#1057843 – linux: ext4 knowledge corruption in 6.1.64-1

Reply or subscribe to this bug.
Report forwarded
to debian-bugs-dist@lists.debian.org, debian-release@lists.debian.org, carnil@debian.org, adsb@debian.org, Debian Kernel Group <debian-kernel@lists.debian.org>
:Bug#1057843
; Package deal src:linux
.
(Sat, 09 Dec 2023 14:09:03 GMT) (full text, mbox, link).
Acknowledgement despatched
to Salvatore Bonaccorso <carnil@debian.org>
:
New Bug report acquired and forwarded. Copy despatched to debian-release@lists.debian.org, carnil@debian.org, adsb@debian.org, Debian Kernel Group <debian-kernel@lists.debian.org>
.
(Sat, 09 Dec 2023 14:09:03 GMT) (full text, mbox, link).
Message #5 acquired at submit@bugs.debian.org (full text, mbox, reply):
Supply: linux Model: 6.1.64-1 Severity: grave Tags: upstream Justification: causes non-serious knowledge loss X-Debbugs-Cc: debian-release@lists.debian.org, carnil@debian.org, adsb@debian.org Hello I am filling this for visibility. There is likely to be a ext4 knowledge corruption difficulty with the kernel launched within the 12.3 bookworm level launch (which is addressed in 6.1.66 upstream already). The report in regards to the regression and a few particulars: https://lore.kernel.org/stable/20231205122122.dfhhoaswsfscuhc3@quack3/ Regards, Salvatore
Added tag(s) confirmed.
Request was from Salvatore Bonaccorso <carnil@debian.org>
to management@bugs.debian.org
.
(Sat, 09 Dec 2023 14:27:03 GMT) (full text, mbox, link).
Info forwarded
to debian-bugs-dist@lists.debian.org, Debian Kernel Group <debian-kernel@lists.debian.org>
:Bug#1057843
; Package deal src:linux
.
(Sat, 09 Dec 2023 14:36:02 GMT) (full text, mbox, link).
Acknowledgement despatched
to Salvatore Bonaccorso <carnil@debian.org>
:
Additional information acquired and forwarded to record. Copy despatched to Debian Kernel Group <debian-kernel@lists.debian.org>
.
(Sat, 09 Dec 2023 14:36:02 GMT) (full text, mbox, link).
Message #14 acquired at 1057843@bugs.debian.org (full text, mbox, reply):
Working the only check with ext4: # LTP_SINGLE_FS_TYPE=ext4 LTP_DEV_FS_TYPE=ext4 ./preadv03_64 tst_device.c:96: TINFO: Discovered free gadget 0 '/dev/loop0' tst_test.c:1690: TINFO: LTP model: 20230929-194-g5c096b2cf tst_test.c:1574: TINFO: Timeout per run is 0h 00m 30s tst_supported_fs_types.c:149: TINFO: WARNING: testing solely ext4 tst_supported_fs_types.c:90: TINFO: Kernel helps ext4 tst_supported_fs_types.c:55: TINFO: mkfs.ext4 does exist tst_test.c:1650: TINFO: === Testing on ext4 === tst_test.c:1105: TINFO: Formatting /dev/loop0 with ext4 opts='' additional opts='' mke2fs 1.47.0 (5-Feb-2023) tst_test.c:1119: TINFO: Mounting /dev/loop0 to /tmp/LTP_preWBHd7l/mntpoint fstyp=ext4 flags=0 preadv03.c:102: TINFO: Utilizing block dimension 512 preadv03.c:77: TFAIL: Buffer mistaken at 0 have 62 anticipated 61 preadv03.c:77: TFAIL: Buffer mistaken at 0 have 62 anticipated 61 preadv03.c:66: TFAIL: preadv(O_DIRECT) learn 0 bytes, anticipated 512 Abstract: handed 0 failed 3 damaged 0 skipped 0 warnings 0
Added tag(s) fixed-upstream.
Request was from Salvatore Bonaccorso <carnil@debian.org>
to management@bugs.debian.org
.
(Sat, 09 Dec 2023 14:39:03 GMT) (full text, mbox, link).
Info forwarded
to debian-bugs-dist@lists.debian.org, Debian Kernel Group <debian-kernel@lists.debian.org>
:Bug#1057843
; Package deal src:linux
.
(Sat, 09 Dec 2023 14:57:02 GMT) (full text, mbox, link).
Acknowledgement despatched
to Salvatore Bonaccorso <carnil@debian.org>
:
Additional information acquired and forwarded to record. Copy despatched to Debian Kernel Group <debian-kernel@lists.debian.org>
.
(Sat, 09 Dec 2023 14:57:02 GMT) (full text, mbox, link).
Message #21 acquired at 1057843@bugs.debian.org (full text, mbox, reply):
Hello, On Sat, Dec 09, 2023 at 03:07:37PM +0100, Salvatore Bonaccorso wrote: > Supply: linux > Model: 6.1.64-1 > Severity: grave > Tags: upstream > Justification: causes non-serious knowledge loss > X-Debbugs-Cc: debian-release@lists.debian.org, carnil@debian.org, adsb@debian.org > > Hello > > I am filling this for visibility. > > There is likely to be a ext4 knowledge corruption difficulty with the kernel launched > within the 12.3 bookworm level launch (which is addressed in 6.1.66 > upstream already). > > The report in regards to the regression and a few particulars: > > https://lore.kernel.org/stable/20231205122122.dfhhoaswsfscuhc3@quack3/ 6.1.66 upstream fixes the difficulty: # uname -a Linux bookworm-amd64 6.1.0-15-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.66-1 (2023-12-06) x86_64 GNU/Linux # LTP_SINGLE_FS_TYPE=ext4 LTP_DEV_FS_TYPE=ext4 ./preadv03_64 tst_device.c:96: TINFO: Discovered free gadget 0 '/dev/loop0' tst_test.c:1690: TINFO: LTP model: 20230929-194-g5c096b2cf tst_test.c:1574: TINFO: Timeout per run is 0h 00m 30s tst_supported_fs_types.c:149: TINFO: WARNING: testing solely ext4 tst_supported_fs_types.c:90: TINFO: Kernel helps ext4 tst_supported_fs_types.c:55: TINFO: mkfs.ext4 does exist tst_test.c:1650: TINFO: === Testing on ext4 === tst_test.c:1105: TINFO: Formatting /dev/loop0 with ext4 opts='' additional opts='' mke2fs 1.47.0 (5-Feb-2023) tst_test.c:1119: TINFO: Mounting /dev/loop0 to /tmp/LTP_preGGYjTj/mntpoint fstyp=ext4 flags=0 preadv03.c:102: TINFO: Utilizing block dimension 512 preadv03.c:87: TPASS: preadv(O_DIRECT) learn 512 bytes efficiently with content material 'a' expectedly preadv03.c:87: TPASS: preadv(O_DIRECT) learn 512 bytes efficiently with content material 'a' expectedly preadv03.c:87: TPASS: preadv(O_DIRECT) learn 512 bytes efficiently with content material 'b' expectedly Abstract: handed 3 failed 0 damaged 0 skipped 0 warnings 0 Regards, Salvatore
Added tag(s) pending.
Request was from Salvatore Bonaccorso <carnil@debian.org>
to management@bugs.debian.org
.
(Sat, 09 Dec 2023 16:27:03 GMT) (full text, mbox, link).
Reply despatched
to Salvatore Bonaccorso <carnil@debian.org>
:
You will have taken duty.
(Sat, 09 Dec 2023 18:00:08 GMT) (full text, mbox, link).
Notification despatched
to Salvatore Bonaccorso <carnil@debian.org>
:
Bug acknowledged by developer.
(Sat, 09 Dec 2023 18:00:08 GMT) (full text, mbox, link).
Message #28 acquired at 1057843-close@bugs.debian.org (full text, mbox, reply):
Supply: linux Supply-Model: 6.1.66-1 Achieved: Salvatore Bonaccorso <carnil@debian.org> We imagine that the bug you reported is mounted within the newest model of linux, which is because of be put in within the Debian FTP archive. A abstract of the modifications between this model and the earlier one is connected. Thanks for reporting the bug, which is able to now be closed. Should you have additional feedback please deal with them to 1057843@bugs.debian.org, and the maintainer will reopen the bug report if acceptable. Debian distribution upkeep software program pp. Salvatore Bonaccorso <carnil@debian.org> (provider of up to date linux package deal) (This message was generated robotically at their request; should you imagine that there's a drawback with it please contact the archive directors by mailing ftpmaster@ftp-master.debian.org) -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Format: 1.8 Date: Sat, 09 Dec 2023 16:48:39 +0100 Supply: linux Structure: supply Model: 6.1.66-1 Distribution: bookworm Urgency: medium Maintainer: Debian Kernel Group <debian-kernel@lists.debian.org> Modified-By: Salvatore Bonaccorso <carnil@debian.org> Closes: 1032104 1057790 1057843 Modifications: linux (6.1.66-1) bookworm; urgency=medium . * New upstream steady replace: https://www.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.65 - afs: Repair afs_server_list to be cleaned up with RCU - afs: Make error on cell lookup failure according to OpenAFS - [arm64,armhf] drm/panel: easy: Repair Innolux G101ICE-L01 bus flags - [arm64,armhf] drm/panel: easy: Repair Innolux G101ICE-L01 timings - wireguard: use DEV_STATS_INC() - ata: pata_isapnp: Add lacking error examine for devm_ioport_map() - [x86] drm/i915: don't clear GT desk on error path - [arm64,armhf] drm/rockchip: vop: Repair coloration for RGB888/BGR888 format on VOP full - HID: repair HID gadget useful resource race between HID core and debugging help - ipv4: Right/silence an endian warning in __ip_do_redirect - internet: usb: ax88179_178a: repair failed operations throughout ax88179_reset - internet/smc: keep away from knowledge corruption attributable to decline - [armhf] arm/xen: repair xen_vcpu_info allocation alignment - [amd64,arm64] amd-xgbe: deal with corner-case throughout sfp hotplug - [amd64,arm64] amd-xgbe: deal with the corner-case throughout tx completion - [amd64,arm64] amd-xgbe: propagate the right pace and duplex standing - afs: Return ENOENT if no cell DNS document will be discovered - afs: Repair file locking on R/O volumes to function in native mode - mm,kfence: decouple kfence from web page granularity mapping judgement - i40e: use ERR_PTR error print in i40e messages - i40e: Repair including unsupported cloud filters - [arm64] USB: dwc3: qcom: repair useful resource leaks on probe deferral - [arm64] USB: dwc3: qcom: repair ACPI platform gadget leak - lockdep: Repair block chain corruption - cifs: minor cleanup of some headers - smb3: permit dumping session and tcon id to enhance stats evaluation and debugging - cifs: print final replace time for interface record - cifs: distribute channels throughout interfaces based mostly on pace - cifs: account for major channel within the interface record - cifs: repair leak of iface for major channel - ext4: add a brand new helper to examine if es should be stored - ext4: issue out __es_alloc_extent() and __es_free_extent() - ext4: use pre-allocated es in __es_insert_extent() - ext4: use pre-allocated es in __es_remove_extent() - ext4: utilizing nofail preallocation in ext4_es_remove_extent() - ext4: utilizing nofail preallocation in ext4_es_insert_delayed_block() - ext4: utilizing nofail preallocation in ext4_es_insert_extent() - ext4: repair slab-use-after-free in ext4_es_insert_extent() - ext4: ensure that allocate pending entry not fail - NFSD: Repair "begin of NFS reply" pointer handed to nfsd_cache_update() - NFSD: Repair checksum mismatches within the duplicate reply cache - ACPI: useful resource: Skip IRQ override on ASUS ExpertBook B1402CVA - swiotlb-xen: present the "max_mapping_size" technique - bcache: substitute a mistaken IS_ERR() by IS_ERR_OR_NULL() in btree_gc_coalesce() - md: repair bi_status reporting in md_end_clone_io - bcache: fixup multi-threaded bch_sectors_dirty_init() wake-up race - io_uring/fs: contemplate link->flags when getting path for LINKAT - [s390x] dasd: defend gadget queue in opposition to concurrent entry - USB: serial: choice: add Luat Air72*U sequence merchandise - hv_netvsc: repair race of netvsc and VF register_netdevice - hv_netvsc: Repair race of register_netdevice_notifier and VF register - hv_netvsc: Mark VF as slave earlier than exposing it to user-mode - dm-delay: repair a race between delay_presuspend and delay_bio - bcache: examine return worth from btree_node_alloc_replacement() - bcache: stop potential division by zero error - bcache: fixup init soiled knowledge errors - bcache: fixup lock c->root error - USB: serial: choice: add Fibocom L7xx modules - USB: serial: choice: repair FM101R-GL defines - USB: serial: choice: do not declare interface 4 for ZTE MF290 - usb: typec: tcpm: Skip laborious reset when in error restoration - [arm64,armhf] USB: dwc2: write HCINT with INTMASK utilized - [arm64,armhf] usb: dwc3: Repair default mode initialization - [arm64,armhf] usb: dwc3: set the dma max_seg_size - [arm64] USB: dwc3: qcom: repair software program node leak on probe errors - [arm64] USB: dwc3: qcom: repair wakeup after probe deferral - io_uring: repair off-by one bvec index https://www.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.66 - cifs: Repair FALLOC_FL_ZERO_RANGE by setting i_size if EOF moved - cifs: Repair FALLOC_FL_INSERT_RANGE by setting i_size after EOF moved - smb: shopper: report right st_size for SMB and NFS symlinks - pinctrl: keep away from reload of p state in record iteration - firewire: core: repair attainable reminiscence leak in create_units() - mmc: sdhci-pci-gli: Disable LPM throughout initialization - mmc: cqhci: Improve restoration halt timeout - mmc: cqhci: Warn of halt or job clear failure - mmc: cqhci: Repair job clearing in CQE error restoration - mmc: block: Retry instructions in CQE error restoration - mmc: block: Don't lose cache flush throughout CQE error restoration - mmc: block: You should definitely wait whereas busy in CQE error restoration - ALSA: hda: Disable power-save on KONTRON SinglePC - ALSA: hda/realtek: Headset Mic VREF to 100% - ALSA: hda/realtek: Add supported ALC257 for ChromeOS - dm-verity: align struct dm_verity_fec_io correctly - scsi: Change SCSI gadget boolean fields to single bit flags - scsi: sd: Repair system begin for ATA units - drm/amd: Allow PCIe PME from D3 - drm/amdgpu: Pressure order between a learn and write to the identical deal with - drm/amd/show: Embody udelay when ready for INBOX0 ACK - drm/amd/show: Take away min_dst_y_next_start examine for Z8 - drm/amd/show: Use DRAM pace from validation for dummy p-state - drm/amd/show: Replace min Z8 residency time to 2100 for DCN314 - drm/amd/show: repair ABM disablement - dm verity: initialize fec io earlier than liberating it - dm verity: do not carry out FEC for failed readahead IO - nvme: examine for legitimate nvme_identify_ns() earlier than utilizing it - [x86] cpufreq/amd-pstate: Repair the return worth of amd_pstate_fast_switch() - dma-buf: repair examine in dma_resv_add_fence - bcache: revert changing IS_ERR_OR_NULL with IS_ERR - [amd64] iommu/vt-d: Add MTL to quirk record to skip TE disabling - [powerpc*] KVM: PPC: Book3S HV: Repair KVM_RUN clobbering FP/VEC consumer registers - [powerpc*] Do not clobber f0/vs0 throughout fp|altivec register save (Closes: #1032104) - btrfs: add dmesg output for first mount and final unmount of a filesystem - btrfs: ref-verify: repair reminiscence leaks in btrfs_ref_tree_mod() - btrfs: repair off-by-one when checking chunk map consists of logical deal with - btrfs: ship: guarantee send_fd is writable - btrfs: make error messages extra clear when getting a bit map - btrfs: repair 64bit compat ship ioctl arguments not initializing model member - Enter: xpad - add HyperX Clutch Gladiate Help - wifi: cfg80211: repair CQM for non-range use - [arm64,armhf] USB: xhci-plat: repair legacy PHY double init - USB: core: Change configuration warnings to notices - usb: config: repair iteration difficulty in 'usb_get_bos_descriptor()' - ipv4: igmp: repair refcnt uaf difficulty when receiving igmp question packet - [arm64] dpaa2-eth: improve the wanted headroom to account for alignment - internet: stmmac: xgmac: Disable FPE MMC interrupts - r8169: stop potential impasse in rtl8169_close - [x86] KVM: x86: Repair lapic timer interrupt misplaced after loading a snapshot. - PCI: Lengthen reset delay for VideoPropulsion Torrent QN16e card - spi: Repair null dereference on droop - drm/amd/show: Restore rptr/wptr for DMCUB as workaround - drm/amd/show: Guard in opposition to invalid RPTR/WPTR being set - [armhf] cpufreq: imx6q: do not warn for disabling a non-existing frequency - [armhf] cpufreq: imx6q: Do not disable 792 Mhz OPP unnecessarily - [amd64] iommu/vt-d: Omit devTLB invalidation requests when TES=0 - [amd64] iommu/vt-d: Allocate pasid desk in gadget probe path - [amd64] iommu/vt-d: Add device_block_translation() helper - [amd64] iommu/vt-d: Disable PCI ATS in legacy passthrough mode - [amd64] iommu/vt-d: Make context clearing according to context mapping - drm/amd/pm: repair a memleak in aldebaran_tables_init - mmc: core: add helpers mmc_regulator_enable/disable_vqmmc - mmc: sdhci-sprd: Repair vqmmc not shutting down after the cardboard was pulled - drm/amd/show: Increase kernel doc for DC - drm/amd/show: clear code-style points in dcn30_set_mpc_shaper_3dlut - drm/amd/show: Repair the delta clamping for shaper LUT - drm/amd/show: Repair MPCC 1DLUT programming - r8169: disable ASPM in case of tx timeout - r8169: repair impasse on RTL8125 in jumbo mtu mode (Closes: #1057790) - xen: Permit platform PCI interrupt to be shared - xen: simplify evtchn_do_upcall() name maze - [x86] xen: repair percpu vcpu_info allocation - [x86] apic/msi: Repair misconfigured non-maskable MSI quirk - iomap: replace ki_pos a bit of later in iomap_dio_complete (Closes: #1057843) . [ Salvatore Bonaccorso ] * Bump ABI to fifteen * [rt] Replace to six.1.66-rt19 Checksums-Sha1: 75d087cfc109545360cb85628ca2068a656f58bf 290924 linux_6.1.66-1.dsc bec397772676a6f06e2fde4c2328bcdc0beaea06 137491816 linux_6.1.66.orig.tar.xz af301e107224490faeda530343f96f17c0684f71 1580256 linux_6.1.66-1.debian.tar.xz 348664cfbc75b567cf6256557f0c3d09fc8f7997 7065 linux_6.1.66-1_source.buildinfo Checksums-Sha256: 769e7dae9fe82bc29b55eb606fda6b0cc6221294a2ba3d7e7fcc78de84683f61 290924 linux_6.1.66-1.dsc 71edfafca7cd8eaae26a8d29c1091bd39cfc2d196de38570b159c94f0d3e1ad5 137491816 linux_6.1.66.orig.tar.xz 907af29ab9ff38594df378ba5c60619110389b92ba2a09a21adc0382f76e0240 1580256 linux_6.1.66-1.debian.tar.xz 342dddfc2e87c8e8211e130e19253e23c1f5952036f29b4c9a7f9273c074d94a 7065 linux_6.1.66-1_source.buildinfo Recordsdata: 505a701b509c26df837dc4a6ff0f0844 290924 kernel elective linux_6.1.66-1.dsc ea8af00da46c0f6392324e83554102a2 137491816 kernel elective linux_6.1.66.orig.tar.xz 307f584ef754457e172ad498979cba6c 1580256 kernel elective linux_6.1.66-1.debian.tar.xz 2f1f205c788b2a730a7e7a89145dbf7d 7065 kernel elective linux_6.1.66-1_source.buildinfo -----BEGIN PGP SIGNATURE----- iQKmBAEBCgCQFiEERkRAmAjBceBVMd3uBUy48xNDz0QFAmV0kZ5fFIAAAAAALgAo aXNzdWVyLWZwckBub3RhdGlvbnMub3BlbnBncC5maWZ0aGhvcnNlbWFuLm5ldDQ2 NDQ0MDk4MDhDMTcxRTA1NTMxRERFRTA1NENCOEYzMTM0M0NGNDQSHGNhcm5pbEBk ZWJpYW4ub3JnAAoJEAVMuPMTQ89ErmUP/RZZN1ZMRy2nqQWPqXDrMUci6fOWcUCj ZPAH/B9D9/+FYyRylJz8NENzJj82ZfNAUdKNMG0odp85pe6UOwvi6qMVIrsM3UrB AQEbzEA0CS8XSTD2qdmOe2EaWHi0LvnopIQRGFy5Z3auQOk+EADbibRi2qZjM0Te +Y1++jfXXLjbY5o+2oJMzaxZLBYRBcQizWfRKWjX4tmwcAZppABFShB5XEBlbeKx 8TZ6HhacpW3UImc1CTi8wFeO3T27bn4HQ9mLeW5vLh8bzbB3wnzmFzOn2wJj8Qf8 lRe6+oolFewZ57hBJjvDoK4CwQ9hapwPMJpAaG1NpE4QXIAAapb3OD5DP8vpsoTQ bORrs3kvDRCA1ooCmGaEW7JBlolLl/mMxrIR4yGMzkXnA37UmAJ/PfmAJJO81J+H cAhHdicTGrs6KQfNBBQzHC9vBJZNJ3Z5PpAXUncGQUDRueEz9kJXdwzZDkOf5DCC JHrBXn7EZQLANXTeCg0za6/29VwURbSmUWXLk914sHb4xEPO/ivOpZAT8baB0JTV hF8J6rzTyqsbv7gOFgOgrCCCvh5XU9Nvl7W3mRIHwiCdBKr39goL2V1S08kt0bmS RLW+46T5U3Flcr5UGoIoEoBsav16aaHfaQ7mx182GGoAJqUf6tM4PuC4mDt1W4pQ 4o6djxs85N9S =gQo4 -----END PGP SIGNATURE-----
Info forwarded
to debian-bugs-dist@lists.debian.org, Debian Kernel Group <debian-kernel@lists.debian.org>
:Bug#1057843
; Package deal src:linux
.
(Solar, 10 Dec 2023 08:00:05 GMT) (full text, mbox, link).
Acknowledgement despatched
to Stephan Verbücheln <verbuecheln@posteo.de>
:
Additional information acquired and forwarded to record. Copy despatched to Debian Kernel Group <debian-kernel@lists.debian.org>
.
(Solar, 10 Dec 2023 08:00:05 GMT) (full text, mbox, link).
Message #33 acquired at 1057843@bugs.debian.org (full text, mbox, reply):
Are there any tips for affected customers who already up to date earlier than they bought the warning? Fascinating questions for affected customers: - Is it protected to imagine that different filesystems (like BTRFS) usually are not affected? - Does this trigger filesystem corruption or solely file corruption? - Does this have an effect on metadata or solely file contents? - Is there a method to detect corrupted information? - If metadata shouldn't be affected, is it protected to imagine that each one information with a modification date older than the replace are high-quality? - Does it assist to close down providers (comparable to Apache) or the entire machine till the repair is obtainable?
Ship a report that this bug log contains spam.
Debian bug monitoring system administrator <owner@bugs.debian.org>.
Final modified:
Solar Dec 10 08:17:21 2023;
Machine Title:
bembo
Debbugs is free software program and licensed underneath the phrases of the GNU
Public License model 2. The present model will be obtained
from https://bugs.debian.org/debbugs-source/.
Copyright © 1999 Darren O. Benham,
1997,2003 nCipher Company Ltd,
1994-97 Ian Jackson,
2005-2017 Don Armstrong, and lots of different contributors.