Commit 39ba1dc8 authored by Christoph Hellwig's avatar Christoph Hellwig Committed by Long Li
Browse files

btrfs: zoned: introduce a zone_info struct in btrfs_load_block_group_zone_info

mainline inclusion
from mainline-v6.6-rc5
commit 15c12fcc50a1b12a747f8b6ec05cdb18c537a4d1
category: bugfix
bugzilla: https://gitee.com/src-openeuler/kernel/issues/IACZX0
CVE: CVE-2024-39496

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=15c12fcc50a1b12a747f8b6ec05cdb18c537a4d1



--------------------------------

Add a new zone_info structure to hold per-zone information in
btrfs_load_block_group_zone_info and prepare for breaking out helpers
from it.

Reviewed-by: default avatarJohannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: default avatarChristoph Hellwig <hch@lst.de>
Reviewed-by: default avatarDavid Sterba <dsterba@suse.com>
Signed-off-by: default avatarDavid Sterba <dsterba@suse.com>
Signed-off-by: default avatarLong Li <leo.lilong@huawei.com>
parent de51c3bc
Loading
Loading
Loading
Loading
+37 −47
Original line number Diff line number Diff line
@@ -1282,6 +1282,12 @@ static int calculate_alloc_pointer(struct btrfs_block_group *cache,
	return ret;
}

struct zone_info {
	u64 physical;
	u64 capacity;
	u64 alloc_offset;
};

int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
{
	struct btrfs_fs_info *fs_info = cache->fs_info;
@@ -1291,12 +1297,10 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
	struct btrfs_device *device;
	u64 logical = cache->start;
	u64 length = cache->length;
	struct zone_info *zone_info = NULL;
	int ret;
	int i;
	unsigned int nofs_flag;
	u64 *alloc_offsets = NULL;
	u64 *caps = NULL;
	u64 *physical = NULL;
	unsigned long *active = NULL;
	u64 last_alloc = 0;
	u32 num_sequential = 0, num_conventional = 0;
@@ -1328,20 +1332,8 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
		goto out;
	}

	alloc_offsets = kcalloc(map->num_stripes, sizeof(*alloc_offsets), GFP_NOFS);
	if (!alloc_offsets) {
		ret = -ENOMEM;
		goto out;
	}

	caps = kcalloc(map->num_stripes, sizeof(*caps), GFP_NOFS);
	if (!caps) {
		ret = -ENOMEM;
		goto out;
	}

	physical = kcalloc(map->num_stripes, sizeof(*physical), GFP_NOFS);
	if (!physical) {
	zone_info = kcalloc(map->num_stripes, sizeof(*zone_info), GFP_NOFS);
	if (!zone_info) {
		ret = -ENOMEM;
		goto out;
	}
@@ -1353,20 +1345,21 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
	}

	for (i = 0; i < map->num_stripes; i++) {
		struct zone_info *info = &zone_info[i];
		bool is_sequential;
		struct blk_zone zone;
		struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
		int dev_replace_is_ongoing = 0;

		device = map->stripes[i].dev;
		physical[i] = map->stripes[i].physical;
		info->physical = map->stripes[i].physical;

		if (device->bdev == NULL) {
			alloc_offsets[i] = WP_MISSING_DEV;
			info->alloc_offset = WP_MISSING_DEV;
			continue;
		}

		is_sequential = btrfs_dev_is_sequential(device, physical[i]);
		is_sequential = btrfs_dev_is_sequential(device, info->physical);
		if (is_sequential)
			num_sequential++;
		else
@@ -1380,7 +1373,7 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
			__set_bit(i, active);

		if (!is_sequential) {
			alloc_offsets[i] = WP_CONVENTIONAL;
			info->alloc_offset = WP_CONVENTIONAL;
			continue;
		}

@@ -1388,25 +1381,25 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
		 * This zone will be used for allocation, so mark this zone
		 * non-empty.
		 */
		btrfs_dev_clear_zone_empty(device, physical[i]);
		btrfs_dev_clear_zone_empty(device, info->physical);

		down_read(&dev_replace->rwsem);
		dev_replace_is_ongoing = btrfs_dev_replace_is_ongoing(dev_replace);
		if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL)
			btrfs_dev_clear_zone_empty(dev_replace->tgtdev, physical[i]);
			btrfs_dev_clear_zone_empty(dev_replace->tgtdev, info->physical);
		up_read(&dev_replace->rwsem);

		/*
		 * The group is mapped to a sequential zone. Get the zone write
		 * pointer to determine the allocation offset within the zone.
		 */
		WARN_ON(!IS_ALIGNED(physical[i], fs_info->zone_size));
		WARN_ON(!IS_ALIGNED(info->physical, fs_info->zone_size));
		nofs_flag = memalloc_nofs_save();
		ret = btrfs_get_dev_zone(device, physical[i], &zone);
		ret = btrfs_get_dev_zone(device, info->physical, &zone);
		memalloc_nofs_restore(nofs_flag);
		if (ret == -EIO || ret == -EOPNOTSUPP) {
			ret = 0;
			alloc_offsets[i] = WP_MISSING_DEV;
			info->alloc_offset = WP_MISSING_DEV;
			continue;
		} else if (ret) {
			goto out;
@@ -1421,27 +1414,26 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
			goto out;
		}

		caps[i] = (zone.capacity << SECTOR_SHIFT);
		info->capacity = (zone.capacity << SECTOR_SHIFT);

		switch (zone.cond) {
		case BLK_ZONE_COND_OFFLINE:
		case BLK_ZONE_COND_READONLY:
			btrfs_err(fs_info,
		"zoned: offline/readonly zone %llu on device %s (devid %llu)",
				  physical[i] >> device->zone_info->zone_size_shift,
				  info->physical >> device->zone_info->zone_size_shift,
				  rcu_str_deref(device->name), device->devid);
			alloc_offsets[i] = WP_MISSING_DEV;
			info->alloc_offset = WP_MISSING_DEV;
			break;
		case BLK_ZONE_COND_EMPTY:
			alloc_offsets[i] = 0;
			info->alloc_offset = 0;
			break;
		case BLK_ZONE_COND_FULL:
			alloc_offsets[i] = caps[i];
			info->alloc_offset = info->capacity;
			break;
		default:
			/* Partially used zone */
			alloc_offsets[i] =
					((zone.wp - zone.start) << SECTOR_SHIFT);
			info->alloc_offset = ((zone.wp - zone.start) << SECTOR_SHIFT);
			__set_bit(i, active);
			break;
		}
@@ -1468,15 +1460,15 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)

	switch (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
	case 0: /* single */
		if (alloc_offsets[0] == WP_MISSING_DEV) {
		if (zone_info[0].alloc_offset == WP_MISSING_DEV) {
			btrfs_err(fs_info,
			"zoned: cannot recover write pointer for zone %llu",
				physical[0]);
				zone_info[0].physical);
			ret = -EIO;
			goto out;
		}
		cache->alloc_offset = alloc_offsets[0];
		cache->zone_capacity = caps[0];
		cache->alloc_offset = zone_info[0].alloc_offset;
		cache->zone_capacity = zone_info[0].capacity;
		if (test_bit(0, active))
			set_bit(BLOCK_GROUP_FLAG_ZONE_IS_ACTIVE, &cache->runtime_flags);
		break;
@@ -1486,21 +1478,21 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
			ret = -EINVAL;
			goto out;
		}
		if (alloc_offsets[0] == WP_MISSING_DEV) {
		if (zone_info[0].alloc_offset == WP_MISSING_DEV) {
			btrfs_err(fs_info,
			"zoned: cannot recover write pointer for zone %llu",
				physical[0]);
				zone_info[0].physical);
			ret = -EIO;
			goto out;
		}
		if (alloc_offsets[1] == WP_MISSING_DEV) {
		if (zone_info[1].alloc_offset == WP_MISSING_DEV) {
			btrfs_err(fs_info,
			"zoned: cannot recover write pointer for zone %llu",
				physical[1]);
				zone_info[1].physical);
			ret = -EIO;
			goto out;
		}
		if (alloc_offsets[0] != alloc_offsets[1]) {
		if (zone_info[0].alloc_offset != zone_info[1].alloc_offset) {
			btrfs_err(fs_info,
			"zoned: write pointer offset mismatch of zones in DUP profile");
			ret = -EIO;
@@ -1516,8 +1508,8 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
				set_bit(BLOCK_GROUP_FLAG_ZONE_IS_ACTIVE,
					&cache->runtime_flags);
		}
		cache->alloc_offset = alloc_offsets[0];
		cache->zone_capacity = min(caps[0], caps[1]);
		cache->alloc_offset = zone_info[0].alloc_offset;
		cache->zone_capacity = min(zone_info[0].capacity, zone_info[1].capacity);
		break;
	case BTRFS_BLOCK_GROUP_RAID1:
	case BTRFS_BLOCK_GROUP_RAID0:
@@ -1570,9 +1562,7 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
		cache->physical_map = NULL;
	}
	bitmap_free(active);
	kfree(physical);
	kfree(caps);
	kfree(alloc_offsets);
	kfree(zone_info);
	free_extent_map(em);

	return ret;