exts < ifp->if_nextents)return -EFBIG ; -if (XFS_TEST_ERROR (false, mp,XFS_ERRTAG_REDUCE_MAX_IEXTENTS ) && - nr_exts >10 ) +if (XFS_TEST_ERROR (mp,XFS_ERRTAG_REDUCE_MAX_IEXTENTS ) && nr_exts >10 )return -EFBIG ;if (nr_exts >xfs_iext_max_nextents (has_large, whichfork)) {
fs/xfs/libxfs/xfs_refcount.c +2 -3
@@ -1073,8 +1073,7 @@ xfs_refcount_still_have_space(* refcount continue update "error" has been injected. */ if (cur->bc_refc.nr_ops >2 && -XFS_TEST_ERROR (false, cur->bc_mp, -XFS_ERRTAG_REFCOUNT_CONTINUE_UPDATE )) +XFS_TEST_ERROR (cur->bc_mp,XFS_ERRTAG_REFCOUNT_CONTINUE_UPDATE ))return false;if (cur->bc_refc.nr_ops ==0 )
@@ -1353,7 +1352,7 @@ xfs_refcount_finish_one(trace_xfs_refcount_deferred (mp, ri); -if (XFS_TEST_ERROR (false, mp,XFS_ERRTAG_REFCOUNT_FINISH_ONE )) +if (XFS_TEST_ERROR (mp,XFS_ERRTAG_REFCOUNT_FINISH_ONE ))return -EIO ;/*
fs/xfs/libxfs/xfs_rmap.c +1 -1
@@ -2579,7 +2579,7 @@ xfs_rmap_finish_one(trace_xfs_rmap_deferred (mp, ri); -if (XFS_TEST_ERROR (false, mp,XFS_ERRTAG_RMAP_FINISH_ONE )) +if (XFS_TEST_ERROR (mp,XFS_ERRTAG_RMAP_FINISH_ONE ))return -EIO ;/*
fs/xfs/scrub/cow_repair.c +1 -1
@@ -297,7 +297,7 @@ xrep_cow_find_bad(* on the debugging knob, replace everything in the CoW fork. */ if ((sc->sm->sm_flags &XFS_SCRUB_IFLAG_FORCE_REBUILD ) || -XFS_TEST_ERROR (false, sc->mp,XFS_ERRTAG_FORCE_SCRUB_REPAIR )) { +XFS_TEST_ERROR (sc->mp,XFS_ERRTAG_FORCE_SCRUB_REPAIR )) { error =xrep_cow_mark_file_range (xc, xc->irec.br_startblock, xc->irec.br_blockcount);if (error)
fs/xfs/scrub/repair.c +1 -1
@@ -990,7 +990,7 @@ xrep_will_attempt(return true;/* Let debug users force us into the repair routines. */ -if (XFS_TEST_ERROR (false, sc->mp,XFS_ERRTAG_FORCE_SCRUB_REPAIR )) +if (XFS_TEST_ERROR (sc->mp,XFS_ERRTAG_FORCE_SCRUB_REPAIR ))return true;/* Metadata is corrupt or failed cross-referencing. */
fs/xfs/xfs_attr_item.c +1 -1
@@ -490,7 +490,7 @@ xfs_attr_finish_item(/* Reset trans after EAGAIN cycle since the transaction is new */ args->trans = tp; -if (XFS_TEST_ERROR (false, args->dp->i_mount,XFS_ERRTAG_LARP )) { +if (XFS_TEST_ERROR (args->dp->i_mount,XFS_ERRTAG_LARP )) { error = -EIO ; goto out; }
fs/xfs/xfs_buf.c +2 -2
@@ -1498,7 +1498,7 @@ xfs_buf_bio_end_io(if (!bio->bi_status && (bp->b_flags &XBF_WRITE ) && (bp->b_flags &XBF_ASYNC ) && -XFS_TEST_ERROR (false, bp->b_mount,XFS_ERRTAG_BUF_IOERROR )) +XFS_TEST_ERROR (bp->b_mount,XFS_ERRTAG_BUF_IOERROR )) bio->bi_status =BLK_STS_IOERR ;/*
@@ -2451,7 +2451,7 @@ void xfs_buf_set_ref(struct xfs_buf *bp,* This allows userspace to disrupt buffer caching for debug/testing * purposes. */ -if (XFS_TEST_ERROR (false, bp->b_mount,XFS_ERRTAG_BUF_LRU_REF )) +if (XFS_TEST_ERROR (bp->b_mount,XFS_ERRTAG_BUF_LRU_REF )) lru_ref =0 ;atomic_set (&bp->b_lru_ref, lru_ref);
fs/xfs/xfs_error.c +2 -3
@@ -292,7 +292,6 @@ xfs_errortag_enabled(bool xfs_errortag_test (struct xfs_mount *mp, -const char *expression,const char *file,int line,unsigned int error_tag)
@@ -318,8 +317,8 @@ xfs_errortag_test(return false;xfs_warn_ratelimited (mp, -"Injecting error (%s) at file %s, line %d, on filesystem \"%s\"" , - expression, file, line, mp->m_super->s_id); +"Injecting error at file %s, line %d, on filesystem \"%s\"" , + file, line, mp->m_super->s_id);return true; }
fs/xfs/xfs_error.h +5 -5
@@ -41,10 +41,10 @@ extern void xfs_inode_verifier_error(str#ifdef DEBUG extern int xfs_errortag_init (struct xfs_mount *mp);extern void xfs_errortag_del (struct xfs_mount *mp); -extern bool xfs_errortag_test (struct xfs_mount *mp,const char *expression, -const char *file,int line,unsigned int error_tag); -#define XFS_TEST_ERROR (expr, mp, tag) \ - ((expr) || xfs_errortag_test((mp), #expr, __FILE__, __LINE__, (tag))) +bool xfs_errortag_test (struct xfs_mount *mp,const char *file,int line, +unsigned int error_tag); +#define XFS_TEST_ERROR (mp, tag) \ + xfs_errortag_test((mp), __FILE__, __LINE__, (tag))bool xfs_errortag_enabled (struct xfs_mount *mp,unsigned int tag);#define XFS_ERRORTAG_DELAY (mp, tag) \ do { \
@@ -66,7 +66,7 @@ extern int xfs_errortag_clearall(struct#else #define xfs_errortag_init (mp) (0)#define xfs_errortag_del (mp) -#define XFS_TEST_ERROR (expr, mp, tag) (expr) +#define XFS_TEST_ERROR (mp, tag) (false)#define XFS_ERRORTAG_DELAY (mp, tag) ((void)0)#define xfs_errortag_set (mp, tag, val) (ENOSYS)#define xfs_errortag_add (mp, tag) (ENOSYS)
fs/xfs/xfs_inode.c +13 -15
@@ -2367,37 +2367,35 @@ xfs_iflush(* error handling as the caller will shutdown and fail the buffer. */ error = -EFSCORRUPTED ; -if (XFS_TEST_ERROR (dip->di_magic !=cpu_to_be16 (XFS_DINODE_MAGIC ), - mp,XFS_ERRTAG_IFLUSH_1 )) { +if (dip->di_magic !=cpu_to_be16 (XFS_DINODE_MAGIC ) || +XFS_TEST_ERROR (mp,XFS_ERRTAG_IFLUSH_1 )) {xfs_alert_tag (mp,XFS_PTAG_IFLUSH ,"%s: Bad inode %llu magic number 0x%x, ptr " PTR_FMT , __func__, ip->i_ino,be16_to_cpu (dip->di_magic), dip); goto flush_out; }if (S_ISREG (VFS_I (ip)->i_mode)) { -if (XFS_TEST_ERROR ( - ip->i_df.if_format !=XFS_DINODE_FMT_EXTENTS && - ip->i_df.if_format !=XFS_DINODE_FMT_BTREE , - mp,XFS_ERRTAG_IFLUSH_3 )) { +if ((ip->i_df.if_format !=XFS_DINODE_FMT_EXTENTS && + ip->i_df.if_format !=XFS_DINODE_FMT_BTREE ) || +XFS_TEST_ERROR (mp,XFS_ERRTAG_IFLUSH_3 )) {xfs_alert_tag (mp,XFS_PTAG_IFLUSH ,"%s: Bad regular inode %llu, ptr " PTR_FMT , __func__, ip->i_ino, ip); goto flush_out; } }else if (S_ISDIR (VFS_I (ip)->i_mode)) { -if (XFS_TEST_ERROR ( - ip->i_df.if_format !=XFS_DINODE_FMT_EXTENTS && - ip->i_df.if_format !=XFS_DINODE_FMT_BTREE && - ip->i_df.if_format !=XFS_DINODE_FMT_LOCAL , - mp,XFS_ERRTAG_IFLUSH_4 )) { +if ((ip->i_df.if_format !=XFS_DINODE_FMT_EXTENTS && + ip->i_df.if_format !=XFS_DINODE_FMT_BTREE && + ip->i_df.if_format !=XFS_DINODE_FMT_LOCAL ) || +XFS_TEST_ERROR (mp,XFS_ERRTAG_IFLUSH_4 )) {xfs_alert_tag (mp,XFS_PTAG_IFLUSH ,"%s: Bad directory inode %llu, ptr " PTR_FMT , __func__, ip->i_ino, ip); goto flush_out; } } -if (XFS_TEST_ERROR (ip->i_df.if_nextents +xfs_ifork_nextents (&ip->i_af) > - ip->i_nblocks, mp,XFS_ERRTAG_IFLUSH_5 )) { +if (ip->i_df.if_nextents +xfs_ifork_nextents (&ip->i_af) > + ip->i_nblocks ||XFS_TEST_ERROR (mp,XFS_ERRTAG_IFLUSH_5 )) { xfs_alert_tag(mp, XFS_PTAG_IFLUSH, "%s: detected corrupt incore inode %llu, " "total extents = %llu nblocks = %lld, ptr "PTR_FMT,
@@ -2406,8 +2404,8 @@ xfs_iflush( ip->i_nblocks, ip); goto flush_out; } -if (XFS_TEST_ERROR (ip->i_forkoff > mp->m_sb.sb_inodesize, - mp,XFS_ERRTAG_IFLUSH_6 )) { +if (ip->i_forkoff > mp->m_sb.sb_inodesize || +XFS_TEST_ERROR (mp,XFS_ERRTAG_IFLUSH_6 )) {xfs_alert_tag (mp,XFS_PTAG_IFLUSH ,"%s: bad inode %llu, forkoff 0x%x, ptr " PTR_FMT , __func__, ip->i_ino, ip->i_forkoff, ip);
fs/xfs/xfs_iomap.c +1 -1
@@ -993,7 +993,7 @@ xfs_buffered_write_iomap_begin(return error;if (XFS_IS_CORRUPT (mp, !xfs_ifork_has_extents (&ip->i_df)) || -XFS_TEST_ERROR (false, mp,XFS_ERRTAG_BMAPIFORMAT )) { +XFS_TEST_ERROR (mp,XFS_ERRTAG_BMAPIFORMAT )) {xfs_bmap_mark_sick (ip,XFS_DATA_FORK ); error = -EFSCORRUPTED ; goto out_unlock;
fs/xfs/xfs_log.c +4 -4
@@ -968,8 +968,8 @@ xfs_log_unmount_write(* counters will be recalculated. Refer to xlog_check_unmount_rec for * more details. */ -if (XFS_TEST_ERROR (xfs_fs_has_sickness (mp,XFS_SICK_FS_COUNTERS ), mp, -XFS_ERRTAG_FORCE_SUMMARY_RECALC )) { +if (xfs_fs_has_sickness (mp,XFS_SICK_FS_COUNTERS ) || +XFS_TEST_ERROR (mp,XFS_ERRTAG_FORCE_SUMMARY_RECALC )) {xfs_alert (mp,"%s: will fix summary counters at next mount" , __func__);return ;
@@ -1239,7 +1239,7 @@ xlog_ioend_work(/* * Race to shutdown the filesystem if we see an error. */ -if (XFS_TEST_ERROR (error, log->l_mp,XFS_ERRTAG_IODONE_IOERR )) { +if (error ||XFS_TEST_ERROR (log->l_mp,XFS_ERRTAG_IODONE_IOERR )) {xfs_alert (log->l_mp,"log I/O error %d" , error);xlog_force_shutdown (log,SHUTDOWN_LOG_IO_ERROR ); }
@@ -1848,7 +1848,7 @@ xlog_sync(* detects the bad CRC and attempts to recover. */ #ifdef DEBUG -if (XFS_TEST_ERROR (false, log->l_mp,XFS_ERRTAG_LOG_BAD_CRC )) { +if (XFS_TEST_ERROR (log->l_mp,XFS_ERRTAG_LOG_BAD_CRC )) { iclog->ic_header.h_crc &=cpu_to_le32 (0xAAAAAAAA ); iclog->ic_fail_crc = true; xfs_warn(log->l_mp,
fs/xfs/xfs_trans_ail.c +1 -1
@@ -385,7 +385,7 @@ xfsaild_push_item(* If log item pinning is enabled, skip the push and track the item as * pinned. This can help induce head-behind-tail conditions. */ -if (XFS_TEST_ERROR (false, ailp->ail_log->l_mp,XFS_ERRTAG_LOG_ITEM_PIN )) +if (XFS_TEST_ERROR (ailp->ail_log->l_mp,XFS_ERRTAG_LOG_ITEM_PIN ))return XFS_ITEM_PINNED ;/*
Greg Kroah-Hartman
maintainerPATCH 37/204 xfs: fix error returns in CoW fork repair
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yingjie Gao <[email protected]>
[ Upstream commit fcf4faba9f986b3bb528da11913c9ec5d6e8f689 ]
xrep_cow_find_bad() returns success after the cleanup labels even if AG setup, btree queries, or bitmap updates failed. This can make repair continue with an incomplete bad-file-offset bitmap instead of stopping at the original error.
The force-rebuild path has a related cleanup problem. If xrep_cow_mark_file_range() fails, the function returns directly and skips the scrub AG context and perag cleanup.
Let the force-rebuild path fall through to the existing cleanup code and return the saved error after cleanup.
--- fs/xfs/scrub/cow_repair.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-)
fs/xfs/scrub/cow_repair.c +2 -5
@@ -297,18 +297,15 @@ xrep_cow_find_bad(* on the debugging knob, replace everything in the CoW fork. */ if ((sc->sm->sm_flags &XFS_SCRUB_IFLAG_FORCE_REBUILD ) || -XFS_TEST_ERROR (sc->mp,XFS_ERRTAG_FORCE_SCRUB_REPAIR )) { +XFS_TEST_ERROR (sc->mp,XFS_ERRTAG_FORCE_SCRUB_REPAIR )) error =xrep_cow_mark_file_range (xc, xc->irec.br_startblock, xc->irec.br_blockcount); -if (error) -return error; - } out_sa:xchk_ag_free (sc, &sc->sa); out_pag:xfs_perag_put (pag); -return 0 ; +return error; }/*
Greg Kroah-Hartman
maintainerPATCH 38/204 Revert "net: bonding: fix use-after-free in bond_xmit_broadcast()"
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kevin Berry <[email protected]>
This reverts commit 3453882f36c40d2339267093676585a89808a73d.
There are two versions of this use-after-free fix commit: this one, which was written to avoid taking a dependency on ce7a381697cb3 ("net:
bonding: add broadcast_neighbor option for 802.3ad"), and the original, simpler version 2884bf72fb8f ("net: bonding: fix use-after-free in bond_xmit_broadcast()"), which implicitly depends on the slave counting changes in ce7a381697cb3. In both the 6.1 and 6.6 stable branches, commit ce7a381697cb3 was included as a stable dep of c4f050ce06c56 ("bonding: 3ad: implement proper RCU rules for port->aggregator"), and the original version of this fix was subsequently applied.
For consistency, and to be able to apply both bug fixes, we should revert this commit, apply the series for ce7a381697cb3 ("net: bonding:
add broadcast_neighbor option for 802.3ad"), and then apply the original version of this fix, 2884bf72fb8f ("net: bonding: fix use-after-free in bond_xmit_broadcast()").
--- drivers/net/bonding/bond_main.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-)
drivers/net/bonding/bond_main.c +4 -8
@@ -5328,22 +5328,18 @@ static netdev_tx_t bond_xmit_broadcast(s struct net_device *bond_dev) {struct bonding *bond =netdev_priv (bond_dev); -struct bond_up_slave *slaves; +struct slave *slave =NULL ; +struct list_head *iter;bool xmit_suc = false;bool skb_used = false; -int slaves_count, i; - slaves =rcu_dereference (bond->all_slaves); - - slaves_count = slaves ?READ_ONCE (slaves->count) :0 ; -for (i =0 ; i < slaves_count; i++) { -struct slave *slave = slaves->arr[i]; +bond_for_each_slave_rcu (bond, slave, iter) {struct sk_buff *skb2;if (!(bond_slave_is_up (slave) && slave->link ==BOND_LINK_UP ))continue ; -if (i +1 == slaves_count) { +if (bond_is_last_slave (bond, slave)) { skb2 = skb; skb_used = true; }else {
Greg Kroah-Hartman
maintainerPATCH 39/204 net: bonding: add broadcast_neighbor option for 802.3ad
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tonghao Zhang <[email protected]>
[ Upstream commit ce7a381697cb3958ffe0b45e5028ac69444e9288 ]
Stacking technology is a type of technology used to expand ports on Ethernet switches. It is widely used as a common access method in large-scale Internet data center architectures. Years of practice have proved that stacking technology has advantages and disadvantages in high-reliability network architecture scenarios. For instance, in stacking networking arch, conventional switch system upgrades require multiple stacked devices to restart at the same time. Therefore, it is inevitable that the business will be interrupted for a while. It is for this reason that "no-stacking" in data centers has become a trend. Additionally, when the stacking link connecting the switches fails or is abnormal, the stack will split. Although it is not common, it still happens in actual operation. The problem is that after the split, it is equivalent to two switches with the same configuration appearing in the network, causing network configuration conflicts and ultimately interrupting the services carried by the stacking system.
To improve network stability, "non-stacking" solutions have been increasingly adopted, particularly by public cloud providers and tech companies like Alibaba, Tencent, and Didi. "non-stacking" is a method of mimicing switch stacking that convinces a LACP peer, bonding in this case, connected to a set of "non-stacked" switches that all of its ports are connected to a single switch (i.e., LACP aggregator), as if those switches were stacked. This enables the LACP peer's ports to aggregate together, and requires
(a) special switch configuration, described in the linked article, and (b) modifications to the bonding 802.3ad (LACP) mode to send all ARP/ND packets across all ports of the active aggregator.
Note that, with multiple aggregators, the current broadcast mode logic will send only packets to the selected aggregator(s).
+-----------+ +-----------+
| switch1 | | switch2 |
+-----------+ +-----------+
^ ^
| |
+-----------------+
| bond4 lacp |
+-----------------+
| |
| NIC1 | NIC2
+-----------------+
| server |
+-----------------+- https://www.ruijie.com/fr-fr/support/tech-gallery/de-stack-data-center-network-architecture/
--- Documentation/networking/bonding.rst | 6 +++ drivers/net/bonding/bond_main.c | 66 +++++++++++++++++++++++++++++++---- drivers/net/bonding/bond_options.c | 42 ++++++++++++++++++++++ include/net/bond_options.h | 1 include/net/bonding.h | 3 + 5 files changed, 112 insertions(+), 6 deletions(-)
Documentation/networking/bonding.rst +6 -0
@@ -562,6 +562,12 @@ lacp_rate The default is slow. +broadcast_neighbor + + Option specifying whether to broadcast ARP/ND packets to all + active slaves. This option has no effect in modes other than + 802.3ad mode. The default is off (0). + max_bonds Specifies the number of bonding devices to create for this
drivers/net/bonding/bond_main.c +60 -6
117 lines not on this page · raw message
drivers/net/bonding/bond_options.c +42 -0
69 lines not on this page · raw message
include/net/bond_options.h +1 -0
@@ -77,6 +77,7 @@ enum {BOND_OPT_NS_TARGETS ,BOND_OPT_PRIO ,BOND_OPT_COUPLED_CONTROL , +BOND_OPT_BROADCAST_NEIGH , BOND_OPT_LAST };
include/net/bonding.h +3 -0
@@ -115,6 +115,8 @@ static inline int is_netpoll_tx_blocked(#define is_netpoll_tx_blocked (dev) (0)#endif +DECLARE_STATIC_KEY_FALSE (bond_bcast_neigh_enabled); +struct bond_params {int mode;int xmit_policy;
@@ -149,6 +151,7 @@ struct bond_params {struct in6_addr ns_targets[BOND_MAX_NS_TARGETS ];#endif int coupled_control; +int broadcast_neighbor;/* 2 bytes of padding : see ether_addr_equal_64bits() */ u8 ad_actor_system[ETH_ALEN +2 ];
Greg Kroah-Hartman
maintainerPATCH 40/204 bonding: add support for per-port LACP actor priority
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hangbin Liu <[email protected]>
[ Upstream commit 6b6dc81ee7e8ca87c71a533e1d69cf96a4f1e986 ]
Introduce a new netlink attribute 'actor_port_prio' to allow setting the LACP actor port priority on a per-slave basis. This extends the existing bonding infrastructure to support more granular control over LACP negotiations.
The priority value is embedded in LACPDU packets and will be used by subsequent patches to influence aggregator selection policies.
--- Documentation/networking/bonding.rst | 9 ++++++++ drivers/net/bonding/bond_3ad.c | 4 +++ drivers/net/bonding/bond_netlink.c | 16 +++++++++++++++ drivers/net/bonding/bond_options.c | 36 +++++++++++++++++++++++++++++++++++ include/net/bond_3ad.h | 1 include/net/bond_options.h | 1 include/uapi/linux/if_link.h | 1 7 files changed, 68 insertions(+)
Documentation/networking/bonding.rst +9 -0
@@ -193,6 +193,15 @@ ad_actor_sys_prio This parameter has effect only in 802.3ad mode and is available through SysFs interface. +actor_port_prio + + In an AD system, this specifies the port priority. The allowed range + is 1 - 65535. If the value is not specified, it takes 255 as the + default value. + + This parameter has effect only in 802.3ad mode and is available through + netlink interface. + ad_actor_system In an AD system, this specifies the mac-address for the actor in
drivers/net/bonding/bond_3ad.c +4 -0
@@ -436,6 +436,7 @@ static void __ad_actor_update_port(struc port->actor_system =BOND_AD_INFO (bond).system.sys_mac_addr; port->actor_system_priority =BOND_AD_INFO (bond).system.sys_priority; + port->actor_port_priority =SLAVE_AD_INFO (port->slave)->port_priority; }/* Conversions */
@@ -2195,6 +2196,9 @@ void bond_3ad_bind_slave(struct slave *sad_initialize_port (port, &bond->params); +/* Port priority is initialized. Update it to slave's ad info */ +SLAVE_AD_INFO (slave)->port_priority = port->actor_port_priority; + port->slave = slave; port->actor_port_number =SLAVE_AD_INFO (slave)->id;/* key is determined according to the link speed, duplex and
drivers/net/bonding/bond_netlink.c +16 -0
@@ -28,6 +28,7 @@ static size_t bond_get_slave_size(constnla_total_size (sizeof (u8)) +/* IFLA_BOND_SLAVE_AD_ACTOR_OPER_PORT_STATE */ nla_total_size (sizeof (u16)) +/* IFLA_BOND_SLAVE_AD_PARTNER_OPER_PORT_STATE */ nla_total_size (sizeof (s32)) +/* IFLA_BOND_SLAVE_PRIO */ +nla_total_size (sizeof (u16)) +/* IFLA_BOND_SLAVE_ACTOR_PORT_PRIO */ 0 ; }
@@ -77,6 +78,10 @@ static int bond_fill_slave_info(struct s ad_port->partner_oper.port_state)) goto nla_put_failure; } + +if (nla_put_u16 (skb,IFLA_BOND_SLAVE_ACTOR_PORT_PRIO , +SLAVE_AD_INFO (slave)->port_priority)) + goto nla_put_failure; }return 0 ;
@@ -129,6 +134,7 @@ static const struct nla_policy bond_polistatic const struct nla_policy bond_slave_policy[IFLA_BOND_SLAVE_MAX +1 ] = { [IFLA_BOND_SLAVE_QUEUE_ID ] = { .type =NLA_U16 }, [IFLA_BOND_SLAVE_PRIO ] = { .type =NLA_S32 }, + [IFLA_BOND_SLAVE_ACTOR_PORT_PRIO ] = { .type =NLA_U16 }, }; static int bond_validate(struct nlattr *tb[], struct nlattr *data[],
@@ -178,6 +184,16 @@ static int bond_slave_changelink(structif (err)return err; } + +if (data[IFLA_BOND_SLAVE_ACTOR_PORT_PRIO ]) { +u16 ad_prio =nla_get_u16 (data[IFLA_BOND_SLAVE_ACTOR_PORT_PRIO ]); + +bond_opt_slave_initval (&newval, &slave_dev, ad_prio); + err =__bond_opt_set (bond,BOND_OPT_ACTOR_PORT_PRIO , &newval, + data[IFLA_BOND_SLAVE_ACTOR_PORT_PRIO ], extack); +if (err) +return err; + }return 0 ; }
drivers/net/bonding/bond_options.c +36 -0
@@ -79,6 +79,8 @@ static int bond_option_tlb_dynamic_lb_se const struct bond_opt_value *newval);static int bond_option_ad_actor_sys_prio_set (struct bonding *bond,const struct bond_opt_value *newval); +static int bond_option_actor_port_prio_set (struct bonding *bond, +const struct bond_opt_value *newval);static int bond_option_ad_actor_system_set (struct bonding *bond,const struct bond_opt_value *newval); static int bond_option_ad_user_port_key_set(struct bonding *bond,
@@ -223,6 +225,13 @@ static const struct bond_opt_value bond_ { NULL, -1, 0}, }; +static const struct bond_opt_value bond_actor_port_prio_tbl[] = { + {"minval" ,0 ,BOND_VALFLAG_MIN }, + {"maxval" ,65535 ,BOND_VALFLAG_MAX }, + {"default" ,255 ,BOND_VALFLAG_DEFAULT }, + {NULL ,-1 ,0 }, +}; +static const struct bond_opt_value bond_ad_user_port_key_tbl[] = { {"minval" ,0 ,BOND_VALFLAG_MIN |BOND_VALFLAG_DEFAULT }, { "maxval", 1023, BOND_VALFLAG_MAX},
@@ -484,6 +493,13 @@ static const struct bond_option bond_opt .values = bond_ad_actor_sys_prio_tbl, .set = bond_option_ad_actor_sys_prio_set, }, + [BOND_OPT_ACTOR_PORT_PRIO] = { + .id = BOND_OPT_ACTOR_PORT_PRIO, + .name = "actor_port_prio", + .unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_8023AD)), + .values = bond_actor_port_prio_tbl, + .set = bond_option_actor_port_prio_set, + }, [BOND_OPT_AD_ACTOR_SYSTEM] = { .id = BOND_OPT_AD_ACTOR_SYSTEM, .name = "ad_actor_system",
@@ -1817,6 +1833,26 @@ static int bond_option_ad_actor_sys_priobond_3ad_update_ad_actor_settings (bond);return 0 ; +} + +static int bond_option_actor_port_prio_set (struct bonding *bond, +const struct bond_opt_value *newval) +{ +struct slave *slave; + + slave =bond_slave_get_rtnl (newval->slave_dev); +if (!slave) { +netdev_dbg (bond->dev,"%s called on NULL slave\n" , __func__); +return -ENODEV ; + } + +netdev_dbg (newval->slave_dev,"Setting actor_port_prio to %llu\n" , + newval->value); + +SLAVE_AD_INFO (slave)->port_priority = newval->value; +bond_3ad_update_ad_actor_settings (bond); + +return 0 ; } static int bond_option_ad_actor_system_set(struct bonding *bond,
include/net/bond_3ad.h +1 -0
@@ -274,6 +274,7 @@ struct ad_slave_info {struct port port;/* 802.3ad port structure */ struct bond_3ad_stats stats;u16 id; +u16 port_priority; };static inline const char *bond_3ad_churn_desc (churn_state_t state)
include/net/bond_options.h +1 -0
@@ -78,6 +78,7 @@ enum {BOND_OPT_PRIO ,BOND_OPT_COUPLED_CONTROL ,BOND_OPT_BROADCAST_NEIGH , +BOND_OPT_ACTOR_PORT_PRIO , BOND_OPT_LAST };
include/uapi/linux/if_link.h +1 -0
@@ -1551,6 +1551,7 @@ enum {IFLA_BOND_SLAVE_AD_ACTOR_OPER_PORT_STATE ,IFLA_BOND_SLAVE_AD_PARTNER_OPER_PORT_STATE ,IFLA_BOND_SLAVE_PRIO , +IFLA_BOND_SLAVE_ACTOR_PORT_PRIO , __IFLA_BOND_SLAVE_MAX, };
Greg Kroah-Hartman
maintainerPATCH 41/204 bonding: print churn state via netlink
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hangbin Liu <[email protected]>
[ Upstream commit 4916f2e2f3fc9aef289fcd07949301e5c29094c2 ]
Currently, the churn state is printed only in sysfs. Add netlink support so users could get the state via netlink.
--- drivers/net/bonding/bond_netlink.c | 9 +++++++++ include/uapi/linux/if_link.h | 2 ++ 2 files changed, 11 insertions(+)
drivers/net/bonding/bond_netlink.c +9 -0
@@ -29,6 +29,8 @@ static size_t bond_get_slave_size(constnla_total_size (sizeof (u16)) +/* IFLA_BOND_SLAVE_AD_PARTNER_OPER_PORT_STATE */ nla_total_size (sizeof (s32)) +/* IFLA_BOND_SLAVE_PRIO */ nla_total_size (sizeof (u16)) +/* IFLA_BOND_SLAVE_ACTOR_PORT_PRIO */ +nla_total_size (sizeof (u8)) +/* IFLA_BOND_SLAVE_AD_CHURN_ACTOR_STATE */ +nla_total_size (sizeof (u8)) +/* IFLA_BOND_SLAVE_AD_CHURN_PARTNER_STATE */ 0 ; }
@@ -77,6 +79,13 @@ static int bond_fill_slave_info(struct s IFLA_BOND_SLAVE_AD_PARTNER_OPER_PORT_STATE, ad_port->partner_oper.port_state)) goto nla_put_failure; + +if (nla_put_u8 (skb,IFLA_BOND_SLAVE_AD_CHURN_ACTOR_STATE , + ad_port->sm_churn_actor_state)) + goto nla_put_failure; +if (nla_put_u8 (skb,IFLA_BOND_SLAVE_AD_CHURN_PARTNER_STATE , + ad_port->sm_churn_partner_state)) + goto nla_put_failure; } if (nla_put_u16(skb, IFLA_BOND_SLAVE_ACTOR_PORT_PRIO,
include/uapi/linux/if_link.h +2 -0
@@ -1552,6 +1552,8 @@ enum {IFLA_BOND_SLAVE_AD_PARTNER_OPER_PORT_STATE ,IFLA_BOND_SLAVE_PRIO ,IFLA_BOND_SLAVE_ACTOR_PORT_PRIO , +IFLA_BOND_SLAVE_AD_CHURN_ACTOR_STATE , +IFLA_BOND_SLAVE_AD_CHURN_PARTNER_STATE , __IFLA_BOND_SLAVE_MAX, };
Greg Kroah-Hartman
maintainerPATCH 42/204 bonding: 3ad: implement proper RCU rules for port->aggregator
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit c4f050ce06c56cfb5993268af4a5cb66ed1cd04e ]
syzbot found a data-race in bond_3ad_get_active_agg_info / bond_3ad_state_machine_handler [1] which hints at lack of proper RCU implementation.
Add __rcu qualifier to port->aggregator, and add proper RCU API.
[1]
BUG: KCSAN: data-race in bond_3ad_get_active_agg_info / bond_3ad_state_machine_handler
write to 0xffff88813cf5c4b0 of 8 bytes by task 36 on cpu 0:
ad_port_selection_logic drivers/net/bonding/bond_3ad.c:1659 [inline] bond_3ad_state_machine_handler+0x9d5/0x2d60 drivers/net/bonding/bond_3ad.c:2569 process_one_work kernel/workqueue.c:3302 [inline] process_scheduled_works+0x4f0/0x9c0 kernel/workqueue.c:3385 worker_thread+0x58a/0x780 kernel/workqueue.c:3466 kthread+0x22a/0x280 kernel/kthread.c:436 ret_from_fork+0x146/0x330 arch/x86/kernel/process.c:158 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
read to 0xffff88813cf5c4b0 of 8 bytes by task 22063 on cpu 1:
__bond_3ad_get_active_agg_info drivers/net/bonding/bond_3ad.c:2858 [inline] bond_3ad_get_active_agg_info+0x8c/0x230 drivers/net/bonding/bond_3ad.c:2881 bond_fill_info+0xe0f/0x10f0 drivers/net/bonding/bond_netlink.c:853 rtnl_link_info_fill net/core/rtnetlink.c:906 [inline] rtnl_link_fill+0x1d7/0x4e0 net/core/rtnetlink.c:927 rtnl_fill_ifinfo+0xf8e/0x1380 net/core/rtnetlink.c:2168 rtmsg_ifinfo_build_skb+0x11c/0x1b0 net/core/rtnetlink.c:4453 rtmsg_ifinfo_event net/core/rtnetlink.c:4486 [inline] rtmsg_ifinfo+0x6d/0x110 net/core/rtnetlink.c:4495 __dev_notify_flags+0x76/0x390 net/core/dev.c:9790 netif_change_flags+0xac/0xd0 net/core/dev.c:9823 do_setlink+0x905/0x2950 net/core/rtnetlink.c:3180 rtnl_group_changelink net/core/rtnetlink.c:3813 [inline] __rtnl_newlink net/core/rtnetlink.c:3981 [inline] rtnl_newlink+0xf55/0x1400 net/core/rtnetlink.c:4109 rtnetlink_rcv_msg+0x64b/0x720 net/core/rtnetlink.c:6995 netlink_rcv_skb+0x123/0x220 net/netlink/af_netlink.c:2550 rtnetlink_rcv+0x1c/0x30 net/core/rtnetlink.c:7022 netlink_unicast_kernel net/netlink/af_netlink.c:1318 [inline] netlink_unicast+0x5a8/0x680 net/netlink/af_netlink.c:1344 netlink_sendmsg+0x5c8/0x6f0 net/netlink/af_netlink.c:1894 sock_sendmsg_nosec net/socket.c:787 [inline] __sock_sendmsg net/socket.c:802 [inline] ____sys_sendmsg+0x563/0x5b0 net/socket.c:2698 ___sys_sendmsg+0x195/0x1e0 net/socket.c:2752 __sys_sendmsg net/socket.c:2784 [inline] __do_sys_sendmsg net/socket.c:2789 [inline] __se_sys_sendmsg net/socket.c:2787 [inline] __x64_sys_sendmsg+0xd4/0x160 net/socket.c:2787 x64_sys_call+0x194c/0x3020 arch/x86/include/generated/asm/syscalls_64.h:47 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x12c/0x3b0 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f
value changed: 0x0000000000000000 -> 0xffff88813cf5c400
Reported by Kernel Concurrency Sanitizer on:
CPU: 1 UID: 0 PID: 22063 Comm: syz.0.31122 Tainted: G W syzkaller #0 PREEMPT(full) Tainted: [W]=WARN Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/18/2026
--- drivers/net/bonding/bond_3ad.c | 109 ++++++++++++++++++--------------- drivers/net/bonding/bond_main.c | 8 +- drivers/net/bonding/bond_netlink.c | 16 +++- drivers/net/bonding/bond_procfs.c | 3 drivers/net/bonding/bond_sysfs_slave.c | 17 +++-- include/net/bond_3ad.h | 2 6 files changed, 89 insertions(+), 66 deletions(-)
drivers/net/bonding/bond_3ad.c +60 -49
@@ -991,6 +991,7 @@ static int ad_marker_send(struct port *pstatic void ad_mux_machine (struct port *port,bool *update_slave_arr) {struct bonding *bond =__get_bond_by_port (port); +struct aggregator *aggregator;mux_states_t last_state;/* keep current State Machine state to compare later if it was
@@ -998,6 +999,7 @@ static void ad_mux_machine(struct port **/ last_state = port->sm_mux_state; + aggregator =rcu_dereference (port->aggregator);if (port->sm_vars &AD_PORT_BEGIN ) { port->sm_mux_state =AD_MUX_DETACHED ; }else {
@@ -1017,7 +1019,7 @@ static void ad_mux_machine(struct port ** cycle to update ready variable, we check * READY_N and update READY here */ -__set_agg_ports_ready (port->aggregator,__agg_ports_are_ready (port->aggregator)); +__set_agg_ports_ready (aggregator,__agg_ports_are_ready (aggregator)); port->sm_mux_state =AD_MUX_DETACHED ;break ; }
@@ -1032,7 +1034,7 @@ static void ad_mux_machine(struct port ** update ready variable, we check READY_N and update * READY here */ -__set_agg_ports_ready (port->aggregator,__agg_ports_are_ready (port->aggregator)); +__set_agg_ports_ready (aggregator,__agg_ports_are_ready (aggregator));/* if the wait_while_timer expired, and the port is * in READY state, move to ATTACHED state
@@ -1048,7 +1050,7 @@ static void ad_mux_machine(struct port *if ((port->sm_vars &AD_PORT_SELECTED ) && (port->partner_oper.port_state &LACP_STATE_SYNCHRONIZATION ) && !__check_agg_selection_timer (port)) { -if (port->aggregator->is_active) { +if (aggregator->is_active) {int state =AD_MUX_COLLECTING_DISTRIBUTING ;if (!bond->params.coupled_control)
@@ -1064,9 +1066,9 @@ static void ad_mux_machine(struct port ** cycle to update ready variable, we check * READY_N and update READY here */ -__set_agg_ports_ready (port->aggregator,__agg_ports_are_ready (port->aggregator)); +__set_agg_ports_ready (aggregator,__agg_ports_are_ready (aggregator)); port->sm_mux_state =AD_MUX_DETACHED ; - }else if (port->aggregator->is_active) { + }else if (aggregator->is_active) { port->actor_oper_port_state |=LACP_STATE_SYNCHRONIZATION ; }
@@ -1077,7 +1079,7 @@ static void ad_mux_machine(struct port ** sure that a collecting distributing * port in an active aggregator is enabled */ -if (port->aggregator->is_active && +if (aggregator->is_active && !__port_is_collecting_distributing (port)) {__enable_port (port); *update_slave_arr = true;
@@ -1096,7 +1098,7 @@ static void ad_mux_machine(struct port **/ struct slave *slave = port->slave; -if (port->aggregator->is_active && +if (aggregator->is_active &&bond_is_slave_rx_disabled (slave)) {ad_enable_collecting (port); *update_slave_arr = true;
@@ -1116,8 +1118,8 @@ static void ad_mux_machine(struct port ** sure that a collecting distributing * port in an active aggregator is enabled */ -if (port->aggregator && - port->aggregator->is_active && +if (aggregator && + aggregator->is_active && !__port_is_collecting_distributing (port)) {__enable_port (port); *update_slave_arr = true;
@@ -1149,7 +1151,7 @@ static void ad_mux_machine(struct port * port->sm_mux_timer_counter =__ad_timer_to_ticks (AD_WAIT_WHILE_TIMER ,0 );break ;case AD_MUX_ATTACHED : -if (port->aggregator->is_active) +if (aggregator->is_active) port->actor_oper_port_state |=LACP_STATE_SYNCHRONIZATION ;else
@@ -1522,9 +1524,9 @@ static void ad_port_selection_logic(stru bond =__get_bond_by_port (port);/* if the port is connected to other aggregator, detach it */ -if (port->aggregator) { + temp_aggregator =rcu_dereference (port->aggregator); +if (temp_aggregator) {/* detach the port from its former aggregator */ - temp_aggregator = port->aggregator;for (curr_port = temp_aggregator->lag_ports; curr_port; last_port = curr_port, curr_port = curr_port->next_port_in_aggregator) {
@@ -1547,7 +1549,7 @@ static void ad_port_selection_logic(stru/* clear the port's relations to this * aggregator */ - port->aggregator =NULL ; +RCU_INIT_POINTER (port->aggregator,NULL ); port->next_port_in_aggregator =NULL ; port->actor_port_aggregator_identifier =0 ;
@@ -1570,7 +1572,7 @@ static void ad_port_selection_logic(stru port->slave->bond->dev->name, port->slave->dev->name, port->actor_port_number, - port->aggregator->aggregator_identifier); + temp_aggregator->aggregator_identifier); } }/* search on all aggregators for a suitable aggregator for this port */
@@ -1594,15 +1596,15 @@ static void ad_port_selection_logic(stru ) ) {/* attach to the founded aggregator */ - port->aggregator = aggregator; +rcu_assign_pointer (port->aggregator, aggregator); port->actor_port_aggregator_identifier = - port->aggregator->aggregator_identifier; + aggregator->aggregator_identifier; port->next_port_in_aggregator = aggregator->lag_ports; - port->aggregator->num_of_ports++; + aggregator->num_of_ports++; aggregator->lag_ports = port;slave_dbg (bond->dev, slave->dev,"Port %d joined LAG %d (existing LAG)\n" , port->actor_port_number, - port->aggregator->aggregator_identifier); + aggregator->aggregator_identifier);/* mark this port as selected */ port->sm_vars |=AD_PORT_SELECTED ;
@@ -1617,39 +1619,40 @@ static void ad_port_selection_logic(struif (!found) {if (free_aggregator) {/* assign port a new aggregator */ - port->aggregator = free_aggregator; port->actor_port_aggregator_identifier = - port->aggregator->aggregator_identifier; + free_aggregator->aggregator_identifier;/* update the new aggregator's parameters * if port was responsed from the end-user */ if (port->actor_oper_port_key &AD_DUPLEX_KEY_MASKS )/* if port is full duplex */ - port->aggregator->is_individual = false; + free_aggregator->is_individual = false;else - port->aggregator->is_individual = true; + free_aggregator->is_individual = true; - port->aggregator->actor_admin_aggregator_key = + free_aggregator->actor_admin_aggregator_key = port->actor_admin_port_key; - port->aggregator->actor_oper_aggregator_key = + free_aggregator->actor_oper_aggregator_key = port->actor_oper_port_key; - port->aggregator->partner_system = + free_aggregator->partner_system = port->partner_oper.system; - port->aggregator->partner_system_priority = + free_aggregator->partner_system_priority = port->partner_oper.system_priority; - port->aggregator->partner_oper_aggregator_key = port->partner_oper.key; - port->aggregator->receive_state =1 ; - port->aggregator->transmit_state =1 ; - port->aggregator->lag_ports = port; - port->aggregator->num_of_ports++; + free_aggregator->partner_oper_aggregator_key = port->partner_oper.key; + free_aggregator->receive_state =1 ; + free_aggregator->transmit_state =1 ; + free_aggregator->lag_ports = port; + free_aggregator->num_of_ports++; + +rcu_assign_pointer (port->aggregator, free_aggregator);/* mark this port as selected */ port->sm_vars |=AD_PORT_SELECTED ;slave_dbg (bond->dev, port->slave->dev,"Port %d joined LAG %d (new LAG)\n" , port->actor_port_number, - port->aggregator->aggregator_identifier); + free_aggregator->aggregator_identifier); }else { slave_err(bond->dev, port->slave->dev, "Port %d did not find a suitable aggregator\n",
@@ -1661,13 +1664,12 @@ static void ad_port_selection_logic(stru* in all aggregator's ports, else set ready=FALSE in all * aggregator's ports */ -__set_agg_ports_ready (port->aggregator, -__agg_ports_are_ready (port->aggregator)); + aggregator =rcu_dereference (port->aggregator); +__set_agg_ports_ready (aggregator,__agg_ports_are_ready (aggregator)); - aggregator =__get_first_agg (port); -ad_agg_selection_logic (aggregator, update_slave_arr); +ad_agg_selection_logic (__get_first_agg (port), update_slave_arr); -if (!port->aggregator->is_active) +if (!aggregator->is_active) port->actor_oper_port_state &= ~LACP_STATE_SYNCHRONIZATION ; }
@@ -2021,13 +2023,15 @@ static void ad_initialize_port(struct po*/ static void ad_enable_collecting (struct port *port) { -if (port->aggregator->is_active) { +struct aggregator *aggregator =rcu_dereference (port->aggregator); + +if (aggregator->is_active) {struct slave *slave = port->slave;slave_dbg (slave->bond->dev, slave->dev,"Enabling collecting on port %d (LAG %d)\n" , port->actor_port_number, - port->aggregator->aggregator_identifier); + aggregator->aggregator_identifier);__enable_collecting_port (port); } }
@@ -2039,11 +2043,13 @@ static void ad_enable_collecting(struct*/ static void ad_disable_distributing (struct port *port,bool *update_slave_arr) { -if (port->aggregator &&__agg_has_partner (port->aggregator)) { +struct aggregator *aggregator =rcu_dereference (port->aggregator); + +if (aggregator &&__agg_has_partner (aggregator)) {slave_dbg (port->slave->bond->dev, port->slave->dev,"Disabling distributing on port %d (LAG %d)\n" , port->actor_port_number, - port->aggregator->aggregator_identifier); + aggregator->aggregator_identifier);__disable_distributing_port (port);/* Slave array needs an update */ *update_slave_arr = true;
@@ -2060,11 +2066,13 @@ static void ad_disable_distributing(strustatic void ad_enable_collecting_distributing (struct port *port,bool *update_slave_arr) { -if (port->aggregator->is_active) { +struct aggregator *aggregator =rcu_dereference (port->aggregator); + +if (aggregator->is_active) {slave_dbg (port->slave->bond->dev, port->slave->dev,"Enabling port %d (LAG %d)\n" , port->actor_port_number, - port->aggregator->aggregator_identifier); + aggregator->aggregator_identifier);__enable_port (port);/* Slave array needs update */ *update_slave_arr = true;
@@ -2079,11 +2087,13 @@ static void ad_enable_collecting_distribstatic void ad_disable_collecting_distributing (struct port *port,bool *update_slave_arr) { -if (port->aggregator &&__agg_has_partner (port->aggregator)) { +struct aggregator *aggregator =rcu_dereference (port->aggregator); + +if (aggregator &&__agg_has_partner (aggregator)) {slave_dbg (port->slave->bond->dev, port->slave->dev,"Disabling port %d (LAG %d)\n" , port->actor_port_number, - port->aggregator->aggregator_identifier); + aggregator->aggregator_identifier);__disable_port (port);/* Slave array needs an update */ *update_slave_arr = true;
@@ -2323,7 +2333,7 @@ void bond_3ad_unbind_slave(struct slave*/ for (temp_port = aggregator->lag_ports; temp_port; temp_port = temp_port->next_port_in_aggregator) { - temp_port->aggregator = new_aggregator; +rcu_assign_pointer (temp_port->aggregator, new_aggregator); temp_port->actor_port_aggregator_identifier = new_aggregator->aggregator_identifier; }
@@ -2792,15 +2802,16 @@ out:int __bond_3ad_get_active_agg_info (struct bonding *bond,struct ad_info *ad_info) { -struct aggregator *aggregator =NULL ; +struct aggregator *aggregator =NULL , *tmp;struct list_head *iter;struct slave *slave;struct port *port;bond_for_each_slave_rcu (bond, slave, iter) { port = &(SLAVE_AD_INFO (slave)->port); -if (port->aggregator && port->aggregator->is_active) { - aggregator = port->aggregator; + tmp =rcu_dereference (port->aggregator); +if (tmp && tmp->is_active) { + aggregator = tmp;break ; } }
drivers/net/bonding/bond_main.c +5 -3
@@ -1470,7 +1470,7 @@ static void bond_poll_controller(structif (BOND_MODE (bond) ==BOND_MODE_8023AD ) {struct aggregator *agg = -SLAVE_AD_INFO (slave)->port.aggregator; +rcu_dereference (SLAVE_AD_INFO (slave)->port.aggregator);if (agg && agg->aggregator_identifier != ad_info.aggregator_id)
@@ -5244,15 +5244,16 @@ int bond_update_slave_arr(struct bondingspin_unlock_bh (&bond->mode_lock); agg_id = ad_info.aggregator_id; } +rcu_read_lock ();bond_for_each_slave (bond, slave, iter) {if (skipslave == slave)continue ; all_slaves->arr[all_slaves->count++] = slave;if (BOND_MODE (bond) ==BOND_MODE_8023AD ) { -struct aggregator *agg; +const struct aggregator *agg; - agg =SLAVE_AD_INFO (slave)->port.aggregator; + agg =rcu_dereference (SLAVE_AD_INFO (slave)->port.aggregator);if (!agg || agg->aggregator_identifier != agg_id)continue ; }
@@ -5264,6 +5265,7 @@ int bond_update_slave_arr(struct bonding usable_slaves->arr[usable_slaves->count++] = slave; } +rcu_read_unlock ();bond_set_slave_arr (bond, usable_slaves, all_slaves);return ret;
drivers/net/bonding/bond_netlink.c +10 -6
@@ -66,27 +66,29 @@ static int bond_fill_slave_info(struct sconst struct port *ad_port; ad_port = &SLAVE_AD_INFO (slave)->port; - agg =SLAVE_AD_INFO (slave)->port.aggregator; +rcu_read_lock (); + agg =rcu_dereference (SLAVE_AD_INFO (slave)->port.aggregator);if (agg) {if (nla_put_u16 (skb,IFLA_BOND_SLAVE_AD_AGGREGATOR_ID , agg->aggregator_identifier)) - goto nla_put_failure; + goto nla_put_failure_rcu;if (nla_put_u8 (skb,IFLA_BOND_SLAVE_AD_ACTOR_OPER_PORT_STATE , ad_port->actor_oper_port_state)) - goto nla_put_failure; + goto nla_put_failure_rcu;if (nla_put_u16 (skb,IFLA_BOND_SLAVE_AD_PARTNER_OPER_PORT_STATE , ad_port->partner_oper.port_state)) - goto nla_put_failure; + goto nla_put_failure_rcu;if (nla_put_u8 (skb,IFLA_BOND_SLAVE_AD_CHURN_ACTOR_STATE , ad_port->sm_churn_actor_state)) - goto nla_put_failure; + goto nla_put_failure_rcu;if (nla_put_u8 (skb,IFLA_BOND_SLAVE_AD_CHURN_PARTNER_STATE , ad_port->sm_churn_partner_state)) - goto nla_put_failure; + goto nla_put_failure_rcu; } +rcu_read_unlock ();if (nla_put_u16 (skb,IFLA_BOND_SLAVE_ACTOR_PORT_PRIO ,SLAVE_AD_INFO (slave)->port_priority))
@@ -95,6 +97,8 @@ static int bond_fill_slave_info(struct sreturn 0 ; +nla_put_failure_rcu: +rcu_read_unlock (); nla_put_failure:return -EMSGSIZE ; }
drivers/net/bonding/bond_procfs.c +2 -1
@@ -187,6 +187,7 @@ static void bond_info_show_master(struct } } +/* Note: runs under rcu_read_lock() */ static void bond_info_show_slave (struct seq_file *seq,const struct slave *slave) {
@@ -213,7 +214,7 @@ static void bond_info_show_slave(structif (BOND_MODE (bond) ==BOND_MODE_8023AD ) {const struct port *port = &SLAVE_AD_INFO (slave)->port; -const struct aggregator *agg = port->aggregator; +const struct aggregator *agg =rcu_dereference (port->aggregator);if (agg) { seq_printf(seq, "Aggregator ID: %d\n",
drivers/net/bonding/bond_sysfs_slave.c +11 -6
@@ -62,10 +62,15 @@ static ssize_t ad_aggregator_id_show(strconst struct aggregator *agg;if (BOND_MODE (slave->bond) ==BOND_MODE_8023AD ) { - agg =SLAVE_AD_INFO (slave)->port.aggregator; -if (agg) -return sysfs_emit (buf,"%d\n" , - agg->aggregator_identifier); +rcu_read_lock (); + agg =rcu_dereference (SLAVE_AD_INFO (slave)->port.aggregator); +if (agg) { +ssize_t res =sysfs_emit (buf,"%d\n" , + agg->aggregator_identifier); +rcu_read_unlock (); +return res; + } +rcu_read_unlock (); }return sysfs_emit (buf,"N/A\n" );
@@ -78,7 +83,7 @@ static ssize_t ad_actor_oper_port_state_if (BOND_MODE (slave->bond) ==BOND_MODE_8023AD ) { ad_port = &SLAVE_AD_INFO (slave)->port; -if (ad_port->aggregator) +if (rcu_access_pointer (ad_port->aggregator))return sysfs_emit (buf,"%u\n" , ad_port->actor_oper_port_state); }
@@ -93,7 +98,7 @@ static ssize_t ad_partner_oper_port_statif (BOND_MODE (slave->bond) ==BOND_MODE_8023AD ) { ad_port = &SLAVE_AD_INFO (slave)->port; -if (ad_port->aggregator) +if (rcu_access_pointer (ad_port->aggregator))return sysfs_emit (buf,"%u\n" , ad_port->partner_oper.port_state); }
include/net/bond_3ad.h +1 -1
@@ -242,7 +242,7 @@ typedef struct port {churn_state_t sm_churn_actor_state;churn_state_t sm_churn_partner_state;struct slave *slave;/* pointer to the bond slave that this port belongs to */ -struct aggregator *aggregator;/* pointer to an aggregator that this port related to */ +struct aggregator __rcu *aggregator;/* pointer to an aggregator that this port related to */ struct port *next_port_in_aggregator;/* Next port on the linked list of the parent aggregator */ u32 transaction_id;/* continuous number for identification of Marker PDU's; */ struct lacpdu lacpdu;/* the lacpdu that will be sent for this port */
Greg Kroah-Hartman
maintainerPATCH 43/204 net: bonding: fix use-after-free in bond_xmit_broadcast()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xiang Mei <[email protected]>
[ Upstream commit 2884bf72fb8f03409e423397319205de48adca16 ]
bond_xmit_broadcast() reuses the original skb for the last slave (determined by bond_is_last_slave()) and clones it for others. Concurrent slave enslave/release can mutate the slave list during RCU-protected iteration, changing which slave is "last" mid-loop. This causes the original skb to be double-consumed (double-freed).
Replace the racy bond_is_last_slave() check with a simple index comparison (i + 1 == slaves_count) against the pre-snapshot slave count taken via READ_ONCE() before the loop. This preserves the zero-copy optimization for the last slave while making the "last" determination stable against concurrent list mutations.
The UAF can trigger the following crash:
==================================================================
BUG: KASAN: slab-use-after-free in skb_clone Read of size 8 at addr ffff888100ef8d40 by task exploit/147 CPU: 1 UID: 0 PID: 147 Comm: exploit Not tainted 7.0.0-rc3+ #4 PREEMPTLAZY Call Trace: <TASK> dump_stack_lvl (lib/dump_stack.c:123) print_report (mm/kasan/report.c:379 mm/kasan/report.c:482) kasan_report (mm/kasan/report.c:597) skb_clone (include/linux/skbuff.h:1724 include/linux/skbuff.h:1792 include/linux/skbuff.h:3396 net/core/skbuff.c:2108) bond_xmit_broadcast (drivers/net/bonding/bond_main.c:5334) bond_start_xmit (drivers/net/bonding/bond_main.c:5567 drivers/net/bonding/bond_main.c:5593) dev_hard_start_xmit (include/linux/netdevice.h:5325 include/linux/netdevice.h:5334 net/core/dev.c:3871 net/core/dev.c:3887) __dev_queue_xmit (include/linux/netdevice.h:3601 net/core/dev.c:4838) ip6_finish_output2 (include/net/neighbour.h:540 include/net/neighbour.h:554 net/ipv6/ip6_output.c:136) ip6_finish_output (net/ipv6/ip6_output.c:208 net/ipv6/ip6_output.c:219) ip6_output (net/ipv6/ip6_output.c:250) ip6_send_skb (net/ipv6/ip6_output.c:1985) udp_v6_send_skb (net/ipv6/udp.c:1442) udpv6_sendmsg (net/ipv6/udp.c:1733) __sys_sendto (net/socket.c:730 net/socket.c:742 net/socket.c:2206) __x64_sys_sendto (net/socket.c:2209) do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) </TASK> Allocated by task 147: Freed by task 147: The buggy address belongs to the object at ffff888100ef8c80 which belongs to the cache skbuff_head_cache of size 224
The buggy address is located 192 bytes inside of
freed 224-byte region [ffff888100ef8c80, ffff888100ef8d60)
Memory state around the buggy address:
ffff888100ef8c00: fb fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc
ffff888100ef8c80: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff888100ef8d00: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
^
ffff888100ef8d80: fc fc fc fc fc fc fc fc fa fb fb fb fb fb fb fb
ffff888100ef8e00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================--- drivers/net/bonding/bond_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/bonding/bond_main.c +1 -1
@@ -5391,7 +5391,7 @@ static netdev_tx_t bond_xmit_broadcast(sif (!(bond_slave_is_up (slave) && slave->link ==BOND_LINK_UP ))continue ; -if (bond_is_last_slave (bond, slave)) { +if (i +1 == slaves_count) { skb2 = skb; skb_used = true; }else {
Greg Kroah-Hartman
maintainerPATCH 44/204 bonding: fix NULL pointer dereference in actor_port_prio setting
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hangbin Liu <[email protected]>
[ Upstream commit 067bf016e99ad72aa4ff869d6dec1fd62a9c6202 ]
Liang reported an issue where setting a slave’s actor_port_prio to predefined values such as 0, 255, or 65535 would cause a system crash.
The problem occurs because in bond_opt_parse(), when the provided value matches a predefined table entry, the function returns that table entry, which does not contain slave information. Later, in bond_option_actor_port_prio_set(), calling bond_slave_get_rtnl() leads to a NULL pointer dereference.
Since actor_port_prio is defined as a u16 and initialized to the default value of 255 in ad_initialize_port(), there is no need for the bond_actor_port_prio_tbl. Using the BOND_OPTFLAG_RAWVAL flag is sufficient.
--- drivers/net/bonding/bond_options.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-)
drivers/net/bonding/bond_options.c +1 -8
@@ -225,13 +225,6 @@ static const struct bond_opt_value bond_ { NULL, -1, 0}, }; -static const struct bond_opt_value bond_actor_port_prio_tbl[] = { - {"minval" ,0 ,BOND_VALFLAG_MIN }, - {"maxval" ,65535 ,BOND_VALFLAG_MAX }, - {"default" ,255 ,BOND_VALFLAG_DEFAULT }, - {NULL ,-1 ,0 }, -}; -static const struct bond_opt_value bond_ad_user_port_key_tbl[] = { {"minval" ,0 ,BOND_VALFLAG_MIN |BOND_VALFLAG_DEFAULT }, { "maxval", 1023, BOND_VALFLAG_MAX},
@@ -497,7 +490,7 @@ static const struct bond_option bond_opt .id = BOND_OPT_ACTOR_PORT_PRIO, .name = "actor_port_prio", .unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_8023AD)), - .values = bond_actor_port_prio_tbl, + .flags = BOND_OPTFLAG_RAWVAL, .set = bond_option_actor_port_prio_set, }, [BOND_OPT_AD_ACTOR_SYSTEM] = {
Greg Kroah-Hartman
maintainerPATCH 45/204 staging: rtl8723bs: fix buffer over-read in rtw_update_protection
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Salman Alghamdi <[email protected]>
[ Upstream commit 514ab98364595007d4557ecc85d7e5f012c504d3 ]
rtw_update_protection() is called with a pointer offset into the ies buffer but the full ie_length is passed, causing a potential buffer over-read.
--- drivers/staging/rtl8723bs/core/rtw_mlme.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/staging/rtl8723bs/core/rtw_mlme.c +8 -2
@@ -449,8 +449,11 @@ static void update_current_network(strucif ((check_fwstate (pmlmepriv, _FW_LINKED) == true) && (is_same_network (&(pmlmepriv->cur_network.network), pnetwork,0 ))) {update_network (&(pmlmepriv->cur_network.network), pnetwork, adapter, true); +if (pmlmepriv->cur_network.network.ie_length <sizeof (struct ndis_802_11_fix_ie )) +return ; +rtw_update_protection (adapter, (pmlmepriv->cur_network.network.ies) +sizeof (struct ndis_802_11_fix_ie ), - pmlmepriv->cur_network.network.ie_length); + pmlmepriv->cur_network.network.ie_length -sizeof (struct ndis_802_11_fix_ie )); } }
@@ -1070,8 +1073,11 @@ static void rtw_joinbss_update_network(sbreak ; } +if (cur_network->network.ie_length <sizeof (struct ndis_802_11_fix_ie )) +return ; +rtw_update_protection (padapter, (cur_network->network.ies) +sizeof (struct ndis_802_11_fix_ie ), - (cur_network->network.ie_length)); + (cur_network->network.ie_length -sizeof (struct ndis_802_11_fix_ie )));rtw_update_ht_cap (padapter, cur_network->network.ies, cur_network->network.ie_length, (u8 ) cur_network->network.configuration.ds_config); }
Greg Kroah-Hartman
maintainerPATCH 46/204 fhandle: fix UAF due to unlocked ->mnt_ns read in may_decode_fh()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jann Horn <[email protected]>
[ Upstream commit 40ab6644b99685755f740b872c00ef40d9aa870e ]
may_decode_fh() accesses mount::mnt_ns without holding any locks; that means the mount can concurrently be unmounted, and the mnt_namespace can concurrently be freed after an RCU grace period.
This race can happens as follows, assuming that the mount point was created by open_tree(..., OPEN_TREE_CLONE):
thread 1 thread 2 RCU
__do_sys_open_by_handle_at
do_handle_open
handle_to_path
may_decode_fh
is_mounted
[mount::mnt_ns access]
[mount::mnt_ns access]
__do_sys_close
fput_close_sync
__fput
dissolve_on_fput
umount_tree
class_namespace_excl_destructor
namespace_unlock
free_mnt_ns
mnt_ns_tree_remove
call_rcu(mnt_ns_release_rcu)
mnt_ns_release_rcu
mnt_ns_release
kfree
[mnt_namespace::user_ns access] **UAF**Fix it by taking rcu_read_lock() around the mount::mnt_ns access, like in __prepend_path().
Additionally, document the semantics of mount::mnt_ns, and use WRITE_ONCE() for writers that can race with lockless readers.
This bug is unreachable unless one of the following is set:
- CONFIG_PREEMPTION
- CONFIG_RCU_STRICT_GRACE_PERIOD
because it requires an RCU grace period to happen during a syscall without an explicit preemption.
This doesn't seem to have interesting security impact; worst-case, it could leak the result of an integer comparison to userspace (from the level check in cap_capable()), cause an endless loop, or crash the kernel by dereferencing an invalid address.
--- fs/fhandle.c | 16 ++++++++++++++-- fs/mount.h | 10 +++++++++- fs/namespace.c | 6 +++--- 3 files changed, 26 insertions(+), 6 deletions(-)
fs/fhandle.c +14 -2
@@ -242,6 +242,19 @@ static int do_handle_to_path(struct filereturn 0 ; } +static bool capable_wrt_mount (struct mount *mount) +{ +struct mnt_namespace *mnt_ns; + +/* +* For ->mnt_ns access. +* The following READ_ONCE() is semantically rcu_dereference(). +*/ +guard (rcu)(); + mnt_ns =READ_ONCE (mount->mnt_ns); +return ns_capable (mnt_ns->user_ns,CAP_SYS_ADMIN ); +} +/* * Allow relaxed permissions of file handles if the caller has the * ability to mount the filesystem or create a bind-mount of the
@@ -273,8 +286,7 @@ static inline bool may_decode_fh(structif (ns_capable (root->mnt->mnt_sb->s_user_ns,CAP_SYS_ADMIN )) ctx->flags =HANDLE_CHECK_PERMS ;else if (is_mounted (root->mnt) && -ns_capable (real_mount (root->mnt)->mnt_ns->user_ns, -CAP_SYS_ADMIN ) && +capable_wrt_mount (real_mount (root->mnt)) && !has_locked_children (real_mount (root->mnt), root->dentry)) ctx->flags =HANDLE_CHECK_PERMS |HANDLE_CHECK_SUBTREE ;else
fs/mount.h +9 -1
@@ -58,7 +58,15 @@ struct mount {struct list_head mnt_slave_list;/* list of slave mounts */ struct list_head mnt_slave;/* slave list entry */ struct mount *mnt_master;/* slave is on master->mnt_slave_list */ -struct mnt_namespace *mnt_ns;/* containing namespace */ +/* +* Containing namespace (active or deactivating, non-refcounted). +* Normally protected by namespace_sem. +* Can also be accessed locklessly under RCU. RCU readers can't rely on +* the namespace still being active, but implicitly hold a passive +* reference (because an RCU delay happens between a namespace being +* deactivated and the corresponding passive refcount drop). +*/ +struct mnt_namespace *mnt_ns;struct mountpoint *mnt_mp;/* where is it mounted */ union {struct hlist_node mnt_mp_list;/* list mounts with the same mountpoint */
fs/namespace.c +3 -3
@@ -1132,7 +1132,7 @@ static void mnt_add_to_ns(struct mnt_namstruct rb_node *parent =NULL ;WARN_ON (mnt_ns_attached (mnt)); - mnt->mnt_ns = ns; +WRITE_ONCE (mnt->mnt_ns, ns);while (*link) { parent = *link;if (mnt->mnt_id_unique <node_to_mount (parent)->mnt_id_unique)
@@ -1493,7 +1493,7 @@ EXPORT_SYMBOL(mntget);void mnt_make_shortterm (struct vfsmount *mnt) {if (mnt) -real_mount (mnt)->mnt_ns =NULL ; +WRITE_ONCE (real_mount (mnt)->mnt_ns,NULL ); }/**
@@ -1805,7 +1805,7 @@ static void umount_tree(struct mount *mn ns->nr_mounts--;__touch_mnt_namespace (ns); } - p->mnt_ns =NULL ; +WRITE_ONCE (p->mnt_ns,NULL );if (how &UMOUNT_SYNC ) p->mnt.mnt_flags |=MNT_SYNC_UMOUNT ;
Greg Kroah-Hartman
maintainerPATCH 47/204 Drivers: hv: vmbus: Improve the logic of reserving fb_mmio on Gen2 VMs
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dexuan Cui <[email protected]>
[ Upstream commit 016a25e4b0df4d77e7c258edee4aaf982e4ee809 ]
If vmbus_reserve_fb() in the kdump/kexec kernel fails to properly reserve the framebuffer MMIO range (which is below 4GB) due to a Gen2 VM's screen.lfb_base being zero [1], there is an MMIO conflict between the drivers hyperv-drm and pci-hyperv: when the driver pci-hyperv's hv_allocate_config_window() calls vmbus_allocate_mmio() to get an MMIO range, typically it gets a 32-bit MMIO range that overlaps with the framebuffer MMIO range, and later hv_pci_enter_d0() fails with an error message "PCI Pass-through VSP failed D0 Entry with status" since the host thinks that PCI devices must not use MMIO space that the host has assigned to the framebuffer.
This is especially an issue if pci-hyperv is built-in and hyperv-drm is built as a module. Consequently, the kdump/kexec kernel fails to detect PCI devices via pci-hyperv, and may fail to mount the root file system, which may reside in a NVMe disk. The issue described here has existed for SR-IOV VF NICs since day one of the pci-hyperv driver, and has been worked around on x64 when possible. With the recent introduction of ARM64 VMs that boot from NVMe, there is no workaround, so we need a formal fix.
On Gen2 VMs, if the screen.lfb_base is 0 in the kdump/kexec kernel [1], fall back to the low MMIO base, which should be equal to the framebuffer MMIO base [2] (the statement is true according to my testing on x64 Windows Server 2016, and on x64 and ARM64 Windows Server 2025 and on Azure. I checked with the Hyper-V team and they said the statement should continue to be true for Gen2 VMs). In the first kernel, screen.lfb_base is not 0; if the user specifies a very high resolution, it's not enough to only reserve 8MB: let's always reserve half of the space below 4GB, but cap the reservation to 128MB, which is the required framebuffer size of the highest resolution 7680*4320 supported by Hyper-V.
While at it, fix the comparison "end > VTPM_BASE_ADDRESS" by changing the > to >=. Here the 'end' is an inclusive end (typically, it's 0xFFFF_FFFF for the low MMIO range).
Note: vmbus_reserve_fb() now also reserves an MMIO range at the beginning of the low MMIO range on CVMs, which have no framebuffers (the 'screen.lfb_base' in vmbus_reserve_fb() is 0 for CVMs), just in case the host might treat the beginning of the low MMIO range specially [3]. BTW, the OpenHCL kernel is not affected by the change, because that kernel boots with DeviceTree rather than ACPI (so vmbus_reserve_fb() won't run there), and there is no framebuffer device for that kernel.
Note: normally Gen1 VMs don't have the MMIO conflict issue because the framebuffer MMIO range (which is hardcoded to base=4GB-128MB and size=64MB for Gen1 VMs by the host) is always reported via the legacy PCI graphics device's BAR, so the kdump/kexec kernel can reserve the 64MB MMIO range; however, if the VM is configured to use a very high resolution and the required framebuffer size exceeds 64MB (AFAIK, in practice, this isn't a typical configuration by users), the hyperv-drm driver may need to allocate an MMIO range above 4GB and change the framebuffer MMIO location to the allocated MMIO range -- in this case, there can still be issues [4] which can't be easily fixed: any possible affected Gen1 users would have to use a resolution whose framebuffer size is <= 64MB, or switch to Gen2 VMs.
[1] https://lore.kernel.org/all/SA1PR21MB692176C1BC53BFC9EAE5CF8EBF51A@SA1PR21MB6921.namprd21.prod.outlook.com/
[2] https://lore.kernel.org/all/SA1PR21MB69218F955B62DFF62E3E88D2BF222@SA1PR21MB6921.namprd21.prod.outlook.com/
[3] https://lore.kernel.org/all/SN6PR02MB415726B17D5A6027CD1717E8D4342@SN6PR02MB4157.namprd02.prod.outlook.com/
[4] https://lore.kernel.org/all/SA1PR21MB69213486F821CA5A2C793C81BF342@SA1PR21MB6921.namprd21.prod.outlook.com/
--- drivers/hv/vmbus_drv.c | 29 ++++++++++++++++++++++++++--- 1 file changed, 26 insertions(+), 3 deletions(-)
drivers/hv/vmbus_drv.c +26 -3
@@ -2205,8 +2205,8 @@ static acpi_status vmbus_walk_resources(return AE_NO_MEMORY ;/* If this range overlaps the virtual TPM, truncate it. */ -if (end >VTPM_BASE_ADDRESS && start <VTPM_BASE_ADDRESS ) - end =VTPM_BASE_ADDRESS ; +if (end >=VTPM_BASE_ADDRESS && start <VTPM_BASE_ADDRESS ) + end =VTPM_BASE_ADDRESS -1 ; new_res->name ="hyperv mmio" ; new_res->flags =IORESOURCE_MEM ;
@@ -2273,6 +2273,7 @@ static void vmbus_mmio_remove(void)static void __maybe_unused vmbus_reserve_fb (void ) {resource_size_t start =0 , size; +resource_size_t low_mmio_base;struct pci_dev *pdev;if (efi_enabled (EFI_BOOT )) {
@@ -2280,6 +2281,24 @@ static void __maybe_unused vmbus_reserveif (IS_ENABLED (CONFIG_SYSFB )) { start = screen_info.lfb_base; size =max_t (__u32, screen_info.lfb_size,0x800000 ); + + low_mmio_base = hyperv_mmio->start; +if (!low_mmio_base ||upper_32_bits (low_mmio_base) || + (start && start < low_mmio_base)) { +pr_warn ("Unexpected low mmio base %pa\n" , &low_mmio_base); + }else { +/* +* If the kdump/kexec or CVM kernel's lfb_base +* is 0, fall back to the low mmio base. +*/ +if (!start) + start = low_mmio_base; +/* +* Reserve half of the space below 4GB for high +* resolutions, but cap the reservation to 128MB. +*/ + size =min ((SZ_4G - start) /2 ,SZ_128M ); + } } }else {/* Gen1 VM: get FB base from PCI */
@@ -2300,8 +2319,10 @@ static void __maybe_unused vmbus_reservepci_dev_put (pdev); } -if (!start) +if (!start) { +pr_warn ("Unexpected framebuffer mmio base of zero\n" );return ; + }/* * Make a claim for the frame buffer in the resource tree under the
@@ -2311,6 +2332,8 @@ static void __maybe_unused vmbus_reserve*/ for (; !fb_mmio && (size >=0x100000 ); size >>=1 ) fb_mmio =__request_region (hyperv_mmio, start, size, fb_mmio_name,0 ); + +pr_info ("hv_mmio=%pR,%pR fb=%pR\n" , hyperv_mmio, hyperv_mmio->sibling, fb_mmio); }/**
Greg Kroah-Hartman
maintainerPATCH 48/204 hv: utils: handle and propagate errors in kvp_register
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thorsten Blum <[email protected]>
[ Upstream commit 3fcf923302a8f5c0dc3af3d2ca2657cb5fae4297 ]
Make kvp_register() return an error code instead of silently ignoring failures, and propagate the error from kvp_handle_handshake() instead of returning success.
This propagates both kzalloc_obj() and hvutil_transport_send() failures to kvp_handle_handshake() and thus to kvp_on_msg().
--- drivers/hv/hv_kvp.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-)
drivers/hv/hv_kvp.c +14 -13
@@ -93,7 +93,7 @@ static void kvp_send_key(struct work_strstatic void kvp_respond_to_host (struct hv_kvp_msg *msg,int error);static void kvp_timeout_func (struct work_struct *dummy);static void kvp_host_handshake_func (struct work_struct *dummy); -static void kvp_register (int ); +static int kvp_register (int ); static DECLARE_DELAYED_WORK(kvp_timeout_work, kvp_timeout_func); static DECLARE_DELAYED_WORK(kvp_host_handshake_work, kvp_host_handshake_func);
@@ -127,24 +127,26 @@ static void kvp_register_done(void)hv_poll_channel (kvp_transaction.recv_channel, kvp_poll_wrapper); } -static void +static int kvp_register (int reg_value) {struct hv_kvp_msg *kvp_msg;char *version; +int ret; kvp_msg =kzalloc (sizeof (*kvp_msg),GFP_KERNEL ); +if (!kvp_msg) +return -ENOMEM ; -if (kvp_msg) { - version = kvp_msg->body.kvp_register.version; - kvp_msg->kvp_hdr.operation = reg_value; -strcpy (version,HV_DRV_VERSION ); - -hvutil_transport_send (hvt, kvp_msg,sizeof (*kvp_msg), - kvp_register_done); -kfree (kvp_msg); - } + version = kvp_msg->body.kvp_register.version; + kvp_msg->kvp_hdr.operation = reg_value; +strcpy (version,HV_DRV_VERSION ); + + ret =hvutil_transport_send (hvt, kvp_msg,sizeof (*kvp_msg), + kvp_register_done); +kfree (kvp_msg); +return ret; }static void kvp_timeout_func (struct work_struct *dummy)
@@ -186,9 +188,8 @@ static int kvp_handle_handshake(struct h*/ pr_debug ("KVP: userspace daemon ver. %d connected\n" , msg->kvp_hdr.operation); -kvp_register (dm_reg_value); -return 0 ; +return kvp_register (dm_reg_value); }
Greg Kroah-Hartman
maintainerPATCH 49/204 futex/requeue: Prevent NULL pointer dereference in remove_waiter() on self-deadlock
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ji'an Zhou <[email protected]>
[ Upstream commit 74e144274af39935b0f410c0ee4d2b91c3730414 ]
When FUTEX_CMP_REQUEUE_PI requeues a non-top waiter that already owns the target PI futex, task_blocks_on_rt_mutex() returns -EDEADLK before setting waiter->task.
The subsequent remove_waiter() in rt_mutex_start_proxy_lock() dereferences the NULL waiter->task, causing a kernel crash.
Add a self-deadlock check for non-top waiters before calling rt_mutex_start_proxy_lock(), analogous to the top-waiter check in futex_lock_pi_atomic().
--- kernel/futex/requeue.c | 6 ++++++ 1 file changed, 6 insertions(+)
kernel/futex/requeue.c +6 -0
@@ -633,6 +633,12 @@ retry_private:continue ; } +/* Self-deadlock: non-top waiter already owns the PI futex. */ +if (rt_mutex_owner (&pi_state->pi_mutex) == this->task) { + ret = -EDEADLK ; +break ; + } + ret =rt_mutex_start_proxy_lock (&pi_state->pi_mutex, this->rt_waiter, this->task);
Greg Kroah-Hartman
maintainerPATCH 50/204 locking/mutex: Remove wakeups from under mutex::wait_lock
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Peter Zijlstra <[email protected]>
[ Upstream commit 894d1b3db41cf7e6ae0304429a1747b3c3f390bc ]
In preparation to nest mutex::wait_lock under rq::lock we need to remove wakeups from under it.
Do this by utilizing wake_qs to defer the wakeup until after the lock is dropped.
[Heavily changed after 55f036ca7e74 ("locking: WW mutex cleanup") and 08295b3b5bee ("locking: Implement an algorithm choice for Wound-Wait mutexes")]
[jstultz: rebased to mainline, added extra wake_up_q & init
to avoid hangs, similar to Connor's rework of this patch]
Stable-dep-of: 40a25d59e85b ("locking/rtmutex: Skip remove_waiter() when waiter is not enqueued")
--- kernel/futex/pi.c | 6 +++- kernel/locking/mutex.c | 16 +++++++++--- kernel/locking/rtmutex.c | 51 +++++++++++++++++++++++++++++----------- kernel/locking/rtmutex_api.c | 12 +++++++-- kernel/locking/rtmutex_common.h | 3 +- kernel/locking/rwbase_rt.c | 8 +++++- kernel/locking/rwsem.c | 4 +-- kernel/locking/spinlock_rt.c | 5 ++- kernel/locking/ww_mutex.h | 30 ++++++++++++++--------- 9 files changed, 96 insertions(+), 39 deletions(-)
kernel/futex/pi.c +5 -1
@@ -922,6 +922,7 @@ int futex_lock_pi(u32 __user *uaddr, unsstruct rt_mutex_waiter rt_waiter;struct futex_hash_bucket *hb;struct futex_q q = futex_q_init; +DEFINE_WAKE_Q (wake_q);int res, ret;if (!IS_ENABLED (CONFIG_FUTEX_PI ))
@@ -1019,8 +1020,11 @@ retry_private:* such that futex_unlock_pi() is guaranteed to observe the waiter when * it sees the futex_q::pi_state. */ - ret =__rt_mutex_start_proxy_lock (&q.pi_state->pi_mutex, &rt_waiter, current); + ret =__rt_mutex_start_proxy_lock (&q.pi_state->pi_mutex, &rt_waiter, current, &wake_q); +preempt_disable ();raw_spin_unlock_irq (&q.pi_state->pi_mutex.wait_lock); +wake_up_q (&wake_q); +preempt_enable ();if (ret) {if (ret ==1 )
kernel/locking/mutex.c +12 -4
@@ -575,6 +575,7 @@ __mutex_lock_common(struct mutex *lock, struct lockdep_map *nest_lock, unsigned long ip, struct ww_acquire_ctx *ww_ctx, const bool use_ww_ctx) { +DEFINE_WAKE_Q (wake_q );struct mutex_waiter waiter;struct ww_mutex *ww;int ret;
@@ -625,7 +626,7 @@ __mutex_lock_common(struct mutex *lock,*/ if (__mutex_trylock (lock)) {if (ww_ctx) -__ww_mutex_check_waiters (lock, ww_ctx); +__ww_mutex_check_waiters (lock, ww_ctx, &wake_q); goto skip_wait; }
@@ -645,7 +646,7 @@ __mutex_lock_common(struct mutex *lock,* Add in stamp order, waking up waiters that must kill * themselves. */ - ret =__ww_mutex_add_waiter (&waiter, lock, ww_ctx); + ret =__ww_mutex_add_waiter (&waiter, lock, ww_ctx, &wake_q);if (ret) goto err_early_kill; }
@@ -681,6 +682,10 @@ __mutex_lock_common(struct mutex *lock, }raw_spin_unlock (&lock->wait_lock); +/* Make sure we do wakeups before calling schedule */ +wake_up_q (&wake_q); +wake_q_init (&wake_q); +schedule_preempt_disabled (); first =__mutex_waiter_is_first (lock, &waiter);
@@ -714,7 +719,7 @@ acquired:*/ if (!ww_ctx->is_wait_die && !__mutex_waiter_is_first (lock, &waiter)) -__ww_mutex_check_waiters (lock, ww_ctx); +__ww_mutex_check_waiters (lock, ww_ctx, &wake_q); }__mutex_remove_waiter (lock, &waiter);
@@ -730,6 +735,7 @@ skip_wait:ww_mutex_lock_acquired (ww, ww_ctx);raw_spin_unlock (&lock->wait_lock); +wake_up_q (&wake_q);preempt_enable ();return 0 ;
@@ -741,6 +747,7 @@ err_early_kill:raw_spin_unlock (&lock->wait_lock);debug_mutex_free_waiter (&waiter);mutex_release (&lock->dep_map, ip); +wake_up_q (&wake_q);preempt_enable ();return ret; }
@@ -951,9 +958,10 @@ static noinline void __sched __mutex_unlif (owner &MUTEX_FLAG_HANDOFF )__mutex_handoff (lock, next); +preempt_disable ();raw_spin_unlock (&lock->wait_lock); -wake_up_q (&wake_q); +preempt_enable (); }#ifndef CONFIG_DEBUG_LOCK_ALLOC
kernel/locking/rtmutex.c +37 -14
157 lines not on this page · raw message
kernel/locking/rtmutex_api.c +9 -3
@@ -275,6 +275,7 @@ void __sched rt_mutex_proxy_unlock(struc* @lock: the rt_mutex to take * @waiter: the pre-initialized rt_mutex_waiter * @task: the task to prepare +* @wake_q: the wake_q to wake tasks after we release the wait_lock * * Starts the rt_mutex acquire; it enqueues the @waiter and does deadlock * detection. It does not wait, see rt_mutex_wait_proxy_lock() for that.
@@ -291,7 +292,8 @@ void __sched rt_mutex_proxy_unlock(struc*/ int __sched __rt_mutex_start_proxy_lock(struct rt_mutex_base *lock,struct rt_mutex_waiter *waiter, -struct task_struct *task) +struct task_struct *task, +struct wake_q_head *wake_q) {int ret;
@@ -302,7 +304,7 @@ int __sched __rt_mutex_start_proxy_lock(/* We enforce deadlock detection for futexes */ ret =task_blocks_on_rt_mutex (lock, waiter, task,NULL , -RT_MUTEX_FULL_CHAINWALK ); +RT_MUTEX_FULL_CHAINWALK , wake_q);if (ret && !rt_mutex_owner (lock)) {/*
@@ -341,12 +343,16 @@ int __sched rt_mutex_start_proxy_lock(st struct task_struct *task) {int ret; +DEFINE_WAKE_Q (wake_q);raw_spin_lock_irq (&lock->wait_lock); - ret =__rt_mutex_start_proxy_lock (lock, waiter, task); + ret =__rt_mutex_start_proxy_lock (lock, waiter, task, &wake_q);if (unlikely (ret))remove_waiter (lock, waiter); +preempt_disable ();raw_spin_unlock_irq (&lock->wait_lock); +wake_up_q (&wake_q); +preempt_enable ();return ret; }
kernel/locking/rtmutex_common.h +2 -1
@@ -83,7 +83,8 @@ extern void rt_mutex_init_proxy_locked(sextern void rt_mutex_proxy_unlock (struct rt_mutex_base *lock);extern int __rt_mutex_start_proxy_lock (struct rt_mutex_base *lock,struct rt_mutex_waiter *waiter, -struct task_struct *task); +struct task_struct *task, +struct wake_q_head *);extern int rt_mutex_start_proxy_lock (struct rt_mutex_base *lock,struct rt_mutex_waiter *waiter,struct task_struct *task);
kernel/locking/rwbase_rt.c +7 -1
@@ -69,6 +69,7 @@ static int __sched __rwbase_read_lock(st unsigned int state) {struct rt_mutex_base *rtm = &rwb->rtmutex; +DEFINE_WAKE_Q (wake_q);int ret;rwbase_pre_schedule ();
@@ -110,7 +111,7 @@ static int __sched __rwbase_read_lock(st* For rwlocks this returns 0 unconditionally, so the below * !ret conditionals are optimized out. */ - ret =rwbase_rtmutex_slowlock_locked (rtm, state); + ret =rwbase_rtmutex_slowlock_locked (rtm, state, &wake_q);/* * On success the rtmutex is held, so there can't be a writer
@@ -121,7 +122,12 @@ static int __sched __rwbase_read_lock(st*/ if (!ret)atomic_inc (&rwb->readers); + +preempt_disable ();raw_spin_unlock_irq (&rtm->wait_lock); +wake_up_q (&wake_q); +preempt_enable (); +if (!ret)rwbase_rtmutex_unlock (rtm);
kernel/locking/rwsem.c +2 -2
@@ -1413,8 +1413,8 @@ static inline void __downgrade_write(str#define rwbase_rtmutex_lock_state (rtm, state) \ __rt_mutex_lock(rtm, state) -#define rwbase_rtmutex_slowlock_locked (rtm, state) \ - __rt_mutex_slowlock_locked(rtm, NULL, state) +#define rwbase_rtmutex_slowlock_locked (rtm, state, wq) \ + __rt_mutex_slowlock_locked(rtm, NULL, state, wq)#define rwbase_rtmutex_unlock (rtm) \ __rt_mutex_unlock(rtm)
kernel/locking/spinlock_rt.c +3 -2
@@ -162,9 +162,10 @@ rwbase_rtmutex_lock_state(struct rt_mute } static __always_inline int -rwbase_rtmutex_slowlock_locked (struct rt_mutex_base *rtm,unsigned int state) +rwbase_rtmutex_slowlock_locked (struct rt_mutex_base *rtm,unsigned int state, +struct wake_q_head *wake_q) { -rtlock_slowlock_locked (rtm); +rtlock_slowlock_locked (rtm, wake_q);return 0 ; }
kernel/locking/ww_mutex.h +19 -11
@@ -275,7 +275,7 @@ __ww_ctx_less(struct ww_acquire_ctx *a,*/ static bool __ww_mutex_die (struct MUTEX *lock,struct MUTEX_WAITER *waiter, -struct ww_acquire_ctx *ww_ctx) +struct ww_acquire_ctx *ww_ctx,struct wake_q_head *wake_q) {if (!ww_ctx->is_wait_die)return false;
@@ -284,7 +284,7 @@ __ww_mutex_die(struct MUTEX *lock, struc#ifndef WW_RT debug_mutex_wake_waiter (lock, waiter);#endif -wake_up_process (waiter->task); +wake_q_add (wake_q, waiter->task); }return true;
@@ -299,7 +299,8 @@ __ww_mutex_die(struct MUTEX *lock, struc*/ static bool __ww_mutex_wound (struct MUTEX *lock,struct ww_acquire_ctx *ww_ctx, -struct ww_acquire_ctx *hold_ctx) +struct ww_acquire_ctx *hold_ctx, +struct wake_q_head *wake_q) {struct task_struct *owner =__ww_mutex_owner (lock);
@@ -331,7 +332,7 @@ static bool __ww_mutex_wound(struct MUTE* wakeup pending to re-read the wounded state. */ if (owner != current) -wake_up_process (owner); +wake_q_add (wake_q, owner);return true; }
@@ -352,7 +353,8 @@ static bool __ww_mutex_wound(struct MUTE* The current task must not be on the wait list. */ static void -__ww_mutex_check_waiters (struct MUTEX *lock,struct ww_acquire_ctx *ww_ctx) +__ww_mutex_check_waiters (struct MUTEX *lock,struct ww_acquire_ctx *ww_ctx, +struct wake_q_head *wake_q) {struct MUTEX_WAITER *cur;
@@ -364,8 +366,8 @@ __ww_mutex_check_waiters(struct MUTEX *lif (!cur->ww_ctx)continue ; -if (__ww_mutex_die (lock, cur, ww_ctx) || -__ww_mutex_wound (lock, cur->ww_ctx, ww_ctx)) +if (__ww_mutex_die (lock, cur, ww_ctx, wake_q) || +__ww_mutex_wound (lock, cur->ww_ctx, ww_ctx, wake_q))break ; } }
@@ -377,6 +379,8 @@ __ww_mutex_check_waiters(struct MUTEX *l static __always_inline voidww_mutex_set_context_fastpath (struct ww_mutex *lock,struct ww_acquire_ctx *ctx) { +DEFINE_WAKE_Q (wake_q); +ww_mutex_lock_acquired (lock, ctx);/*
@@ -405,8 +409,11 @@ ww_mutex_set_context_fastpath(struct ww_* die or wound us. */ lock_wait_lock (&lock->base); -__ww_mutex_check_waiters (&lock->base, ctx); +__ww_mutex_check_waiters (&lock->base, ctx, &wake_q); +preempt_disable ();unlock_wait_lock (&lock->base); +wake_up_q (&wake_q); +preempt_enable (); }static __always_inline int
@@ -488,7 +495,8 @@ __ww_mutex_check_kill(struct MUTEX *lockstatic inline int __ww_mutex_add_waiter (struct MUTEX_WAITER *waiter,struct MUTEX *lock, -struct ww_acquire_ctx *ww_ctx) +struct ww_acquire_ctx *ww_ctx, +struct wake_q_head *wake_q) {struct MUTEX_WAITER *cur, *pos =NULL ;bool is_wait_die;
@@ -532,7 +540,7 @@ __ww_mutex_add_waiter(struct MUTEX_WAITE pos = cur;/* Wait-Die: ensure younger waiters die. */ -__ww_mutex_die (lock, cur, ww_ctx); +__ww_mutex_die (lock, cur, ww_ctx, wake_q); }__ww_waiter_add (lock, waiter, pos);
@@ -550,7 +558,7 @@ __ww_mutex_add_waiter(struct MUTEX_WAITE* such that either we or the fastpath will wound @ww->ctx. */ smp_mb (); -__ww_mutex_wound (lock, ww_ctx, ww->ctx); +__ww_mutex_wound (lock, ww_ctx, ww->ctx, wake_q); }return 0 ;
Greg Kroah-Hartman
maintainerPATCH 51/204 locking/rtmutex: Skip remove_waiter() when waiter is not enqueued
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Davidlohr Bueso <[email protected]>
[ Upstream commit 40a25d59e85b3c8709ac2424d44f65610467871e ]
syzbot triggered the following splat in remove_waiter() via FUTEX_CMP_REQUEUE_PI:
KASAN: null-ptr-deref in range [0x0000000000000a88-0x0000000000000a8f] class_raw_spinlock_constructor remove_waiter+0x159/0x1200 kernel/locking/rtmutex.c:1561 rt_mutex_start_proxy_lock+0x103/0x120 futex_requeue+0x10e4/0x20d0 __x64_sys_futex+0x34f/0x4d0
task_blocks_on_rt_mutex() does not arm the waiter upon deadlock detection, leaving waiter->task nil, where 3bfdc63936dd ("rtmutex: Use waiter::task instead of current in remove_waiter()") made this fatal.
Furthermore, rt_mutex_start_proxy_lock() should not be calling into remove_waiter() upon a successfully grabbing the rtmutex. 1a1fb985f2e2 ("futex: Handle early deadlock return correctly"), moved the remove_waiter() out of __rt_mutex_start_proxy_lock() (where 'ret' was only ever 0 or < 0) into the wrapper. Tighten this check to account for try_to_take_rt_mutex().
--- kernel/locking/rtmutex.c | 3 +++ kernel/locking/rtmutex_api.c | 2 +- 2 files changed, 4 insertions(+), 1 deletion(-)
kernel/locking/rtmutex.c +3 -0
@@ -1550,6 +1550,9 @@ static void __sched remove_waiter(structlockdep_assert_held (&lock->wait_lock); +if (!waiter_task)/* never enqueued */ +return ; + scoped_guard(raw_spinlock, &waiter_task->pi_lock) {rt_mutex_dequeue (lock, waiter); waiter_task->pi_blocked_on =NULL ;
kernel/locking/rtmutex_api.c +1 -1
@@ -347,7 +347,7 @@ int __sched rt_mutex_start_proxy_lock(straw_spin_lock_irq (&lock->wait_lock); ret =__rt_mutex_start_proxy_lock (lock, waiter, task, &wake_q); -if (unlikely (ret)) +if (unlikely (ret <0 ))remove_waiter (lock, waiter);preempt_disable ();raw_spin_unlock_irq (&lock->wait_lock);
Greg Kroah-Hartman
maintainerPATCH 52/204 phonet: Pass ifindex to fill_addr().
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 08a9572be36819b5d9011604edfa5db6c5062a7a ]
We will convert addr_doit() and getaddr_dumpit() to RCU, both of which call fill_addr().
The former will call phonet_address_notify() outside of RCU due to GFP_KERNEL, so dev will not be available in fill_addr().
Let's pass ifindex directly to fill_addr().
Stable-dep-of: 71de0177b28d ("net: phonet: free phonet_device after RCU grace period")
--- net/phonet/pn_netlink.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)
net/phonet/pn_netlink.c +7 -6
@@ -19,7 +19,7 @@/* Device address handling */ -static int fill_addr (struct sk_buff *skb,struct net_device *dev,u8 addr, +static int fill_addr (struct sk_buff *skb,u32 ifindex,u8 addr,u32 portid,u32 seq,int event);void phonet_address_notify (int event,struct net_device *dev,u8 addr)
@@ -31,7 +31,8 @@ void phonet_address_notify(int event, st nla_total_size(1), GFP_KERNEL);if (skb ==NULL ) goto errout; - err =fill_addr (skb, dev, addr,0 ,0 , event); + + err =fill_addr (skb, dev->ifindex, addr,0 ,0 , event);if (err <0 ) {WARN_ON (err == -EMSGSIZE );kfree_skb (skb);
@@ -92,8 +93,8 @@ static int addr_doit(struct sk_buff *skbreturn err; } -static int fill_addr (struct sk_buff *skb,struct net_device *dev,u8 addr, -u32 portid,u32 seq,int event) +static int fill_addr (struct sk_buff *skb,u32 ifindex,u8 addr, +u32 portid,u32 seq,int event) {struct ifaddrmsg *ifm;struct nlmsghdr *nlh;
@@ -107,7 +108,7 @@ static int fill_addr(struct sk_buff *skb ifm->ifa_prefixlen =0 ; ifm->ifa_flags =IFA_F_PERMANENT ; ifm->ifa_scope =RT_SCOPE_LINK ; - ifm->ifa_index = dev->ifindex; + ifm->ifa_index = ifindex;if (nla_put_u8 (skb,IFA_LOCAL , addr)) goto nla_put_failure;nlmsg_end (skb, nlh);
@@ -140,7 +141,7 @@ static int getaddr_dumpit(struct sk_buffif (addr_idx++ < addr_start_idx)continue ; -if (fill_addr (skb, pnd->netdev, addr <<2 , +if (fill_addr (skb, pnd->netdev->ifindex, addr <<2 ,NETLINK_CB (cb->skb).portid, cb->nlh->nlmsg_seq,RTM_NEWADDR ) <0 ) goto out;
Greg Kroah-Hartman
maintainerPATCH 53/204 phonet: Pass net and ifindex to phonet_address_notify().
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 68ed5c38b512b734caf3da1f87db4a99fcfe3002 ]
Currently, phonet_address_notify() fetches netns and ifindex from dev.
Once addr_doit() is converted to RCU, phonet_address_notify() will be called outside of RCU due to GFP_KERNEL, and dev will be unavailable there.
Let's pass net and ifindex to phonet_address_notify().
Stable-dep-of: 71de0177b28d ("net: phonet: free phonet_device after RCU grace period")
--- include/net/phonet/pn_dev.h | 2 +- net/phonet/pn_dev.c | 10 +++++++--- net/phonet/pn_netlink.c | 12 ++++++------ 3 files changed, 14 insertions(+), 10 deletions(-)
include/net/phonet/pn_dev.h +1 -1
@@ -38,7 +38,7 @@ int phonet_address_add(struct net_deviceint phonet_address_del (struct net_device *dev,u8 addr);u8 phonet_address_get (struct net_device *dev,u8 addr);int phonet_address_lookup (struct net *net,u8 addr); -void phonet_address_notify (int event,struct net_device *dev,u8 addr); +void phonet_address_notify (struct net *net,int event,u32 ifindex,u8 addr);int phonet_route_add (struct net_device *dev,u8 daddr);int phonet_route_del (struct net_device *dev,u8 daddr);
net/phonet/pn_dev.c +7 -3
@@ -98,10 +98,13 @@ static void phonet_device_destroy(structmutex_unlock (&pndevs->lock);if (pnd) { +struct net *net =dev_net (dev); +u32 ifindex = dev->ifindex;u8 addr;for_each_set_bit (addr, pnd->addrs,64 ) -phonet_address_notify (RTM_DELADDR , dev, addr); +phonet_address_notify (net,RTM_DELADDR , ifindex, addr); +kfree (pnd); } }
@@ -244,8 +247,9 @@ static int phonet_device_autoconf(struct ret =phonet_address_add (dev, req.ifr_phonet_autoconf.device);if (ret)return ret; -phonet_address_notify (RTM_NEWADDR , dev, - req.ifr_phonet_autoconf.device); + +phonet_address_notify (dev_net (dev),RTM_NEWADDR , dev->ifindex, + req.ifr_phonet_autoconf.device);return 0 ; }
net/phonet/pn_netlink.c +6 -6
@@ -22,7 +22,7 @@static int fill_addr (struct sk_buff *skb,u32 ifindex,u8 addr,u32 portid,u32 seq,int event); -void phonet_address_notify (int event,struct net_device *dev,u8 addr) +void phonet_address_notify (struct net *net,int event,u32 ifindex,u8 addr) {struct sk_buff *skb;int err = -ENOBUFS ;
@@ -32,17 +32,17 @@ void phonet_address_notify(int event, stif (skb ==NULL ) goto errout; - err =fill_addr (skb, dev->ifindex, addr,0 ,0 , event); + err =fill_addr (skb, ifindex, addr,0 ,0 , event);if (err <0 ) {WARN_ON (err == -EMSGSIZE );kfree_skb (skb); goto errout; } -rtnl_notify (skb,dev_net (dev),0 , -RTNLGRP_PHONET_IFADDR ,NULL ,GFP_KERNEL ); + +rtnl_notify (skb, net,0 ,RTNLGRP_PHONET_IFADDR ,NULL ,GFP_KERNEL );return ; errout: -rtnl_set_sk_err (dev_net (dev),RTNLGRP_PHONET_IFADDR , err); +rtnl_set_sk_err (net,RTNLGRP_PHONET_IFADDR , err); }static const struct nla_policy ifa_phonet_policy[IFA_MAX +1 ] = {
@@ -89,7 +89,7 @@ static int addr_doit(struct sk_buff *skbelse err =phonet_address_del (dev, pnaddr);if (!err) -phonet_address_notify (nlh->nlmsg_type, dev, pnaddr); +phonet_address_notify (net, nlh->nlmsg_type, ifm->ifa_index, pnaddr);return err; }
Greg Kroah-Hartman
maintainerPATCH 54/204 net: phonet: free phonet_device after RCU grace period
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Santosh Kalluri <[email protected]>
[ Upstream commit 71de0177b28da751f407581a4515cf4d762f6296 ]
phonet_device_destroy() removes a phonet_device from the per-net device list with list_del_rcu(), but frees it immediately. RCU readers walking the same list can still hold a pointer to the object after it has been removed, leading to a slab-use-after-free.
Use kfree_rcu(), matching the lifetime rule already used by phonet_address_del() for the same object type.
--- net/phonet/pn_dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/phonet/pn_dev.c +1 -1
@@ -105,7 +105,7 @@ static void phonet_device_destroy(structfor_each_set_bit (addr, pnd->addrs,64 )phonet_address_notify (net,RTM_DELADDR , ifindex, addr); -kfree (pnd); +kfree_rcu (pnd, rcu); } }
Greg Kroah-Hartman
maintainerPATCH 55/204 rxrpc: Fix the ACK parser to extract the SACK table for parsing
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Howells <[email protected]>
[ Upstream commit 333b6d5bb9f87827ac2639c737bf9613dbae7253 ]
Fix modification of the received skbuff in rxrpc_input_soft_acks() and a potential incorrect access of the buffer in a fragmented UDP packet (the packet would probably have to be deliberately pre-generated as fragmented) when AF_RXRPC tries to extract the contents of the SACK table by copying out the contents of the SACK table into a buffer before attempting to parse
AF_RXRPC assumes that it can just call skb_condense() and then validly access the SACK table from skb->data and that it will be a flat buffer - but skb_condense() can silently fail to do anything under some circumstances.
Note that whilst rxrpc_input_soft_acks() should be able to parse extended ACKs, the rest of AF_RXRPC doesn't currently support that.
Further, there's then no need to call skb_condense() in rxrpc_input_ack(), so don't.
--- net/rxrpc/input.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-)
net/rxrpc/input.c +16 -5
@@ -775,9 +775,23 @@ static void rxrpc_input_soft_acks(struct rxrpc_seq_t since) {struct rxrpc_skb_priv *sp =rxrpc_skb (skb); -unsigned int i, old_nacks =0 ; +unsigned int i, old_nacks =0 , nsack;rxrpc_seq_t lowest_nak = seq + sp->ack.nr_acks; -u8 *acks = skb->data +sizeof (struct rxrpc_wire_header ) +sizeof (struct rxrpc_ackpacket ); +u8 sack[256 ]__aligned(sizeof(unsigned long)) ; +u8 *acks = sack; + +/* AF_RXRPC assumes that it can access the SACK table directly from +* skb->data as a flat buffer, but the skb may be non-linear (e.g. a +* fragmented UDP packet) and skb_condense() can silently fail to +* linearise it. Copy the SACK table out into a local buffer before +* parsing it. +*/ +memset (sack,0 ,sizeof (sack)); + nsack =umin (sp->ack.nr_acks,256 ); +if (skb_copy_bits (skb, +sizeof (struct rxrpc_wire_header ) +sizeof (struct rxrpc_ackpacket ), + sack, nsack) <0 ) +return ;for (i =0 ; i < sp->ack.nr_acks; i++) {if (acks[i] ==RXRPC_ACK_TYPE_ACK ) {
@@ -934,9 +948,6 @@ static void rxrpc_input_ack(struct rxrpc skb_copy_bits(skb, ioffset, &trailer, sizeof(trailer)) < 0)return rxrpc_proto_abort (call,0 , rxrpc_badmsg_short_ack_trailer); -if (nr_acks >0 ) -skb_condense (skb); -if (call->cong_last_nack) { since =rxrpc_input_check_prev_ack (call, &summary, first_soft_ack);rxrpc_free_skb (call->cong_last_nack, rxrpc_skb_put_last_nack);
Greg Kroah-Hartman
maintainerPATCH 56/204 fuse: re-lock request before replacing page cache folio
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Joanne Koong <[email protected]>
[ Upstream commit a078484921052d0badd827fcc2770b5cfc1d4120 ]
fuse_try_move_folio() unlocks the request on entry but does not re-lock it on the success path. This means fuse_chan_abort() can end the request and free the fuse_io_args (eg fuse_readpages_end()) while the subsequent copy chain logic after fuse_try_move_folio() accesses the fuse_io_args, leading to use-after-free issues.
Fix this by calling lock_request() before replace_page_cache_folio(). This ensures the request is locked on the success path which will prevent the fuse_io_args from being freed while the later copying logic runs, and also ensures that the ap->folios[i]->mapping is never null since ap->folios[i] will always point to the newfolio after replace_page_cache_folio().
--- fs/fuse/dev.c | 19 +++++-------------- 1 file changed, 5 insertions(+), 14 deletions(-)
fs/fuse/dev.c +5 -14
@@ -891,6 +891,10 @@ static int fuse_try_move_page(struct fusif (WARN_ON (folio_test_mlocked (oldfolio))) goto out_fallback_unlock; + err =lock_request (cs->req); +if (err) + goto out_fallback_unlock; +replace_page_cache_folio (oldfolio, newfolio);folio_get (newfolio);
@@ -904,20 +908,7 @@ static int fuse_try_move_page(struct fus*/ pipe_buf_release (cs->pipe, buf); - err =0 ; -spin_lock (&cs->req->waitq.lock); -if (test_bit (FR_ABORTED , &cs->req->flags)) - err = -ENOENT ; -else - *pagep = &newfolio->page; -spin_unlock (&cs->req->waitq.lock); - -if (err) { -folio_unlock (newfolio); -folio_put (newfolio); - goto out_put_old; - } - + *pagep = &newfolio->page;folio_unlock (oldfolio);/* Drop ref for ap->pages[] array */ folio_put (oldfolio);
Greg Kroah-Hartman
maintainerPATCH 57/204 scripts/sorttable: Remove unused macro defines
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 28b24394c6e9a3166fcb4480cba054562526657c ]
The code of sorttable.h was copied from the recordmcount.h which defined a bunch of Elf MACROs so that they could be used between 32bit and 64bit functions. But there's several MACROs that sorttable.h does not use but was copied over. Remove them to clean up the code.
--- scripts/sorttable.h | 27 --------------------------- 1 file changed, 27 deletions(-)
scripts/sorttable.h +0 -27
@@ -27,19 +27,10 @@#undef Elf_Ehdr#undef Elf_Shdr#undef Elf_Rel -#undef Elf_Rela#undef Elf_Sym -#undef ELF_R_SYM -#undef Elf_r_sym -#undef ELF_R_INFO -#undef Elf_r_info -#undef ELF_ST_BIND#undef ELF_ST_TYPE -#undef fn_ELF_R_SYM -#undef fn_ELF_R_INFO#undef uint_t#undef _r -#undef _w#ifdef SORTTABLE_64 # define extable_ent_size 16
@@ -52,19 +43,10 @@# define Elf_Ehdr Elf64_Ehdr# define Elf_Shdr Elf64_Shdr# define Elf_Rel Elf64_Rel -# define Elf_Rela Elf64_Rela# define Elf_Sym Elf64_Sym -# define ELF_R_SYM ELF64_R_SYM -# define Elf_r_sym Elf64_r_sym -# define ELF_R_INFO ELF64_R_INFO -# define Elf_r_info Elf64_r_info -# define ELF_ST_BIND ELF64_ST_BIND# define ELF_ST_TYPE ELF64_ST_TYPE -# define fn_ELF_R_SYM fn_ELF64_R_SYM -# define fn_ELF_R_INFO fn_ELF64_R_INFO# define uint_t uint64_t# define _r r8 -# define _w w8#else # define extable_ent_size 8# define compare_extable compare_extable_32
@@ -76,19 +58,10 @@# define Elf_Ehdr Elf32_Ehdr# define Elf_Shdr Elf32_Shdr# define Elf_Rel Elf32_Rel -# define Elf_Rela Elf32_Rela# define Elf_Sym Elf32_Sym -# define ELF_R_SYM ELF32_R_SYM -# define Elf_r_sym Elf32_r_sym -# define ELF_R_INFO ELF32_R_INFO -# define Elf_r_info Elf32_r_info -# define ELF_ST_BIND ELF32_ST_BIND# define ELF_ST_TYPE ELF32_ST_TYPE -# define fn_ELF_R_SYM fn_ELF32_R_SYM -# define fn_ELF_R_INFO fn_ELF32_R_INFO# define uint_t uint32_t# define _r r -# define _w w#endif #if defined(SORTTABLE_64 ) && defined(UNWINDER_ORC_ENABLED )
Greg Kroah-Hartman
maintainerPATCH 58/204 scripts/sorttable: Remove unused write functions
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 4f48a28b37d594dab38092514a42ae9f4b781553 ]
The code of sorttable.h was copied from the recordmcount.h which defined various write functions for different sizes (2, 4, 8 byte lengths). But sorttable only uses the 4 byte writes. Remove the extra versions as they are not used.
--- scripts/sorttable.c | 26 -------------------------- 1 file changed, 26 deletions(-)
scripts/sorttable.c +0 -26
@@ -68,8 +68,6 @@ static uint32_t (*r)(const uint32_t *);static uint16_t (*r2)(const uint16_t *);static uint64_t (*r8)(const uint64_t *);static void (*w)(uint32_t ,uint32_t *); -static void (*w2)(uint16_t ,uint16_t *); -static void (*w8)(uint64_t ,uint64_t *);typedef void (*table_sort_t )(char *,int );/*
@@ -146,31 +144,11 @@ static void wbe(uint32_t val, uint32_t *put_unaligned_be32 (val, x); } -static void w2be (uint16_t val,uint16_t *x) -{ -put_unaligned_be16 (val, x); -} - -static void w8be (uint64_t val,uint64_t *x) -{ -put_unaligned_be64 (val, x); -} -static void wle (uint32_t val,uint32_t *x) {put_unaligned_le32 (val, x); } -static void w2le (uint16_t val,uint16_t *x) -{ -put_unaligned_le16 (val, x); -} - -static void w8le (uint64_t val,uint64_t *x) -{ -put_unaligned_le64 (val, x); -} -/* * Move reserved section indices SHN_LORESERVE..SHN_HIRESERVE out of * the way to -256..-1, to avoid conflicting with real section
@@ -277,16 +255,12 @@ static int do_file(char const *const fna r2 = r2le; r8 = r8le; w = wle; - w2 = w2le; - w8 = w8le;break ;case ELFDATA2MSB : r = rbe; r2 = r2be; r8 = r8be; w = wbe; - w2 = w2be; - w8 = w8be;break ;default : fprintf(stderr, "unrecognized ELF data encoding %d: %s\n",
Greg Kroah-Hartman
maintainerPATCH 59/204 scripts/sorttable: Remove unneeded Elf_Rel
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 6f2c2f93a190467cebd6ebd03feb49514fead5ca ]
The code had references to initialize the Elf_Rel relocation tables, but it was never used. Remove it.
--- scripts/sorttable.h | 23 ++--------------------- 1 file changed, 2 insertions(+), 21 deletions(-)
scripts/sorttable.h +2 -21
@@ -26,7 +26,6 @@#undef Elf_Addr#undef Elf_Ehdr#undef Elf_Shdr -#undef Elf_Rel#undef Elf_Sym#undef ELF_ST_TYPE#undef uint_t
@@ -42,7 +41,6 @@# define Elf_Addr Elf64_Addr# define Elf_Ehdr Elf64_Ehdr# define Elf_Shdr Elf64_Shdr -# define Elf_Rel Elf64_Rel# define Elf_Sym Elf64_Sym# define ELF_ST_TYPE ELF64_ST_TYPE# define uint_t uint64_t
@@ -57,7 +55,6 @@# define Elf_Addr Elf32_Addr# define Elf_Ehdr Elf32_Ehdr# define Elf_Shdr Elf32_Shdr -# define Elf_Rel Elf32_Rel# define Elf_Sym Elf32_Sym# define ELF_ST_TYPE ELF32_ST_TYPE# define uint_t uint32_t
@@ -248,14 +245,10 @@ static int do_sort(Elf_Ehdr *ehdr,Elf32_Word *symtab_shndx =NULL ;Elf_Sym *sort_needed_sym =NULL ;Elf_Shdr *sort_needed_sec; -Elf_Rel *relocs =NULL ; -int relocs_size =0 ;uint32_t *sort_needed_loc;const char *secstrings;const char *strtab;char *extab_image; -int extab_index =0 ; -int i;int idx;unsigned int shnum;unsigned int shstrndx;
@@ -279,23 +272,15 @@ static int do_sort(Elf_Ehdr *ehdr,if (shnum ==SHN_UNDEF ) shnum =_r (&shdr[0 ].sh_size); -for (i =0 , s = shdr; s < shdr + shnum; i++, s++) { +for (s = shdr; s < shdr + shnum; s++) { idx =r (&s->sh_name); -if (!strcmp (secstrings + idx,"__ex_table" )) { +if (!strcmp (secstrings + idx,"__ex_table" )) extab_sec = s; - extab_index = i; - }if (!strcmp (secstrings + idx,".symtab" )) symtab_sec = s;if (!strcmp (secstrings + idx,".strtab" )) strtab_sec = s; -if ((r (&s->sh_type) ==SHT_REL || -r (&s->sh_type) ==SHT_RELA ) && -r (&s->sh_info) == extab_index) { - relocs = (void *)ehdr +_r (&s->sh_offset); - relocs_size =_r (&s->sh_size); - }if (r (&s->sh_type) ==SHT_SYMTAB_SHNDX ) symtab_shndx = (Elf32_Word *)((const char *)ehdr +_r (&s->sh_offset));
@@ -397,10 +382,6 @@ static int do_sort(Elf_Ehdr *ehdr, extable_ent_size, compare_extable); } -/* If there were relocations, we no longer need them. */ -if (relocs) -memset (relocs,0 , relocs_size); -/* find the flag main_extable_sort_needed */ for (sym = (void *)ehdr + _r(&symtab_sec->sh_offset); sym < sym + _r(&symtab_sec->sh_size) / sizeof(Elf_Sym);
Greg Kroah-Hartman
maintainerPATCH 60/204 scripts/sorttable: Have the ORC code use the _r() functions to read
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 66990c003306c240d570b3ba274ec4f68cf18c91 ]
The ORC code reads the section information directly from the file. This currently works because the default read function is for 64bit little endian machines. But if for some reason that ever changes, this will break. Instead of having a surprise breakage, use the _r() functions that will read the values from the file properly.
--- scripts/sorttable.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
scripts/sorttable.h +4 -4
@@ -299,14 +299,14 @@ static int do_sort(Elf_Ehdr *ehdr,#if defined(SORTTABLE_64 ) && defined(UNWINDER_ORC_ENABLED )/* locate the ORC unwind tables */ if (!strcmp (secstrings + idx,".orc_unwind_ip" )) { - orc_ip_size = s->sh_size; + orc_ip_size =_r (&s->sh_size); g_orc_ip_table = (int *)((void *)ehdr + - s->sh_offset); +_r (&s->sh_offset)); }if (!strcmp (secstrings + idx,".orc_unwind" )) { - orc_size = s->sh_size; + orc_size =_r (&s->sh_size); g_orc_table = (struct orc_entry *)((void *)ehdr + - s->sh_offset); +_r (&s->sh_offset)); }#endif }/* for loop */
Greg Kroah-Hartman
maintainerPATCH 61/204 scripts/sorttable: Make compare_extable() into two functions
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 7ffc0d0819f438779ed592e2e2e3576f43ce14f0 ]
Instead of having the compare_extable() part of the sorttable.h header where it get's defined twice, since it is a very simple function, just define it twice in sorttable.c, and then it can use the proper read functions for the word size and endianess and the Elf_Addr macro can be removed from sorttable.h.
Also add a micro optimization. Instead of:
if (a < b)
return -1;
if (a > b)
return 1;
return 0;That can be shorten to:
if (a < b)
return -1;
return a > b;
--- scripts/sorttable.c | 20 ++++++++++++++++++++ scripts/sorttable.h | 14 -------------- 2 files changed, 20 insertions(+), 14 deletions(-)
scripts/sorttable.c +20 -0
@@ -173,6 +173,26 @@ static inline unsigned int get_secindex(return r (&symtab_shndx_start[sym_offs]); } +static int compare_extable_32 (const void *a,const void *b) +{ +Elf32_Addr av =r (a); +Elf32_Addr bv =r (b); + +if (av < bv) +return -1 ; +return av > bv; +} + +static int compare_extable_64 (const void *a,const void *b) +{ +Elf64_Addr av =r8 (a); +Elf64_Addr bv =r8 (b); + +if (av < bv) +return -1 ; +return av > bv; +} +/* 32 bit and 64 bit are very similar */ #include "sorttable.h" #define SORTTABLE_64
scripts/sorttable.h +0 -14
@@ -23,7 +23,6 @@#undef sort_mcount_loc#undef elf_mcount_loc#undef do_sort -#undef Elf_Addr#undef Elf_Ehdr#undef Elf_Shdr#undef Elf_Sym
@@ -38,7 +37,6 @@# define sort_mcount_loc sort_mcount_loc_64# define elf_mcount_loc elf_mcount_loc_64# define do_sort do_sort_64 -# define Elf_Addr Elf64_Addr# define Elf_Ehdr Elf64_Ehdr# define Elf_Shdr Elf64_Shdr# define Elf_Sym Elf64_Sym
@@ -52,7 +50,6 @@# define sort_mcount_loc sort_mcount_loc_32# define elf_mcount_loc elf_mcount_loc_32# define do_sort do_sort_32 -# define Elf_Addr Elf32_Addr# define Elf_Ehdr Elf32_Ehdr# define Elf_Shdr Elf32_Shdr# define Elf_Sym Elf32_Sym
@@ -160,17 +157,6 @@ static void *sort_orctable(void *arg) }#endif -static int compare_extable (const void *a,const void *b) -{ -Elf_Addr av =_r (a); -Elf_Addr bv =_r (b); - -if (av < bv) -return -1 ; -if (av > bv) -return 1 ; -return 0 ; -}#ifdef MCOUNT_SORT_ENABLED pthread_t mcount_sort_thread;
Greg Kroah-Hartman
maintainerPATCH 62/204 scripts/sorttable: Convert Elf_Ehdr to union
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 157fb5b3cfd2cb5950314f926a76e567fc1921c5 ]
In order to remove the double #include of sorttable.h for 64 and 32 bit to create duplicate functions for both, replace the Elf_Ehdr macro with a union that defines both Elf64_Ehdr and Elf32_Ehdr, with field e64 for the 64bit version, and e32 for the 32bit version.
Then a macro etype can be used instead to get to the proper value.
This will eventually be replaced with just single functions that can handle both 32bit and 64bit ELF parsing.
--- scripts/sorttable.c | 36 ++++++++++++++++++++---------------- scripts/sorttable.h | 12 ++++++------ 2 files changed, 26 insertions(+), 22 deletions(-)
scripts/sorttable.c +20 -16
@@ -64,6 +64,11 @@#define EM_LOONGARCH 258#endif +typedef union { +Elf32_Ehdr e32; +Elf64_Ehdr e64; +}Elf_Ehdr ; +static uint32_t (*r)(const uint32_t *);static uint16_t (*r2)(const uint16_t *);static uint64_t (*r8)(const uint64_t *);
@@ -266,10 +271,10 @@ static void sort_relative_table_with_datstatic int do_file (char const *const fname,void *addr) {int rc =-1 ; -Elf32_Ehdr *ehdr = addr; +Elf_Ehdr *ehdr = addr;table_sort_t custom_sort =NULL ; -switch (ehdr->e_ident[EI_DATA ]) { +switch (ehdr->e32.e_ident[EI_DATA ]) {case ELFDATA2LSB : r = rle; r2 = r2le;
@@ -284,18 +289,18 @@ static int do_file(char const *const fnabreak ;default :fprintf (stderr,"unrecognized ELF data encoding %d: %s\n" , - ehdr->e_ident[EI_DATA ], fname); + ehdr->e32.e_ident[EI_DATA ], fname);return -1 ; } -if (memcmp (ELFMAG , ehdr->e_ident,SELFMAG ) !=0 || - (r2 (&ehdr->e_type) !=ET_EXEC &&r2 (&ehdr->e_type) !=ET_DYN ) || - ehdr->e_ident[EI_VERSION ] !=EV_CURRENT ) { +if (memcmp (ELFMAG , ehdr->e32.e_ident,SELFMAG ) !=0 || + (r2 (&ehdr->e32.e_type) !=ET_EXEC &&r2 (&ehdr->e32.e_type) !=ET_DYN ) || + ehdr->e32.e_ident[EI_VERSION ] !=EV_CURRENT ) {fprintf (stderr,"unrecognized ET_EXEC/ET_DYN file %s\n" , fname);return -1 ; } -switch (r2 (&ehdr->e_machine)) { +switch (r2 (&ehdr->e32.e_machine)) {case EM_386 :case EM_AARCH64 :case EM_LOONGARCH :
@@ -318,14 +323,14 @@ static int do_file(char const *const fnabreak ;default :fprintf (stderr,"unrecognized e_machine %d %s\n" , -r2 (&ehdr->e_machine), fname); +r2 (&ehdr->e32.e_machine), fname);return -1 ; } -switch (ehdr->e_ident[EI_CLASS ]) { +switch (ehdr->e32.e_ident[EI_CLASS ]) {case ELFCLASS32 : -if (r2 (&ehdr->e_ehsize) !=sizeof (Elf32_Ehdr) || -r2 (&ehdr->e_shentsize) !=sizeof (Elf32_Shdr)) { +if (r2 (&ehdr->e32.e_ehsize) !=sizeof (Elf32_Ehdr) || +r2 (&ehdr->e32.e_shentsize) !=sizeof (Elf32_Shdr)) {fprintf (stderr,"unrecognized ET_EXEC/ET_DYN file: %s\n" , fname);break ;
@@ -334,20 +339,19 @@ static int do_file(char const *const fnabreak ;case ELFCLASS64 : { -Elf64_Ehdr *const ghdr = (Elf64_Ehdr *)ehdr; -if (r2 (&ghdr->e_ehsize) !=sizeof (Elf64_Ehdr) || -r2 (&ghdr->e_shentsize) !=sizeof (Elf64_Shdr)) { +if (r2 (&ehdr->e64.e_ehsize) !=sizeof (Elf64_Ehdr) || +r2 (&ehdr->e64.e_shentsize) !=sizeof (Elf64_Shdr)) {fprintf (stderr,"unrecognized ET_EXEC/ET_DYN file: %s\n" , fname);break ; } - rc =do_sort_64 (ghdr, fname, custom_sort); + rc =do_sort_64 (ehdr, fname, custom_sort); }break ;default :fprintf (stderr,"unrecognized ELF class %d %s\n" , - ehdr->e_ident[EI_CLASS ], fname); + ehdr->e32.e_ident[EI_CLASS ], fname);break ; }
scripts/sorttable.h +6 -6
@@ -23,12 +23,12 @@#undef sort_mcount_loc#undef elf_mcount_loc#undef do_sort -#undef Elf_Ehdr#undef Elf_Shdr#undef Elf_Sym#undef ELF_ST_TYPE#undef uint_t#undef _r +#undef etype#ifdef SORTTABLE_64 # define extable_ent_size 16
@@ -37,12 +37,12 @@# define sort_mcount_loc sort_mcount_loc_64# define elf_mcount_loc elf_mcount_loc_64# define do_sort do_sort_64 -# define Elf_Ehdr Elf64_Ehdr# define Elf_Shdr Elf64_Shdr# define Elf_Sym Elf64_Sym# define ELF_ST_TYPE ELF64_ST_TYPE# define uint_t uint64_t# define _r r8 +# define etype e64#else # define extable_ent_size 8# define compare_extable compare_extable_32
@@ -50,12 +50,12 @@# define sort_mcount_loc sort_mcount_loc_32# define elf_mcount_loc elf_mcount_loc_32# define do_sort do_sort_32 -# define Elf_Ehdr Elf32_Ehdr# define Elf_Shdr Elf32_Shdr# define Elf_Sym Elf32_Sym# define ELF_ST_TYPE ELF32_ST_TYPE# define uint_t uint32_t# define _r r +# define etype e32#endif #if defined(SORTTABLE_64 ) && defined(UNWINDER_ORC_ENABLED )
@@ -222,7 +222,7 @@ static int do_sort(Elf_Ehdr *ehdr, table_sort_t custom_sort) {int rc =-1 ; -Elf_Shdr *s, *shdr = (Elf_Shdr *)((char *)ehdr +_r (&ehdr->e_shoff)); +Elf_Shdr *s, *shdr = (Elf_Shdr *)((char *)ehdr +_r (&ehdr->etype.e_shoff));Elf_Shdr *strtab_sec =NULL ;Elf_Shdr *symtab_sec =NULL ;Elf_Shdr *extab_sec =NULL ;
@@ -249,12 +249,12 @@ static int do_sort(Elf_Ehdr *ehdr,unsigned int orc_num_entries =0 ;#endif - shstrndx =r2 (&ehdr->e_shstrndx); + shstrndx =r2 (&ehdr->etype.e_shstrndx);if (shstrndx ==SHN_XINDEX ) shstrndx =r (&shdr[0 ].sh_link); secstrings = (const char *)ehdr +_r (&shdr[shstrndx].sh_offset); - shnum =r2 (&ehdr->e_shnum); + shnum =r2 (&ehdr->etype.e_shnum);if (shnum ==SHN_UNDEF ) shnum =_r (&shdr[0 ].sh_size);
Greg Kroah-Hartman
maintainerPATCH 63/204 scripts/sorttable: Replace Elf_Shdr Macro with a union
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 545f6cf8f4c9a268e0bab2637f1d279679befdbf ]
In order to remove the double #include of sorttable.h for 64 and 32 bit to create duplicate functions for both, replace the Elf_Shdr macro with a union that defines both Elf64_Shdr and Elf32_Shdr, with field e64 for the 64bit version, and e32 for the 32bit version.
It can then use the macro etype to get the proper value.
This will eventually be replaced with just single functions that can handle both 32bit and 64bit ELF parsing.
--- scripts/sorttable.c | 10 +++++++ scripts/sorttable.h | 74 ++++++++++++++++++++++++++++------------------------ 2 files changed, 51 insertions(+), 33 deletions(-)
scripts/sorttable.c +10 -0
@@ -69,6 +69,11 @@ typedef union {Elf64_Ehdr e64; } Elf_Ehdr; +typedef union { +Elf32_Shdr e32; +Elf64_Shdr e64; +}Elf_Shdr ; +static uint32_t (*r)(const uint32_t *);static uint16_t (*r2)(const uint16_t *);static uint64_t (*r8)(const uint64_t *);
@@ -198,6 +203,11 @@ static int compare_extable_64(const voidreturn av > bv; } +static inline void *get_index (void *start,int entsize,int index) +{ +return start + (entsize * index); +} +/* 32 bit and 64 bit are very similar */ #include "sorttable.h" #define SORTTABLE_64
scripts/sorttable.h +41 -33
170 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 64/204 scripts/sorttable: Convert Elf_Sym MACRO over to a union
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 200d015e73b4da69bcd8212a7c58695452b12bad ]
In order to remove the double #include of sorttable.h for 64 and 32 bit to create duplicate functions for both, replace the Elf_Sym macro with a union that defines both Elf64_Sym and Elf32_Sym, with field e64 for the 64bit version, and e32 for the 32bit version.
It can then use the macro etype to get the proper value.
This will eventually be replaced with just single functions that can handle both 32bit and 64bit ELF parsing.
--- scripts/sorttable.c | 5 +++++ scripts/sorttable.h | 25 ++++++++++++++----------- 2 files changed, 19 insertions(+), 11 deletions(-)
scripts/sorttable.c +5 -0
@@ -74,6 +74,11 @@ typedef union {Elf64_Shdr e64; } Elf_Shdr; +typedef union { +Elf32_Sym e32; +Elf64_Sym e64; +}Elf_Sym ; +static uint32_t (*r)(const uint32_t *);static uint16_t (*r2)(const uint16_t *);static uint64_t (*r8)(const uint64_t *);
scripts/sorttable.h +14 -11
@@ -23,7 +23,6 @@#undef sort_mcount_loc#undef elf_mcount_loc#undef do_sort -#undef Elf_Sym#undef ELF_ST_TYPE#undef uint_t#undef _r
@@ -36,7 +35,6 @@# define sort_mcount_loc sort_mcount_loc_64# define elf_mcount_loc elf_mcount_loc_64# define do_sort do_sort_64 -# define Elf_Sym Elf64_Sym# define ELF_ST_TYPE ELF64_ST_TYPE# define uint_t uint64_t# define _r r8
@@ -48,7 +46,6 @@# define sort_mcount_loc sort_mcount_loc_32# define elf_mcount_loc elf_mcount_loc_32# define do_sort do_sort_32 -# define Elf_Sym Elf32_Sym# define ELF_ST_TYPE ELF32_ST_TYPE# define uint_t uint32_t# define _r r
@@ -230,10 +227,13 @@ static int do_sort(Elf_Ehdr *ehdr,Elf_Sym *sort_needed_sym =NULL ;Elf_Shdr *sort_needed_sec;uint32_t *sort_needed_loc; +void *sym_start; +void *sym_end;const char *secstrings;const char *strtab;char *extab_image;int sort_need_index; +int symentsize;int shentsize;int idx;int i;
@@ -376,12 +376,15 @@ static int do_sort(Elf_Ehdr *ehdr, }/* find the flag main_extable_sort_needed */ -for (sym = (void *)ehdr +_r (&symtab_sec->etype.sh_offset); - sym < sym +_r (&symtab_sec->etype.sh_size) /sizeof (Elf_Sym); - sym++) { -if (ELF_ST_TYPE (sym->st_info) !=STT_OBJECT ) + sym_start = (void *)ehdr +_r (&symtab_sec->etype.sh_offset); + sym_end = sym_start +_r (&symtab_sec->etype.sh_size); + symentsize =_r (&symtab_sec->etype.sh_entsize); + +for (sym = sym_start; (void *)sym + symentsize < sym_end; + sym = (void *)sym + symentsize) { +if (ELF_ST_TYPE (sym->etype.st_info) !=STT_OBJECT )continue ; -if (!strcmp (strtab +r (&sym->st_name), +if (!strcmp (strtab +r (&sym->etype.st_name),"main_extable_sort_needed" )) { sort_needed_sym = sym;break ;
@@ -395,13 +398,13 @@ static int do_sort(Elf_Ehdr *ehdr, goto out; } - sort_need_index =get_secindex (r2 (&sym->st_shndx), - sort_needed_sym - symtab, + sort_need_index =get_secindex (r2 (&sym->etype.st_shndx), + ((void *)sort_needed_sym - (void *)symtab) / symentsize, symtab_shndx); sort_needed_sec =get_index (shdr_start, shentsize, sort_need_index); sort_needed_loc = (void *)ehdr +_r (&sort_needed_sec->etype.sh_offset) + -_r (&sort_needed_sym->st_value) - +_r (&sort_needed_sym->etype.st_value) -_r (&sort_needed_sec->etype.sh_addr);/* extable has been sorted, clear the flag */
Greg Kroah-Hartman
maintainerPATCH 65/204 scripts/sorttable: Add helper functions for Elf_Ehdr
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 1dfb59a228dde59ad7d99b2fa2104e90004995c7 ]
In order to remove the double #include of sorttable.h for 64 and 32 bit to create duplicate functions, add helper functions for Elf_Ehdr. This will create a function pointer for each helper that will get assigned to the appropriate function to handle either the 64bit or 32bit version.
This also moves the _r()/r() wrappers for the Elf_Ehdr references that handle endian and size differences between the different architectures, into the helper function and out of the open code which is more error prone.
--- scripts/sorttable.c | 25 +++++++++++++++++++++++++ scripts/sorttable.h | 20 ++++++++++++++++---- 2 files changed, 41 insertions(+), 4 deletions(-)
scripts/sorttable.c +25 -0
@@ -85,6 +85,31 @@ static uint64_t (*r8)(const uint64_t *);static void (*w)(uint32_t ,uint32_t *);typedef void (*table_sort_t )(char *,int ); +static uint64_t ehdr64_shoff (Elf_Ehdr *ehdr) +{ +return r8 (&ehdr->e64.e_shoff); +} + +static uint64_t ehdr32_shoff (Elf_Ehdr *ehdr) +{ +return r (&ehdr->e32.e_shoff); +} + +#define EHDR_HALF (fn_name) \ +static uint16_t ehdr64_##fn_name(Elf_Ehdr *ehdr) \ +{ \ + return r2(&ehdr->e64.e_##fn_name); \ +} \ + \ +static uint16_t ehdr32_##fn_name(Elf_Ehdr *ehdr) \ +{ \ + return r2(&ehdr->e32.e_##fn_name); \ +} + +EHDR_HALF(shentsize) +EHDR_HALF(shstrndx) +EHDR_HALF(shnum) + /* * Get the whole file as a programming convenience in order to avoid * malloc+lseek+read+free of many pieces. If successful, then mmap
scripts/sorttable.h +16 -4
@@ -27,6 +27,10 @@#undef uint_t#undef _r#undef etype +#undef ehdr_shoff +#undef ehdr_shentsize +#undef ehdr_shstrndx +#undef ehdr_shnum#ifdef SORTTABLE_64 # define extable_ent_size 16
@@ -39,6 +43,10 @@# define uint_t uint64_t# define _r r8# define etype e64 +# define ehdr_shoff ehdr64_shoff +# define ehdr_shentsize ehdr64_shentsize +# define ehdr_shstrndx ehdr64_shstrndx +# define ehdr_shnum ehdr64_shnum#else # define extable_ent_size 8# define compare_extable compare_extable_32
@@ -50,6 +58,10 @@# define uint_t uint32_t# define _r r# define etype e32 +# define ehdr_shoff ehdr32_shoff +# define ehdr_shentsize ehdr32_shentsize +# define ehdr_shstrndx ehdr32_shstrndx +# define ehdr_shnum ehdr32_shnum#endif #if defined(SORTTABLE_64 ) && defined(UNWINDER_ORC_ENABLED )
@@ -250,16 +262,16 @@ static int do_sort(Elf_Ehdr *ehdr,unsigned int orc_num_entries =0 ;#endif - shdr_start = (Elf_Shdr *)((char *)ehdr +_r (&ehdr->etype.e_shoff)); - shentsize =r2 (&ehdr->etype.e_shentsize); + shdr_start = (Elf_Shdr *)((char *)ehdr +ehdr_shoff (ehdr)); + shentsize =ehdr_shentsize (ehdr); - shstrndx =r2 (&ehdr->etype.e_shstrndx); + shstrndx =ehdr_shstrndx (ehdr);if (shstrndx ==SHN_XINDEX ) shstrndx =r (&shdr_start->etype.sh_link); string_sec =get_index (shdr_start, shentsize, shstrndx); secstrings = (const char *)ehdr +_r (&string_sec->etype.sh_offset); - shnum =r2 (&ehdr->etype.e_shnum); + shnum =ehdr_shnum (ehdr);if (shnum ==SHN_UNDEF ) shnum =_r (&shdr_start->etype.sh_size);
Greg Kroah-Hartman
maintainerPATCH 66/204 scripts/sorttable: Add helper functions for Elf_Shdr
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 67afb7f504400e5b4e5ff895459fbb3eb63d4450 ]
In order to remove the double #include of sorttable.h for 64 and 32 bit to create duplicate functions, add helper functions for Elf_Shdr. This will create a function pointer for each helper that will get assigned to the appropriate function to handle either the 64bit or 32bit version.
This also moves the _r()/r() wrappers for the Elf_Shdr references that handle endian and size differences between the different architectures, into the helper function and out of the open code which is more error prone.
--- scripts/sorttable.c | 42 +++++++++++++++++++++++++++++++++ scripts/sorttable.h | 66 +++++++++++++++++++++++++++++++++------------------- 2 files changed, 85 insertions(+), 23 deletions(-)
scripts/sorttable.c +42 -0
@@ -110,6 +110,48 @@ EHDR_HALF(shentsize) EHDR_HALF(shstrndx) EHDR_HALF(shnum) +#define SHDR_WORD (fn_name) \ +static uint32_t shdr64_##fn_name(Elf_Shdr *shdr) \ +{ \ + return r(&shdr->e64.sh_##fn_name); \ +} \ + \ +static uint32_t shdr32_##fn_name(Elf_Shdr *shdr) \ +{ \ + return r(&shdr->e32.sh_##fn_name); \ +} + +#define SHDR_ADDR (fn_name) \ +static uint64_t shdr64_##fn_name(Elf_Shdr *shdr) \ +{ \ + return r8(&shdr->e64.sh_##fn_name); \ +} \ + \ +static uint64_t shdr32_##fn_name(Elf_Shdr *shdr) \ +{ \ + return r(&shdr->e32.sh_##fn_name); \ +} + +#define SHDR_WORD (fn_name) \ +static uint32_t shdr64_##fn_name(Elf_Shdr *shdr) \ +{ \ + return r(&shdr->e64.sh_##fn_name); \ +} \ + \ +static uint32_t shdr32_##fn_name(Elf_Shdr *shdr) \ +{ \ + return r(&shdr->e32.sh_##fn_name); \ +} + +SHDR_ADDR(addr) +SHDR_ADDR(offset) +SHDR_ADDR(size) +SHDR_ADDR(entsize) + +SHDR_WORD(link) +SHDR_WORD(name) +SHDR_WORD(type) +/* * Get the whole file as a programming convenience in order to avoid * malloc+lseek+read+free of many pieces. If successful, then mmap
scripts/sorttable.h +43 -23
142 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 67/204 scripts/sorttable: Add helper functions for Elf_Sym
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 17bed33ac12f011f4695059960e1b1d6457229a7 ]
In order to remove the double #include of sorttable.h for 64 and 32 bit to create duplicate functions, add helper functions for Elf_Sym. This will create a function pointer for each helper that will get assigned to the appropriate function to handle either the 64bit or 32bit version.
This also removes the last references of etype and _r() macros from the sorttable.h file as their references are now just defined in the appropriate architecture version of the helper functions. All read functions now exist in the helper functions which makes it easier to maintain, as the helper functions define the necessary architecture sizes.
--- scripts/sorttable.c | 47 +++++++++++++++++++++++++++++++++++++++++++++++ scripts/sorttable.h | 30 ++++++++++++++++-------------- 2 files changed, 63 insertions(+), 14 deletions(-)
scripts/sorttable.c +47 -0
@@ -152,6 +152,53 @@ SHDR_WORD(link) SHDR_WORD(name) SHDR_WORD(type) +#define SYM_ADDR (fn_name) \ +static uint64_t sym64_##fn_name(Elf_Sym *sym) \ +{ \ + return r8(&sym->e64.st_##fn_name); \ +} \ + \ +static uint64_t sym32_##fn_name(Elf_Sym *sym) \ +{ \ + return r(&sym->e32.st_##fn_name); \ +} + +#define SYM_WORD (fn_name) \ +static uint32_t sym64_##fn_name(Elf_Sym *sym) \ +{ \ + return r(&sym->e64.st_##fn_name); \ +} \ + \ +static uint32_t sym32_##fn_name(Elf_Sym *sym) \ +{ \ + return r(&sym->e32.st_##fn_name); \ +} + +#define SYM_HALF (fn_name) \ +static uint16_t sym64_##fn_name(Elf_Sym *sym) \ +{ \ + return r2(&sym->e64.st_##fn_name); \ +} \ + \ +static uint16_t sym32_##fn_name(Elf_Sym *sym) \ +{ \ + return r2(&sym->e32.st_##fn_name); \ +} + +static uint8_t sym64_type (Elf_Sym *sym) +{ +return ELF64_ST_TYPE (sym->e64.st_info); +} + +static uint8_t sym32_type (Elf_Sym *sym) +{ +return ELF32_ST_TYPE (sym->e32.st_info); +} + +SYM_ADDR(value) +SYM_WORD(name) +SYM_HALF(shndx) + /* * Get the whole file as a programming convenience in order to avoid * malloc+lseek+read+free of many pieces. If successful, then mmap
scripts/sorttable.h +16 -14
@@ -23,10 +23,7 @@#undef sort_mcount_loc#undef elf_mcount_loc#undef do_sort -#undef ELF_ST_TYPE#undef uint_t -#undef _r -#undef etype#undef ehdr_shoff#undef ehdr_shentsize#undef ehdr_shstrndx
@@ -38,6 +35,10 @@#undef shdr_name#undef shdr_type#undef shdr_entsize +#undef sym_type +#undef sym_name +#undef sym_value +#undef sym_shndx#ifdef SORTTABLE_64 # define extable_ent_size 16
@@ -46,10 +47,7 @@# define sort_mcount_loc sort_mcount_loc_64# define elf_mcount_loc elf_mcount_loc_64# define do_sort do_sort_64 -# define ELF_ST_TYPE ELF64_ST_TYPE# define uint_t uint64_t -# define _r r8 -# define etype e64# define ehdr_shoff ehdr64_shoff# define ehdr_shentsize ehdr64_shentsize# define ehdr_shstrndx ehdr64_shstrndx
@@ -61,6 +59,10 @@# define shdr_name shdr64_name# define shdr_type shdr64_type# define shdr_entsize shdr64_entsize +# define sym_type sym64_type +# define sym_name sym64_name +# define sym_value sym64_value +# define sym_shndx sym64_shndx#else # define extable_ent_size 8# define compare_extable compare_extable_32
@@ -68,10 +70,7 @@# define sort_mcount_loc sort_mcount_loc_32# define elf_mcount_loc elf_mcount_loc_32# define do_sort do_sort_32 -# define ELF_ST_TYPE ELF32_ST_TYPE# define uint_t uint32_t -# define _r r -# define etype e32# define ehdr_shoff ehdr32_shoff# define ehdr_shentsize ehdr32_shentsize# define ehdr_shstrndx ehdr32_shstrndx
@@ -83,6 +82,10 @@# define shdr_name shdr32_name# define shdr_type shdr32_type# define shdr_entsize shdr32_entsize +# define sym_type sym32_type +# define sym_name sym32_name +# define sym_value sym32_value +# define sym_shndx sym32_shndx#endif #if defined(SORTTABLE_64 ) && defined(UNWINDER_ORC_ENABLED )
@@ -414,9 +417,9 @@ static int do_sort(Elf_Ehdr *ehdr,for (sym = sym_start; (void *)sym + symentsize < sym_end; sym = (void *)sym + symentsize) { -if (ELF_ST_TYPE (sym->etype.st_info) !=STT_OBJECT ) +if (sym_type (sym) !=STT_OBJECT )continue ; -if (!strcmp (strtab +r (&sym->etype.st_name), +if (!strcmp (strtab +sym_name (sym),"main_extable_sort_needed" )) { sort_needed_sym = sym;break ;
@@ -430,14 +433,13 @@ static int do_sort(Elf_Ehdr *ehdr, goto out; } - sort_need_index =get_secindex (r2 (&sym->etype.st_shndx), + sort_need_index =get_secindex (sym_shndx (sym), ((void *)sort_needed_sym - (void *)symtab) / symentsize, symtab_shndx); sort_needed_sec =get_index (shdr_start, shentsize, sort_need_index); sort_needed_loc = (void *)ehdr +shdr_offset (sort_needed_sec) + -_r (&sort_needed_sym->etype.st_value) - -shdr_addr (sort_needed_sec); +sym_value (sort_needed_sym) -shdr_addr (sort_needed_sec);/* extable has been sorted, clear the flag */ w (0 , sort_needed_loc);
Greg Kroah-Hartman
maintainerPATCH 68/204 scripts/sorttable: Use uint64_t for mcount sorting
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 1b649e6ab8dc9188d82c64069493afe66ca0edad ]
The mcount sorting defines uint_t to uint64_t on 64bit architectures and uint32_t on 32bit architectures. It can work with just using uint64_t as that will hold the values of both, and they are not used to point into the ELF file.
sizeof(uint_t) is used for defining the size of the mcount_loc section. Instead of using a type, define long_size and use that instead. This will allow the header code to be moved into the C file as generic functions and not need to include sorttable.h twice, once for 64bit and once for 32bit.
--- scripts/sorttable.h | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-)
scripts/sorttable.h +11 -11
@@ -23,7 +23,6 @@#undef sort_mcount_loc#undef elf_mcount_loc#undef do_sort -#undef uint_t#undef ehdr_shoff#undef ehdr_shentsize#undef ehdr_shstrndx
@@ -39,6 +38,7 @@#undef sym_name#undef sym_value#undef sym_shndx +#undef long_size#ifdef SORTTABLE_64 # define extable_ent_size 16
@@ -47,7 +47,6 @@# define sort_mcount_loc sort_mcount_loc_64# define elf_mcount_loc elf_mcount_loc_64# define do_sort do_sort_64 -# define uint_t uint64_t# define ehdr_shoff ehdr64_shoff# define ehdr_shentsize ehdr64_shentsize# define ehdr_shstrndx ehdr64_shstrndx
@@ -63,6 +62,7 @@# define sym_name sym64_name# define sym_value sym64_value# define sym_shndx sym64_shndx +# define long_size 8#else # define extable_ent_size 8# define compare_extable compare_extable_32
@@ -70,7 +70,6 @@# define sort_mcount_loc sort_mcount_loc_32# define elf_mcount_loc elf_mcount_loc_32# define do_sort do_sort_32 -# define uint_t uint32_t# define ehdr_shoff ehdr32_shoff# define ehdr_shentsize ehdr32_shentsize# define ehdr_shstrndx ehdr32_shstrndx
@@ -86,6 +85,7 @@# define sym_name sym32_name# define sym_value sym32_value# define sym_shndx sym32_shndx +# define long_size 4#endif #if defined(SORTTABLE_64 ) && defined(UNWINDER_ORC_ENABLED )
@@ -193,25 +193,25 @@ pthread_t mcount_sort_thread;struct elf_mcount_loc {Elf_Ehdr *ehdr;Elf_Shdr *init_data_sec; -uint_t start_mcount_loc; -uint_t stop_mcount_loc; +uint64_t start_mcount_loc; +uint64_t stop_mcount_loc; };/* Sort the addresses stored between __start_mcount_loc to __stop_mcount_loc in vmlinux */ static void *sort_mcount_loc (void *arg) {struct elf_mcount_loc *emloc = (struct elf_mcount_loc *)arg; -uint_t offset = emloc->start_mcount_loc -shdr_addr (emloc->init_data_sec) +uint64_t offset = emloc->start_mcount_loc -shdr_addr (emloc->init_data_sec) +shdr_offset (emloc->init_data_sec); -uint_t count = emloc->stop_mcount_loc - emloc->start_mcount_loc; +uint64_t count = emloc->stop_mcount_loc - emloc->start_mcount_loc;unsigned char *start_loc = (void *)emloc->ehdr + offset; -qsort (start_loc, count/sizeof (uint_t),sizeof (uint_t), compare_extable); +qsort (start_loc, count/long_size, long_size, compare_extable);return NULL ; }/* Get the address of __start_mcount_loc and __stop_mcount_loc in System.map */ -static void get_mcount_loc (uint_t *_start,uint_t *_stop) +static void get_mcount_loc (uint64_t *_start,uint64_t *_stop) {FILE *file_start, *file_stop;char start_buff[20 ];
@@ -277,8 +277,8 @@ static int do_sort(Elf_Ehdr *ehdr,unsigned int shstrndx;#ifdef MCOUNT_SORT_ENABLED struct elf_mcount_loc mstruct = {0 }; -uint_t _start_mcount_loc =0 ; -uint_t _stop_mcount_loc =0 ; +uint64_t _start_mcount_loc =0 ; +uint64_t _stop_mcount_loc =0 ;#endif #if defined(SORTTABLE_64 ) && defined(UNWINDER_ORC_ENABLED )unsigned int orc_ip_size =0 ;
Greg Kroah-Hartman
maintainerPATCH 69/204 scripts/sorttable: Move code from sorttable.h into sorttable.c
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 58d87678a0f46c6120904b4326aaf5ebf4454c69 ]
Instead of having the main code live in a header file and included twice with MACROs that define the Elf structures for 64 bit or 32 bit, move the code in the C file now that the Elf structures are defined in a union that has both. All accesses to the Elf structure fields are done through helper function pointers. If the file being parsed if for a 64 bit architecture, all the helper functions point to the 64 bit versions to retrieve the Elf fields. The same is true if the architecture is 32 bit, where the function pointers will point to the 32 bit helper functions.
Note, when the value of a field can be either 32 bit or 64 bit, a 64 bit is always returned, as it works for the 32 bit code as well.
This makes the code easier to read and maintain, and it now all exists in sorttable.c and sorttable.h may be removed.
--- scripts/sorttable.c | 473 +++++++++++++++++++++++++++++++++++++++++++++++++- scripts/sorttable.h | 485 ---------------------------------------------------- 2 files changed, 460 insertions(+), 498 deletions(-) delete mode 100644 scripts/sorttable.h
scripts/sorttable.c +460 -13
@@ -327,10 +327,423 @@ static inline void *get_index(void *starreturn start + (entsize * index); } -/* 32 bit and 64 bit are very similar */ -#include "sorttable.h" -#define SORTTABLE_64 -#include "sorttable.h" + +static int (*compare_extable)(const void *a,const void *b); +static uint64_t (*ehdr_shoff)(Elf_Ehdr *ehdr); +static uint16_t (*ehdr_shstrndx)(Elf_Ehdr *ehdr); +static uint16_t (*ehdr_shentsize)(Elf_Ehdr *ehdr); +static uint16_t (*ehdr_shnum)(Elf_Ehdr *ehdr); +static uint64_t (*shdr_addr)(Elf_Shdr *shdr); +static uint64_t (*shdr_offset)(Elf_Shdr *shdr); +static uint64_t (*shdr_size)(Elf_Shdr *shdr); +static uint64_t (*shdr_entsize)(Elf_Shdr *shdr); +static uint32_t (*shdr_link)(Elf_Shdr *shdr); +static uint32_t (*shdr_name)(Elf_Shdr *shdr); +static uint32_t (*shdr_type)(Elf_Shdr *shdr); +static uint8_t (*sym_type)(Elf_Sym *sym); +static uint32_t (*sym_name)(Elf_Sym *sym); +static uint64_t (*sym_value)(Elf_Sym *sym); +static uint16_t (*sym_shndx)(Elf_Sym *sym); + +static int extable_ent_size; +static int long_size; + + +#ifdef UNWINDER_ORC_ENABLED +/* ORC unwinder only support X86_64 */ +#include <asm/orc_types.h> + +#define ERRSTR_MAXSZ 256 + +static char g_err[ERRSTR_MAXSZ ]; +static int *g_orc_ip_table; +static struct orc_entry *g_orc_table; + +static pthread_t orc_sort_thread; + +static inline unsigned long orc_ip (const int *ip) +{ +return (unsigned long )ip + *ip; +} + +static int orc_sort_cmp (const void *_a,const void *_b) +{ +struct orc_entry *orc_a, *orc_b; +const int *a = g_orc_ip_table + *(int *)_a; +const int *b = g_orc_ip_table + *(int *)_b; +unsigned long a_val =orc_ip (a); +unsigned long b_val =orc_ip (b); + +if (a_val > b_val) +return 1 ; +if (a_val < b_val) +return -1 ; + +/* +* The "weak" section terminator entries need to always be on the left +* to ensure the lookup code skips them in favor of real entries. +* These terminator entries exist to handle any gaps created by +* whitelisted .o files which didn't get objtool generation. +*/ + orc_a = g_orc_table + (a - g_orc_ip_table); + orc_b = g_orc_table + (b - g_orc_ip_table); +if (orc_a->type ==ORC_TYPE_UNDEFINED && orc_b->type ==ORC_TYPE_UNDEFINED ) +return 0 ; +return orc_a->type ==ORC_TYPE_UNDEFINED ?-1 :1 ; +} + +static void *sort_orctable (void *arg) +{ +int i; +int *idxs =NULL ; +int *tmp_orc_ip_table =NULL ; +struct orc_entry *tmp_orc_table =NULL ; +unsigned int *orc_ip_size = (unsigned int *)arg; +unsigned int num_entries = *orc_ip_size /sizeof (int ); +unsigned int orc_size = num_entries *sizeof (struct orc_entry ); + + idxs = (int *)malloc (*orc_ip_size); +if (!idxs) { +snprintf (g_err,ERRSTR_MAXSZ ,"malloc idxs: %s" , +strerror (errno)); +pthread_exit (g_err); + } + + tmp_orc_ip_table = (int *)malloc (*orc_ip_size); +if (!tmp_orc_ip_table) { +snprintf (g_err,ERRSTR_MAXSZ ,"malloc tmp_orc_ip_table: %s" , +strerror (errno)); +pthread_exit (g_err); + } + + tmp_orc_table = (struct orc_entry *)malloc (orc_size); +if (!tmp_orc_table) { +snprintf (g_err,ERRSTR_MAXSZ ,"malloc tmp_orc_table: %s" , +strerror (errno)); +pthread_exit (g_err); + } + +/* initialize indices array, convert ip_table to absolute address */ +for (i =0 ; i < num_entries; i++) { + idxs[i] = i; + tmp_orc_ip_table[i] = g_orc_ip_table[i] + i *sizeof (int ); + } +memcpy (tmp_orc_table, g_orc_table, orc_size); + +qsort (idxs, num_entries,sizeof (int ), orc_sort_cmp); + +for (i =0 ; i < num_entries; i++) { +if (idxs[i] == i) +continue ; + +/* convert back to relative address */ + g_orc_ip_table[i] = tmp_orc_ip_table[idxs[i]] - i *sizeof (int ); + g_orc_table[i] = tmp_orc_table[idxs[i]]; + } + +free (idxs); +free (tmp_orc_ip_table); +free (tmp_orc_table); +pthread_exit (NULL ); +} +#endif + +#ifdef MCOUNT_SORT_ENABLED +static pthread_t mcount_sort_thread; + +struct elf_mcount_loc { +Elf_Ehdr *ehdr; +Elf_Shdr *init_data_sec; +uint64_t start_mcount_loc; +uint64_t stop_mcount_loc; +}; + +/* Sort the addresses stored between __start_mcount_loc to __stop_mcount_loc in vmlinux */ +static void *sort_mcount_loc (void *arg) +{ +struct elf_mcount_loc *emloc = (struct elf_mcount_loc *)arg; +uint64_t offset = emloc->start_mcount_loc -shdr_addr (emloc->init_data_sec) + +shdr_offset (emloc->init_data_sec); +uint64_t count = emloc->stop_mcount_loc - emloc->start_mcount_loc; +unsigned char *start_loc = (void *)emloc->ehdr + offset; + +qsort (start_loc, count/long_size, long_size, compare_extable); +return NULL ; +} + +/* Get the address of __start_mcount_loc and __stop_mcount_loc in System.map */ +static void get_mcount_loc (uint64_t *_start,uint64_t *_stop) +{ +FILE *file_start, *file_stop; +char start_buff[20 ]; +char stop_buff[20 ]; +int len =0 ; + + file_start =popen (" grep start_mcount System.map | awk '{print $1}' " ,"r" ); +if (!file_start) { +fprintf (stderr,"get start_mcount_loc error!" ); +return ; + } + + file_stop =popen (" grep stop_mcount System.map | awk '{print $1}' " ,"r" ); +if (!file_stop) { +fprintf (stderr,"get stop_mcount_loc error!" ); +pclose (file_start); +return ; + } + +while (fgets (start_buff,sizeof (start_buff), file_start) !=NULL ) { + len =strlen (start_buff); + start_buff[len -1 ] ='\0' ; + } + *_start =strtoul (start_buff,NULL ,16 ); + +while (fgets (stop_buff,sizeof (stop_buff), file_stop) !=NULL ) { + len =strlen (stop_buff); + stop_buff[len -1 ] ='\0' ; + } + *_stop =strtoul (stop_buff,NULL ,16 ); + +pclose (file_start); +pclose (file_stop); +} +#endif +static int do_sort (Elf_Ehdr *ehdr, +char const *const fname, +table_sort_t custom_sort) +{ +int rc =-1 ; +Elf_Shdr *shdr_start; +Elf_Shdr *strtab_sec =NULL ; +Elf_Shdr *symtab_sec =NULL ; +Elf_Shdr *extab_sec =NULL ; +Elf_Shdr *string_sec; +Elf_Sym *sym; +const Elf_Sym *symtab; +Elf32_Word *symtab_shndx =NULL ; +Elf_Sym *sort_needed_sym =NULL ; +Elf_Shdr *sort_needed_sec; +uint32_t *sort_needed_loc; +void *sym_start; +void *sym_end; +const char *secstrings; +const char *strtab; +char *extab_image; +int sort_need_index; +int symentsize; +int shentsize; +int idx; +int i; +unsigned int shnum; +unsigned int shstrndx; +#ifdef MCOUNT_SORT_ENABLED +struct elf_mcount_loc mstruct = {0 }; +uint64_t _start_mcount_loc =0 ; +uint64_t _stop_mcount_loc =0 ; +#endif +#ifdef UNWINDER_ORC_ENABLED +unsigned int orc_ip_size =0 ; +unsigned int orc_size =0 ; +unsigned int orc_num_entries =0 ; +#endif + + shdr_start = (Elf_Shdr *)((char *)ehdr +ehdr_shoff (ehdr)); + shentsize =ehdr_shentsize (ehdr); + + shstrndx =ehdr_shstrndx (ehdr); +if (shstrndx ==SHN_XINDEX ) + shstrndx =shdr_link (shdr_start); + string_sec =get_index (shdr_start, shentsize, shstrndx); + secstrings = (const char *)ehdr +shdr_offset (string_sec); + + shnum =ehdr_shnum (ehdr); +if (shnum ==SHN_UNDEF ) + shnum =shdr_size (shdr_start); + +for (i =0 ; i < shnum; i++) { +Elf_Shdr *shdr =get_index (shdr_start, shentsize, i); + + idx =shdr_name (shdr); +if (!strcmp (secstrings + idx,"__ex_table" )) + extab_sec = shdr; +if (!strcmp (secstrings + idx,".symtab" )) + symtab_sec = shdr; +if (!strcmp (secstrings + idx,".strtab" )) + strtab_sec = shdr; + +if (shdr_type (shdr) ==SHT_SYMTAB_SHNDX ) + symtab_shndx = (Elf32_Word *)((const char *)ehdr + +shdr_offset (shdr)); + +#ifdef MCOUNT_SORT_ENABLED +/* locate the .init.data section in vmlinux */ +if (!strcmp (secstrings + idx,".init.data" )) { +get_mcount_loc (&_start_mcount_loc, &_stop_mcount_loc); + mstruct.ehdr = ehdr; + mstruct.init_data_sec = shdr; + mstruct.start_mcount_loc = _start_mcount_loc; + mstruct.stop_mcount_loc = _stop_mcount_loc; + } +#endif + +#ifdef UNWINDER_ORC_ENABLED +/* locate the ORC unwind tables */ +if (!strcmp (secstrings + idx,".orc_unwind_ip" )) { + orc_ip_size =shdr_size (shdr); + g_orc_ip_table = (int *)((void *)ehdr + +shdr_offset (shdr)); + } +if (!strcmp (secstrings + idx,".orc_unwind" )) { + orc_size =shdr_size (shdr); + g_orc_table = (struct orc_entry *)((void *)ehdr + +shdr_offset (shdr)); + } +#endif + }/* for loop */ + +#ifdef UNWINDER_ORC_ENABLED +if (!g_orc_ip_table || !g_orc_table) { +fprintf (stderr, +"incomplete ORC unwind tables in file: %s\n" , fname); + goto out; + } + + orc_num_entries = orc_ip_size /sizeof (int ); +if (orc_ip_size %sizeof (int ) !=0 || + orc_size %sizeof (struct orc_entry ) !=0 || + orc_num_entries != orc_size /sizeof (struct orc_entry )) { +fprintf (stderr, +"inconsistent ORC unwind table entries in file: %s\n" , + fname); + goto out; + } + +/* create thread to sort ORC unwind tables concurrently */ +if (pthread_create (&orc_sort_thread,NULL , + sort_orctable, &orc_ip_size)) { +fprintf (stderr, +"pthread_create orc_sort_thread failed '%s': %s\n" , +strerror (errno), fname); + goto out; + } +#endif + +#ifdef MCOUNT_SORT_ENABLED +if (!mstruct.init_data_sec || !_start_mcount_loc || !_stop_mcount_loc) { +fprintf (stderr, +"incomplete mcount's sort in file: %s\n" , + fname); + goto out; + } + +/* create thread to sort mcount_loc concurrently */ +if (pthread_create (&mcount_sort_thread,NULL , &sort_mcount_loc, &mstruct)) { +fprintf (stderr, +"pthread_create mcount_sort_thread failed '%s': %s\n" , +strerror (errno), fname); + goto out; + } +#endif +if (!extab_sec) { +fprintf (stderr,"no __ex_table in file: %s\n" , fname); + goto out; + } + +if (!symtab_sec) { +fprintf (stderr,"no .symtab in file: %s\n" , fname); + goto out; + } + +if (!strtab_sec) { +fprintf (stderr,"no .strtab in file: %s\n" , fname); + goto out; + } + + extab_image = (void *)ehdr +shdr_offset (extab_sec); + strtab = (const char *)ehdr +shdr_offset (strtab_sec); + symtab = (const Elf_Sym *)((const char *)ehdr +shdr_offset (symtab_sec)); + +if (custom_sort) { +custom_sort (extab_image,shdr_size (extab_sec)); + }else { +int num_entries =shdr_size (extab_sec) / extable_ent_size; +qsort (extab_image, num_entries, + extable_ent_size, compare_extable); + } + +/* find the flag main_extable_sort_needed */ + sym_start = (void *)ehdr +shdr_offset (symtab_sec); + sym_end = sym_start +shdr_size (symtab_sec); + symentsize =shdr_entsize (symtab_sec); + +for (sym = sym_start; (void *)sym + symentsize < sym_end; + sym = (void *)sym + symentsize) { +if (sym_type (sym) !=STT_OBJECT ) +continue ; +if (!strcmp (strtab +sym_name (sym), +"main_extable_sort_needed" )) { + sort_needed_sym = sym; +break ; + } + } + +if (!sort_needed_sym) { +fprintf (stderr, +"no main_extable_sort_needed symbol in file: %s\n" , + fname); + goto out; + } + + sort_need_index =get_secindex (sym_shndx (sym), + ((void *)sort_needed_sym - (void *)symtab) / symentsize, + symtab_shndx); + sort_needed_sec =get_index (shdr_start, shentsize, sort_need_index); + sort_needed_loc = (void *)ehdr + +shdr_offset (sort_needed_sec) + +sym_value (sort_needed_sym) -shdr_addr (sort_needed_sec); + +/* extable has been sorted, clear the flag */ +w (0 , sort_needed_loc); + rc =0 ; + +out: +#ifdef UNWINDER_ORC_ENABLED +if (orc_sort_thread) { +void *retval =NULL ; +/* wait for ORC tables sort done */ + rc =pthread_join (orc_sort_thread, &retval); +if (rc) { +fprintf (stderr, +"pthread_join failed '%s': %s\n" , +strerror (errno), fname); + }else if (retval) { + rc =-1 ; +fprintf (stderr, +"failed to sort ORC tables '%s': %s\n" , + (char *)retval, fname); + } + } +#endif + +#ifdef MCOUNT_SORT_ENABLED +if (mcount_sort_thread) { +void *retval =NULL ; +/* wait for mcount sort done */ + rc =pthread_join (mcount_sort_thread, &retval); +if (rc) { +fprintf (stderr, +"pthread_join failed '%s': %s\n" , +strerror (errno), fname); + }else if (retval) { + rc =-1 ; +fprintf (stderr, +"failed to sort mcount '%s': %s\n" , + (char *)retval, fname); + } + } +#endif +return rc; +}static int compare_relative_table (const void *a,const void *b) {
@@ -399,7 +812,6 @@ static void sort_relative_table_with_datstatic int do_file (char const *const fname,void *addr) { -int rc =-1 ;Elf_Ehdr *ehdr = addr;table_sort_t custom_sort =NULL ;
@@ -462,29 +874,64 @@ static int do_file(char const *const fna r2(&ehdr->e32.e_shentsize) != sizeof(Elf32_Shdr)) {fprintf (stderr,"unrecognized ET_EXEC/ET_DYN file: %s\n" , fname); -break ; +return -1 ; } - rc =do_sort_32 (ehdr, fname, custom_sort); + + compare_extable = compare_extable_32; + ehdr_shoff = ehdr32_shoff; + ehdr_shentsize = ehdr32_shentsize; + ehdr_shstrndx = ehdr32_shstrndx; + ehdr_shnum = ehdr32_shnum; + shdr_addr = shdr32_addr; + shdr_offset = shdr32_offset; + shdr_link = shdr32_link; + shdr_size = shdr32_size; + shdr_name = shdr32_name; + shdr_type = shdr32_type; + shdr_entsize = shdr32_entsize; + sym_type = sym32_type; + sym_name = sym32_name; + sym_value = sym32_value; + sym_shndx = sym32_shndx; + long_size =4 ; + extable_ent_size =8 ;break ;case ELFCLASS64 : - {if (r2 (&ehdr->e64.e_ehsize) !=sizeof (Elf64_Ehdr) ||r2 (&ehdr->e64.e_shentsize) !=sizeof (Elf64_Shdr)) {fprintf (stderr,"unrecognized ET_EXEC/ET_DYN file: %s\n" , fname); -break ; - } - rc =do_sort_64 (ehdr, fname, custom_sort); +return -1 ; } + + compare_extable = compare_extable_64; + ehdr_shoff = ehdr64_shoff; + ehdr_shentsize = ehdr64_shentsize; + ehdr_shstrndx = ehdr64_shstrndx; + ehdr_shnum = ehdr64_shnum; + shdr_addr = shdr64_addr; + shdr_offset = shdr64_offset; + shdr_link = shdr64_link; + shdr_size = shdr64_size; + shdr_name = shdr64_name; + shdr_type = shdr64_type; + shdr_entsize = shdr64_entsize; + sym_type = sym64_type; + sym_name = sym64_name; + sym_value = sym64_value; + sym_shndx = sym64_shndx; + long_size =8 ; + extable_ent_size =16 ; +break ;default :fprintf (stderr,"unrecognized ELF class %d %s\n" , ehdr->e32.e_ident[EI_CLASS ], fname); -break ; +return -1 ; } -return rc; +return do_sort (ehdr, fname, custom_sort); }int main (int argc,char *argv[])
scripts/sorttable.h deleted +0 -485
485 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 70/204 scripts/sorttable: Get start/stop_mcount_loc from ELF file directly
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 4acda8edefa1ce66d3de845f1c12745721cd14c3 ]
The get_mcount_loc() does a cheesy trick to find the start_mcount_loc and stop_mcount_loc values. That trick is:
file_start = popen(" grep start_mcount System.map | awk '{print $1}' ", "r");
and
file_stop = popen(" grep stop_mcount System.map | awk '{print $1}' ", "r");
Those values are stored in the Elf symbol table. Use that to capture those values. Using the symbol table is more efficient and more robust. The above could fail if another variable had "start_mcount" or "stop_mcount" as part of its name.
--- scripts/sorttable.c | 95 ++++++++++++++++++++++++---------------------------- 1 file changed, 45 insertions(+), 50 deletions(-)
scripts/sorttable.c +45 -50
@@ -472,42 +472,41 @@ static void *sort_mcount_loc(void *arg) }/* Get the address of __start_mcount_loc and __stop_mcount_loc in System.map */ -static void get_mcount_loc (uint64_t *_start,uint64_t *_stop) +static void get_mcount_loc (struct elf_mcount_loc *emloc,Elf_Shdr *symtab_sec, +const char *strtab) { -FILE *file_start, *file_stop; -char start_buff[20 ]; -char stop_buff[20 ]; -int len =0 ; +Elf_Sym *sym, *end_sym; +int symentsize =shdr_entsize (symtab_sec); +int found =0 ; + + sym = (void *)emloc->ehdr +shdr_offset (symtab_sec); + end_sym = (void *)sym +shdr_size (symtab_sec); + +while (sym < end_sym) { +if (!strcmp (strtab +sym_name (sym),"__start_mcount_loc" )) { + emloc->start_mcount_loc =sym_value (sym); +if (++found ==2 ) +break ; + }else if (!strcmp (strtab +sym_name (sym),"__stop_mcount_loc" )) { + emloc->stop_mcount_loc =sym_value (sym); +if (++found ==2 ) +break ; + } + sym = (void *)sym + symentsize; + } - file_start =popen (" grep start_mcount System.map | awk '{print $1}' " ,"r" ); -if (!file_start) { +if (!emloc->start_mcount_loc) {fprintf (stderr,"get start_mcount_loc error!" );return ; } - file_stop =popen (" grep stop_mcount System.map | awk '{print $1}' " ,"r" ); -if (!file_stop) { +if (!emloc->stop_mcount_loc) {fprintf (stderr,"get stop_mcount_loc error!" ); -pclose (file_start);return ; } - -while (fgets (start_buff,sizeof (start_buff), file_start) !=NULL ) { - len =strlen (start_buff); - start_buff[len -1 ] ='\0' ; - } - *_start =strtoul (start_buff,NULL ,16 ); - -while (fgets (stop_buff,sizeof (stop_buff), file_stop) !=NULL ) { - len =strlen (stop_buff); - stop_buff[len -1 ] ='\0' ; - } - *_stop =strtoul (stop_buff,NULL ,16 ); - -pclose (file_start); -pclose (file_stop); }#endif +static int do_sort (Elf_Ehdr *ehdr,char const *const fname,table_sort_t custom_sort)
@@ -538,8 +537,6 @@ static int do_sort(Elf_Ehdr *ehdr,unsigned int shstrndx;#ifdef MCOUNT_SORT_ENABLED struct elf_mcount_loc mstruct = {0 }; -uint64_t _start_mcount_loc =0 ; -uint64_t _stop_mcount_loc =0 ;#endif #ifdef UNWINDER_ORC_ENABLED unsigned int orc_ip_size =0 ;
@@ -577,13 +574,8 @@ static int do_sort(Elf_Ehdr *ehdr,#ifdef MCOUNT_SORT_ENABLED /* locate the .init.data section in vmlinux */ -if (!strcmp (secstrings + idx,".init.data" )) { -get_mcount_loc (&_start_mcount_loc, &_stop_mcount_loc); - mstruct.ehdr = ehdr; +if (!strcmp (secstrings + idx,".init.data" )) mstruct.init_data_sec = shdr; - mstruct.start_mcount_loc = _start_mcount_loc; - mstruct.stop_mcount_loc = _stop_mcount_loc; - }#endif #ifdef UNWINDER_ORC_ENABLED
@@ -627,23 +619,6 @@ static int do_sort(Elf_Ehdr *ehdr, goto out; }#endif - -#ifdef MCOUNT_SORT_ENABLED -if (!mstruct.init_data_sec || !_start_mcount_loc || !_stop_mcount_loc) { -fprintf (stderr, -"incomplete mcount's sort in file: %s\n" , - fname); - goto out; - } - -/* create thread to sort mcount_loc concurrently */ -if (pthread_create (&mcount_sort_thread,NULL , &sort_mcount_loc, &mstruct)) { -fprintf (stderr, -"pthread_create mcount_sort_thread failed '%s': %s\n" , -strerror (errno), fname); - goto out; - } -#endif if (!extab_sec) {fprintf (stderr,"no __ex_table in file: %s\n" , fname); goto out;
@@ -663,6 +638,26 @@ static int do_sort(Elf_Ehdr *ehdr, strtab = (const char *)ehdr +shdr_offset (strtab_sec); symtab = (const Elf_Sym *)((const char *)ehdr +shdr_offset (symtab_sec)); +#ifdef MCOUNT_SORT_ENABLED + mstruct.ehdr = ehdr; +get_mcount_loc (&mstruct, symtab_sec, strtab); + +if (!mstruct.init_data_sec || !mstruct.start_mcount_loc || !mstruct.stop_mcount_loc) { +fprintf (stderr, +"incomplete mcount's sort in file: %s\n" , + fname); + goto out; + } + +/* create thread to sort mcount_loc concurrently */ +if (pthread_create (&mcount_sort_thread,NULL , &sort_mcount_loc, &mstruct)) { +fprintf (stderr, +"pthread_create mcount_sort_thread failed '%s': %s\n" , +strerror (errno), fname); + goto out; + } +#endif +if (custom_sort) {custom_sort (extab_image,shdr_size (extab_sec)); }else {
Greg Kroah-Hartman
maintainerPATCH 71/204 scripts/sorttable: Use a structure of function pointers for elf helpers
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 1e5f6771c247b28135307058d2cfe3b0153733dc ]
Instead of having a series of function pointers that gets assigned to the Elf64 or Elf32 versions, put them all into a single structure and use that. Add the helper function that chooses the structure into the macros that build the different versions of the elf functions.
--- scripts/sorttable.c | 175 +++++++++++++++++++++++++++++++++++----------------- 1 file changed, 118 insertions(+), 57 deletions(-)
scripts/sorttable.c +118 -57
@@ -85,6 +85,25 @@ static uint64_t (*r8)(const uint64_t *);static void (*w)(uint32_t ,uint32_t *);typedef void (*table_sort_t )(char *,int ); +static struct elf_funcs { +int (*compare_extable)(const void *a,const void *b); +uint64_t (*ehdr_shoff)(Elf_Ehdr *ehdr); +uint16_t (*ehdr_shstrndx)(Elf_Ehdr *ehdr); +uint16_t (*ehdr_shentsize)(Elf_Ehdr *ehdr); +uint16_t (*ehdr_shnum)(Elf_Ehdr *ehdr); +uint64_t (*shdr_addr)(Elf_Shdr *shdr); +uint64_t (*shdr_offset)(Elf_Shdr *shdr); +uint64_t (*shdr_size)(Elf_Shdr *shdr); +uint64_t (*shdr_entsize)(Elf_Shdr *shdr); +uint32_t (*shdr_link)(Elf_Shdr *shdr); +uint32_t (*shdr_name)(Elf_Shdr *shdr); +uint32_t (*shdr_type)(Elf_Shdr *shdr); +uint8_t (*sym_type)(Elf_Sym *sym); +uint32_t (*sym_name)(Elf_Sym *sym); +uint64_t (*sym_value)(Elf_Sym *sym); +uint16_t (*sym_shndx)(Elf_Sym *sym); +} e; +static uint64_t ehdr64_shoff (Elf_Ehdr *ehdr) {return r8 (&ehdr->e64.e_shoff);
@@ -95,6 +114,11 @@ static uint64_t ehdr32_shoff(Elf_Ehdr *ereturn r (&ehdr->e32.e_shoff); } +static uint64_t ehdr_shoff (Elf_Ehdr *ehdr) +{ +return e.ehdr_shoff (ehdr); +} +#define EHDR_HALF (fn_name) \ static uint16_t ehdr64_##fn_name(Elf_Ehdr *ehdr) \ { \
@@ -104,6 +128,11 @@ static uint16_t ehdr64_##fn_name(Elf_Ehd static uint16_t ehdr32_##fn_name(Elf_Ehdr *ehdr) \ { \ return r2(&ehdr->e32.e_##fn_name); \ +} \ + \ +static uint16_t ehdr_##fn_name(Elf_Ehdr *ehdr) \ +{ \ + return e.ehdr_##fn_name(ehdr); \ }EHDR_HALF (shentsize)
@@ -119,6 +148,11 @@ static uint32_t shdr64_##fn_name(Elf_Shd static uint32_t shdr32_##fn_name(Elf_Shdr *shdr) \ { \ return r(&shdr->e32.sh_##fn_name); \ +} \ + \ +static uint32_t shdr_##fn_name(Elf_Shdr *shdr) \ +{ \ + return e.shdr_##fn_name(shdr); \ }#define SHDR_ADDR (fn_name) \
@@ -130,6 +164,11 @@ static uint64_t shdr64_##fn_name(Elf_Shd static uint64_t shdr32_##fn_name(Elf_Shdr *shdr) \ { \ return r(&shdr->e32.sh_##fn_name); \ +} \ + \ +static uint64_t shdr_##fn_name(Elf_Shdr *shdr) \ +{ \ + return e.shdr_##fn_name(shdr); \ }#define SHDR_WORD (fn_name) \
@@ -141,6 +180,10 @@ static uint32_t shdr64_##fn_name(Elf_Shd static uint32_t shdr32_##fn_name(Elf_Shdr *shdr) \ { \ return r(&shdr->e32.sh_##fn_name); \ +} \ +static uint32_t shdr_##fn_name(Elf_Shdr *shdr) \ +{ \ + return e.shdr_##fn_name(shdr); \ }SHDR_ADDR (addr)
@@ -161,6 +204,11 @@ static uint64_t sym64_##fn_name(Elf_Sym static uint64_t sym32_##fn_name(Elf_Sym *sym) \ { \ return r(&sym->e32.st_##fn_name); \ +} \ + \ +static uint64_t sym_##fn_name(Elf_Sym *sym) \ +{ \ + return e.sym_##fn_name(sym); \ }#define SYM_WORD (fn_name) \
@@ -172,6 +220,11 @@ static uint32_t sym64_##fn_name(Elf_Sym static uint32_t sym32_##fn_name(Elf_Sym *sym) \ { \ return r(&sym->e32.st_##fn_name); \ +} \ + \ +static uint32_t sym_##fn_name(Elf_Sym *sym) \ +{ \ + return e.sym_##fn_name(sym); \ }#define SYM_HALF (fn_name) \
@@ -183,6 +236,11 @@ static uint16_t sym64_##fn_name(Elf_Sym static uint16_t sym32_##fn_name(Elf_Sym *sym) \ { \ return r2(&sym->e32.st_##fn_name); \ +} \ + \ +static uint16_t sym_##fn_name(Elf_Sym *sym) \ +{ \ + return e.sym_##fn_name(sym); \ }static uint8_t sym64_type (Elf_Sym *sym)
@@ -195,6 +253,11 @@ static uint8_t sym32_type(Elf_Sym *sym)return ELF32_ST_TYPE (sym->e32.st_info); } +static uint8_t sym_type (Elf_Sym *sym) +{ +return e.sym_type (sym); +} + SYM_ADDR(value) SYM_WORD(name) SYM_HALF(shndx)
@@ -322,29 +385,16 @@ static int compare_extable_64(const voidreturn av > bv; } +static int compare_extable (const void *a,const void *b) +{ +return e.compare_extable (a, b); +} +static inline void *get_index (void *start,int entsize,int index) {return start + (entsize * index); } - -static int (*compare_extable)(const void *a,const void *b); -static uint64_t (*ehdr_shoff)(Elf_Ehdr *ehdr); -static uint16_t (*ehdr_shstrndx)(Elf_Ehdr *ehdr); -static uint16_t (*ehdr_shentsize)(Elf_Ehdr *ehdr); -static uint16_t (*ehdr_shnum)(Elf_Ehdr *ehdr); -static uint64_t (*shdr_addr)(Elf_Shdr *shdr); -static uint64_t (*shdr_offset)(Elf_Shdr *shdr); -static uint64_t (*shdr_size)(Elf_Shdr *shdr); -static uint64_t (*shdr_entsize)(Elf_Shdr *shdr); -static uint32_t (*shdr_link)(Elf_Shdr *shdr); -static uint32_t (*shdr_name)(Elf_Shdr *shdr); -static uint32_t (*shdr_type)(Elf_Shdr *shdr); -static uint8_t (*sym_type)(Elf_Sym *sym); -static uint32_t (*sym_name)(Elf_Sym *sym); -static uint64_t (*sym_value)(Elf_Sym *sym); -static uint16_t (*sym_shndx)(Elf_Sym *sym); -static int extable_ent_size;static int long_size;
@@ -864,7 +914,30 @@ static int do_file(char const *const fna }switch (ehdr->e32.e_ident[EI_CLASS ]) { -case ELFCLASS32 : +case ELFCLASS32 : { +struct elf_funcs efuncs = { + .compare_extable = compare_extable_32, + .ehdr_shoff = ehdr32_shoff, + .ehdr_shentsize = ehdr32_shentsize, + .ehdr_shstrndx = ehdr32_shstrndx, + .ehdr_shnum = ehdr32_shnum, + .shdr_addr = shdr32_addr, + .shdr_offset = shdr32_offset, + .shdr_link = shdr32_link, + .shdr_size = shdr32_size, + .shdr_name = shdr32_name, + .shdr_type = shdr32_type, + .shdr_entsize = shdr32_entsize, + .sym_type = sym32_type, + .sym_name = sym32_name, + .sym_value = sym32_value, + .sym_shndx = sym32_shndx, + }; + + e = efuncs; + long_size =4 ; + extable_ent_size =8 ; +if (r2 (&ehdr->e32.e_ehsize) !=sizeof (Elf32_Ehdr) ||r2 (&ehdr->e32.e_shentsize) !=sizeof (Elf32_Shdr)) { fprintf(stderr,
@@ -872,26 +945,32 @@ static int do_file(char const *const fnareturn -1 ; } - compare_extable = compare_extable_32; - ehdr_shoff = ehdr32_shoff; - ehdr_shentsize = ehdr32_shentsize; - ehdr_shstrndx = ehdr32_shstrndx; - ehdr_shnum = ehdr32_shnum; - shdr_addr = shdr32_addr; - shdr_offset = shdr32_offset; - shdr_link = shdr32_link; - shdr_size = shdr32_size; - shdr_name = shdr32_name; - shdr_type = shdr32_type; - shdr_entsize = shdr32_entsize; - sym_type = sym32_type; - sym_name = sym32_name; - sym_value = sym32_value; - sym_shndx = sym32_shndx; - long_size =4 ; - extable_ent_size =8 ; + }break ; -case ELFCLASS64 : +case ELFCLASS64 : { +struct elf_funcs efuncs = { + .compare_extable = compare_extable_64, + .ehdr_shoff = ehdr64_shoff, + .ehdr_shentsize = ehdr64_shentsize, + .ehdr_shstrndx = ehdr64_shstrndx, + .ehdr_shnum = ehdr64_shnum, + .shdr_addr = shdr64_addr, + .shdr_offset = shdr64_offset, + .shdr_link = shdr64_link, + .shdr_size = shdr64_size, + .shdr_name = shdr64_name, + .shdr_type = shdr64_type, + .shdr_entsize = shdr64_entsize, + .sym_type = sym64_type, + .sym_name = sym64_name, + .sym_value = sym64_value, + .sym_shndx = sym64_shndx, + }; + + e = efuncs; + long_size =8 ; + extable_ent_size =16 ; +if (r2 (&ehdr->e64.e_ehsize) !=sizeof (Elf64_Ehdr) ||r2 (&ehdr->e64.e_shentsize) !=sizeof (Elf64_Shdr)) { fprintf(stderr,
@@ -900,25 +979,7 @@ static int do_file(char const *const fnareturn -1 ; } - compare_extable = compare_extable_64; - ehdr_shoff = ehdr64_shoff; - ehdr_shentsize = ehdr64_shentsize; - ehdr_shstrndx = ehdr64_shstrndx; - ehdr_shnum = ehdr64_shnum; - shdr_addr = shdr64_addr; - shdr_offset = shdr64_offset; - shdr_link = shdr64_link; - shdr_size = shdr64_size; - shdr_name = shdr64_name; - shdr_type = shdr64_type; - shdr_entsize = shdr64_entsize; - sym_type = sym64_type; - sym_name = sym64_name; - sym_value = sym64_value; - sym_shndx = sym64_shndx; - long_size =8 ; - extable_ent_size =16 ; - + }break ;default : fprintf(stderr, "unrecognized ELF class %d %s\n",
Greg Kroah-Hartman
maintainerPATCH 72/204 arm64: scripts/sorttable: Implement sorting mcount_loc at boot for arm64
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit b3d09d06e052e1d754645acea4e4d1e96f81c934 ]
The mcount_loc section holds the addresses of the functions that get patched by ftrace when enabling function callbacks. It can contain tens of thousands of entries. These addresses must be sorted. If they are not sorted at compile time, they are sorted at boot. Sorting at boot does take some time and does have a small impact on boot performance.
x86 and arm32 have the addresses in the mcount_loc section of the ELF file. But for arm64, the section just contains zeros. The .rela.dyn Elf_Rela section holds the addresses and they get patched at boot during the relocation phase.
In order to sort these addresses, the Elf_Rela needs to be updated instead of the location in the binary that holds the mcount_loc section. Have the sorttable code, allocate an array to hold the functions, load the addresses from the Elf_Rela entries, sort them, then put them back in order into the Elf_rela entries so that they will be sorted at boot up without having to sort them during boot up.
--- arch/arm64/Kconfig | 1 scripts/sorttable.c | 185 +++++++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 183 insertions(+), 3 deletions(-)
arch/arm64/Kconfig +1 -0
@@ -212,6 +212,7 @@ config ARM64 if DYNAMIC_FTRACE_WITH_ARGS select HAVE_SAMPLE_FTRACE_DIRECT select HAVE_SAMPLE_FTRACE_DIRECT_MULTI + select HAVE_BUILDTIME_MCOUNT_SORT select HAVE_EFFICIENT_UNALIGNED_ACCESS select HAVE_GUP_FAST select HAVE_FTRACE_MCOUNT_RECORD
scripts/sorttable.c +182 -3
280 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 73/204 scripts/sorttable: Have mcount rela sort use direct values
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit a0265659322540d656727b9e132edfb6f06b6c1a ]
The mcount_loc sorting for when the values are stored in the Elf_Rela entries uses the compare_extable() function to do the compares in the qsort(). That function does handle byte swapping if the machine being compiled for is a different endian than the host machine. But the sort_relocs() function sorts an array that pulled in the values from the Elf_Rela section and has already done the swapping.
Create two new compare functions that will sort the direct values. One will sort 32 bit values and the other will sort the 64 bit value. One of these will be assigned to a compare_values function pointer and that will be used for sorting the Elf_Rela mcount values.
--- scripts/sorttable.c | 26 +++++++++++++++++++++++++- 1 file changed, 25 insertions(+), 1 deletion(-)
scripts/sorttable.c +25 -1
@@ -552,6 +552,28 @@ static void *sort_orctable(void *arg)#ifdef MCOUNT_SORT_ENABLED +static int compare_values_64 (const void *a,const void *b) +{ +uint64_t av = *(uint64_t *)a; +uint64_t bv = *(uint64_t *)b; + +if (av < bv) +return -1 ; +return av > bv; +} + +static int compare_values_32 (const void *a,const void *b) +{ +uint32_t av = *(uint32_t *)a; +uint32_t bv = *(uint32_t *)b; + +if (av < bv) +return -1 ; +return av > bv; +} + +static int (*compare_values)(const void *a,const void *b); +/* Only used for sorting mcount table */ static void rela_write_addend (Elf_Rela *rela,uint64_t val) {
@@ -583,6 +605,8 @@ static void *sort_relocs(Elf_Ehdr *ehdr,void *vals;void *ptr; + compare_values = long_size ==4 ? compare_values_32 : compare_values_64; + shdr_start = (Elf_Shdr *)((char *)ehdr +ehdr_shoff (ehdr)); shentsize =ehdr_shentsize (ehdr);
@@ -640,7 +664,7 @@ static void *sort_relocs(Elf_Ehdr *ehdr, } } count = ptr - vals; -qsort (vals, count / long_size, long_size, compare_extable); +qsort (vals, count / long_size, long_size, compare_values); ptr = vals;for (int i =0 ; i < shnum; i++) {
Greg Kroah-Hartman
maintainerPATCH 74/204 scripts/sorttable: Always use an array for the mcount_loc sorting
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 5fb964f5ba53afda0e2b6dbc00b8205461ffe04a ]
The sorting of the mcount_loc section is done directly to the section for x86 and arm32 but it uses a separate array for arm64 as arm64 has the values for the mcount_loc stored in the rela sections of the vmlinux ELF file.
In order to use the same code to remove weak functions, always use a separate array to do the sorting. This requires splitting up the filling of the array into one function and the placing the contents of the array back into the rela sections or into the mcount_loc section into a separate file.
--- scripts/sorttable.c | 122 ++++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 90 insertions(+), 32 deletions(-)
scripts/sorttable.c +90 -32
@@ -594,31 +594,19 @@ struct elf_mcount_loc {uint64_t stop_mcount_loc; }; -/* Sort the relocations not the address itself */ -static void *sort_relocs (Elf_Ehdr *ehdr,uint64_t start_loc,uint64_t size) +/* Fill the array with the content of the relocs */ +static int fill_relocs (void *ptr,uint64_t size,Elf_Ehdr *ehdr,uint64_t start_loc) {Elf_Shdr *shdr_start;Elf_Rela *rel;unsigned int shnum; -unsigned int count; +unsigned int count =0 ;int shentsize; -void *vals; -void *ptr; - - compare_values = long_size ==4 ? compare_values_32 : compare_values_64; +void *array_end = ptr + size; shdr_start = (Elf_Shdr *)((char *)ehdr +ehdr_shoff (ehdr)); shentsize =ehdr_shentsize (ehdr); - vals =malloc (long_size * size); -if (!vals) { -snprintf (m_err,ERRSTR_MAXSZ ,"Failed to allocate sort array" ); -pthread_exit (m_err); -return NULL ; - } - - ptr = vals; - shnum =ehdr_shnum (ehdr);if (shnum ==SHN_UNDEF ) shnum =shdr_size (shdr_start);
@@ -637,22 +625,18 @@ static void *sort_relocs(Elf_Ehdr *ehdr,uint64_t offset =rela_offset (rel);if (offset >= start_loc && offset < start_loc + size) { -if (ptr + long_size > vals + size) { -free (vals); +if (ptr + long_size > array_end) {snprintf (m_err,ERRSTR_MAXSZ ,"Too many relocations" ); -pthread_exit (m_err); -return NULL ; +return -1 ; }/* Make sure this has the correct type */ if (rela_info (rel) != rela_type) { -free (vals);snprintf (m_err,ERRSTR_MAXSZ ,"rela has type %lx but expected %lx\n" , (long )rela_info (rel), rela_type); -pthread_exit (m_err); -return NULL ; +return -1 ; }if (long_size ==4 )
@@ -660,13 +644,28 @@ static void *sort_relocs(Elf_Ehdr *ehdr,else *(uint64_t *)ptr =rela_addend (rel); ptr += long_size; + count++; } } } - count = ptr - vals; -qsort (vals, count / long_size, long_size, compare_values); +return count; +} + +/* Put the sorted vals back into the relocation elements */ +static void replace_relocs (void *ptr,uint64_t size,Elf_Ehdr *ehdr,uint64_t start_loc) +{ +Elf_Shdr *shdr_start; +Elf_Rela *rel; +unsigned int shnum; +int shentsize; + + shdr_start = (Elf_Shdr *)((char *)ehdr +ehdr_shoff (ehdr)); + shentsize =ehdr_shentsize (ehdr); + + shnum =ehdr_shnum (ehdr); +if (shnum ==SHN_UNDEF ) + shnum =shdr_size (shdr_start); - ptr = vals;for (int i =0 ; i < shnum; i++) {Elf_Shdr *shdr =get_index (shdr_start, shentsize, i);void *end;
@@ -689,8 +688,32 @@ static void *sort_relocs(Elf_Ehdr *ehdr, } } } -free (vals); -return NULL ; +} + +static int fill_addrs (void *ptr,uint64_t size,void *addrs) +{ +void *end = ptr + size; +int count =0 ; + +for (; ptr < end; ptr += long_size, addrs += long_size, count++) { +if (long_size ==4 ) + *(uint32_t *)ptr =r (addrs); +else + *(uint64_t *)ptr =r8 (addrs); + } +return count; +} + +static void replace_addrs (void *ptr,uint64_t size,void *addrs) +{ +void *end = ptr + size; + +for (; ptr < end; ptr += long_size, addrs += long_size) { +if (long_size ==4 ) +w (*(uint32_t *)ptr, addrs); +else +w8 (*(uint64_t *)ptr, addrs); + } }/* Sort the addresses stored between __start_mcount_loc to __stop_mcount_loc in vmlinux */
@@ -699,14 +722,49 @@ static void *sort_mcount_loc(void *arg)struct elf_mcount_loc *emloc = (struct elf_mcount_loc *)arg;uint64_t offset = emloc->start_mcount_loc -shdr_addr (emloc->init_data_sec) +shdr_offset (emloc->init_data_sec); -uint64_t count = emloc->stop_mcount_loc - emloc->start_mcount_loc; +uint64_t size = emloc->stop_mcount_loc - emloc->start_mcount_loc;unsigned char *start_loc = (void *)emloc->ehdr + offset; +Elf_Ehdr *ehdr = emloc->ehdr; +void *e_msg =NULL ; +void *vals; +int count; + + vals =malloc (long_size * size); +if (!vals) { +snprintf (m_err,ERRSTR_MAXSZ ,"Failed to allocate sort array" ); +pthread_exit (m_err); + }if (sort_reloc) -return sort_relocs (emloc->ehdr, emloc->start_mcount_loc, count); + count =fill_relocs (vals, size, ehdr, emloc->start_mcount_loc); +else + count =fill_addrs (vals, size, start_loc); + +if (count <0 ) { + e_msg = m_err; + goto out; + } + +if (count != size / long_size) { +snprintf (m_err,ERRSTR_MAXSZ ,"Expected %u mcount elements but found %u\n" , + (int )(size / long_size), count); + e_msg = m_err; + goto out; + } + + compare_values = long_size ==4 ? compare_values_32 : compare_values_64; + +qsort (vals, count, long_size, compare_values); + +if (sort_reloc) +replace_relocs (vals, size, ehdr, emloc->start_mcount_loc); +else +replace_addrs (vals, size, start_loc); + +out: +free (vals); -qsort (start_loc, count/long_size, long_size, compare_extable); -return NULL ; +pthread_exit (e_msg); }/* Get the address of __start_mcount_loc and __stop_mcount_loc in System.map */
Greg Kroah-Hartman
maintainerPATCH 75/204 scripts/sorttable: Zero out weak functions in mcount_loc table
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit ef378c3b8233855497a414b9d67bf22592c928a4 ]
When a function is annotated as "weak" and is overridden, the code is not removed. If it is traced, the fentry/mcount location in the weak function will be referenced by the "__mcount_loc" section. This will then be added to the available_filter_functions list. Since only the address of the functions are listed, to find the name to show, a search of kallsyms is used.
Since kallsyms will return the function by simply finding the function that the address is after but before the next function, an address of a weak function will show up as the function before it. This is because kallsyms does not save names of weak functions. This has caused issues in the past, as now the traced weak function will be listed in available_filter_functions with the name of the function before it.
At best, this will cause the previous function's name to be listed twice. At worse, if the previous function was marked notrace, it will now show up as a function that can be traced. Note that it only shows up that it can be traced but will not be if enabled, which causes confusion.
https://lore.kernel.org/all/[email protected]/
The commit b39181f7c6907 ("ftrace: Add FTRACE_MCOUNT_MAX_OFFSET to avoid adding weak function") was a workaround to this by checking the function address before printing its name. If the address was too far from the function given by the name then instead of printing the name it would
print: __ftrace_invalid_address___<invalid-offset>
The real issue is that these invalid addresses are listed in the ftrace table look up which available_filter_functions is derived from. A place holder must be listed in that file because set_ftrace_filter may take a series of indexes into that file instead of names to be able to do O(1) lookups to enable filtering (many tools use this method).
Even if kallsyms saved the size of the function, it does not remove the need of having these place holders. The real solution is to not add a weak function into the ftrace table in the first place.
To solve this, the sorttable.c code that sorts the mcount regions during the build is modified to take a "nm -S vmlinux" input, sort it, and any function listed in the mcount_loc section that is not within a boundary of the function list given by nm is considered a weak function and is zeroed out.
Note, this does not mean they will remain zero when booting as KASLR will still shift those addresses. To handle this, the entries in the mcount_loc section will be ignored if they are zero or match the kaslr_offset() value.
Before:
~# grep __ftrace_invalid_address___ /sys/kernel/tracing/available_filter_functions | wc -l 551
After:
~# grep __ftrace_invalid_address___ /sys/kernel/tracing/available_filter_functions | wc -l 0
--- kernel/trace/ftrace.c | 6 +- scripts/link-vmlinux.sh | 4 + scripts/sorttable.c | 128 +++++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 134 insertions(+), 4 deletions(-)
kernel/trace/ftrace.c +5 -1
@@ -7048,6 +7048,7 @@ static int ftrace_process_locs(struct mounsigned long count;unsigned long *p;unsigned long addr; +unsigned long kaslr;unsigned long flags =0 ;/* Shut up gcc */ int ret = -ENOMEM ;
@@ -7096,6 +7097,9 @@ static int ftrace_process_locs(struct mo ftrace_pages->next = start_pg; } +/* For zeroed locations that were shifted for core kernel */ + kaslr = !mod ?kaslr_offset () :0 ; + p = start; pg = start_pg;while (p < end) {
@@ -7107,7 +7111,7 @@ static int ftrace_process_locs(struct mo* object files to satisfy alignments. * Skip any NULL pointers. */ -if (!addr) { +if (!addr || addr == kaslr) { skipped++;continue ; }
scripts/link-vmlinux.sh +3 -1
@@ -173,12 +173,14 @@ mksysmap() sorttable() { - ${objtree}/scripts/sorttable ${1} + ${NM} -S ${1} > .tmp_vmlinux.nm-sort + ${objtree}/scripts/sorttable -s .tmp_vmlinux.nm-sort ${1} } cleanup() { rm -f .btf.* + rm -f .tmp_vmlinux.nm-sort rm -f System.map rm -f vmlinux rm -f vmlinux.map
scripts/sorttable.c +126 -2
158 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 76/204 ftrace: Update the mcount_loc check of skipped entries
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 4a3efc6baff931da9a85c6d2e42c87bd9a827399 ]
Now that weak functions turn into skipped entries, update the check to make sure the amount that was allocated would fit both the entries that were allocated as well as those that were skipped.
--- kernel/trace/ftrace.c | 23 ++++++++++++++++++++++- 1 file changed, 22 insertions(+), 1 deletion(-)
kernel/trace/ftrace.c +22 -1
@@ -7155,7 +7155,28 @@ static int ftrace_process_locs(struct mo/* We should have used all pages unless we skipped some */ if (pg_unuse) { -WARN_ON (!skipped); +unsigned long pg_remaining, remaining =0 ; +unsigned long skip; + +/* Count the number of entries unused and compare it to skipped. */ + pg_remaining = (ENTRIES_PER_PAGE << pg->order) - pg->index; + +if (!WARN (skipped < pg_remaining,"Extra allocated pages for ftrace" )) { + + skip = skipped - pg_remaining; + +for (pg = pg_unuse; pg; pg = pg->next) + remaining +=1 << pg->order; + + skip =DIV_ROUND_UP (skip,ENTRIES_PER_PAGE ); + +/* +* Check to see if the number of pages remaining would +* just fit the number of entries skipped. +*/ +WARN (skip != remaining,"Extra allocated pages for ftrace: %lu with %lu skipped" , + remaining, skipped); + }/* Need to synchronize with ftrace_location_range() */ synchronize_rcu ();ftrace_free_pages (pg_unuse);
Greg Kroah-Hartman
maintainerPATCH 77/204 ftrace: Have ftrace pages output reflect freed pages
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 264143c4e54412095f4b615e65bf736fc3c60af0 ]
The amount of memory that ftrace uses to save the descriptors to manage the functions it can trace is shown at output. But if there are a lot of functions that are skipped because they were weak or the architecture added holes into the tables, then the extra pages that were allocated are freed. But these freed pages are not reflected in the numbers shown, and they can even be inconsistent with what is reported:
ftrace: allocating 57482 entries in 225 pages
ftrace: allocated 224 pages with 3 groups
The above shows the number of original entries that are in the mcount_loc section and the pages needed to save them (225), but the second output reflects the number of pages that were actually used. The two should be consistent as:
ftrace: allocating 56739 entries in 224 pages
ftrace: allocated 224 pages with 3 groups
The above also shows the accurate number of entires that were actually stored and does not include the entries that were removed.
--- kernel/trace/ftrace.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-)
kernel/trace/ftrace.c +12 -3
@@ -7050,6 +7050,7 @@ static int ftrace_process_locs(struct mounsigned long addr;unsigned long kaslr;unsigned long flags =0 ;/* Shut up gcc */ +unsigned long pages;int ret = -ENOMEM ; count = end - start;
@@ -7057,6 +7058,8 @@ static int ftrace_process_locs(struct moif (!count)return 0 ; + pages =DIV_ROUND_UP (count,ENTRIES_PER_PAGE ); +/* * Sorting mcount in vmlinux at build time depend on * CONFIG_BUILDTIME_MCOUNT_SORT, while mcount loc in
@@ -7168,6 +7171,8 @@ static int ftrace_process_locs(struct mofor (pg = pg_unuse; pg; pg = pg->next) remaining +=1 << pg->order; + pages -= remaining; + skip =DIV_ROUND_UP (skip,ENTRIES_PER_PAGE );/*
@@ -7181,6 +7186,13 @@ static int ftrace_process_locs(struct mosynchronize_rcu ();ftrace_free_pages (pg_unuse); } + +if (!mod) { + count -= skipped; +pr_info ("ftrace: allocating %ld entries in %ld pages\n" , + count, pages); + } +return ret; }
@@ -7835,9 +7847,6 @@ void __init ftrace_init(void) goto failed; } -pr_info ("ftrace: allocating %ld entries in %ld pages\n" , - count,DIV_ROUND_UP (count,ENTRIES_PER_PAGE )); - ret =ftrace_process_locs (NULL , __start_mcount_loc, __stop_mcount_loc);
Greg Kroah-Hartman
maintainerPATCH 78/204 ftrace: Do not over-allocate ftrace memory
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Guenter Roeck <[email protected]>
[ Upstream commit be55257fab181b93af38f8c4b1b3cb453a78d742 ]
The pg_remaining calculation in ftrace_process_locs() assumes that ENTRIES_PER_PAGE multiplied by 2^order equals the actual capacity of the allocated page group. However, ENTRIES_PER_PAGE is PAGE_SIZE / ENTRY_SIZE (integer division). When PAGE_SIZE is not a multiple of ENTRY_SIZE (e.g. 4096 / 24 = 170 with remainder 16), high-order allocations (like 256 pages) have significantly more capacity than 256 * 170. This leads to pg_remaining being underestimated, which in turn makes skip (derived from skipped - pg_remaining) larger than expected, causing the WARN(skip != remaining) to trigger.
Extra allocated pages for ftrace: 2 with 654 skipped
WARNING: CPU: 0 PID: 0 at kernel/trace/ftrace.c:7295 ftrace_process_locs+0x5bf/0x5e0
A similar problem in ftrace_allocate_records() can result in allocating too many pages. This can trigger the second warning in ftrace_process_locs().
Extra allocated pages for ftrace
WARNING: CPU: 0 PID: 0 at kernel/trace/ftrace.c:7276 ftrace_process_locs+0x548/0x580
Use the actual capacity of a page group to determine the number of pages to allocate. Have ftrace_allocate_pages() return the number of allocated pages to avoid having to calculate it. Use the actual page group capacity when validating the number of unused pages due to skipped entries. Drop the definition of ENTRIES_PER_PAGE since it is no longer used.
--- kernel/trace/ftrace.c | 29 +++++++++++++++-------------- 1 file changed, 15 insertions(+), 14 deletions(-)
kernel/trace/ftrace.c +15 -14
@@ -1122,7 +1122,6 @@ struct ftrace_page { };#define ENTRY_SIZE sizeof(struct dyn_ftrace) -#define ENTRIES_PER_PAGE (PAGE_SIZE / ENTRY_SIZE)static struct ftrace_page *ftrace_pages_start;static struct ftrace_page *ftrace_pages;
@@ -3754,7 +3753,8 @@ static int ftrace_update_code(struct modreturn 0 ; } -static int ftrace_allocate_records (struct ftrace_page *pg,int count) +static int ftrace_allocate_records (struct ftrace_page *pg,int count, +unsigned long *num_pages) {int order;int pages;
@@ -3764,7 +3764,7 @@ static int ftrace_allocate_records(strucreturn -EINVAL ;/* We want to fill as much as possible, with no empty pages */ - pages =DIV_ROUND_UP (count,ENTRIES_PER_PAGE ); + pages =DIV_ROUND_UP (count *ENTRY_SIZE ,PAGE_SIZE ); order =fls (pages) -1 ; again:
@@ -3779,6 +3779,7 @@ static int ftrace_allocate_records(struc } ftrace_number_of_pages +=1 << order; + *num_pages +=1 << order; ftrace_number_of_groups++; cnt = (PAGE_SIZE << order) /ENTRY_SIZE ;
@@ -3807,12 +3808,14 @@ static void ftrace_free_pages(struct ftr }static struct ftrace_page * -ftrace_allocate_pages (unsigned long num_to_init) +ftrace_allocate_pages (unsigned long num_to_init,unsigned long *num_pages) {struct ftrace_page *start_pg;struct ftrace_page *pg;int cnt; + *num_pages =0 ; +if (!num_to_init)return NULL ;
@@ -3826,7 +3829,7 @@ ftrace_allocate_pages(unsigned long num_* waste as little space as possible. */ for (;;) { - cnt =ftrace_allocate_records (pg, num_to_init); + cnt =ftrace_allocate_records (pg, num_to_init, num_pages);if (cnt <0 ) goto free_pages;
@@ -7058,8 +7061,6 @@ static int ftrace_process_locs(struct moif (!count)return 0 ; - pages =DIV_ROUND_UP (count,ENTRIES_PER_PAGE ); -/* * Sorting mcount in vmlinux at build time depend on * CONFIG_BUILDTIME_MCOUNT_SORT, while mcount loc in
@@ -7072,7 +7073,7 @@ static int ftrace_process_locs(struct motest_is_sorted (start, count); } - start_pg =ftrace_allocate_pages (count); + start_pg =ftrace_allocate_pages (count, &pages);if (!start_pg)return -ENOMEM ;
@@ -7159,27 +7160,27 @@ static int ftrace_process_locs(struct mo/* We should have used all pages unless we skipped some */ if (pg_unuse) {unsigned long pg_remaining, remaining =0 ; -unsigned long skip; +long skip;/* Count the number of entries unused and compare it to skipped. */ - pg_remaining = (ENTRIES_PER_PAGE << pg->order) - pg->index; + pg_remaining = (PAGE_SIZE << pg->order) /ENTRY_SIZE - pg->index;if (!WARN (skipped < pg_remaining,"Extra allocated pages for ftrace" )) { skip = skipped - pg_remaining; -for (pg = pg_unuse; pg; pg = pg->next) +for (pg = pg_unuse; pg && skip >0 ; pg = pg->next) { remaining +=1 << pg->order; + skip -= (PAGE_SIZE << pg->order) /ENTRY_SIZE ; + } pages -= remaining; - skip =DIV_ROUND_UP (skip,ENTRIES_PER_PAGE ); -/* * Check to see if the number of pages remaining would * just fit the number of entries skipped. */ -WARN (skip != remaining,"Extra allocated pages for ftrace: %lu with %lu skipped" , +WARN (pg || skip >0 ,"Extra allocated pages for ftrace: %lu with %lu skipped" , remaining, skipped); }/* Need to synchronize with ftrace_location_range() */
Greg Kroah-Hartman
maintainerPATCH 79/204 ftrace: Test mcount_loc addr before calling ftrace_call_addr()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 6eeca746fa5f1dd03c6ee05cb03f5eb1ddda1c81 ]
The addresses in the mcount_loc can be zeroed and then moved by KASLR making them invalid addresses. ftrace_call_addr() for ARM 64 expects a valid address to kernel text. If the addr read from the mcount_loc section is invalid, it must not call ftrace_call_addr(). Move the addr check before calling ftrace_call_addr() in ftrace_process_locs().
--- kernel/trace/ftrace.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
kernel/trace/ftrace.c +5 -1
@@ -7108,7 +7108,9 @@ static int ftrace_process_locs(struct mo pg = start_pg;while (p < end) {unsigned long end_offset; - addr =ftrace_call_adjust (*p++); + + addr = *p++; +/* * Some architecture linkers will pad between * the different mcount_loc sections of different
@@ -7120,6 +7122,8 @@ static int ftrace_process_locs(struct mocontinue ; } + addr =ftrace_call_adjust (addr); + end_offset = (pg->index+1 ) *sizeof (pg->records[0 ]);if (end_offset >PAGE_SIZE << pg->order) {/* We should have allocated enough */
Greg Kroah-Hartman
maintainerPATCH 80/204 ftrace: Check against is_kernel_text() instead of kaslr_offset()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit da0f622b344be769ed61e7c1caf95cd0cdb47964 ]
As kaslr_offset() is architecture dependent and also may not be defined by all architectures, when zeroing out unused weak functions, do not check against kaslr_offset(), but instead check if the address is within the kernel text sections. If KASLR added a shift to the zeroed out function, it would still not be located in the kernel text. This is a more robust way to test if the text is valid or not.
--- kernel/trace/ftrace.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-)
kernel/trace/ftrace.c +12 -5
@@ -7051,7 +7051,6 @@ static int ftrace_process_locs(struct mounsigned long count;unsigned long *p;unsigned long addr; -unsigned long kaslr;unsigned long flags =0 ;/* Shut up gcc */ unsigned long pages;int ret = -ENOMEM ;
@@ -7101,9 +7100,6 @@ static int ftrace_process_locs(struct mo ftrace_pages->next = start_pg; } -/* For zeroed locations that were shifted for core kernel */ - kaslr = !mod ?kaslr_offset () :0 ; - p = start; pg = start_pg;while (p < end) {
@@ -7117,7 +7113,18 @@ static int ftrace_process_locs(struct mo* object files to satisfy alignments. * Skip any NULL pointers. */ -if (!addr || addr == kaslr) { +if (!addr) { + skipped++; +continue ; + } + +/* +* If this is core kernel, make sure the address is in core +* or inittext, as weak functions get zeroed and KASLR can +* move them to something other than zero. It just will not +* move it to an area where kernel text is. +*/ +if (!mod && !(is_kernel_text (addr) ||is_kernel_inittext (addr))) { skipped++;continue ; }
Greg Kroah-Hartman
maintainerPATCH 81/204 scripts/sorttable: Use normal sort if theres no relocs in the mcount section
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit 46514b3c2c17c67cefe84b0c1a59e0aaf6093131 ]
When ARM 64 is compiled with gcc, the mcount_loc section will be filled with zeros and the addresses will be located in the Elf_Rela sections. To sort the mcount_loc section, the addresses from the Elf_Rela need to be placed into an array and that is sorted.
But when ARM 64 is compiled with clang, it does it the same way as other architectures and leaves the addresses as is in the mcount_loc section.
To handle both cases, ARM 64 will first try to sort the Elf_Rela section, and if it doesn't find any functions, it will then fall back to the sorting of the addresses in the mcount_loc section itself.
--- scripts/sorttable.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
scripts/sorttable.c +7 -2
@@ -827,9 +827,14 @@ static void *sort_mcount_loc(void *arg)pthread_exit (m_err); } -if (sort_reloc) +if (sort_reloc) { count =fill_relocs (vals, size, ehdr, emloc->start_mcount_loc); -else +/* gcc may use relocs to save the addresses, but clang does not. */ +if (!count) { + count =fill_addrs (vals, size, start_loc); + sort_reloc =0 ; + } + }else count =fill_addrs (vals, size, start_loc);if (count <0 ) {
Greg Kroah-Hartman
maintainerPATCH 82/204 scripts/sorttable: Allow matches to functions before function entry
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
[ Upstream commit dc208c69c033d3caba0509da1ae065d2b5ff165f ]
ARM 64 uses -fpatchable-function-entry=4,2 which adds padding before the function and the addresses in the mcount_loc point there instead of the function entry that is returned by nm. In order to find a function from nm to make sure it's not an unused weak function, the entries in the mcount_loc section needs to match the entries from nm. Since it can be an instruction before the entry, add a before_func variable that ARM 64 can set to 8, and if the mcount_loc entry is within 8 bytes of the nm function entry, then it will be considered a match.
--- scripts/sorttable.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
scripts/sorttable.c +6 -1
@@ -611,13 +611,16 @@ static int add_field(uint64_t addr, uintreturn 0 ; } +/* Used for when mcount/fentry is before the function entry */ +static int before_func; +/* Only return match if the address lies inside the function size */ static int cmp_func_addr (const void *K ,const void *A ) {uint64_t key = *(const uint64_t *)K ;const struct func_info *a =A ; -if (key < a->addr) +if (key + before_func < a->addr)return -1 ;return key >= a->addr + a->size; }
@@ -1253,6 +1256,8 @@ static int do_file(char const *const fna#ifdef MCOUNT_SORT_ENABLED sort_reloc = true; rela_type =0x403 ; +/* arm64 uses patchable function entry placing before function */ + before_func =8 ;#endif /* fallthrough */ case EM_386 :
Greg Kroah-Hartman
maintainerPATCH 83/204 scripts/sorttable: Fix endianness handling in build-time mcount sort
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vasily Gorbik <[email protected]>
[ Upstream commit 023f124a64174c47e18340ded7e2a39b96eb9523 ]
Kernel cross-compilation with BUILDTIME_MCOUNT_SORT produces zeroed mcount values if the build-host endianness does not match the ELF file endianness.
The mcount values array is converted from ELF file
endianness to build-host endianness during initialization in fill_relocs()/fill_addrs(). Avoid extra conversion of these values during weak-function zeroing; otherwise, they do not match nm-parsed addresses and all mcount values are zeroed out.
--- scripts/sorttable.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
scripts/sorttable.c +1 -1
@@ -857,7 +857,7 @@ static void *sort_mcount_loc(void *arg)for (void *ptr = vals; ptr < vals + size; ptr += long_size) {uint64_t key; - key = long_size ==4 ?r ((uint32_t *)ptr) :r8 ((uint64_t *)ptr); + key = long_size ==4 ? *(uint32_t *)ptr : *(uint64_t *)ptr;if (!find_func (key)) {if (long_size ==4 ) *(uint32_t *)ptr =0 ;
Greg Kroah-Hartman
maintainerPATCH 84/204 net: ipv6: Make udp_tunnel6_xmit_skb() void
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Petr Machata <[email protected]>
commit 6a7d88ca15f73c5c570c372238f71d63da1fda55 upstream.
The function always returns zero, thus the return value does not carry any signal. Just make it void.
Most callers already ignore the return value. However:
- Refold arguments of the call from sctp_v6_xmit() so that they fit into the 80-column limit.
- tipc_udp_xmit() initializes err from the return value, but that should already be always zero at that point. So there's no practical change, but elision of the assignment prompts a couple more tweaks to clean up the function.
--- include/net/udp_tunnel.h | 14 +++++++------- net/ipv6/ip6_udp_tunnel.c | 15 +++++++-------- net/sctp/ipv6.c | 7 ++++--- net/tipc/udp_media.c | 10 +++++----- 4 files changed, 23 insertions(+), 23 deletions(-)
include/net/udp_tunnel.h +7 -7
@@ -152,13 +152,13 @@ void udp_tunnel_xmit_skb(struct rtable * __be16 df, __be16 src_port, __be16 dst_port, bool xnet, bool nocheck); -int udp_tunnel6_xmit_skb (struct dst_entry *dst,struct sock *sk, -struct sk_buff *skb, -struct net_device *dev, -const struct in6_addr *saddr, -const struct in6_addr *daddr, -__u8 prio,__u8 ttl,__be32 label, -__be16 src_port,__be16 dst_port,bool nocheck); +void udp_tunnel6_xmit_skb (struct dst_entry *dst,struct sock *sk, +struct sk_buff *skb, +struct net_device *dev, +const struct in6_addr *saddr, +const struct in6_addr *daddr, +__u8 prio,__u8 ttl,__be32 label, +__be16 src_port,__be16 dst_port,bool nocheck);void udp_tunnel_sock_release (struct socket *sock);
net/ipv6/ip6_udp_tunnel.c +7 -8
@@ -74,13 +74,13 @@ error: }EXPORT_SYMBOL_GPL (udp_sock_create6); -int udp_tunnel6_xmit_skb (struct dst_entry *dst,struct sock *sk, -struct sk_buff *skb, -struct net_device *dev, -const struct in6_addr *saddr, -const struct in6_addr *daddr, -__u8 prio,__u8 ttl,__be32 label, -__be16 src_port,__be16 dst_port,bool nocheck) +void udp_tunnel6_xmit_skb (struct dst_entry *dst,struct sock *sk, +struct sk_buff *skb, +struct net_device *dev, +const struct in6_addr *saddr, +const struct in6_addr *daddr, +__u8 prio,__u8 ttl,__be32 label, +__be16 src_port,__be16 dst_port,bool nocheck) {struct udphdr *uh;struct ipv6hdr *ip6h;
@@ -109,7 +109,6 @@ int udp_tunnel6_xmit_skb(struct dst_entr ip6h->saddr = *saddr;ip6tunnel_xmit (sk, skb, dev); -return 0 ; }EXPORT_SYMBOL_GPL (udp_tunnel6_xmit_skb);
net/sctp/ipv6.c +4 -3
@@ -261,9 +261,10 @@ static int sctp_v6_xmit(struct sk_buff *skb_set_inner_ipproto (skb,IPPROTO_SCTP ); label =ip6_make_flowlabel (sock_net (sk), skb, fl6->flowlabel, true, fl6); -return udp_tunnel6_xmit_skb (dst, sk, skb,NULL , &fl6->saddr, - &fl6->daddr, tclass,ip6_dst_hoplimit (dst), - label,sctp_sk (sk)->udp_port, t->encap_port, false); +udp_tunnel6_xmit_skb (dst, sk, skb,NULL , &fl6->saddr, &fl6->daddr, + tclass,ip6_dst_hoplimit (dst), label, +sctp_sk (sk)->udp_port, t->encap_port, false); +return 0 ; }/* Returns the dst cache entry for the given source and destination ip
net/tipc/udp_media.c +5 -5
@@ -172,7 +172,7 @@ static int tipc_udp_xmit(struct net *net struct udp_media_addr *dst, struct dst_cache *cache) { struct dst_entry *ndst; -int ttl, err =0 ; +int ttl, err;local_bh_disable (); ndst =dst_cache_get (cache);
@@ -217,13 +217,13 @@ static int tipc_udp_xmit(struct net *net dst_cache_set_ip6(cache, ndst, &fl6.saddr); } ttl =ip6_dst_hoplimit (ndst); - err =udp_tunnel6_xmit_skb (ndst, ub->ubsock->sk, skb,NULL , - &src->ipv6, &dst->ipv6,0 , ttl,0 , - src->port, dst->port, false); +udp_tunnel6_xmit_skb (ndst, ub->ubsock->sk, skb,NULL , + &src->ipv6, &dst->ipv6,0 , ttl,0 , + src->port, dst->port, false);#endif }local_bh_enable (); -return err; +return 0 ; tx_error:local_bh_enable ();
Greg Kroah-Hartman
maintainerPATCH 85/204 sctp: disable BH before calling udp_tunnel_xmit_skb()
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xin Long <[email protected]>
commit 2cd7e6971fc2787408ceef17906ea152791448cf upstream.
udp_tunnel_xmit_skb() / udp_tunnel6_xmit_skb() are expected to run with BH disabled. After commit 6f1a9140ecda ("add xmit recursion limit to tunnel xmit functions"), on the path:
udp(6)_tunnel_xmit_skb() -> ip(6)tunnel_xmit()
dev_xmit_recursion_inc()/dec() must stay balanced on the same CPU.
Without local_bh_disable(), the context may move between CPUs, which can break the inc/dec pairing. This may lead to incorrect recursion level detection and cause packets to be dropped in ip(6)_tunnel_xmit() or __dev_queue_xmit().
Fix it by disabling BH around both IPv4 and IPv6 SCTP UDP xmit paths.
In my testing, after enabling the SCTP over UDP:
# ip net exec ha sysctl -w net.sctp.udp_port=9899 # ip net exec ha sysctl -w net.sctp.encap_port=9899 # ip net exec hb sysctl -w net.sctp.udp_port=9899 # ip net exec hb sysctl -w net.sctp.encap_port=9899
# ip net exec ha iperf3 -s
- without this patch:
# ip net exec hb iperf3 -c 192.168.0.1 --sctp [ 5] 0.00-10.00 sec 37.2 MBytes 31.2 Mbits/sec sender [ 5] 0.00-10.00 sec 37.1 MBytes 31.1 Mbits/sec receiver
- with this patch:
# ip net exec hb iperf3 -c 192.168.0.1 --sctp [ 5] 0.00-10.00 sec 3.14 GBytes 2.69 Gbits/sec sender [ 5] 0.00-10.00 sec 3.14 GBytes 2.69 Gbits/sec receiver
--- net/sctp/ipv6.c | 2 ++ net/sctp/protocol.c | 2 ++ 2 files changed, 4 insertions(+)
net/sctp/ipv6.c +2 -0
@@ -261,9 +261,11 @@ static int sctp_v6_xmit(struct sk_buff *skb_set_inner_ipproto (skb,IPPROTO_SCTP ); label =ip6_make_flowlabel (sock_net (sk), skb, fl6->flowlabel, true, fl6); +local_bh_disable ();udp_tunnel6_xmit_skb (dst, sk, skb,NULL , &fl6->saddr, &fl6->daddr, tclass,ip6_dst_hoplimit (dst), label,sctp_sk (sk)->udp_port, t->encap_port, false); +local_bh_enable ();return 0 ; }
net/sctp/protocol.c +2 -0
@@ -1086,9 +1086,11 @@ static inline int sctp_v4_xmit(struct skskb_reset_inner_mac_header (skb);skb_reset_inner_transport_header (skb);skb_set_inner_ipproto (skb,IPPROTO_SCTP ); +local_bh_disable ();udp_tunnel_xmit_skb (dst_rtable (dst), sk, skb, fl4->saddr, fl4->daddr, dscp,ip4_dst_hoplimit (dst), df,sctp_sk (sk)->udp_port, t->encap_port, false, false); +local_bh_enable ();return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 86/204 iio: light: veml6075: add bounds check to veml6075_it_ms index
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sam Daly <[email protected]>
commit 307dc4240bd41852d9e0912921e298160db1c109 upstream.
veml6075_it_ms has 5 elements but VEML6075_CONF_IT can yield values 0-7. If it returns a value >= 5, this causes an out-of-bounds array access. Add a bounds check and return -EINVAL if the index is out of range.
The problem values are reserved so should never be read from the register. Hence this is hardening against fault device, missprogramming or bus corruption.
--- drivers/iio/light/veml6075.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/iio/light/veml6075.c +6 -2
@@ -100,7 +100,7 @@ static const struct iio_chan_spec veml60static int veml6075_request_measurement (struct veml6075_data *data) { -int ret, conf, int_time; +int ret, conf, int_time, int_index; ret =regmap_read (data->regmap,VEML6075_CMD_CONF , &conf);if (ret <0 )
@@ -117,7 +117,11 @@ static int veml6075_request_measurement(* time for all possible configurations. Using a 1.50 factor simplifies * operations and ensures reliability under all circumstances. */ - int_time = veml6075_it_ms[FIELD_GET (VEML6075_CONF_IT , conf)]; + int_index =FIELD_GET (VEML6075_CONF_IT , conf); +if (int_index >=ARRAY_SIZE (veml6075_it_ms)) +return -EINVAL ; + + int_time = veml6075_it_ms[int_index];msleep (int_time + (int_time /2 ));/* shutdown again, data registers are still accessible */
Greg Kroah-Hartman
maintainerPATCH 87/204 iio: adc: ti-ads1298: add bounds check to pga_settings index
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sam Daly <[email protected]>
commit 95e8a48d7a85d4226934020e57815a3316d3a14b upstream.
ads1298_pga_settings has 7 elements but ADS1298_MASK_CH_PGA can yield values 0-7. If it yields a value >= 7, this causes an out-of-bounds array access. Add a bounds check and return -EINVAL if the index is out of range.
Note that the remaining value b111 is reserved so should not be seen in a correctly functioning system.
--- drivers/iio/adc/ti-ads1298.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/iio/adc/ti-ads1298.c +6 -1
@@ -279,6 +279,7 @@ static const u8 ads1298_pga_settings[] =static int ads1298_get_scale (struct ads1298_private *priv,int channel,int *val,int *val2) { +unsigned int pga_idx;int ret;unsigned int regval;u8 gain;
@@ -302,7 +303,11 @@ static int ads1298_get_scale(struct ads1if (ret)return ret; - gain = ads1298_pga_settings[FIELD_GET (ADS1298_MASK_CH_PGA , regval)]; + pga_idx =FIELD_GET (ADS1298_MASK_CH_PGA , regval); +if (pga_idx >=ARRAY_SIZE (ads1298_pga_settings)) +return -EINVAL ; + + gain = ads1298_pga_settings[pga_idx]; *val /= gain;/* Full scale is VREF / gain */ *val2 =ADS1298_BITS_PER_SAMPLE -1 ;/* Signed, hence the -1 */
Greg Kroah-Hartman
maintainerPATCH 88/204 vc_screen: fix null-ptr-deref in vcs_notifier() during concurrent vcs_write
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yi Yang <[email protected]>
commit a287620312dc6dcb9a093417a0e589bf30fcf38a upstream.
A KASAN null-ptr-deref was observed in vcs_notifier():
BUG: KASAN: null-ptr-deref in vcs_notifier+0x98/0x130
Read of size 2 at addr qmp_cmd_name: qmp_capabilities, arguments: {}The issue is a race condition in vcs_write(). When the console_lock is temporarily dropped (to copy data from userspace), the vc_data pointer obtained from vcs_vc() may become stale. After re-acquiring the lock, vcs_vc() is called again to re-validate the pointer. If the vc has been deallocated in the meantime, vcs_vc() returns NULL, and the while loop breaks (with written > 0). However, after the loop, vcs_scr_updated(vc) is still called with the now-NULL vc pointer, leading to a null pointer dereference in the notifier chain (vcs_notifier dereferences param->vc).
Fix this by adding a NULL check for vc before calling vcs_scr_updated().
--- drivers/tty/vt/vc_screen.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/tty/vt/vc_screen.c +1 -1
@@ -699,7 +699,7 @@ vcs_write(struct file *file, const char } *ppos += written; ret = written; -if (written) +if (written && vc)vcs_scr_updated (vc); unlock_out:
Greg Kroah-Hartman
maintainerPATCH 89/204 serial: qcom_geni: Fix RX DMA stall when SE_DMA_RX_LEN_IN is zero
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Viken Dadhaniya <[email protected]>
commit b93062b6d8a1b2d9bad235cac25558a909819026 upstream.
In qcom_geni_serial_handle_rx_dma(), geni_se_rx_dma_unprep() clears port->rx_dma_addr before SE_DMA_RX_LEN_IN is read. If the register is zero, for example when the RX stale counter fires on an idle line, the handler returns without calling geni_se_rx_dma_prep().
The next RX DMA interrupt then hits the !port->rx_dma_addr guard and returns immediately, so the RX DMA buffer is never rearmed and later input is lost.
Keep the handler on the rearm path when rx_in is zero. Warn about the unexpected zero-length DMA completion, skip received-data handling, and always call geni_se_rx_dma_prep().
--- drivers/tty/serial/qcom_geni_serial.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-)
drivers/tty/serial/qcom_geni_serial.c +3 -6
@@ -867,12 +867,9 @@ static void qcom_geni_serial_handle_rx_d port->rx_dma_addr =0 ; rx_in =readl (uport->membase +SE_DMA_RX_LEN_IN ); -if (!rx_in) { -dev_warn (uport->dev,"serial engine reports 0 RX bytes in!\n" ); -return ; - } - -if (!drop) +if (!rx_in) +dev_warn_ratelimited (uport->dev,"serial engine reports 0 RX bytes in!\n" ); +else if (!drop)handle_rx_uart (uport, rx_in); ret = geni_se_rx_dma_prep(&port->se, port->rx_buf,
Greg Kroah-Hartman
maintainerPATCH 90/204 ksmbd: reject non-VALID session in compound request branch
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gil Portnoy <[email protected]>
commit 609ca17d869d04ba249e32cdcbf13c0b1c66f43c upstream.
smb2_check_user_session() takes a shortcut for any operation that is not the first in a COMPOUND request: it reuses work->sess (the session bound by the first operation) and validates only the SessionId, then returns "valid". It never re-checks work->sess->state == SMB2_SESSION_VALID, and a SessionId of 0xFFFFFFFFFFFFFFFF (ULLONG_MAX, the MS-SMB2 related-operation value) skips even the id comparison. The standalone path (ksmbd_session_lookup_all() plus the SESSION_SETUP state machine) does enforce the VALID state; the compound branch bypasses all of it.
A SESSION_SETUP carrying only an NTLM Type-1 (NtLmNegotiate) blob publishes a fresh SMB2_SESSION_IN_PROGRESS session whose sess->user is still NULL (->user is assigned later, by ntlm_authenticate()). Used as operation 1 of a COMPOUND with operation 2 = TREE_CONNECT (related, SessionId=ULLONG_MAX, \\host\IPC$), the tree-connect then runs on that IN_PROGRESS session and reaches ksmbd_ipc_tree_connect_request(), which dereferences user_name(sess->user) with sess->user == NULL (transport_ipc.c:687/701/704) -> remote NULL-pointer dereference and a kernel Oops that wedges the ksmbd worker for all clients.
Reject any non-first compound operation that lands on a session which is not SMB2_SESSION_VALID, mirroring the validity the standalone lookup path enforces. SESSION_SETUP itself legitimately runs on an IN_PROGRESS session, but it is never carried as a non-first compound operation, so multi-leg authentication is unaffected by this check.
--- fs/smb/server/smb2pdu.c | 5 +++++ 1 file changed, 5 insertions(+)
fs/smb/server/smb2pdu.c +5 -0
@@ -612,6 +612,11 @@ int smb2_check_user_session(struct ksmbd sess_id, work->sess->id);return -EINVAL ; } +if (work->sess->state !=SMB2_SESSION_VALID ) { +pr_err ("compound request on a non-valid session (state %d)\n" , + work->sess->state); +return -EINVAL ; + }return 1 ; }
Greg Kroah-Hartman
maintainerPATCH 91/204 media: vidtv: fix NULL pointer dereference in vidtv_mux_push_si
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ruslan Valiyev <[email protected]>
commit 7d8bf3d8f91073f4db347ed3aa6302b56107499c upstream.
syzbot reported a general protection fault invidtv_psi_ts_psi_write_into [1].
vidtv_mux_get_pid_ctx() can return NULL, but vidtv_mux_push_si() does not check for this before dereferencing the returned pointer to access the continuity counter. This leads to a general protection fault when accessing a near-NULL address.
The root cause is that vidtv_mux_pid_ctx_init() does not check the return value of vidtv_mux_create_pid_ctx_once() for PMT section PIDs. If the allocation fails, the PID context is never created, but init returns success. The subsequent vidtv_mux_push_si() call then gets NULL from vidtv_mux_get_pid_ctx() and crashes.
Fix both the root cause (add error check in vidtv_mux_pid_ctx_init for PMT PIDs) and add defensive NULL checks in vidtv_mux_push_si for all vidtv_mux_get_pid_ctx() calls.
[1]
Oops: general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] SMP KASAN PTI KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] Workqueue: events vidtv_mux_tick RIP: 0010:vidtv_psi_ts_psi_write_into+0x54a/0xbc0 drivers/media/test-drivers/vidtv/vidtv_psi.c:197 Call Trace: <TASK> vidtv_psi_table_header_write_into drivers/media/test-drivers/vidtv/vidtv_psi.c:799 [inline] vidtv_psi_pmt_write_into+0x3b2/0xa70 drivers/media/test-drivers/vidtv/vidtv_psi.c:1231 vidtv_mux_push_si+0x932/0xe80 drivers/media/test-drivers/vidtv/vidtv_mux.c:196 vidtv_mux_tick+0xe9b/0x1480 drivers/media/test-drivers/vidtv/vidtv_mux.c:408
--- drivers/media/test-drivers/vidtv/vidtv_mux.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/media/test-drivers/vidtv/vidtv_mux.c +7 -1
@@ -101,7 +101,8 @@ static int vidtv_mux_pid_ctx_init(struct/* add a ctx for all PMT sections */ while (p) { pid =vidtv_psi_get_pat_program_pid (p); -vidtv_mux_create_pid_ctx_once (m, pid); +if (!vidtv_mux_create_pid_ctx_once (m, pid)) + goto free; p = p->next; }
@@ -170,6 +171,9 @@ static u32 vidtv_mux_push_si(struct vidt nit_ctx =vidtv_mux_get_pid_ctx (m,VIDTV_NIT_PID ); eit_ctx =vidtv_mux_get_pid_ctx (m,VIDTV_EIT_PID ); +if (!pat_ctx || !sdt_ctx || !nit_ctx || !eit_ctx) +return 0 ; + pat_args.offset = m->mux_buf_offset; pat_args.continuity_counter = &pat_ctx->cc;
@@ -186,6 +190,8 @@ static u32 vidtv_mux_push_si(struct vidt } pmt_ctx =vidtv_mux_get_pid_ctx (m, pmt_pid); +if (!pmt_ctx) +continue ; pmt_args.offset = m->mux_buf_offset; pmt_args.pmt = m->si.pmt_secs[i];
Greg Kroah-Hartman
maintainerPATCH 92/204 virtiofs: fix UAF on submount umount
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Miklos Szeredi <[email protected]>
commit 06b41351779e9289e8785694ade9042ae85e41ea upstream.
iput() called from fuse_release_end() can Oops if the super block has already been destroyed. Normally this is prevented by waiting for num_waiting to go down to zero before commencing with super block shutdown.
This only works, however, for the last submount instance, as the wait counter is per connection, not per superblock.
Revert to using synchronous release requests for the auto_submounts case, which is virtiofs only at this time.
--- fs/fuse/file.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
fs/fuse/file.c +7 -1
@@ -373,8 +373,14 @@ void fuse_file_release(struct inode *ino* aio and closes the fd before the aio completes. Since aio takes its * own ref to the file, the IO completion has to drop the ref, which is * how the fuse server can end up closing its clients' files. +* +* Exception is virtio-fs, which is not affected by the above (server is +* on host, cannot close open files in guest). Virtio-fs needs sync +* release, because the num_waiting mechanism to wait for all requests +* before commencing with fs shutdown doesn't work if submounts are +* used. */ -fuse_file_put (ff, false); +fuse_file_put (ff, ff->fm->fc->auto_submounts); }void fuse_release_common (struct file *file,bool isdir)
Greg Kroah-Hartman
maintainerPATCH 93/204 KVM: x86: Fix shadow paging use-after-free due to unexpected role
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Paolo Bonzini <[email protected]>
commit 81ccda30b4e83d8f5cc4fd50503c44e3a33abfeb upstream.
Commit 0cb2af2ea66ad ("KVM: x86: Fix shadow paging use-after-free due to unexpected GFN") fixed a shadow paging mismatch between stored and computed GFNs; the bug could be triggered by changing a PDE mapping from outside the guest, and then deleting a memslot. The rmap_remove() call would miss entries created after the PDE change because the GFN of the leaf SPTE does not match the GFN of the struct kvm_mmu_page.
A similar hole however remains if the modified PDE points to a non-leaf page. In this case the gfn can be made to match, but the role does not
match: the original large 2MB page creates a kvm_mmu_page with direct=1, while the new 4KB needs a kvm_mmu_page with direct=0. However, kvm_mmu_get_child_sp() does not compare the role, and therefore reuses the page.
The next step is installing a leaf (4KB) SPTE on the new path which records an rmap entry under the gfn resolved by the walk. But when that child is zapped its parent kvm_mmu_page has direct=1 and kvm_mmu_page_get_gfn() computes the gfn for the 4KB page as sp->gfn + index instead of using sp->shadowed_translation[] (or sp->gfns[] in older kernels). It therefore fails to remove the recorded entry.
When the memslot is dropped the shadow page is freed but the rmap entry survives, as in the scenario that was already fixed. Code that later walks that gfn (dirty logging, MMU notifier invalidation, and so on) dereferences an sptep that lies in the freed page, causing the use-after-free.
--- arch/x86/kvm/mmu/mmu.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)
arch/x86/kvm/mmu/mmu.c +6 -4
@@ -2329,13 +2329,15 @@ static struct kvm_mmu_page *kvm_mmu_get_child_sp(struct kvm_vcpu *vcpu, u64 *sptep, gfn_t gfn, bool direct, unsigned int access) { - union kvm_mmu_page_role role; + union kvm_mmu_page_role role = kvm_mmu_child_role(sptep, direct, access); -if (is_shadow_present_pte (*sptep) && !is_large_pte (*sptep) && -spte_to_child_sp (*sptep) &&spte_to_child_sp (*sptep)->gfn == gfn) +if (is_shadow_present_pte (*sptep) && + !is_large_pte (*sptep) && +spte_to_child_sp (*sptep) && +spte_to_child_sp (*sptep)->gfn == gfn && +spte_to_child_sp (*sptep)->role.word == role.word)return ERR_PTR (-EEXIST ); - role =kvm_mmu_child_role (sptep, direct, access);return kvm_mmu_get_shadow_page (vcpu, gfn, role); }
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 94/204 KVM: x86/mmu: Ensure hugepage is in by slot before checking max mapping level
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sean Christopherson <[email protected]>
commit ef057cbf825e03b63f6edf5980f96abf3c53089d upstream.
When recovering hugepages in the shadow MMU, verify that the base gfn of the shadow page is actually contained within the target memslot, *before* querying the max mapping level given the shadow page's gfn. Failure to pre-check the validity of the gfn can lead to an out-of-bounds access to the slot's lpage_info (which typically manifests as a host #PF because the lpage_info is vmalloc'd) if the guest creates a hugepage mapping (in its PTEs) that extends "below" the bounds of a memslot.
When faulting in memory for a guest, and the size of the guest mapping is greater than KVM's (current) max mapping, then KVM will create a "direct" shadow page (direct in that there are no gPTEs to shadow, and so the target gfn is a direct calculation given the base gfn of the shadow page). The hugepage recovery flow looks for such direct shadow pages, as forcing 4KiB mappings when dirty logging generates the guest > host mapping size case. When the 4KiB restriction is lifted, then KVM can replace the shadow page with a hugepage.
But if KVM originally used a smaller mapping than the guest because the range of memory covered by the guest hugepage exceeds the bounds of a memslot, then KVM will link a direct shadow page with a gfn that is outside the bounds of the memslot being used to fault in memory. The rmap entry added for the leaf mapping is correct and within bounds, but the gfn of the leaf SPTE's parent shadow page will be out of bounds.
BUG: unable to handle page fault for address: ffffc90000806ffc #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 100000067 P4D 100000067 PUD 1002a7067 PMD 10612f067 PTE 0 Oops: Oops: 0000 [#1] SMP CPU: 13 UID: 1000 PID: 757 Comm: mmu_stress_test Not tainted 7.1.0-rc1-48ce1e26eace-x86_pir_to_irr_comments-vm #341 PREEMPT Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015 RIP: 0010:kvm_mmu_max_mapping_level+0x79/0x2b0 [kvm] Call Trace: <TASK> kvm_mmu_recover_huge_pages+0x21b/0x320 [kvm] kvm_set_memslot+0x1ee/0x590 [kvm] kvm_set_memory_region.part.0+0x3a1/0x4d0 [kvm] kvm_vm_ioctl+0x9bf/0x15d0 [kvm] __x64_sys_ioctl+0x8a/0xd0 do_syscall_64+0xb7/0xbb0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 RIP: 0033:0x7f21c0f1a9bf </TASK>
Don't bother pre-checking the bounds of the potential hugepage, i.e. don't check that e.g. sp->gfn + KVM_PAGES_PER_HPAGE(sp->role.level + 1) is also within the memslot, as the checks performed by kvm_mmu_max_mapping_level() are a superset of the basic bounds checks. I.e. pre-checking the full range would be a dubious micro-optimization.
--- arch/x86/kvm/mmu/mmu.c | 18 ++++++++++++------ include/linux/kvm_host.h | 7 ++++++- 2 files changed, 18 insertions(+), 7 deletions(-)
arch/x86/kvm/mmu/mmu.c +12 -6
@@ -6952,13 +6952,19 @@ static bool kvm_mmu_zap_collapsible_spte(struct kvm *kvm, sp =sptep_to_sp (sptep);/* -* We cannot do huge page mapping for indirect shadow pages, -* which are found on the last rmap (level = 1) when not using -* tdp; such shadow pages are synced with the page table in -* the guest, and the guest page table is using 4K page size -* mapping if the indirect sp has level = 1. +* Direct shadow page can be replaced by a hugepage if the host +* mapping level allows it and the memslot maps all of the host +* hugepage. Note! If the memslot maps only part of the +* hugepage, sp->gfn may be below slot->base_gfn, and querying +* the max mapping level would cause an out-of-bounds lpage_info +* access. So the gfn bounds check *must* be done first. +* +* Indirect shadow pages are created when the guest page tables +* are using 4K pages. Since the host mapping is always +* constrained by the page size in the guest, indirect shadow +* pages are never collapsible. */ -if (sp->role.direct && +if (sp->role.direct &&is_gfn_in_memslot (slot, sp->gfn) && sp->role.level <kvm_mmu_max_mapping_level (kvm, slot, sp->gfn,PG_LEVEL_NUM )) {kvm_zap_one_rmap_spte (kvm, rmap_head, sptep);
include/linux/kvm_host.h +6 -1
@@ -1745,6 +1745,11 @@ int kvm_request_irq_source_id(struct kvm *kvm);void kvm_free_irq_source_id (struct kvm *kvm,int irq_source_id);bool kvm_arch_irqfd_allowed (struct kvm *kvm,struct kvm_irqfd *args); +static inline bool is_gfn_in_memslot (const struct kvm_memory_slot *slot,gfn_t gfn) +{ +return gfn >= slot->base_gfn && gfn < slot->base_gfn + slot->npages; +} +/* * Returns a pointer to the memslot if it contains gfn. * Otherwise returns NULL.
@@ -1755,7 +1760,7 @@ try_get_memslot(struct kvm_memory_slot *slot, gfn_t gfn)if (!slot)return NULL ; -if (gfn >= slot->base_gfn && gfn < slot->base_gfn + slot->npages) +if (is_gfn_in_memslot (slot, gfn))return slot;else return NULL ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 95/204 Revert "PCI: qcom: Advertise Hotplug Slot Capability with no Command Completion support"
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
This reverts commit 480c94d3affbc11b9e98ca223a9fa19d90b84fbb.
--- drivers/pci/controller/dwc/pcie-qcom.c | 17 ++++++----------- 1 file changed, 6 insertions(+), 11 deletions(-)
drivers/pci/controller/dwc/pcie-qcom.c +6 -11
@@ -329,20 +329,15 @@ static void qcom_pcie_clear_aspm_l0s(struct dw_pcie *pci)dw_pcie_dbi_ro_wr_dis (pci); } -static void qcom_pcie_set_slot_nccs (struct dw_pcie *pci) +static void qcom_pcie_clear_hpc (struct dw_pcie *pci) {u16 offset =dw_pcie_find_capability (pci,PCI_CAP_ID_EXP );u32 val;dw_pcie_dbi_ro_wr_en (pci); -/* -* Qcom PCIe Root Ports do not support generating command completion -* notifications for the Hot-Plug commands. So set the NCCS field to -* avoid waiting for the completions. -*/ val =readl (pci->dbi_base + offset +PCI_EXP_SLTCAP ); - val |=PCI_EXP_SLTCAP_NCCS ; + val &= ~PCI_EXP_SLTCAP_HPC ;writel (val, pci->dbi_base + offset +PCI_EXP_SLTCAP );dw_pcie_dbi_ro_wr_dis (pci);
@@ -537,7 +532,7 @@ static int qcom_pcie_post_init_2_1_0(struct qcom_pcie *pcie)writel (CFG_BRIDGE_SB_INIT , pci->dbi_base +AXI_MSTR_RESP_COMP_CTRL1 ); -qcom_pcie_set_slot_nccs (pcie->pci); +qcom_pcie_clear_hpc (pcie->pci);return 0 ; }
@@ -617,7 +612,7 @@ static int qcom_pcie_post_init_1_0_0(struct qcom_pcie *pcie)writel (val, pcie->parf +PARF_AXI_MSTR_WR_ADDR_HALT ); } -qcom_pcie_set_slot_nccs (pcie->pci); +qcom_pcie_clear_hpc (pcie->pci);return 0 ; }
@@ -710,7 +705,7 @@ static int qcom_pcie_post_init_2_3_2(struct qcom_pcie *pcie) val |=EN ;writel (val, pcie->parf +PARF_AXI_MSTR_WR_ADDR_HALT_V2 ); -qcom_pcie_set_slot_nccs (pcie->pci); +qcom_pcie_clear_hpc (pcie->pci);return 0 ; }
@@ -1014,7 +1009,7 @@ static int qcom_pcie_post_init_2_7_0(struct qcom_pcie *pcie)writel (WR_NO_SNOOP_OVERIDE_EN |RD_NO_SNOOP_OVERIDE_EN , pcie->parf +PARF_NO_SNOOP_OVERIDE ); -qcom_pcie_set_slot_nccs (pcie->pci); +qcom_pcie_clear_hpc (pcie->pci);return 0 ; }
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 96/204 KVM: SEV: Ignore MMIO requests of length 0
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sean Christopherson <[email protected]>
commit 1aa8a6dc7dac8b83234b53518311bf78231f4fa5 upstream.
Explicitly ignore MMIO requests of length '0', so that setting up the software scratch area (and other code) doesn't have to worry about underflowing the length, and to allow for special casing '0' in the future.
Message-ID: <[email protected]>
--- arch/x86/kvm/svm/sev.c | 6 ++++++ 1 file changed, 6 insertions(+)
arch/x86/kvm/svm/sev.c +6 -0
@@ -4347,6 +4347,9 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) exit_code =kvm_ghcb_get_sw_exit_code (control);switch (exit_code) {case SVM_VMGEXIT_MMIO_READ : +if (!control->exit_info_2) +return 1 ; + ret =setup_vmgexit_scratch (svm, true, control->exit_info_2);if (ret)break ;
@@ -4357,6 +4360,9 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) svm->sev_es.ghcb_sa);break ;case SVM_VMGEXIT_MMIO_WRITE : +if (!control->exit_info_2) +return 1 ; + ret =setup_vmgexit_scratch (svm, false, control->exit_info_2);if (ret)break ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 97/204 KVM: SEV: Reject MMIO requests larger than 8 bytes with GHCB v2+
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sean Christopherson <[email protected]>
commit dcf1b2d4b0564a27e4ca7c654871aab4f9620046 upstream.
When using GHCB v2+, reject MMIO requests that are larger than 8 bytes. Per the GHCB spec:
SW_EXITINFO2 must be less than or equal to 0x7fffffff for version 1 and less than or equal to 0x8 for all other versions.
Message-ID: <[email protected]>
--- arch/x86/kvm/svm/sev.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+)
arch/x86/kvm/svm/sev.c +14 -0
@@ -4350,6 +4350,13 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu)if (!control->exit_info_2)return 1 ; +if (to_kvm_sev_info (vcpu->kvm)->ghcb_version >=2 && + control->exit_info_2 >8 ) { +ghcb_set_sw_exit_info_1 (svm->sev_es.ghcb,2 ); +ghcb_set_sw_exit_info_2 (svm->sev_es.ghcb,GHCB_ERR_INVALID_INPUT ); +return 1 ; + } + ret =setup_vmgexit_scratch (svm, true, control->exit_info_2);if (ret)break ;
@@ -4363,6 +4370,13 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu)if (!control->exit_info_2)return 1 ; +if (to_kvm_sev_info (vcpu->kvm)->ghcb_version >=2 && + control->exit_info_2 >8 ) { +ghcb_set_sw_exit_info_1 (svm->sev_es.ghcb,2 ); +ghcb_set_sw_exit_info_2 (svm->sev_es.ghcb,GHCB_ERR_INVALID_INPUT ); +return 1 ; + } + ret =setup_vmgexit_scratch (svm, false, control->exit_info_2);if (ret)break ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 98/204 KVM: SEV: Ignore Port I/O requests of length 0
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sean Christopherson <[email protected]>
commit 3988bd2723de407ae90fa7a6f6029b4e60238c58 upstream.
Explicitly ignore Port I/O requests of length '0' (or count '0'), so that setting up the software scratch area (and other code) doesn't have to worry about underflowing the length, and to allow for WARNing on trying to configure the scratch area with len==0.
Message-ID: <[email protected]>
--- arch/x86/kvm/svm/sev.c | 8 ++++++++ 1 file changed, 8 insertions(+)
arch/x86/kvm/svm/sev.c +8 -0
@@ -4459,6 +4459,11 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) control->exit_info_1, control->exit_info_2); ret = -EINVAL ;break ; +case SVM_EXIT_IOIO : +if (!((control->exit_info_1 &SVM_IOIO_SIZE_MASK ) >>SVM_IOIO_SIZE_SHIFT )) +return 1 ; + + fallthrough;default : ret =svm_invoke_exit_handler (vcpu, exit_code); }
@@ -4479,6 +4484,9 @@ int sev_es_string_io(struct vcpu_svm *svm, int size, unsigned int port, int in)if (unlikely (check_mul_overflow (count, size, &bytes)))return -EINVAL ; +if (!bytes) +return 1 ; + r =setup_vmgexit_scratch (svm, in, bytes);if (r)return r;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 99/204 batman-adv: tp_meter: keep unacked list in ascending ordered
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit 5aa8651527ea0b610e7a09fb3b8204c1398b9525 upstream.
When batadv_tp_handle_out_of_order inserts a new entry in the list of unacked (out of order) packets, it searches from the entry with the newest sequence number towards oldest sequence number. If an entry is found which is older than the newly entry, the new entry has to be added after the found one to keep the ascending order.
But for this operation list_add_tail() was used. But this function adds an entry _before_ another one. As result, the list would contain a lot of swapped sequence numbers. The consumer of this list (batadv_tp_ack_unordered()) would then fail to correctly ack packets.
--- net/batman-adv/tp_meter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/batman-adv/tp_meter.c +1 -1
@@ -1325,7 +1325,7 @@ static bool batadv_tp_handle_out_of_order(struct batadv_tp_vars *tp_vars,* one is attached _after_ it. In this way the list is kept in * ascending order */ -list_add_tail (&new->list, &un->list); +list_add (&new->list, &un->list); added = true;break ; }
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 100/204 batman-adv: tp_meter: initialize dup_acks explicitly
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit b2b68b32a715e0328662801576974aa37b942b00 upstream.
When an ack with a sequence number equal to the last_acked is received, the dup_acks counter is increased to decide whether fast retransmit should be performed. Only when the sequence numbers are not equal, the dup_acks is set to the initial value (0).
But if the initial packet would have the sequence number BATADV_TP_FIRST_SEQ, dup_acks would not be initialized and atomic_inc would operate on an undefined starting value. It is therefore required to have it explicitly initialized during the start of the sender session.
--- net/batman-adv/tp_meter.c | 1 + 1 file changed, 1 insertion(+)
net/batman-adv/tp_meter.c +1 -0
@@ -1045,6 +1045,7 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, tp_vars->icmp_uid = icmp_uid; tp_vars->last_sent =BATADV_TP_FIRST_SEQ ; +atomic_set (&tp_vars->dup_acks,0 );atomic_set (&tp_vars->last_acked,BATADV_TP_FIRST_SEQ ); tp_vars->fast_recovery = false; tp_vars->recover =BATADV_TP_FIRST_SEQ ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 101/204 batman-adv: tp_meter: initialize dec_cwnd explicitly
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit febfb1b86224489535312296ecfa3d4bf467f339 upstream.
When batadv_tp_update_cwnd() is called, dec_cwnd is increased. But dec_cwnd is only initialixed (to 0) when a duplicate Ack was received or when cwnd is below the ss_threshold.
Just initialize the cwnd during the initialization to avoid any potential access of uninitialized data.
--- net/batman-adv/tp_meter.c | 2 ++ 1 file changed, 2 insertions(+)
net/batman-adv/tp_meter.c +2 -0
@@ -1055,6 +1055,8 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst,* soft_interface, hence its MTU */ tp_vars->cwnd =BATADV_TP_PLEN *3 ; + tp_vars->dec_cwnd =0 ; +/* at the beginning initialise the SS threshold to the biggest possible * window size, hence the AWND size */
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 102/204 batman-adv: tp_meter: avoid window underflow
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit 765947b81fb54b6ebb0bc1cfe55c0fa399e002b8 upstream.
In batadv_tp_avail(), win_left is calculated with 32-bit unsigned
arithmetic: win_left = win_limit - tp_vars->last_sent;
During Fast Recovery, cwnd is inflated and last_sent advances rapidly. When Fast Recovery ends, cwnd drops abruptly back to ss_threshold. If the newly shrunk win_limit is less than last_sent, the unsigned subtraction will underflow, wrapping to a massive positive value. Instead of returning that the window is full (unavailable), it returns that the sender can continue sending.
To handle this situation, it must be checked whether the windows end sequence number (win_limit) has to be compared with the last sent sequence number. If it would be before the last sent sequence number, then more acks are needed before the transmission can be started again.
--- net/batman-adv/tp_meter.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
net/batman-adv/tp_meter.c +6 -1
@@ -817,10 +817,15 @@ static void batadv_tp_recv_ack(struct batadv_priv *bat_priv,static bool batadv_tp_avail (struct batadv_tp_vars *tp_vars,size_t payload_len) { +u32 last_sent =READ_ONCE (tp_vars->last_sent);u32 win_left, win_limit; win_limit =atomic_read (&tp_vars->last_acked) + tp_vars->cwnd; - win_left = win_limit - tp_vars->last_sent; + +if (batadv_seq_before (last_sent, win_limit)) + win_left = win_limit - last_sent; +else + win_left =0 ;return win_left >= payload_len; }
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 103/204 batman-adv: tp_meter: avoid divide-by-zero for dec_cwnd
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit 33ccd52f3cc9ed46ce395199f89aa3234dc83314 upstream.
The cwnd is always MSS <= cwnd <= 0x20000000. But the calculation in batadv_tp_update_cwnd() assumes unsigned 32 bit arithmetics.
((mss * 8) ** 2) / (cwnd * 8)
In case cwnd is actually 0x20000000, it will be shifted by 3 bit to the left end up at 0x100000000 or U32_MAX + 1. It will therefore wrap around and be 0 - resulting in:
((mss * 8) ** 2) / 0
This is of course invalid and cannot be calculated. The calculation should must be simplified to avoid this overflow:
(mss ** 2) * 8 / cwnd
It will keep the precision enhancement from the scaling (by 8) but avoid the overflow in the divisor.
In theory, there could still be an overflow in the dividend. It is at the moment fixed to BATADV_TP_PLEN in batadv_tp_recv_ack() - so it is not an imminent problem. But allowing it to use the whole u32 bit range, would mean that it can still use up to 67 bits. To keep this calculation safe for 32 bit arithmetic, mss must never use more than floor((32 - 3) / 2) bits - or in other words: must never be larger than 16383.
--- net/batman-adv/tp_meter.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
net/batman-adv/tp_meter.c +4 -1
@@ -154,9 +154,12 @@ static void batadv_tp_update_cwnd(struct batadv_tp_vars *tp_vars, u32 mss)return ; } +/* prevent overflow in (mss * mss) << 3 */ + mss =min_t (u32, mss, (1U <<14 ) -1 ); +/* increment CWND at least of 1 (section 3.1 of RFC5681) */ tp_vars->dec_cwnd +=max_t (u32,1U <<3 , - ((mss * mss) <<6 ) / (tp_vars->cwnd <<3 )); + ((mss * mss) <<3 ) / tp_vars->cwnd);if (tp_vars->dec_cwnd < (mss <<3 )) {spin_unlock_bh (&tp_vars->cwnd_lock);return ;
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 104/204 batman-adv: tp_meter: fix fast recovery precondition
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit 2b0d08f08ed3b2174f05c43089ec65f3543a025b upstream.
The fast recovery precondition checks if the recover (initialized to BATADV_TP_FIRST_SEQ) is bigger than the received ack. But since recover is only updated when this check is successful, it will never enter the fast recovery mode.
According to RFC6582 Section 3.2 step 2, the check should actually be different:
Quoted 3 lines
When the third duplicate ACK is received, the TCP sender first checks the value of recover to see if the Cumulative Acknowledgment field covers more than recover
The precondition must therefore check if recover is smaller than the received ack - basically swapping the operands of the current check.
--- net/batman-adv/tp_meter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/batman-adv/tp_meter.c +1 -1
@@ -733,7 +733,7 @@ static void batadv_tp_recv_ack(struct batadv_priv *bat_priv,if (atomic_read (&tp_vars->dup_acks) !=3 ) goto out; -if (recv_ack >= tp_vars->recover) +if (tp_vars->recover >= recv_ack) goto out;/* if this is the third duplicate ACK do Fast Retransmit */
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 105/204 batman-adv: tp_meter: handle seqno wrap-around for fast recovery detection
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit f54c85ed42a1b27a516cf2a4728f5a612b799e07 upstream.
The recover variable and the last_sent sequence number are initialized on purpose as a really high value which will wrap-around after the first 2000 bytes. The fast recovery precondition must therefore not use simple integer comparisons but use helpers which are aware of the sequence number wrap-arounds.
--- net/batman-adv/tp_meter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/batman-adv/tp_meter.c +1 -1
@@ -733,7 +733,7 @@ static void batadv_tp_recv_ack(struct batadv_priv *bat_priv,if (atomic_read (&tp_vars->dup_acks) !=3 ) goto out; -if (tp_vars->recover >= recv_ack) +if (!batadv_seq_before (tp_vars->recover, recv_ack)) goto out;/* if this is the third duplicate ACK do Fast Retransmit */
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 106/204 batman-adv: tp_meter: add only finished tp_vars to lists
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit 15ccbf685222274f5add1387af58c2a41a95f81e upstream.
When the receiver variables (aka "session") are initialized, then they are added to the list of sessions before the timer is set up. A RCU protected reader could therefore find the entry and run mod_setup before batadv_tp_init_recv() finished the timer initialization.
The same is true for batadv_tp_start(), which must first initialize the finish_work and the test_length to avoid a similar problem.
--- net/batman-adv/tp_meter.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-)
net/batman-adv/tp_meter.c +9 -9
@@ -1096,21 +1096,21 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, tp_vars->prerandom_offset =0 ;spin_lock_init (&tp_vars->prerandom_lock); -kref_get (&tp_vars->refcount); -hlist_add_head_rcu (&tp_vars->list, &bat_priv->tp_list); -spin_unlock_bh (&bat_priv->tp_list_lock); - tp_vars->test_length = test_length;if (!tp_vars->test_length) tp_vars->test_length =BATADV_TP_DEF_TEST_LENGTH ; +/* init work item for finished tp tests */ +INIT_DELAYED_WORK (&tp_vars->finish_work, batadv_tp_sender_finish); + +kref_get (&tp_vars->refcount); +hlist_add_head_rcu (&tp_vars->list, &bat_priv->tp_list); +spin_unlock_bh (&bat_priv->tp_list_lock); +batadv_dbg (BATADV_DBG_TP_METER , bat_priv,"Meter: starting throughput meter towards %pM (length=%ums)\n" , dst, test_length); -/* init work item for finished tp tests */ -INIT_DELAYED_WORK (&tp_vars->finish_work, batadv_tp_sender_finish); -/* start tp kthread. This way the write() call issued from userspace can * happily return and avoid to block */
@@ -1430,10 +1430,10 @@ batadv_tp_init_recv(struct batadv_priv *bat_priv,INIT_LIST_HEAD (&tp_vars->unacked_list);kref_get (&tp_vars->refcount); -hlist_add_head_rcu (&tp_vars->list, &bat_priv->tp_list); +timer_setup (&tp_vars->timer, batadv_tp_receiver_shutdown,0 );kref_get (&tp_vars->refcount); -timer_setup (&tp_vars->timer, batadv_tp_receiver_shutdown,0 ); +hlist_add_head_rcu (&tp_vars->list, &bat_priv->tp_list);batadv_tp_reset_receiver_timer (tp_vars);
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 107/204 batman-adv: bla: annotate lasttime access with READ/WRITE_ONCE
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit 98b0fb191c878a64cbaebfe231d96d57576acf8c upstream.
The lasttime field for claim, backbone_gw, and loopdetect tracks the jiffies value of the most recent activity and is used to detect timeouts. These accesses are not consistently protected by a lock, so READ_ONCE/WRITE_ONCE must be used to prevent data races caused by compiler optimizations.
--- net/batman-adv/bridge_loop_avoidance.c | 28 +++++++++++++------------- 1 file changed, 14 insertions(+), 14 deletions(-)
net/batman-adv/bridge_loop_avoidance.c +14 -14
@@ -512,7 +512,7 @@ batadv_bla_get_backbone_gw(struct batadv_priv *bat_priv, const u8 *orig,return NULL ; entry->vid = vid; - entry->lasttime = jiffies; +WRITE_ONCE (entry->lasttime, jiffies); entry->crc =BATADV_BLA_CRC_INIT ; entry->bat_priv = bat_priv;spin_lock_init (&entry->crc_lock);
@@ -580,7 +580,7 @@ batadv_bla_update_own_backbone_gw(struct batadv_priv *bat_priv,if (unlikely (!backbone_gw))return ; - backbone_gw->lasttime = jiffies; +WRITE_ONCE (backbone_gw->lasttime, jiffies);batadv_backbone_gw_put (backbone_gw); }
@@ -714,7 +714,7 @@ static void batadv_bla_add_claim(struct batadv_priv *bat_priv,ether_addr_copy (claim->addr, mac);spin_lock_init (&claim->backbone_lock); claim->vid = vid; - claim->lasttime = jiffies; +WRITE_ONCE (claim->lasttime, jiffies);kref_get (&backbone_gw->refcount); claim->backbone_gw = backbone_gw;kref_init (&claim->refcount);
@@ -736,7 +736,7 @@ static void batadv_bla_add_claim(struct batadv_priv *bat_priv,return ; } }else { - claim->lasttime = jiffies; +WRITE_ONCE (claim->lasttime, jiffies);if (claim->backbone_gw == backbone_gw)/* no need to register a new backbone */ goto claim_free_ref;
@@ -769,7 +769,7 @@ static void batadv_bla_add_claim(struct batadv_priv *bat_priv,spin_lock_bh (&backbone_gw->crc_lock); backbone_gw->crc ^=crc16 (0 , claim->addr,ETH_ALEN );spin_unlock_bh (&backbone_gw->crc_lock); - backbone_gw->lasttime = jiffies; +WRITE_ONCE (backbone_gw->lasttime, jiffies); claim_free_ref:batadv_claim_put (claim);
@@ -858,7 +858,7 @@ static bool batadv_handle_announce(struct batadv_priv *bat_priv, u8 *an_addr,return true;/* handle as ANNOUNCE frame */ - backbone_gw->lasttime = jiffies; +WRITE_ONCE (backbone_gw->lasttime, jiffies); crc =ntohs (*((__force __be16 *)(&an_addr[4 ]))); batadv_dbg(BATADV_DBG_BLA, bat_priv,
@@ -1253,7 +1253,7 @@ static void batadv_bla_purge_backbone_gw(struct batadv_priv *bat_priv, int now) head, hash_entry) {if (now) goto purge_now; -if (!batadv_has_timed_out (backbone_gw->lasttime, +if (!batadv_has_timed_out (READ_ONCE (backbone_gw->lasttime),BATADV_BLA_BACKBONE_TIMEOUT ))continue ;
@@ -1334,7 +1334,7 @@ static void batadv_bla_purge_claims(struct batadv_priv *bat_priv, primary_if->net_dev->dev_addr)) goto skip; -if (!batadv_has_timed_out (claim->lasttime, +if (!batadv_has_timed_out (READ_ONCE (claim->lasttime),BATADV_BLA_CLAIM_TIMEOUT )) goto skip;
@@ -1494,7 +1494,7 @@ static void batadv_bla_periodic_work(struct work_struct *work)eth_random_addr (bat_priv->bla.loopdetect_addr); bat_priv->bla.loopdetect_addr[0 ] =0xba ; bat_priv->bla.loopdetect_addr[1 ] =0xbe ; - bat_priv->bla.loopdetect_lasttime = jiffies; +WRITE_ONCE (bat_priv->bla.loopdetect_lasttime, jiffies);atomic_set (&bat_priv->bla.loopdetect_next,BATADV_BLA_LOOPDETECT_PERIODS );
@@ -1515,7 +1515,7 @@ static void batadv_bla_periodic_work(struct work_struct *work) primary_if->net_dev->dev_addr))continue ; - backbone_gw->lasttime = jiffies; +WRITE_ONCE (backbone_gw->lasttime, jiffies);batadv_bla_send_announce (bat_priv, backbone_gw);if (send_loopdetect)
@@ -1900,7 +1900,7 @@ batadv_bla_loopdetect_check(struct batadv_priv *bat_priv, struct sk_buff *skb,/* If the packet came too late, don't forward it on the mesh * but don't consider that as loop. It might be a coincidence. */ -if (batadv_has_timed_out (bat_priv->bla.loopdetect_lasttime, +if (batadv_has_timed_out (READ_ONCE (bat_priv->bla.loopdetect_lasttime),BATADV_BLA_LOOPDETECT_TIMEOUT ))return true;
@@ -2016,7 +2016,7 @@ bool batadv_bla_rx(struct batadv_priv *bat_priv, struct sk_buff *skb,if (own_claim) {/* ... allow it in any case */ - claim->lasttime = jiffies; +WRITE_ONCE (claim->lasttime, jiffies); goto allow; }
@@ -2118,7 +2118,7 @@ bool batadv_bla_tx(struct batadv_priv *bat_priv, struct sk_buff *skb,/* if yes, the client has roamed and we have * to unclaim it. */ -if (batadv_has_timed_out (claim->lasttime,100 )) { +if (batadv_has_timed_out (READ_ONCE (claim->lasttime),100 )) {/* only unclaim if the last claim entry is * older than 100 ms to make sure we really * have a roaming client here.
@@ -2372,7 +2372,7 @@ batadv_bla_backbone_dump_entry(struct sk_buff *msg, u32 portid, backbone_crc = backbone_gw->crc;spin_unlock_bh (&backbone_gw->crc_lock); - msecs =jiffies_to_msecs (jiffies - backbone_gw->lasttime); + msecs =jiffies_to_msecs (jiffies -READ_ONCE (backbone_gw->lasttime));if (is_own)if (nla_put_flag (msg,BATADV_ATTR_BLA_OWN )) {
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 108/204 batman-adv: prevent ELP transmission interval underflow
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit 5e50d4b8ae3ea622122d3c6a38d7f6fe68dfddca upstream.
batadv_v_elp_start_timer() enqeues a delayed work. The time when it starts is randomly chosen between (elp_interval - BATADV_JITTER) and (elp_interval + BATADV_JITTER). The configured elp_interval must therefore be larger or equal to BATADV_JITTER to avoid that it causes an underflow of the unsigned integer. If this would happen, then a "fast" ELP interval would turn into a "day long" delay.
At the same time, it must not be larger than the maximum value the variable can store.
--- net/batman-adv/netlink.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
net/batman-adv/netlink.c +7 -1
@@ -936,9 +936,15 @@ static int batadv_netlink_set_hardif(struct sk_buff *skb,#ifdef CONFIG_BATMAN_ADV_BATMAN_V if (info->attrs[BATADV_ATTR_ELP_INTERVAL ]) { +u32 elp_interval; + attr = info->attrs[BATADV_ATTR_ELP_INTERVAL ]; + elp_interval =nla_get_u32 (attr); + + elp_interval =min_t (u32, elp_interval,INT_MAX ); + elp_interval =max_t (u32, elp_interval,BATADV_JITTER ); -atomic_set (&hard_iface->bat_v.elp_interval,nla_get_u32 (attr)); +atomic_set (&hard_iface->bat_v.elp_interval, elp_interval); }if (info->attrs[BATADV_ATTR_THROUGHPUT_OVERRIDE ]) {
Signature · 1 line
2.53.0
Greg Kroah-Hartman
maintainerPATCH 109/204 batman-adv: tp_meter: initialize last_recv_time during init
In reply to Greg Kroah-Hartman6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sven Eckelmann <[email protected]>
commit 811cb00fa8cdc3f0a7f6eefc000a6888367c8c8f upstream.
The last_recv_time is the most important indicator for a receiver session to figure out whether a session timed out or not. But this information was only initialized after the session was added to the tp_receiver_list and after the timer was started.
In the worst case, the timer (function) could have tried to access this information before the actual initialization was reached. Like rest of the variables of the tp_meter receiver session, this field has to be filled out before any other (parallel running) context has the chance to access it.
--- net/batman-adv/tp_meter.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
net/batman-adv/tp_meter.c +7 -3
@@ -1403,8 +1403,10 @@ batadv_tp_init_recv(struct batadv_priv *bat_priv, tp_vars =batadv_tp_list_find_session (bat_priv, icmp->orig, icmp->session,BATADV_TP_RECEIVER ); -if (tp_vars) +if (tp_vars) { + tp_vars->last_recv_time = jiffies; goto out_unlock; + }if (!atomic_add_unless (&bat_priv->tp_num,1 ,BATADV_TP_MAX_NUM )) { batadv_dbg(BATADV_DBG_TP_METER, bat_priv,
@@ -1432,6 +1434,8 @@ batadv_tp_init_recv(struct batadv_priv *bat_priv,kref_get (&tp_vars->refcount);timer_setup (&tp_vars->timer, batadv_tp_receiver_shutdown,0 ); + tp_vars->last_recv_time = jiffies; +kref_get (&tp_vars->refcount);hlist_add_head_rcu (&tp_vars->list, &bat_priv->tp_list);
@@ -1480,9 +1484,9 @@ static void batadv_tp_recv_msg(struct batadv_priv *bat_priv, icmp->orig); goto out; } - } - tp_vars->last_recv_time = jiffies; + tp_vars->last_recv_time = jiffies; + }/* if the packet is a duplicate, it may be the case that an ACK has been * lost. Resend the ACK
Signature · 1 line
2.53.0