On 3/31/20 10:08 PM, Goffredo Baroncelli wrote:
On 3/31/20 7:09 PM, Andrei Borzenkov wrote:
31.03.2020 20:01, Eli V пишет:
Another option is to put the 12TB drives in an mdadm RAID, and then
use the mdadm raid & the ssd for btrfs RAID1 metadata, with SINGLE
data on the the array.
How do you restrict specific device for metadata only?
I never tried, but I don't think that it would be so complicated.
When BTRFS has to allocate a new chunk, it collects all the available
free spaces on the disks; it sorts all these free spaces on the basis of
criterion like the largest contiguous area and how the disk is full
and pick the top one.
It could be sufficient to add another criteria to the sorting algorithm,
something like that
- if the chunk is a metadata one, an SSD has an higher priority
- if the chunk is a data one, an SSD has a lower priority
So the metadata will have an higher likelihood to be on the SSD,
instead the data will have an higher likelihood to be a NON SSD disk.
Of course this is a soft constraint, when a kind of disk is full, it will
be possible to use the other kind, only with a lower priority.
This is only to give an idea. In order to enable the feature, it must be mounted
with the flag ssd_metadata:
# mount -o ssd_metadata /dev/sdX /mnt/test
(don't try at home !)
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2e9f938508e9..0f3c09cc4863 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1187,6 +1187,7 @@ static inline u32 BTRFS_MAX_XATTR_SIZE(const struct btrfs_fs_info *info)
#define BTRFS_MOUNT_FREE_SPACE_TREE (1 << 26)
#define BTRFS_MOUNT_NOLOGREPLAY (1 << 27)
#define BTRFS_MOUNT_REF_VERIFY (1 << 28)
+#define BTRFS_MOUNT_SSD_METADATA (1 << 29)
#define BTRFS_DEFAULT_COMMIT_INTERVAL (30)
#define BTRFS_DEFAULT_MAX_INLINE (2048)
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index c6557d44907a..d0a5cf496f90 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -346,6 +346,7 @@ enum {
#ifdef CONFIG_BTRFS_FS_REF_VERIFY
Opt_ref_verify,
#endif
+ Opt_ssd_metadata,
Opt_err,
};
@@ -416,6 +417,7 @@ static const match_table_t tokens = {
#ifdef CONFIG_BTRFS_FS_REF_VERIFY
{Opt_ref_verify, "ref_verify"},
#endif
+ {Opt_ssd_metadata, "ssd_metadata"},
{Opt_err, NULL},
};
@@ -853,6 +855,10 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
btrfs_set_opt(info->mount_opt, REF_VERIFY);
break;
#endif
+ case Opt_ssd_metadata:
+ btrfs_set_and_info(info, SSD_METADATA,
+ "enabling ssd_metadata");
+ break;
case Opt_err:
btrfs_info(info, "unrecognized mount option '%s'", p);
ret = -EINVAL;
@@ -1369,6 +1375,8 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
#endif
if (btrfs_test_opt(info, REF_VERIFY))
seq_puts(seq, ",ref_verify");
+ if (btrfs_test_opt(info, SSD_METADATA))
+ seq_puts(seq, ",ssd_metadata");
seq_printf(seq, ",subvolid=%llu",
BTRFS_I(d_inode(dentry))->root->root_key.objectid);
seq_puts(seq, ",subvol=");
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index a8b71ded4d21..43bb5d98a8cb 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -4758,6 +4758,67 @@ static int btrfs_cmp_device_info(const void *a, const void *b)
return 0;
}
+/*
+ * sort the devices in descending order by rotational,
+ * max_avail, total_avail
+ */
+static int btrfs_cmp_device_info_metadata(const void *a, const void *b)
+{
+ const struct btrfs_device_info *di_a = a;
+ const struct btrfs_device_info *di_b = b;
+ const int nrot_a = test_bit(QUEUE_FLAG_NONROT,
+ &(bdev_get_queue(di_a->dev->bdev)->queue_flags));
+
+ const int nrot_b = test_bit(QUEUE_FLAG_NONROT,
+ &(bdev_get_queue(di_b->dev->bdev)->queue_flags));
+
+ /* metadata -> non rotational first */
+ if (nrot_a && !nrot_b)
+ return -1;
+ if (!nrot_a && nrot_b)
+ return 1;
+ if (di_a->max_avail > di_b->max_avail)
+ return -1;
+ if (di_a->max_avail < di_b->max_avail)
+ return 1;
+ if (di_a->total_avail > di_b->total_avail)
+ return -1;
+ if (di_a->total_avail < di_b->total_avail)
+ return 1;
+ return 0;
+}
+
+/*
+ * sort the devices in descending order by !rotational,
+ * max_avail, total_avail
+ */
+static int btrfs_cmp_device_info_data(const void *a, const void *b)
+{
+ const struct btrfs_device_info *di_a = a;
+ const struct btrfs_device_info *di_b = b;
+ const int nrot_a = test_bit(QUEUE_FLAG_NONROT,
+ &(bdev_get_queue(di_a->dev->bdev)->queue_flags));
+ const int nrot_b = test_bit(QUEUE_FLAG_NONROT,
+ &(bdev_get_queue(di_b->dev->bdev)->queue_flags));
+
+ /* data -> non rotational last */
+ if (nrot_a && !nrot_b)
+ return 1;
+ if (!nrot_a && nrot_b)
+ return -1;
+ if (di_a->max_avail > di_b->max_avail)
+ return -1;
+ if (di_a->max_avail < di_b->max_avail)
+ return 1;
+ if (di_a->total_avail > di_b->total_avail)
+ return -1;
+ if (di_a->total_avail < di_b->total_avail)
+ return 1;
+ return 0;
+}
+
+
+
static void check_raid56_incompat_flag(struct btrfs_fs_info *info, u64 type)
{
if (!(type & BTRFS_BLOCK_GROUP_RAID56_MASK))
@@ -4917,9 +4978,17 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
/*
* now sort the devices by hole size / available space
*/
- sort(devices_info, ndevs, sizeof(struct btrfs_device_info),
- btrfs_cmp_device_info, NULL);
-
+ if (((type & BTRFS_BLOCK_GROUP_DATA) &&
+ (type & BTRFS_BLOCK_GROUP_METADATA)) ||
+ !btrfs_test_opt(info, SSD_METADATA))
+ sort(devices_info, ndevs, sizeof(struct btrfs_device_info),
+ btrfs_cmp_device_info, NULL);
+ else if (type & BTRFS_BLOCK_GROUP_DATA)
+ sort(devices_info, ndevs, sizeof(struct btrfs_device_info),
+ btrfs_cmp_device_info_data, NULL);
+ else
+ sort(devices_info, ndevs, sizeof(struct btrfs_device_info),
+ btrfs_cmp_device_info_metadata, NULL);
/*
* Round down to number of usable stripes, devs_increment can be any
* number so we can't use round_down()
Currently, this will make roughly half of the
meta data lookups run at SSD speed, but there is a pending patch to
allow all the metadata reads to go to the SSD. This option is, of
course, only useful for speeding up metadata operations. It can make
large btrfs filesystems feel much more responsive in interactive use
however.
--
gpg @keyserver.linux.it: Goffredo Baroncelli <kreijackATinwind.it>
Key fingerprint BBF5 1610 0B64 DAC6 5F7D 17B2 0EDA 9B37 8B82 E0B5
