kernel.git/fs/bcachefs/ec.c, branch linux-6.15.y

bcachefs: fix extent_has_stripe_ptr()

2025-05-19T02:35:33Z

This wasn't checking indirect extents. Fixes: https://github.com/koverstreet/bcachefs/issues/887 Signed-off-by: Kent Overstreet

bcachefs: add missing sched_annotate_sleep()

2025-05-01T17:54:58Z

00594 ------------[ cut here ]------------ 00594 do not call blocking ops when !TASK_RUNNING; state=2 set at [<000000003e51ef4a>] prepare_to_wait_event+0x5c/0x1c0 00594 WARNING: CPU: 12 PID: 1117 at kernel/sched/core.c:8741 __might_sleep+0x74/0x88 00594 Modules linked in: 00594 CPU: 12 UID: 0 PID: 1117 Comm: umount Not tainted 6.15.0-rc4-ktest-g3a72e369412d #21845 PREEMPT 00594 Hardware name: linux,dummy-virt (DT) 00594 pstate: 60001005 (nZCv daif -PAN -UAO -TCO -DIT +SSBS BTYPE=--) 00594 pc : __might_sleep+0x74/0x88 00594 lr : __might_sleep+0x74/0x88 00594 sp : ffffff80c8d67a90 00594 x29: ffffff80c8d67a90 x28: ffffff80f5903500 x27: 0000000000000000 00594 x26: 0000000000000000 x25: ffffff80cf5002a0 x24: ffffffc087dad000 00594 x23: ffffff80c8d67b40 x22: 0000000000000000 x21: 0000000000000000 00594 x20: 0000000000000242 x19: ffffffc080b92020 x18: 00000000ffffffff 00594 x17: 30303c5b20746120 x16: 74657320323d6574 x15: 617473203b474e49 00594 x14: 0000000000000001 x13: 00000000000c0000 x12: ffffff80facc0000 00594 x11: 0000000000000001 x10: 0000000000000001 x9 : ffffffc0800b0774 00594 x8 : c0000000fffbffff x7 : ffffffc087dac670 x6 : 00000000015fffa8 00594 x5 : ffffff80facbffa8 x4 : ffffff80fbd30b90 x3 : 0000000000000000 00594 x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffffff80f5903500 00594 Call trace: 00594 __might_sleep+0x74/0x88 (P) 00594 __mutex_lock+0x64/0x8d8 00594 mutex_lock_nested+0x28/0x38 00594 bch2_fs_ec_flush+0xf8/0x128 00594 __bch2_fs_read_only+0x54/0x1d8 00594 bch2_fs_read_only+0x3e0/0x438 00594 __bch2_fs_stop+0x5c/0x250 00594 bch2_put_super+0x18/0x28 00594 generic_shutdown_super+0x6c/0x140 00594 bch2_kill_sb+0x1c/0x38 00594 deactivate_locked_super+0x54/0xd0 00594 deactivate_super+0x70/0x90 00594 cleanup_mnt+0xec/0x188 00594 __cleanup_mnt+0x18/0x28 00594 task_work_run+0x90/0xd8 00594 do_notify_resume+0x138/0x148 00594 el0_svc+0x9c/0xa0 00594 el0t_64_sync_handler+0x104/0x130 00594 el0t_64_sync+0x154/0x158 Signed-off-by: Kent Overstreet

bcachefs: Kill btree_iter.trans

2025-04-02T14:24:34Z

This was planned to be done ages ago, now finally completed; there are places where we have quite a few btree_trans objects on the stack, so this reduces stack usage somewhat. Signed-off-by: Kent Overstreet

bcachefs: Split up bch_dev.io_ref

2025-04-02T14:24:34Z

We now have separate per device io_refs for read and write access. This fixes a device removal bug where the discard workers were still running while we're removing alloc info for that device. It's also a bit of hardening; we no longer allow writes to devices that are read-only. Signed-off-by: Kent Overstreet

bcachefs: Fix field spanning write warning

2025-03-31T21:39:10Z

Struct with embedded VLA... memcpy: detected field-spanning write (size 8) of single field "&gc->r.e" at fs/bcachefs/ec.c:465 (size 3) WARNING: CPU: 1 PID: 936 at fs/bcachefs/ec.c:465 bch2_trigger_stripe+0x706/0x730 Modules linked in: CPU: 1 UID: 0 PID: 936 Comm: mount.bcachefs Not tainted 6.14.0-rc6-ktest-00236-gefb0b5c62dbc #55 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 RIP: 0010:bch2_trigger_stripe+0x706/0x730 Code: b4 00 01 b9 03 00 00 00 48 89 fb 48 c7 c7 33 54 da 81 48 89 d6 49 89 d6 48 c7 c2 c3 36 db 81 e8 60 54 c5 ff 48 89 df 4c 89 f2 <0f> 0b e9 5c fd ff ff e8 fe 5e 4e 00 bf 10 00 00 00 48 c7 c6 ff ff RSP: 0018:ffff88817081f680 EFLAGS: 00010246 RAX: f8fe7dd1c56b5600 RBX: ffff888101265368 RCX: 0000000000000027 RDX: 0000000000000008 RSI: 00000000fffbffff RDI: ffff888101265368 RBP: 0000000000000000 R08: 000000000003ffff R09: ffff88817f1fe000 R10: 00000000000bfffd R11: 0000000000000004 R12: ffff8881012652c0 R13: 0000000000000000 R14: 0000000000000008 R15: ffff88817081f6c9 FS: 00007fc428bc7c80(0000) GS:ffff888179280000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007ffd3ee4a038 CR3: 000000010a9bc000 CR4: 0000000000750eb0 PKRU: 55555554 Call Trace: ? __warn+0xce/0x1b0 ? bch2_trigger_stripe+0x706/0x730 ? report_bug+0x11b/0x1a0 ? bch2_trigger_stripe+0x706/0x730 ? handle_bug+0x5e/0x90 ? exc_invalid_op+0x1a/0x50 ? asm_exc_invalid_op+0x1a/0x20 ? bch2_trigger_stripe+0x706/0x730 bch2_gc_mark_key+0x2cf/0x430 bch2_check_allocations+0x1a64/0x1ed0 ? vsnprintf+0x1ad/0x420 ? bch2_check_allocations+0x191f/0x1ed0 bch2_run_recovery_passes+0x13b/0x2b0 bch2_fs_recovery+0x9b7/0x1290 ? __bch2_print+0xb2/0xf0 ? bch2_printbuf_exit+0x1e/0x30 ? print_mount_opts+0x153/0x180 bch2_fs_start+0x274/0x3b0 bch2_fs_get_tree+0x516/0x6e0 vfs_get_tree+0x21/0xa0 do_new_mount+0x153/0x350 __x64_sys_mount+0x16c/0x1f0 do_syscall_64+0x6c/0x140 ? arch_exit_to_user_mode_prepare+0x9/0x40 entry_SYSCALL_64_after_hwframe+0x4b/0x53 Signed-off-by: Kent Overstreet

bcachefs: Don't use designated initializers for disk_accounting_pos

2025-03-30T20:35:13Z

Not all compilers fully initialize these - they're not guaranteed to because of the union shenanigans. Fixes: https://github.com/koverstreet/bcachefs/issues/844 Signed-off-by: Kent Overstreet

bcachefs: Consistent indentation of multiline fsck errors

2025-03-29T02:31:47Z

Add the new helper printbuf_indent_add_nextline(), and use it in __bch2_fsck_err() to centralize setting the indentation of multiline fsck errors. Signed-off-by: Kent Overstreet

bcachefs: EIO cleanup

2025-03-24T13:50:36Z

Replace these with proper private error codes, so that when we get an error message we're not sifting through the entire codebase to see where it came from. Signed-off-by: Kent Overstreet

bcachefs: We no longer read stripes into memory at startup

2025-03-15T01:02:16Z

And the stripes heap gets deleted. Signed-off-by: Kent Overstreet

bcachefs: trace_stripe_create

2025-03-15T01:02:16Z

Add a simple tracepoint for stripe creation, we'll want to expand this later. Signed-off-by: Kent Overstreet