btrfs: add allocator_hint mode

When this mode is enabled, the chunk allocation policy is modified as
follows:

Each disk may have a different tag:
- BTRFS_DEV_ALLOCATION_PREFERRED_METADATA
- BTRFS_DEV_ALLOCATION_METADATA_ONLY
- BTRFS_DEV_ALLOCATION_DATA_ONLY
- BTRFS_DEV_ALLOCATION_PREFERRED_DATA (default)

Where:
- ALLOCATION_PREFERRED_X means that it is preferred to use this disk
  for the X chunk type (the other type may be allowed when the space is
  low)
- ALLOCATION_X_ONLY means that it is used *only* for the X chunk type.
  This means also that it is a preferred choice.

Each time the allocator allocates a chunk of type X, first it takes the
disks tagged as ALLOCATION_X_ONLY or ALLOCATION_PREFERRED_X.

If the space is not enough, it uses also the disks tagged as
ALLOCATION_METADATA_ONLY.

If the space is not enough, it uses also the other disks, with the
exception of the one marked as ALLOCATION_PREFERRED_Y, where Y is the
other type of chunk (i.e. not X).

Co-authored-by: Goffredo Baroncelli <kreijack@inwind.it>
Signed-off-by: Kai Krakow <kai@kaishome.de>

Author: kakra

fs/btrfs/volumes.c

@@ -184,6 +184,21 @@ enum btrfs_raid_types __attribute_const__ btrfs_bg_flags_to_raid_index(u64 flags
 	return BTRFS_BG_FLAG_TO_INDEX(profile);
 }
 
+#ifdef CONFIG_BTRFS_ALLOCATOR_HINTS
+#define BTRFS_DEV_ALLOCATION_MASK ((1ULL << \
+		BTRFS_DEV_ALLOCATION_MASK_BIT_COUNT) - 1)
+#define BTRFS_DEV_ALLOCATION_MASK_COUNT (1ULL << \
+		BTRFS_DEV_ALLOCATION_MASK_BIT_COUNT)
+
+static const char alloc_hint_map[BTRFS_DEV_ALLOCATION_MASK_COUNT] = {
+	[BTRFS_DEV_ALLOCATION_DATA_ONLY] = -1,
+	[BTRFS_DEV_ALLOCATION_PREFERRED_DATA] = 0,
+	[BTRFS_DEV_ALLOCATION_PREFERRED_METADATA] = 1,
+	[BTRFS_DEV_ALLOCATION_METADATA_ONLY] = 2,
+	/* the other values are set to 0 */
+};
+#endif
+
 const char *btrfs_bg_type_to_raid_name(u64 flags)
 {
 	const int index = btrfs_bg_flags_to_raid_index(flags);
@@ -5089,13 +5104,20 @@ static int btrfs_add_system_chunk(struct btrfs_fs_info *fs_info,
 }
 
 /*
- * sort the devices in descending order by max_avail, total_avail
+ * sort the devices in descending order by alloc_hint (optional),
+ * max_avail, total_avail
  */
 static int btrfs_cmp_device_info(const void *a, const void *b)
 {
 	const struct btrfs_device_info *di_a = a;
 	const struct btrfs_device_info *di_b = b;
 
+#ifdef CONFIG_BTRFS_ALLOCATOR_HINTS
+	if (di_a->alloc_hint > di_b->alloc_hint)
+		return -1;
+	if (di_a->alloc_hint < di_b->alloc_hint)
+		return 1;
+#endif
 	if (di_a->max_avail > di_b->max_avail)
 		return -1;
 	if (di_a->max_avail < di_b->max_avail)
@@ -5303,16 +5325,97 @@ static int gather_device_info(struct btrfs_fs_devices *fs_devices,
 		devices_info[ndevs].max_avail = max_avail;
 		devices_info[ndevs].total_avail = total_avail;
 		devices_info[ndevs].dev = device;
+
+#ifdef CONFIG_BTRFS_ALLOCATOR_HINTS
+		if ((ctl->type & BTRFS_BLOCK_GROUP_DATA) &&
+		     (ctl->type & BTRFS_BLOCK_GROUP_METADATA)) {
+			/*
+			 * if mixed bg set all the alloc_hint
+			 * fields to the same value, so the sorting
+			 * is not affected
+			 */
+			devices_info[ndevs].alloc_hint = 0;
+		} else if (ctl->type & BTRFS_BLOCK_GROUP_DATA) {
+			int hint = device->type & BTRFS_DEV_ALLOCATION_MASK;
+
+			/*
+			 * skip BTRFS_DEV_METADATA_ONLY disks
+			 */
+			if (hint == BTRFS_DEV_ALLOCATION_METADATA_ONLY)
+				continue;
+			/*
+			 * if a data chunk must be allocated,
+			 * sort also by hint (data disk
+			 * higher priority)
+			 */
+			devices_info[ndevs].alloc_hint = -alloc_hint_map[hint];
+		} else { /* BTRFS_BLOCK_GROUP_METADATA */
+			int hint = device->type & BTRFS_DEV_ALLOCATION_MASK;
+
+			/*
+			 * skip BTRFS_DEV_DATA_ONLY disks
+			 */
+			if (hint == BTRFS_DEV_ALLOCATION_DATA_ONLY)
+				continue;
+			/*
+			 * if a data chunk must be allocated,
+			 * sort also by hint (metadata hint
+			 * higher priority)
+			 */
+			devices_info[ndevs].alloc_hint = alloc_hint_map[hint];
+		}
+#endif
+
 		++ndevs;
 	}
 	ctl->ndevs = ndevs;
 
+#ifdef CONFIG_BTRFS_ALLOCATOR_HINTS
+	/*
+	 * no devices available
+	 */
+	if (!ndevs)
+		return 0;
+#endif
+
 	/*
 	 * now sort the devices by hole size / available space
 	 */
 	sort(devices_info, ndevs, sizeof(struct btrfs_device_info),
 	     btrfs_cmp_device_info, NULL);
 
+#ifdef CONFIG_BTRFS_ALLOCATOR_HINTS
+	/*
+	 * select the minimum set of disks grouped by hint that
+	 * can host the chunk
+	 */
+	ndevs = 0;
+	while (ndevs < ctl->ndevs) {
+		int hint = devices_info[ndevs++].alloc_hint;
+		while (ndevs < ctl->ndevs &&
+		       devices_info[ndevs].alloc_hint == hint)
+				ndevs++;
+		if (ndevs >= ctl->devs_min)
+			break;
+	}
+
+	BUG_ON(ndevs > ctl->ndevs);
+	ctl->ndevs = ndevs;
+
+	/*
+	 * the next layers require the devices_info ordered by
+	 * max_avail. If we are returing two (or more) different
+	 * group of alloc_hint, this is not always true. So sort
+	 * these gain.
+	 */
+
+	for (int i = 0 ; i < ndevs ; i++)
+		devices_info[i].alloc_hint = 0;
+
+	sort(devices_info, ndevs, sizeof(struct btrfs_device_info),
+	     btrfs_cmp_device_info, NULL);
+#endif
+
 	return 0;
 }
 

fs/btrfs/volumes.h

@@ -599,6 +599,9 @@ struct btrfs_device_info {
 	u64 dev_offset;
 	u64 max_avail;
 	u64 total_avail;
+#ifdef CONFIG_BTRFS_ALLOCATOR_HINTS
+	int alloc_hint;
+#endif
 };
 
 struct btrfs_raid_attr {