Displaying 4 results from an estimated 4 matches for "multi_ret".
Did you mean:
multi_get
2010 Jan 04
0
[RFC 03/12 RESEND PATCH] Btrfs: Reorder __btrfs_map_block to make code more efficient.
...map_block(struct btrfs_mapping_tree *map_tree, int rw,
u64 offset;
u64 stripe_offset;
u64 stripe_nr;
- int stripes_allocated = 8;
- int stripes_required = 1;
int stripe_index;
int i;
int num_stripes;
- int max_errors = 0;
+ int max_errors;
struct btrfs_multi_bio *multi = NULL;
- if (multi_ret && !(rw & (1 << BIO_RW)))
- stripes_allocated = 1;
-again:
- if (multi_ret) {
- multi = kzalloc(btrfs_multi_bio_size(stripes_allocated),
- GFP_NOFS);
- if (!multi)
- return -ENOMEM;
-
- atomic_set(&multi->error, 0);
- }
-
read_lock(&em_tree->lock);
em =...
2009 Aug 05
3
RAID[56] with arbitrary numbers of "parity" stripes.
...ID6)
- ret = 3;
+ else if (map->type & BTRFS_BLOCK_GROUP_RAID56)
+ ret = nr_parity_stripes(map);
else
ret = 1;
free_extent_map(em);
@@ -2734,8 +2716,8 @@ again:
max_errors = 1;
}
}
- if (map->type & (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6)
- && multi_ret && (rw & (1 << BIO_RW) || mirror_num > 1) && raid_map_ret) {
+ if (map->type & BTRFS_BLOCK_GROUP_RAID56 && multi_ret &&
+ (rw & (1 << BIO_RW) || mirror_num > 1) && raid_map_ret) {
/* RAID[56] write or recovery. Retu...
2012 Mar 20
13
[PATCH 0 of 3 v2] PV-GRUB: add support for ext4 and btrfs
Hi,
The following patches add support for ext4 and btrfs to
PV-GRUB. These patches are taken nearly verbatim from those provided
by Fedora and Gentoo.
We''ve been using these patches for the PV-GRUB images available in EC2
for some time now with no problems.
Changes from v1:
- Makefile has been changed to check the exit code from patch
- The btrfs patch has been rebased to apply
2011 Aug 23
40
[PATCH 00/21] [RFC] Btrfs: restriper
Hello,
This patch series adds an initial implementation of restriper (it''s a
clever name for relocation framework that allows to do selective profile
changing and selective balancing with some goodies like pausing/resuming
and reporting progress to the user.
Profile changing is global (per-FS) so far, per-subvolume profiles
require some discussion and can be implemented in future.