Age | Commit message (Collapse) | Author |
|
bch2_btree_node_mem_alloc()
|
|
tracepoint
|
|
BTREE_TRIGGER_WANTS_OLD_AND_NEW
|
|
Also fix the freezable_schedule_timeout() macro.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
|
|
journal entries
|
|
This fixes an observed journal deadlock - real spinlocks aren't safe in
userspace.
|
|
not use lookup_bdev()
|
|
|
|
|
|
calculating fragmentation
|
|
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
|
|
|
|
Signed-off-by: Brett Holman <bholman.devel@gmail.com>
|
|
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
always call bch2_trans_begin()
|
|
|
|
check_dirents()
|
|
features when going read-write
|
|
|
|
bch2_trans_kmalloc()
|
|
end of usable space
|
|
|
|
|
|
This gets us round_up() and round_down()
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
not initializing len/version fields
|
|
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
bch2_btree_and_journal_walk()
|
|
Spinlocks aren't a good idea in userspace, where we can't actually
disable preemption.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
|
|
Imported from the kernel.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
interior btree ptrs
|
|
check_directory_structure()
|
|
PAGE_SIZE uses
|
|
bcachefs assumes kmalloc & krealloc give out allocations that are
naturally aligned, like the kernel allocators do.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
|
|
for interior nodes
|
|
|
|
|
|
every journal write
|
|
initialization
|
|
|
|
this fixes an address sanitizer splat
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
This was causing us some valgrind errors.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
|
|
bitmap in fsck
|
|
read error paths
|
|
alloc btree
|