iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

bcachefs: Don't stop copygc while removing devices

Browse Source 
With the new backpointer based copygc we don't need an explicit copygc
reserve, we're always evacuating buckets one at a time - so this is no
longer needed, and in fact removing it fixes a deadlock in
bch2_dev_allocator_remove().

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This commit is contained in:
Kent Overstreet 2022-10-09 04:26:06 -04:00
parent c9828cea31
commit 15949c5499
1 changed files with 0 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
fs/bcachefs/super.c
View File

@ -1345,19 +1345,11 @@ static bool bch2_fs_may_start(struct bch_fs *c)
static void __bch2_dev_read_only(struct bch_fs *c, struct bch_dev *ca)
{
/*
* Device going read only means the copygc reserve get smaller, so we
* don't want that happening while copygc is in progress:
*/
bch2_copygc_stop(c);
/*
* The allocator thread itself allocates btree nodes, so stop it first:
*/
bch2_dev_allocator_remove(c, ca);
bch2_dev_journal_stop(&c->journal, ca);
bch2_copygc_start(c);
}
static void __bch2_dev_read_write(struct bch_fs *c, struct bch_dev *ca)