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

bcachefs: Fix rereplicate when we already have a cached pointer

Browse Source 
When we need to add more replicas to an extent, it might be the case
that we already have a replica on every device, but some of them are
cached.

This patch fixes a bug where we'd spin on that extent because the write
path fails to find a device we can allocate from: we allow allocating
from devices that already have cached replicas on them, and change
bch2_data_update_index_update() to drop the cached replica if needed.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This commit is contained in:
Kent Overstreet 2023-01-07 22:55:42 -05:00
parent 7c909f654b
commit d7dd3fb84f
1 changed files with 13 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
fs/bcachefs/data_update.c
View File

@ -183,7 +183,17 @@ int bch2_data_update_index_update(struct bch_write_op *op)
/* Add new ptrs: */
extent_for_each_ptr_decode(extent_i_to_s(new), p, entry) {
if (bch2_bkey_has_device(bkey_i_to_s_c(insert), p.ptr.dev)) {
const struct bch_extent_ptr *existing_ptr =
bch2_bkey_has_device(bkey_i_to_s_c(insert), p.ptr.dev);
if (existing_ptr && existing_ptr->cached) {
/*
* We're replacing a cached pointer with a non
* cached pointer:
*/
bch2_bkey_drop_device_noerror(bkey_i_to_s(insert),
existing_ptr->dev);
} else if (existing_ptr) {
/*
* raced with another move op? extent already
* has a pointer to the device we just wrote
@ -334,7 +344,8 @@ int bch2_data_update_init(struct bch_fs *c, struct data_update *m,
p.ptr.cached)
BUG();
if (!((1U << i) & m->data_opts.rewrite_ptrs))
if (!((1U << i) & m->data_opts.rewrite_ptrs) &&
!p.ptr.cached)
bch2_dev_list_add_dev(&m->op.devs_have, p.ptr.dev);
if (((1U << i) & m->data_opts.rewrite_ptrs) &&