aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/tools/perf/scripts/python/export-to-sqlite.py (unfollow)
AgeCommit message (Collapse)AuthorFilesLines
2024-09-23fs/mnt_idmapping: introduce an invalid_mnt_idmapAlexander Mikhalitsyn2-2/+21
Link: https://lore.kernel.org/linux-fsdevel/20240904-baugrube-erhoben-b3c1c49a2645@brauner/ Suggested-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-23fs/fuse: introduce and use fuse_simple_idmap_request() helperAlexander Mikhalitsyn9-46/+60
Let's convert all existing callers properly. No functional changes intended. Suggested-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-23fs/fuse: fix null-ptr-deref when checking SB_I_NOIDMAP flagAlexander Mikhalitsyn1-2/+2
It was reported [1] that on linux-next/fs-next the following crash is reproducible: [ 42.659136] Oops: general protection fault, probably for non-canonical address 0xdffffc000000000b: 0000 [#1] PREEMPT SMP KASAN NOPTI [ 42.660501] fbcon: Taking over console [ 42.660930] KASAN: null-ptr-deref in range [0x0000000000000058-0x000000000000005f] [ 42.661752] CPU: 1 UID: 0 PID: 1589 Comm: dtprobed Not tainted 6.11.0-rc6+ #1 [ 42.662565] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.6.6 08/22/2023 [ 42.663472] RIP: 0010:fuse_get_req+0x36b/0x990 [fuse] [ 42.664046] Code: 48 89 fa 48 c1 ea 03 80 3c 02 00 0f 85 8c 05 00 00 48 b8 00 00 00 00 00 fc ff df 48 8b 6d 08 48 8d 7d 58 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 4d 05 00 00 f6 45 59 20 0f 85 06 03 00 00 48 83 [ 42.666945] RSP: 0018:ffffc900009a7730 EFLAGS: 00010212 [ 42.668837] RAX: dffffc0000000000 RBX: 1ffff92000134eed RCX: ffffffffc20dec9a [ 42.670122] RDX: 000000000000000b RSI: 0000000000000008 RDI: 0000000000000058 [ 42.672154] RBP: 0000000000000000 R08: 0000000000000001 R09: ffffed1022110172 [ 42.672160] R10: ffff888110880b97 R11: ffffc900009a737a R12: 0000000000000001 [ 42.672179] R13: ffff888110880b60 R14: ffff888110880b90 R15: ffff888169973840 [ 42.672186] FS: 00007f28cd21d7c0(0000) GS:ffff8883ef280000(0000) knlGS:0000000000000000 [ 42.672191] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 42.[ CR02: ;32m00007f3237366208 CR3: 0 OK 79e001 CR4: 0000000000770ef0 [ 42.672214] PKRU: 55555554 [ 42.672218] Call Trace: [ 42.672223] <TASK> [ 42.672226] ? die_addr+0x41/0xa0 [ 42.672238] ? exc_general_protection+0x14c/0x230 [ 42.672250] ? asm_exc_general_protection+0x26/0x30 [ 42.672260] ? fuse_get_req+0x77a/0x990 [fuse] [ 42.672281] ? fuse_get_req+0x36b/0x990 [fuse] [ 42.672300] ? kasan_unpoison+0x27/0x60 [ 42.672310] ? __pfx_fuse_get_req+0x10/0x10 [fuse] [ 42.672327] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.672333] ? alloc_pages_mpol_noprof+0x195/0x440 [ 42.672340] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.672345] ? kasan_unpoison+0x27/0x60 [ 42.672350] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.672355] ? __kasan_slab_alloc+0x4d/0x90 [ 42.672362] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.672367] ? __kmalloc_cache_noprof+0x134/0x350 [ 42.672376] fuse_simple_background+0xe7/0x180 [fuse] [ 42.672406] cuse_channel_open+0x540/0x710 [cuse] [ 42.672415] misc_open+0x2a7/0x3a0 [ 42.672424] chrdev_open+0x1ef/0x5f0 [ 42.672432] ? __pfx_chrdev_open+0x10/0x10 [ 42.672439] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.672443] ? security_file_open+0x3bb/0x720 [ 42.672451] do_dentry_open+0x43d/0x1200 [ 42.672459] ? __pfx_chrdev_open+0x10/0x10 [ 42.672468] vfs_open+0x79/0x340 [ 42.672475] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.672482] do_open+0x68c/0x11e0 [ 42.672489] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.672495] ? __pfx_do_open+0x10/0x10 [ 42.672501] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.672506] ? open_last_lookups+0x2a2/0x1370 [ 42.672515] path_openat+0x24f/0x640 [ 42.672522] ? __pfx_path_openat+0x10/0x10 [ 42.723972] ? stack_depot_save_flags+0x45/0x4b0 [ 42.724787] ? __fput+0x43c/0xa70 [ 42.725100] do_filp_open+0x1b3/0x3e0 [ 42.725710] ? poison_slab_object+0x10d/0x190 [ 42.726145] ? __kasan_slab_free+0x33/0x50 [ 42.726570] ? __pfx_do_filp_open+0x10/0x10 [ 42.726981] ? do_syscall_64+0x64/0x170 [ 42.727418] ? entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 42.728018] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.728505] ? do_raw_spin_lock+0x131/0x270 [ 42.728922] ? __pfx_do_raw_spin_lock+0x10/0x10 [ 42.729494] ? do_raw_spin_unlock+0x14c/0x1f0 [ 42.729992] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.730889] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.732178] ? alloc_fd+0x176/0x5e0 [ 42.732585] do_sys_openat2+0x122/0x160 [ 42.732929] ? __pfx_do_sys_openat2+0x10/0x10 [ 42.733448] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.734013] ? __pfx_map_id_up+0x10/0x10 [ 42.734482] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.735529] ? __memcg_slab_free_hook+0x292/0x500 [ 42.736131] __x64_sys_openat+0x123/0x1e0 [ 42.736526] ? __pfx___x64_sys_openat+0x10/0x10 [ 42.737369] ? __x64_sys_close+0x7c/0xd0 [ 42.737717] ? srso_alias_return_thunk+0x5/0xfbef5 [ 42.738192] ? syscall_trace_enter+0x11e/0x1b0 [ 42.738739] do_syscall_64+0x64/0x170 [ 42.739113] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 42.739638] RIP: 0033:0x7f28cd13e87b [ 42.740038] Code: 25 00 00 41 00 3d 00 00 41 00 74 4b 64 8b 04 25 18 00 00 00 85 c0 75 67 44 89 e2 48 89 ee bf 9c ff ff ff b8 01 01 00 00 0f 05 <48> 3d 00 f0 ff ff 0f 87 91 00 00 00 48 8b 54 24 28 64 48 2b 14 25 [ 42.741943] RSP: 002b:00007ffc992546c0 EFLAGS: 00000246 ORIG_RAX: 0000000000000101 [ 42.742951] RAX: ffffffffffffffda RBX: 00007f28cd44f1ee RCX: 00007f28cd13e87b [ 42.743660] RDX: 0000000000000002 RSI: 00007f28cd44f2fa RDI: 00000000ffffff9c [ 42.744518] RBP: 00007f28cd44f2fa R08: 0000000000000000 R09: 0000000000000001 [ 42.745211] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000002 [ 42.745920] R13: 00007f28cd44f2fa R14: 0000000000000000 R15: 0000000000000003 [ 42.746708] </TASK> [ 42.746937] Modules linked in: cuse vfat fat ext4 mbcache jbd2 intel_rapl_msr intel_rapl_common kvm_amd ccp bochs drm_vram_helper kvm drm_ttm_helper ttm pcspkr i2c_piix4 drm_kms_helper i2c_smbus pvpanic_mmio pvpanic joydev sch_fq_codel drm fuse xfs nvme_tcp nvme_fabrics nvme_core sd_mod sg virtio_net net_failover virtio_scsi failover crct10dif_pclmul crc32_pclmul ata_generic pata_acpi ata_piix ghash_clmulni_intel virtio_pci sha512_ssse3 virtio_pci_legacy_dev sha256_ssse3 virtio_pci_modern_dev sha1_ssse3 libata serio_raw dm_multipath btrfs blake2b_generic xor zstd_compress raid6_pq sunrpc dm_mirror dm_region_hash dm_log dm_mod be2iscsi bnx2i cnic uio cxgb4i cxgb4 tls cxgb3i cxgb3 mdio libcxgbi libcxgb qla4xxx iscsi_boot_sysfs iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi qemu_fw_cfg aesni_intel crypto_simd cryptd [ 42.754333] ---[ end trace 0000000000000000 ]--- [ 42.756899] RIP: 0010:fuse_get_req+0x36b/0x990 [fuse] [ 42.757851] Code: 48 89 fa 48 c1 ea 03 80 3c 02 00 0f 85 8c 05 00 00 48 b8 00 00 00 00 00 fc ff df 48 8b 6d 08 48 8d 7d 58 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 4d 05 00 00 f6 45 59 20 0f 85 06 03 00 00 48 83 [ 42.760334] RSP: 0018:ffffc900009a7730 EFLAGS: 00010212 [ 42.760940] RAX: dffffc0000000000 RBX: 1ffff92000134eed RCX: ffffffffc20dec9a [ 42.761697] RDX: 000000000000000b RSI: 0000000000000008 RDI: 0000000000000058 [ 42.763009] RBP: 0000000000000000 R08: 0000000000000001 R09: ffffed1022110172 [ 42.763920] R10: ffff888110880b97 R11: ffffc900009a737a R12: 0000000000000001 [ 42.764839] R13: ffff888110880b60 R14: ffff888110880b90 R15: ffff888169973840 [ 42.765716] FS: 00007f28cd21d7c0(0000) GS:ffff8883ef280000(0000) knlGS:0000000000000000 [ 42.766890] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 42.767828] CR2: 00007f3237366208 CR3: 000000012c79e001 CR4: 0000000000770ef0 [ 42.768730] PKRU: 55555554 [ 42.769022] Kernel panic - not syncing: Fatal exception [ 42.770758] Kernel Offset: 0x7200000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff) [ 42.771947] ---[ end Kernel panic - not syncing: Fatal exception ]--- It's obviously CUSE related callstack. For CUSE case, we don't have superblock and our checks for SB_I_NOIDMAP flag does not make any sense. Let's handle this case gracefully. Fixes: aa16880d9f13 ("fuse: add basic infrastructure to support idmappings") Link: https://lore.kernel.org/linux-next/87v7z586py.fsf@debian-BULLSEYE-live-builder-AMD64/ [1] Reported-by: Chandan Babu R <chandanbabu@kernel.org> Reported-by: syzbot+20c7e20cc8f5296dca12@syzkaller.appspotmail.com Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-19fuse: allow O_PATH fd for FUSE_DEV_IOC_BACKING_OPENMiklos Szeredi1-5/+2
Only f_path is used from backing files registered with FUSE_DEV_IOC_BACKING_OPEN, so it makes sense to allow O_PATH descriptors. O_PATH files have an empty f_op, so don't check read_iter/write_iter. Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-18virtio_fs: allow idmapped mountsAlexander Mikhalitsyn1-0/+1
Allow idmapped mounts for virtiofs. It's absolutely safe as for virtiofs we have the same feature negotiation mechanism as for classical fuse filesystems. This does not affect any existing setups anyhow. virtiofsd support: https://gitlab.com/virtio-fs/virtiofsd/-/merge_requests/245 Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: allow idmapped mountsAlexander Mikhalitsyn2-4/+28
Now we have everything in place and we can allow idmapped mounts by setting the FS_ALLOW_IDMAP flag. Notice that real availability of idmapped mounts will depend on the fuse daemon. Fuse daemon have to set FUSE_ALLOW_IDMAP flag in the FUSE_INIT reply. To discuss: - we enable idmapped mounts support only if "default_permissions" mode is enabled, because otherwise we would need to deal with UID/GID mappings in the userspace side OR provide the userspace with idmapped req->in.h.uid/req->in.h.gid values which is not something that we probably want to. Idmapped mounts philosophy is not about faking caller uid/gid. Some extra links and examples: - libfuse support https://github.com/mihalicyn/libfuse/commits/idmap_support - fuse-overlayfs support: https://github.com/mihalicyn/fuse-overlayfs/commits/idmap_support - cephfs-fuse conversion example https://github.com/mihalicyn/ceph/commits/fuse_idmap - glusterfs conversion example https://github.com/mihalicyn/glusterfs/commits/fuse_idmap Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: warn if fuse_access is called when idmapped mounts are allowedAlexander Mikhalitsyn1-0/+8
It is not possible with the current fuse code, but let's protect ourselves from regressions in the future. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: handle idmappings properly in ->write_iter()Alexander Mikhalitsyn1-1/+2
This is needed to properly clear suid/sgid. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: support idmapped ->rename opAlexander Mikhalitsyn1-4/+5
RENAME_WHITEOUT is a special case of ->rename and we need to take idmappings into account there. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: support idmapped ->set_aclAlexander Mikhalitsyn1-2/+2
It's just a matter of adjusting a permission check condition for S_ISGID flag. All the rest is already handled in the generic VFS code. Notice that this permission check is the analog of what we have in posix_acl_update_mode() generic helper, but fuse doesn't use this helper as on the kernel side we don't care about ensuring that POSIX ACL and CHMOD permissions are in sync as it is a responsibility of a userspace daemon to handle that. For the same reason we don't have a calls to posix_acl_chmod(), while most of other filesystem do. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: drop idmap argument from __fuse_get_aclAlexander Mikhalitsyn1-4/+2
We don't need to have idmap in the __fuse_get_acl as we don't have any use for it. In the current POSIX ACL implementation, idmapped mounts are taken into account on the userspace/kernel border (see vfs_set_acl_idmapped_mnt() and vfs_posix_acl_to_xattr()). Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: support idmapped ->setattr opAlexander Mikhalitsyn3-14/+26
Need to translate uid and gid in case of chown(2). Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: support idmapped ->permission inode opAlexander Mikhalitsyn1-2/+2
We only cover the case when "default_permissions" flag is used. A reason for that is that otherwise all the permission checks are done in the userspace and we have to deal with VFS idmapping in the userspace (which is bad), alternatively we have to provide the userspace with idmapped req->in.h.uid/req->in.h.gid which is also not align with VFS idmaps philosophy. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: support idmapped getattr inode opAlexander Mikhalitsyn1-20/+24
We have to: - pass an idmapping to the generic_fillattr() to properly handle UIG/GID mapping for the userspace. - pass -/- to fuse_fillattr() (analog of generic_fillattr() in fuse). Difference between these two is that generic_fillattr() takes all the stat() data from the inode directly, while fuse_fillattr() codepath takes a fresh data just from the userspace reply on the FUSE_GETATTR request. In some cases we can just pass &nop_mnt_idmap, because idmapping won't be used in these codepaths. For example, when 3rd argument of fuse_do_getattr() is NULL then idmap argument is not used. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: support idmap for mkdir/mknod/symlink/create/tmpfileAlexander Mikhalitsyn1-18/+20
We have all the infrastructure in place, we just need to pass an idmapping here. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: support idmapped FUSE_EXT_GROUPSAlexander Mikhalitsyn1-7/+12
We don't need to remap parent_gid, but have to adjust group membership checks and take idmapping into account. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: add an idmap argument to fuse_simple_requestAlexander Mikhalitsyn9-44/+47
If idmap == NULL *and* filesystem daemon declared idmapped mounts support, then uid/gid values in a fuse header will be -1. No functional changes intended. Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04fuse: add basic infrastructure to support idmappingsAlexander Mikhalitsyn3-12/+41
Add some preparational changes in fuse_get_req/fuse_force_creds to handle idmappings. Miklos suggested [1], [2] to change the meaning of in.h.uid/in.h.gid fields when daemon declares support for idmapped mounts. In a new semantic, we fill uid/gid values in fuse header with a id-mapped caller uid/gid (for requests which create new inodes), for all the rest cases we just send -1 to userspace. No functional changes intended. Link: https://lore.kernel.org/all/CAJfpegsVY97_5mHSc06mSw79FehFWtoXT=hhTUK_E-Yhr7OAuQ@mail.gmail.com/ [1] Link: https://lore.kernel.org/all/CAJfpegtHQsEUuFq1k4ZbTD3E1h-GsrN3PWyv7X8cg6sfU_W2Yw@mail.gmail.com/ [2] Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-09-04namespace: introduce SB_I_NOIDMAP flagAlexander Mikhalitsyn2-0/+5
Right now we determine if filesystem support vfs idmappings or not basing on the FS_ALLOW_IDMAP flag presence. This "static" way works perfecly well for local filesystems like ext4, xfs, btrfs, etc. But for network-like filesystems like fuse, cephfs this approach is not ideal, because sometimes proper support of vfs idmaps requires some extensions for the on-wire protocol, which implies that changes have to be made not only in the Linux kernel code but also in the 3rd party components like libfuse, cephfs MDS server and so on. We have seen that issue during our work on cephfs idmapped mounts [1] with Christian, but right now I'm working on the idmapped mounts support for fuse/virtiofs and I think that it is a right time for this extension. [1] 5ccd8530dd7 ("ceph: handle idmapped mounts in create_request_message()") Suggested-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Reviewed-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: use correct name fuse_conn_list in docstringAurelien Aptel1-1/+1
fuse_mount_list doesn't exist, use fuse_conn_list. Signed-off-by: Aurelien Aptel <aaptel@nvidia.com> Reviewed-by: Bernd Schubert <bschubert@ddn.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: add simple request tracepointsJosef Bacik3-0/+140
I've been timing various fuse operations and it's quite annoying to do with kprobes. Add two tracepoints for sending and ending fuse requests to make it easier to debug and time various operations. Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: Bernd Schubert <bschubert@ddn.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: refactor out shared logic in fuse_writepages_fill() and fuse_writepage_locked()Joanne Koong1-46/+57
This change refactors the shared logic in fuse_writepages_fill() and fuse_writepages_locked() into two separate helper functions, fuse_writepage_args_page_fill() and fuse_writepage_args_setup(). No functional changes added. Signed-off-by: Joanne Koong <joannelkoong@gmail.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: move fuse file initialization to wpa allocation timeJoanne Koong1-4/+2
Before this change, wpa->ia.ff is initialized with an acquired reference on the fuse file right before it submits the writeback request. If there are auxiliary writebacks, then the initialization and reference acquisition needs to also be set before we submit the auxiliary writeback request. To make the logic simpler and to pave the way for a subsequent refactoring of fuse_writepages_fill() and fuse_writepage_locked(), this change initializes and acquires wpa->ia.ff when the wpa is allocated. No functional changes added. Signed-off-by: Joanne Koong <joannelkoong@gmail.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: convert fuse_writepages_fill() to use a folio for its tmp pageJoanne Koong1-7/+7
To pave the way for refactoring out the shared logic in fuse_writepages_fill() and fuse_writepage_locked(), this change converts the temporary page in fuse_writepages_fill() to use the folio API. This is similar to the change in commit e0887e095a80 ("fuse: Convert fuse_writepage_locked to take a folio"), which converted the tmp page in fuse_writepage_locked() to use the folio API. inc_node_page_state() is intentionally preserved here instead of converting to node_stat_add_folio() since it is updating the stat of the underlying page and to better maintain API symmetry with dec_node_page_stat() in fuse_writepage_finish_stat(). No functional changes added. Signed-off-by: Joanne Koong <joannelkoong@gmail.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: move initialization of fuse_file to fuse_writepages() instead of in callbackJoanne Koong1-12/+6
Prior to this change, data->ff is checked and if not initialized then initialized in the fuse_writepages_fill() callback, which gets called for every dirty page in the address space mapping. This logic is better placed in the main fuse_writepages() caller where data.ff is initialized before walking the dirty pages. No functional changes added. Signed-off-by: Joanne Koong <joannelkoong@gmail.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: refactor finished writeback stats updates into helper functionJoanne Koong1-17/+14
Move the logic for updating the bdi and page stats for a finished writeback into a separate helper function, where it can be called from both fuse_writepage_finish() and fuse_writepage_add() (in the case where there is already an auxiliary write request for the page). No functional changes added. Suggested by: Jingbo Xu <jefflexu@linux.alibaba.com> Signed-off-by: Joanne Koong <joannelkoong@gmail.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: drop unused fuse_mount arg in fuse_writepage_finish()Joanne Koong1-4/+3
Drop the unused "struct fuse_mount *fm" arg in fuse_writepage_finish(). No functional changes added. Signed-off-by: Joanne Koong <joannelkoong@gmail.com> Reviewed-by: Jingbo Xu <jefflexu@linux.alibaba.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: add fast path for fuse_range_is_writebackyangyun1-3/+3
In some cases, the fi->writepages may be empty. And there is no need to check fi->writepages with spin_lock, which may have an impact on performance due to lock contention. For example, in scenarios where multiple readers read the same file without any writers, or where the page cache is not enabled. Also remove the outdated comment since commit 6b2fb79963fb ("fuse: optimize writepages search") has optimize the situation by replacing list with rb-tree. Signed-off-by: yangyun <yangyun50@huawei.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: cleanup request queuing towards virtiofsMiklos Szeredi3-113/+106
Virtiofs has its own queuing mechanism, but still requests are first queued on fiq->pending to be immediately dequeued and queued onto the virtio queue. The queuing on fiq->pending is unnecessary and might even have some performance impact due to being a contention point. Forget requests are handled similarly. Move the queuing of requests and forgets into the fiq->ops->*. fuse_iqueue_ops are renamed to reflect the new semantics. Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Fixed-by: Jingbo Xu <jefflexu@linux.alibaba.com> Reviewed-by: Jingbo Xu <jefflexu@linux.alibaba.com> Tested-by: Peter-Jan Gootzen <pgootzen@nvidia.com> Reviewed-by: Peter-Jan Gootzen <pgootzen@nvidia.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-29fuse: disable the combination of passthrough and writeback cacheBernd Schubert1-1/+6
Current design and handling of passthrough is without fuse caching and with that FUSE_WRITEBACK_CACHE is conflicting. Fixes: 7dc4e97a4f9a ("fuse: introduce FUSE_PASSTHROUGH capability") Cc: stable@kernel.org # v6.9 Signed-off-by: Bernd Schubert <bschubert@ddn.com> Acked-by: Amir Goldstein <amir73il@gmail.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-28fuse: update stats for pages in dropped aux writeback listJoanne Koong1-1/+7
In the case where the aux writeback list is dropped (e.g. the pages have been truncated or the connection is broken), the stats for its pages and backing device info need to be updated as well. Fixes: e2653bd53a98 ("fuse: fix leaked aux requests") Signed-off-by: Joanne Koong <joannelkoong@gmail.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Cc: <stable@vger.kernel.org> # v5.1 Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-28fuse: clear PG_uptodate when using a stolen pageMiklos Szeredi1-4/+1
Originally when a stolen page was inserted into fuse's page cache by fuse_try_move_page(), it would be marked uptodate. Then fuse_readpages_end() would call SetPageUptodate() again on the already uptodate page. Commit 413e8f014c8b ("fuse: Convert fuse_readpages_end() to use folio_end_read()") changed that by replacing the SetPageUptodate() + unlock_page() combination with folio_end_read(), which does mostly the same, except it sets the uptodate flag with an xor operation, which in the above scenario resulted in the uptodate flag being cleared, which in turn resulted in EIO being returned on the read. Fix by clearing PG_uptodate instead of setting it in fuse_try_move_page(), conforming to the expectation of folio_end_read(). Reported-by: Jürg Billeter <j@bitron.ch> Debugged-by: Matthew Wilcox <willy@infradead.org> Fixes: 413e8f014c8b ("fuse: Convert fuse_readpages_end() to use folio_end_read()") Cc: <stable@vger.kernel.org> # v6.10 Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-28fuse: fix memory leak in fuse_create_openyangyun1-1/+1
The memory of struct fuse_file is allocated but not freed when get_create_ext return error. Fixes: 3e2b6fdbdc9a ("fuse: send security context of inode on file") Cc: stable@vger.kernel.org # v5.17 Signed-off-by: yangyun <yangyun50@huawei.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-28fuse: check aborted connection before adding requests to pending list for resendingJoanne Koong1-0/+9
There is a race condition where inflight requests will not be aborted if they are in the middle of being re-sent when the connection is aborted. If fuse_resend has already moved all the requests in the fpq->processing lists to its private queue ("to_queue") and then the connection starts and finishes aborting, these requests will be added to the pending queue and remain on it indefinitely. Fixes: 760eac73f9f6 ("fuse: Introduce a new notification type for resend pending requests") Signed-off-by: Joanne Koong <joannelkoong@gmail.com> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: Jingbo Xu <jefflexu@linux.alibaba.com> Cc: <stable@vger.kernel.org> # v6.9 Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-08-28fuse: use unsigned type for getxattr/listxattr size truncationJann Horn1-2/+2
The existing code uses min_t(ssize_t, outarg.size, XATTR_LIST_MAX) when parsing the FUSE daemon's response to a zero-length getxattr/listxattr request. On 32-bit kernels, where ssize_t and outarg.size are the same size, this is wrong: The min_t() will pass through any size values that are negative when interpreted as signed. fuse_listxattr() will then return this userspace-supplied negative value, which callers will treat as an error value. This kind of bug pattern can lead to fairly bad security bugs because of how error codes are used in the Linux kernel. If a caller were to convert the numeric error into an error pointer, like so: struct foo *func(...) { int len = fuse_getxattr(..., NULL, 0); if (len < 0) return ERR_PTR(len); ... } then it would end up returning this userspace-supplied negative value cast to a pointer - but the caller of this function wouldn't recognize it as an error pointer (IS_ERR_VALUE() only detects values in the narrow range in which legitimate errno values are), and so it would just be treated as a kernel pointer. I think there is at least one theoretical codepath where this could happen, but that path would involve virtio-fs with submounts plus some weird SELinux configuration, so I think it's probably not a concern in practice. Cc: stable@vger.kernel.org # v4.9 Fixes: 63401ccdb2ca ("fuse: limit xattr returned size") Signed-off-by: Jann Horn <jannh@google.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2024-07-28Linux 6.11-rc1Linus Torvalds1-2/+2
2024-07-28minmax: simplify and clarify min_t()/max_t() implementationLinus Torvalds1-8/+11
This simplifies the min_t() and max_t() macros by no longer making them work in the context of a C constant expression. That means that you can no longer use them for static initializers or for array sizes in type definitions, but there were only a couple of such uses, and all of them were converted (famous last words) to use MIN_T/MAX_T instead. Cc: David Laight <David.Laight@aculab.com> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-07-28minmax: add a few more MIN_T/MAX_T usersLinus Torvalds7-10/+10
Commit 3a7e02c040b1 ("minmax: avoid overly complicated constant expressions in VM code") added the simpler MIN_T/MAX_T macros in order to avoid some excessive expansion from the rather complicated regular min/max macros. The complexity of those macros stems from two issues: (a) trying to use them in situations that require a C constant expression (in static initializers and for array sizes) (b) the type sanity checking and MIN_T/MAX_T avoids both of these issues. Now, in the whole (long) discussion about all this, it was pointed out that the whole type sanity checking is entirely unnecessary for min_t/max_t which get a fixed type that the comparison is done in. But that still leaves min_t/max_t unnecessarily complicated due to worries about the C constant expression case. However, it turns out that there really aren't very many cases that use min_t/max_t for this, and we can just force-convert those. This does exactly that. Which in turn will then allow for much simpler implementations of min_t()/max_t(). All the usual "macros in all upper case will evaluate the arguments multiple times" rules apply. We should do all the same things for the regular min/max() vs MIN/MAX() cases, but that has the added complexity of various drivers defining their own local versions of MIN/MAX, so that needs another level of fixes first. Link: https://lore.kernel.org/all/b47fad1d0cf8449886ad148f8c013dae@AcuMS.aculab.com/ Cc: David Laight <David.Laight@aculab.com> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-07-29kbuild: Fix '-S -c' in x86 stack protector scriptsNathan Chancellor2-2/+2
After a recent change in clang to stop consuming all instances of '-S' and '-c' [1], the stack protector scripts break due to the kernel's use of -Werror=unused-command-line-argument to catch cases where flags are not being properly consumed by the compiler driver: $ echo | clang -o - -x c - -S -c -Werror=unused-command-line-argument clang: error: argument unused during compilation: '-c' [-Werror,-Wunused-command-line-argument] This results in CONFIG_STACKPROTECTOR getting disabled because CONFIG_CC_HAS_SANE_STACKPROTECTOR is no longer set. '-c' and '-S' both instruct the compiler to stop at different stages of the pipeline ('-S' after compiling, '-c' after assembling), so having them present together in the same command makes little sense. In this case, the test wants to stop before assembling because it is looking at the textual assembly output of the compiler for either '%fs' or '%gs', so remove '-c' from the list of arguments to resolve the error. All versions of GCC continue to work after this change, along with versions of clang that do or do not contain the change mentioned above. Cc: stable@vger.kernel.org Fixes: 4f7fd4d7a791 ("[PATCH] Add the -fstack-protector option to the CFLAGS") Fixes: 60a5317ff0f4 ("x86: implement x86_32 stack protector") Link: https://github.com/llvm/llvm-project/commit/6461e537815f7fa68cef06842505353cf5600e9c [1] Signed-off-by: Nathan Chancellor <nathan@kernel.org> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
2024-07-28ubi: Fix ubi_init() ubiblock_exit() section mismatchRichard Weinberger1-1/+1
Since ubiblock_exit() is now called from an init function, the __exit section no longer makes sense. Cc: Ben Hutchings <bwh@kernel.org> Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202407131403.wZJpd8n2-lkp@intel.com/ Signed-off-by: Richard Weinberger <richard@nod.at> Reviewed-by: Zhihao Cheng <chengzhihao1@huawei.com>
2024-07-28kbuild: rpm-pkg: ghost modules.weakdep fileJose Ignacio Tornos Martinez1-1/+1
In the same way as for other similar files, mark as ghost the new file generated by depmod for configured weak dependencies for modules, modules.weakdep, so that although it is not included in the package, claim the ownership on it. Signed-off-by: Jose Ignacio Tornos Martinez <jtornosm@redhat.com> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
2024-07-27hostfs: fix the host directory parse when mounting.Hongbo Li1-10/+55
hostfs not keep the host directory when mounting. When the host directory is none (default), fc->source is used as the host root directory, and this is wrong. Here we use `parse_monolithic` to handle the old mount path for parsing the root directory. For new mount path, The `parse_param` is used for the host directory parse. Reported-and-tested-by: Maciej Żenczykowski <maze@google.com> Fixes: cd140ce9f611 ("hostfs: convert hostfs to use the new mount API") Link: https://lore.kernel.org/all/CANP3RGceNzwdb7w=vPf5=7BCid5HVQDmz1K5kC9JG42+HVAh_g@mail.gmail.com/ Cc: Christian Brauner <brauner@kernel.org> Signed-off-by: Hongbo Li <lihongbo22@huawei.com> Link: https://lore.kernel.org/r/20240725065130.1821964-1-lihongbo22@huawei.com [brauner: minor fixes] Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-07-27fs: don't allow non-init s_user_ns for filesystems without FS_USERNS_MOUNTSeth Forshee (DigitalOcean)1-0/+11
Christian noticed that it is possible for a privileged user to mount most filesystems with a non-initial user namespace in sb->s_user_ns. When fsopen() is called in a non-init namespace the caller's namespace is recorded in fs_context->user_ns. If the returned file descriptor is then passed to a process priviliged in init_user_ns, that process can call fsconfig(fd_fs, FSCONFIG_CMD_CREATE), creating a new superblock with sb->s_user_ns set to the namespace of the process which called fsopen(). This is problematic. We cannot assume that any filesystem which does not set FS_USERNS_MOUNT has been written with a non-initial s_user_ns in mind, increasing the risk for bugs and security issues. Prevent this by returning EPERM from sget_fc() when FS_USERNS_MOUNT is not set for the filesystem and a non-initial user namespace will be used. sget() does not need to be updated as it always uses the user namespace of the current context, or the initial user namespace if SB_SUBMOUNT is set. Fixes: cb50b348c71f ("convenience helpers: vfs_get_super() and sget_fc()") Reported-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Seth Forshee (DigitalOcean) <sforshee@kernel.org> Link: https://lore.kernel.org/r/20240724-s_user_ns-fix-v1-1-895d07c94701@kernel.org Reviewed-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com> Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-07-27ALSA: firewire-lib: fix wrong value as length of header for CIP_NO_HEADER caseTakashi Sakamoto1-2/+1
In a commit 1d717123bb1a ("ALSA: firewire-lib: Avoid -Wflex-array-member-not-at-end warning"), DEFINE_FLEX() macro was used to handle variable length of array for header field in struct fw_iso_packet structure. The usage of macro has a side effect that the designated initializer assigns the count of array to the given field. Therefore CIP_HEADER_QUADLETS (=2) is assigned to struct fw_iso_packet.header, while the original designated initializer assigns zero to all fields. With CIP_NO_HEADER flag, the change causes invalid length of header in isochronous packet for 1394 OHCI IT context. This bug affects all of devices supported by ALSA fireface driver; RME Fireface 400, 800, UCX, UFX, and 802. This commit fixes the bug by replacing it with the alternative version of macro which corresponds no initializer. Cc: stable@vger.kernel.org Fixes: 1d717123bb1a ("ALSA: firewire-lib: Avoid -Wflex-array-member-not-at-end warning") Reported-by: Edmund Raile <edmund.raile@proton.me> Closes: https://lore.kernel.org/r/rrufondjeynlkx2lniot26ablsltnynfaq2gnqvbiso7ds32il@qk4r6xps7jh2/ Reviewed-by: Takashi Iwai <tiwai@suse.de> Link: https://lore.kernel.org/r/20240725155640.128442-1-o-takashi@sakamocchi.jp Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2024-07-27Revert "firewire: Annotate struct fw_iso_packet with __counted_by()"Takashi Sakamoto1-3/+2
This reverts commit d3155742db89df3b3c96da383c400e6ff4d23c25. The header_length field is byte unit, thus it can not express the number of elements in header field. It seems that the argument for counted_by attribute can have no arithmetic expression, therefore this commit just reverts the issued commit. Suggested-by: Gustavo A. R. Silva <gustavoars@kernel.org> Link: https://lore.kernel.org/r/20240725161648.130404-1-o-takashi@sakamocchi.jp Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2024-07-26minmax: avoid overly complicated constant expressions in VM codeLinus Torvalds2-2/+9
The minmax infrastructure is overkill for simple constants, and can cause huge expansions because those simple constants are then used by other things. For example, 'pageblock_order' is a core VM constant, but because it was implemented using 'min_t()' and all the type-checking that involves, it actually expanded to something like 2.5kB of preprocessor noise. And when that simple constant was then used inside other expansions: #define pageblock_nr_pages (1UL << pageblock_order) #define pageblock_start_pfn(pfn) ALIGN_DOWN((pfn), pageblock_nr_pages) and we then use that inside a 'max()' macro: case ISOLATE_SUCCESS: update_cached = false; last_migrated_pfn = max(cc->zone->zone_start_pfn, pageblock_start_pfn(cc->migrate_pfn - 1)); the end result was that one statement expanding to 253kB in size. There are probably other cases of this, but this one case certainly stood out. I've added 'MIN_T()' and 'MAX_T()' macros for this kind of "core simple constant with specific type" use. These macros skip the type checking, and as such need to be very sparingly used only for obvious cases that have active issues like this. Reported-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com> Link: https://lore.kernel.org/all/36aa2cad-1db1-4abf-8dd2-fb20484aabc3@lucifer.local/ Cc: David Laight <David.Laight@aculab.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-07-26minmax: avoid overly complex min()/max() macro arguments in xenLinus Torvalds1-2/+3
We have some very fancy min/max macros that have tons of sanity checking to warn about mixed signedness etc. This is all things that a sane compiler should warn about, but there are no sane compiler interfaces for this, and '-Wsign-compare' is broken [1] and not useful. So then we compensate (some would say over-compensate) by doing the checks manually with some truly horrid macro games. And no, we can't just use __builtin_types_compatible_p(), because the whole question of "does it make sense to compare these two values" is a lot more complicated than that. For example, it makes a ton of sense to compare unsigned values with simple constants like "5", even if that is indeed a signed type. So we have these very strange macros to try to make sensible type checking decisions on the arguments to 'min()' and 'max()'. But that can cause enormous code expansion if the min()/max() macros are used with complicated expressions, and particularly if you nest these things so that you get the first big expansion then expanded again. The xen setup.c file ended up ballooning to over 50MB of preprocessed noise that takes 15s to compile (obviously depending on the build host), largely due to one single line. So let's split that one single line to just be simpler. I think it ends up being more legible to humans too at the same time. Now that single file compiles in under a second. Reported-and-reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com> Link: https://lore.kernel.org/all/c83c17bb-be75-4c67-979d-54eee38774c6@lucifer.local/ Link: https://staticthinking.wordpress.com/2023/07/25/wsign-compare-is-garbage/ [1] Cc: David Laight <David.Laight@aculab.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-07-26nilfs2: handle inconsistent state in nilfs_btnode_create_block()Ryusuke Konishi2-7/+22
Syzbot reported that a buffer state inconsistency was detected in nilfs_btnode_create_block(), triggering a kernel bug. It is not appropriate to treat this inconsistency as a bug; it can occur if the argument block address (the buffer index of the newly created block) is a virtual block number and has been reallocated due to corruption of the bitmap used to manage its allocation state. So, modify nilfs_btnode_create_block() and its callers to treat it as a possible filesystem error, rather than triggering a kernel bug. Link: https://lkml.kernel.org/r/20240725052007.4562-1-konishi.ryusuke@gmail.com Fixes: a60be987d45d ("nilfs2: B-tree node cache") Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com> Reported-by: syzbot+89cc4f2324ed37988b60@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=89cc4f2324ed37988b60 Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-07-26selftests/mm: skip test for non-LPA2 and non-LVA systemsDev Jain1-1/+15
Post my improvement of the test in e4a4ba415419 ("selftests/mm: va_high_addr_switch: dynamically initialize testcases to enable LPA2 testing"): The test begins to fail on 4k and 16k pages, on non-LPA2 systems. To reduce noise in the CI systems, let us skip the test when higher address space is not implemented. Link: https://lkml.kernel.org/r/20240718052504.356517-1-dev.jain@arm.com Fixes: e4a4ba415419 ("selftests/mm: va_high_addr_switch: dynamically initialize testcases to enable LPA2 testing") Signed-off-by: Dev Jain <dev.jain@arm.com> Reviewed-by: Ryan Roberts <ryan.roberts@arm.com> Cc: Anshuman Khandual <anshuman.khandual@arm.com> Cc: Mark Brown <broonie@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-07-26mm/page_alloc: fix pcp->count race between drain_pages_zone() vs __rmqueue_pcplist()Li Zhijian1-7/+11
It's expected that no page should be left in pcp_list after calling zone_pcp_disable() in offline_pages(). Previously, it's observed that offline_pages() gets stuck [1] due to some pages remaining in pcp_list. Cause: There is a race condition between drain_pages_zone() and __rmqueue_pcplist() involving the pcp->count variable. See below scenario: CPU0 CPU1 ---------------- --------------- spin_lock(&pcp->lock); __rmqueue_pcplist() { zone_pcp_disable() { /* list is empty */ if (list_empty(list)) { /* add pages to pcp_list */ alloced = rmqueue_bulk() mutex_lock(&pcp_batch_high_lock) ... __drain_all_pages() { drain_pages_zone() { /* read pcp->count, it's 0 here */ count = READ_ONCE(pcp->count) /* 0 means nothing to drain */ /* update pcp->count */ pcp->count += alloced << order; ... ... spin_unlock(&pcp->lock); In this case, after calling zone_pcp_disable() though, there are still some pages in pcp_list. And these pages in pcp_list are neither movable nor isolated, offline_pages() gets stuck as a result. Solution: Expand the scope of the pcp->lock to also protect pcp->count in drain_pages_zone(), to ensure no pages are left in the pcp list after zone_pcp_disable() [1] https://lore.kernel.org/linux-mm/6a07125f-e720-404c-b2f9-e55f3f166e85@fujitsu.com/ Link: https://lkml.kernel.org/r/20240723064428.1179519-1-lizhijian@fujitsu.com Fixes: 4b23a68f9536 ("mm/page_alloc: protect PCP lists with a spinlock") Signed-off-by: Li Zhijian <lizhijian@fujitsu.com> Reported-by: Yao Xingtao <yaoxt.fnst@fujitsu.com> Reviewed-by: Vlastimil Babka <vbabka@suse.cz> Cc: David Hildenbrand <david@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.