kernel.git/fs/bcachefs/sysfs.c, branch linux-6.15.yHosts the 0x221E linux distro kernel.
https://git.0xinfinity.dev/distro/kernel.git/atom?h=linux-6.15.y2025-04-21T00:01:48Zbcachefs: bch2_copygc_wakeup()2025-04-21T00:01:48ZKent Overstreet2025-04-15T14:26:05Zurn:sha1:10e42b6f25995c6ce7c462e1bcb9684acc0f09a0
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: Don't use designated initializers for disk_accounting_pos2025-03-30T20:35:13ZKent Overstreet2025-03-30T00:58:32Zurn:sha1:393a05a7413aa325a15c6d3b35867843f91f1646
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 <kent.overstreet@linux.dev>
bcachefs: Recovery no longer holds state_lock2025-03-28T15:13:25ZKent Overstreet2025-03-26T15:41:07Zurn:sha1:2dd202dbaf0acfa4af7fcdf258f35866e31f7425
state_lock guards against devices coming or leaving, changing state, or
the filesystem changing between ro <-> rw.
But it's not necessary for running recovery passes, and holding it
blocks asynchronous events that would cause us to go RO or kick out
devices.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: Filesystem discard option now propagates to devices2025-03-24T13:50:35ZKent Overstreet2025-03-13T04:54:10Zurn:sha1:80be08cdb5a82208a0bb67ad93b3fb7447fd2873
the discard option is special, because it's both a filesystem and a
device option.
When set at the filesytsem level, it's supposed to propagate to (if set
persistently via sysfs) or override (if non persistently as a mount
option) the devices - that now works correctly.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: Device state is now a runtime option2025-03-24T13:50:35ZKent Overstreet2025-03-13T04:55:52Zurn:sha1:8d7b7ac367cd0e7f0e496ba6526799994b3c1237
Other options can normally be set at runtime via sysfs, no reason for
this one not to be as well - it just doesn't support the degraded flags
argument this way, that requires the ioctl.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: Setting foreground_target at runtime now triggers rebalance2025-03-24T13:50:35ZKent Overstreet2025-03-13T04:55:23Zurn:sha1:7b84d934a16274eaedfb38cb94b909426048c48e
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: Device options now use standard sysfs code2025-03-24T13:50:35ZKent Overstreet2025-03-11T22:44:25Zurn:sha1:8b294a9b5c14473d6d7e35756d201922ba785042
Device options now use the common code for sysfs, and can superblock
fields (in a struct bch_member).
This replaces BCH_DEV_OPT_SETTERS(), which was weird and easy to miss.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: We no longer read stripes into memory at startup2025-03-15T01:02:16ZKent Overstreet2025-02-11T01:15:40Zurn:sha1:94373026d930b9ed72c8f8f0f3d532e13654fdb1
And the stripes heap gets deleted.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: sysfs internal/trigger_btree_updates2025-03-15T01:02:15ZKent Overstreet2025-02-13T17:46:15Zurn:sha1:24d790a7daa37093679874804842703cf13bf511
Add a debug knob to manually trigger the btree updates worker.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: Move write_points to debugfs2025-03-15T01:02:11ZKent Overstreet2025-01-22T17:07:54Zurn:sha1:78c9c6f6cd25c2aabc29aa8129996031fdfeb31b
this was hitting the sysfs 4k limit
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>