1 files changed, 291 insertions, 72 deletions

diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index b843308e2bc6..1f64c132b387 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -10,6 +10,7 @@
 #include <linux/completion.h>
 #include <linux/bug.h>
 #include <linux/list.h>
+#include <linux/string_choices.h>
 #include <crypto/hash.h>
 #include "messages.h"
 #include "ctree.h"
@@ -25,6 +26,7 @@
 #include "misc.h"
 #include "fs.h"
 #include "accessors.h"
+#include "zoned.h"
 
 /*
  * Structure name                       Path
@@ -160,8 +162,7 @@ static int can_modify_feature(struct btrfs_feature_attr *fa)
 		clear = BTRFS_FEATURE_INCOMPAT_SAFE_CLEAR;
 		break;
 	default:
-		pr_warn("btrfs: sysfs: unknown feature set %d\n",
-				fa->feature_set);
+		btrfs_warn(NULL, "sysfs: unknown feature set %d", fa->feature_set);
 		return 0;
 	}
 
@@ -295,7 +296,7 @@ BTRFS_FEAT_ATTR_INCOMPAT(simple_quota, SIMPLE_QUOTA);
 #ifdef CONFIG_BLK_DEV_ZONED
 BTRFS_FEAT_ATTR_INCOMPAT(zoned, ZONED);
 #endif
-#ifdef CONFIG_BTRFS_DEBUG
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
 /* Remove once support for extent tree v2 is feature complete */
 BTRFS_FEAT_ATTR_INCOMPAT(extent_tree_v2, EXTENT_TREE_V2);
 /* Remove once support for raid stripe tree is feature complete. */
@@ -329,7 +330,7 @@ static struct attribute *btrfs_supported_feature_attrs[] = {
 #ifdef CONFIG_BLK_DEV_ZONED
 	BTRFS_FEAT_ATTR_PTR(zoned),
 #endif
-#ifdef CONFIG_BTRFS_DEBUG
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
 	BTRFS_FEAT_ATTR_PTR(extent_tree_v2),
 	BTRFS_FEAT_ATTR_PTR(raid_stripe_tree),
 #endif
@@ -410,12 +411,17 @@ static ssize_t supported_sectorsizes_show(struct kobject *kobj,
 					  char *buf)
 {
 	ssize_t ret = 0;
+	bool has_output = false;
 
-	/* An artificial limit to only support 4K and PAGE_SIZE */
-	if (PAGE_SIZE > SZ_4K)
-		ret += sysfs_emit_at(buf, ret, "%u ", SZ_4K);
-	ret += sysfs_emit_at(buf, ret, "%lu\n", PAGE_SIZE);
-
+	for (u32 cur = BTRFS_MIN_BLOCKSIZE; cur <= BTRFS_MAX_BLOCKSIZE; cur *= 2) {
+		if (!btrfs_supported_blocksize(cur))
+			continue;
+		if (has_output)
+			ret += sysfs_emit_at(buf, ret, " ");
+		ret += sysfs_emit_at(buf, ret, "%u", cur);
+		has_output = true;
+	}
+	ret += sysfs_emit_at(buf, ret, "\n");
 	return ret;
 }
 BTRFS_ATTR(static_feature, supported_sectorsizes,
@@ -1118,7 +1124,7 @@ static ssize_t btrfs_nodesize_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return sysfs_emit(buf, "%u\n", fs_info->super_copy->nodesize);
+	return sysfs_emit(buf, "%u\n", fs_info->nodesize);
 }
 
 BTRFS_ATTR(, nodesize, btrfs_nodesize_show);
@@ -1128,7 +1134,7 @@ static ssize_t btrfs_sectorsize_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return sysfs_emit(buf, "%u\n", fs_info->super_copy->sectorsize);
+	return sysfs_emit(buf, "%u\n", fs_info->sectorsize);
 }
 
 BTRFS_ATTR(, sectorsize, btrfs_sectorsize_show);
@@ -1137,13 +1143,21 @@ static ssize_t btrfs_commit_stats_show(struct kobject *kobj,
 				       struct kobj_attribute *a, char *buf)
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
+	u64 now = ktime_get_ns();
+	u64 start_time = fs_info->commit_stats.critical_section_start_time;
+	u64 pending = 0;
+
+	if (start_time)
+		pending = now - start_time;
 
 	return sysfs_emit(buf,
 		"commits %llu\n"
+		"cur_commit_ms %llu\n"
 		"last_commit_ms %llu\n"
 		"max_commit_ms %llu\n"
 		"total_commit_ms %llu\n",
 		fs_info->commit_stats.commit_count,
+		div_u64(pending, NSEC_PER_MSEC),
 		div_u64(fs_info->commit_stats.last_commit_dur, NSEC_PER_MSEC),
 		div_u64(fs_info->commit_stats.max_commit_dur, NSEC_PER_MSEC),
 		div_u64(fs_info->commit_stats.total_commit_dur, NSEC_PER_MSEC));
@@ -1175,12 +1189,62 @@ static ssize_t btrfs_commit_stats_store(struct kobject *kobj,
 }
 BTRFS_ATTR_RW(, commit_stats, btrfs_commit_stats_show, btrfs_commit_stats_store);
 
+static ssize_t btrfs_zoned_stats_show(struct kobject *kobj,
+				      struct kobj_attribute *a, char *buf)
+{
+	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
+	struct btrfs_block_group *bg;
+	size_t ret = 0;
+
+
+	if (!btrfs_is_zoned(fs_info))
+		return ret;
+
+	spin_lock(&fs_info->zone_active_bgs_lock);
+	ret += sysfs_emit_at(buf, ret, "active block-groups: %zu\n",
+			     list_count_nodes(&fs_info->zone_active_bgs));
+	spin_unlock(&fs_info->zone_active_bgs_lock);
+
+	mutex_lock(&fs_info->reclaim_bgs_lock);
+	spin_lock(&fs_info->unused_bgs_lock);
+	ret += sysfs_emit_at(buf, ret, "\treclaimable: %zu\n",
+			     list_count_nodes(&fs_info->reclaim_bgs));
+	ret += sysfs_emit_at(buf, ret, "\tunused: %zu\n",
+			     list_count_nodes(&fs_info->unused_bgs));
+	spin_unlock(&fs_info->unused_bgs_lock);
+	mutex_unlock(&fs_info->reclaim_bgs_lock);
+
+	ret += sysfs_emit_at(buf, ret, "\tneed reclaim: %s\n",
+			     str_true_false(btrfs_zoned_should_reclaim(fs_info)));
+
+	if (fs_info->data_reloc_bg)
+		ret += sysfs_emit_at(buf, ret,
+				     "data relocation block-group: %llu\n",
+				     fs_info->data_reloc_bg);
+	if (fs_info->treelog_bg)
+		ret += sysfs_emit_at(buf, ret,
+				     "tree-log block-group: %llu\n",
+				     fs_info->treelog_bg);
+
+	spin_lock(&fs_info->zone_active_bgs_lock);
+	ret += sysfs_emit_at(buf, ret, "active zones:\n");
+	list_for_each_entry(bg, &fs_info->zone_active_bgs, active_bg_list) {
+		ret += sysfs_emit_at(buf, ret,
+				     "\tstart: %llu, wp: %llu used: %llu, reserved: %llu, unusable: %llu\n",
+				     bg->start, bg->alloc_offset, bg->used,
+				     bg->reserved, bg->zone_unusable);
+	}
+	spin_unlock(&fs_info->zone_active_bgs_lock);
+	return ret;
+}
+BTRFS_ATTR(, zoned_stats, btrfs_zoned_stats_show);
+
 static ssize_t btrfs_clone_alignment_show(struct kobject *kobj,
 				struct kobj_attribute *a, char *buf)
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return sysfs_emit(buf, "%u\n", fs_info->super_copy->sectorsize);
+	return sysfs_emit(buf, "%u\n", fs_info->sectorsize);
 }
 
 BTRFS_ATTR(, clone_alignment, btrfs_clone_alignment_show);
@@ -1201,7 +1265,7 @@ static ssize_t quota_override_store(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 	unsigned long knob;
-	int err;
+	int ret;
 
 	if (!fs_info)
 		return -EPERM;
@@ -1209,9 +1273,9 @@ static ssize_t quota_override_store(struct kobject *kobj,
 	if (!capable(CAP_SYS_RESOURCE))
 		return -EPERM;
 
-	err = kstrtoul(buf, 10, &knob);
-	if (err)
-		return err;
+	ret = kstrtoul(buf, 10, &knob);
+	if (ret)
+		return ret;
 	if (knob > 1)
 		return -EINVAL;
 
@@ -1305,7 +1369,74 @@ static ssize_t btrfs_temp_fsid_show(struct kobject *kobj,
 }
 BTRFS_ATTR(, temp_fsid, btrfs_temp_fsid_show);
 
-static const char * const btrfs_read_policy_name[] = { "pid" };
+static const char *btrfs_read_policy_name[] = {
+	"pid",
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
+	"round-robin",
+	"devid",
+#endif
+};
+
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
+
+/* Global module configuration parameters. */
+static char *read_policy;
+char *btrfs_get_mod_read_policy(void)
+{
+	return read_policy;
+}
+
+/* Set perms to 0, disable /sys/module/btrfs/parameter/read_policy interface. */
+module_param(read_policy, charp, 0);
+MODULE_PARM_DESC(read_policy,
+"Global read policy: pid (default), round-robin[:<min_contig_read>], devid[:<devid>]");
+#endif
+
+int btrfs_read_policy_to_enum(const char *str, s64 *value_ret)
+{
+	char param[32];
+	char __maybe_unused *value_str;
+
+	if (!str || strlen(str) == 0)
+		return 0;
+
+	strscpy(param, str);
+
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
+	/* Separate value from input in policy:value format. */
+	value_str = strchr(param, ':');
+	if (value_str) {
+		char *retptr;
+
+		*value_str = 0;
+		value_str++;
+		if (!value_ret)
+			return -EINVAL;
+
+		*value_ret = memparse(value_str, &retptr);
+		/* There could be any trailing typos after the value. */
+		retptr = skip_spaces(retptr);
+		if (*retptr != 0 || *value_ret <= 0)
+			return -EINVAL;
+	}
+#endif
+
+	return sysfs_match_string(btrfs_read_policy_name, param);
+}
+
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
+int __init btrfs_read_policy_init(void)
+{
+	s64 value;
+
+	if (btrfs_read_policy_to_enum(read_policy, &value) == -EINVAL) {
+		btrfs_err(NULL, "invalid read policy or value %s", read_policy);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+#endif
 
 static ssize_t btrfs_read_policy_show(struct kobject *kobj,
 				      struct kobj_attribute *a, char *buf)
@@ -1316,14 +1447,25 @@ static ssize_t btrfs_read_policy_show(struct kobject *kobj,
 	int i;
 
 	for (i = 0; i < BTRFS_NR_READ_POLICY; i++) {
-		if (policy == i)
-			ret += sysfs_emit_at(buf, ret, "%s[%s]",
-					 (ret == 0 ? "" : " "),
-					 btrfs_read_policy_name[i]);
-		else
-			ret += sysfs_emit_at(buf, ret, "%s%s",
-					 (ret == 0 ? "" : " "),
-					 btrfs_read_policy_name[i]);
+		if (ret != 0)
+			ret += sysfs_emit_at(buf, ret, " ");
+
+		if (i == policy)
+			ret += sysfs_emit_at(buf, ret, "[");
+
+		ret += sysfs_emit_at(buf, ret, "%s", btrfs_read_policy_name[i]);
+
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
+		if (i == BTRFS_READ_POLICY_RR)
+			ret += sysfs_emit_at(buf, ret, ":%u",
+					     READ_ONCE(fs_devices->rr_min_contig_read));
+
+		if (i == BTRFS_READ_POLICY_DEVID)
+			ret += sysfs_emit_at(buf, ret, ":%llu",
+					     READ_ONCE(fs_devices->read_devid));
+#endif
+		if (i == policy)
+			ret += sysfs_emit_at(buf, ret, "]");
 	}
 
 	ret += sysfs_emit_at(buf, ret, "\n");
@@ -1336,21 +1478,80 @@ static ssize_t btrfs_read_policy_store(struct kobject *kobj,
 				       const char *buf, size_t len)
 {
 	struct btrfs_fs_devices *fs_devices = to_fs_devs(kobj);
-	int i;
+	int index;
+	s64 value = -1;
 
-	for (i = 0; i < BTRFS_NR_READ_POLICY; i++) {
-		if (sysfs_streq(buf, btrfs_read_policy_name[i])) {
-			if (i != READ_ONCE(fs_devices->read_policy)) {
-				WRITE_ONCE(fs_devices->read_policy, i);
-				btrfs_info(fs_devices->fs_info,
-					   "read policy set to '%s'",
-					   btrfs_read_policy_name[i]);
+	index = btrfs_read_policy_to_enum(buf, &value);
+	if (index < 0)
+		return -EINVAL;
+
+#ifdef CONFIG_BTRFS_EXPERIMENTAL
+	/* If moving from RR then disable collecting fs stats. */
+	if (fs_devices->read_policy == BTRFS_READ_POLICY_RR && index != BTRFS_READ_POLICY_RR)
+		fs_devices->collect_fs_stats = false;
+
+	if (index == BTRFS_READ_POLICY_RR) {
+		if (value != -1) {
+			const u32 sectorsize = fs_devices->fs_info->sectorsize;
+
+			if (!IS_ALIGNED(value, sectorsize)) {
+				u64 temp_value = round_up(value, sectorsize);
+
+				btrfs_debug(fs_devices->fs_info,
+"read_policy: min contig read %lld should be multiple of sectorsize %u, rounded to %llu",
+					  value, sectorsize, temp_value);
+				value = temp_value;
 			}
-			return len;
+		} else {
+			value = BTRFS_DEFAULT_RR_MIN_CONTIG_READ;
 		}
+
+		if (index != READ_ONCE(fs_devices->read_policy) ||
+		    value != READ_ONCE(fs_devices->rr_min_contig_read)) {
+			WRITE_ONCE(fs_devices->read_policy, index);
+			WRITE_ONCE(fs_devices->rr_min_contig_read, value);
+
+			btrfs_info(fs_devices->fs_info, "read policy set to '%s:%lld'",
+				   btrfs_read_policy_name[index], value);
+		}
+
+		fs_devices->collect_fs_stats = true;
+
+		return len;
 	}
 
-	return -EINVAL;
+	if (index == BTRFS_READ_POLICY_DEVID) {
+		if (value != -1) {
+			BTRFS_DEV_LOOKUP_ARGS(args);
+
+			/* Validate input devid. */
+			args.devid = value;
+			if (btrfs_find_device(fs_devices, &args) == NULL)
+				return -EINVAL;
+		} else {
+			/* Set default devid to the devid of the latest device. */
+			value = fs_devices->latest_dev->devid;
+		}
+
+		if (index != READ_ONCE(fs_devices->read_policy) ||
+		    value != READ_ONCE(fs_devices->read_devid)) {
+			WRITE_ONCE(fs_devices->read_policy, index);
+			WRITE_ONCE(fs_devices->read_devid, value);
+
+			btrfs_info(fs_devices->fs_info, "read policy set to '%s:%llu'",
+				   btrfs_read_policy_name[index], value);
+		}
+
+		return len;
+	}
+#endif
+	if (index != READ_ONCE(fs_devices->read_policy)) {
+		WRITE_ONCE(fs_devices->read_policy, index);
+		btrfs_info(fs_devices->fs_info, "read policy set to '%s'",
+			   btrfs_read_policy_name[index]);
+	}
+
+	return len;
 }
 BTRFS_ATTR_RW(, read_policy, btrfs_read_policy_show, btrfs_read_policy_store);
 
@@ -1450,6 +1651,7 @@ static const struct attribute *btrfs_attrs[] = {
 	BTRFS_ATTR_PTR(, bg_reclaim_threshold),
 	BTRFS_ATTR_PTR(, commit_stats),
 	BTRFS_ATTR_PTR(, temp_fsid),
+	BTRFS_ATTR_PTR(, zoned_stats),
 #ifdef CONFIG_BTRFS_EXPERIMENTAL
 	BTRFS_ATTR_PTR(, offload_csum),
 #endif
@@ -1792,16 +1994,35 @@ void btrfs_sysfs_remove_space_info(struct btrfs_space_info *space_info)
 	kobject_put(&space_info->kobj);
 }
 
-static const char *alloc_name(u64 flags)
+static const char *alloc_name(struct btrfs_space_info *space_info)
 {
+	u64 flags = space_info->flags;
+
 	switch (flags) {
 	case BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA:
 		return "mixed";
 	case BTRFS_BLOCK_GROUP_METADATA:
-		return "metadata";
+		switch (space_info->subgroup_id) {
+		case BTRFS_SUB_GROUP_PRIMARY:
+			return "metadata";
+		case BTRFS_SUB_GROUP_TREELOG:
+			return "metadata-treelog";
+		default:
+			WARN_ON_ONCE(1);
+			return "metadata (unknown sub-group)";
+		}
 	case BTRFS_BLOCK_GROUP_DATA:
-		return "data";
+		switch (space_info->subgroup_id) {
+		case BTRFS_SUB_GROUP_PRIMARY:
+			return "data";
+		case BTRFS_SUB_GROUP_DATA_RELOC:
+			return "data-reloc";
+		default:
+			WARN_ON_ONCE(1);
+			return "data (unknown sub-group)";
+		}
 	case BTRFS_BLOCK_GROUP_SYSTEM:
+		ASSERT(space_info->subgroup_id == BTRFS_SUB_GROUP_PRIMARY);
 		return "system";
 	default:
 		WARN_ON(1);
@@ -1813,14 +2034,13 @@ static const char *alloc_name(u64 flags)
  * Create a sysfs entry for a space info type at path
  * /sys/fs/btrfs/UUID/allocation/TYPE
  */
-int btrfs_sysfs_add_space_info_type(struct btrfs_fs_info *fs_info,
-				    struct btrfs_space_info *space_info)
+int btrfs_sysfs_add_space_info_type(struct btrfs_space_info *space_info)
 {
 	int ret;
 
 	ret = kobject_init_and_add(&space_info->kobj, &space_info_ktype,
-				   fs_info->space_info_kobj, "%s",
-				   alloc_name(space_info->flags));
+				   space_info->fs_info->space_info_kobj, "%s",
+				   alloc_name(space_info));
 	if (ret) {
 		kobject_put(&space_info->kobj);
 		return ret;
@@ -2082,7 +2302,7 @@ void btrfs_kobject_uevent(struct block_device *bdev, enum kobject_action action)
 
 	ret = kobject_uevent(&disk_to_dev(bdev->bd_disk)->kobj, action);
 	if (ret)
-		pr_warn("BTRFS: Sending event '%d' to kobject: '%s' (%p): failed\n",
+		btrfs_warn(NULL, "sending event %d to kobject: '%s' (%p): failed",
 			action, kobject_name(&disk_to_dev(bdev->bd_disk)->kobj),
 			&disk_to_dev(bdev->bd_disk)->kobj);
 }
@@ -2125,15 +2345,15 @@ static struct kset *btrfs_kset;
  */
 int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs)
 {
-	int error;
+	int ret;
 
 	init_completion(&fs_devs->kobj_unregister);
 	fs_devs->fsid_kobj.kset = btrfs_kset;
-	error = kobject_init_and_add(&fs_devs->fsid_kobj, &btrfs_ktype, NULL,
-				     "%pU", fs_devs->fsid);
-	if (error) {
+	ret = kobject_init_and_add(&fs_devs->fsid_kobj, &btrfs_ktype, NULL,
+				   "%pU", fs_devs->fsid);
+	if (ret) {
 		kobject_put(&fs_devs->fsid_kobj);
-		return error;
+		return ret;
 	}
 
 	fs_devs->devices_kobj = kobject_create_and_add("devices",
@@ -2159,71 +2379,70 @@ int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs)
 
 int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info)
 {
-	int error;
+	int ret;
 	struct btrfs_fs_devices *fs_devs = fs_info->fs_devices;
 	struct kobject *fsid_kobj = &fs_devs->fsid_kobj;
 
-	error = btrfs_sysfs_add_fs_devices(fs_devs);
-	if (error)
-		return error;
+	ret = btrfs_sysfs_add_fs_devices(fs_devs);
+	if (ret)
+		return ret;
 
-	error = sysfs_create_files(fsid_kobj, btrfs_attrs);
-	if (error) {
+	ret = sysfs_create_files(fsid_kobj, btrfs_attrs);
+	if (ret) {
 		btrfs_sysfs_remove_fs_devices(fs_devs);
-		return error;
+		return ret;
 	}
 
-	error = sysfs_create_group(fsid_kobj,
-				   &btrfs_feature_attr_group);
-	if (error)
+	ret = sysfs_create_group(fsid_kobj, &btrfs_feature_attr_group);
+	if (ret)
 		goto failure;
 
 #ifdef CONFIG_BTRFS_DEBUG
 	fs_info->debug_kobj = kobject_create_and_add("debug", fsid_kobj);
 	if (!fs_info->debug_kobj) {
-		error = -ENOMEM;
+		ret = -ENOMEM;
 		goto failure;
 	}
 
-	error = sysfs_create_files(fs_info->debug_kobj, btrfs_debug_mount_attrs);
-	if (error)
+	ret = sysfs_create_files(fs_info->debug_kobj, btrfs_debug_mount_attrs);
+	if (ret)
 		goto failure;
 #endif
 
 	/* Discard directory */
 	fs_info->discard_kobj = kobject_create_and_add("discard", fsid_kobj);
 	if (!fs_info->discard_kobj) {
-		error = -ENOMEM;
+		ret = -ENOMEM;
 		goto failure;
 	}
 
-	error = sysfs_create_files(fs_info->discard_kobj, discard_attrs);
-	if (error)
+	ret = sysfs_create_files(fs_info->discard_kobj, discard_attrs);
+	if (ret)
 		goto failure;
 
-	error = addrm_unknown_feature_attrs(fs_info, true);
-	if (error)
+	ret = addrm_unknown_feature_attrs(fs_info, true);
+	if (ret)
 		goto failure;
 
-	error = sysfs_create_link(fsid_kobj, &fs_info->sb->s_bdi->dev->kobj, "bdi");
-	if (error)
+	ret = sysfs_create_link(fsid_kobj, &fs_info->sb->s_bdi->dev->kobj, "bdi");
+	if (ret)
 		goto failure;
 
 	fs_info->space_info_kobj = kobject_create_and_add("allocation",
 						  fsid_kobj);
 	if (!fs_info->space_info_kobj) {
-		error = -ENOMEM;
+		ret = -ENOMEM;
 		goto failure;
 	}
 
-	error = sysfs_create_files(fs_info->space_info_kobj, allocation_attrs);
-	if (error)
+	ret = sysfs_create_files(fs_info->space_info_kobj, allocation_attrs);
+	if (ret)
 		goto failure;
 
 	return 0;
 failure:
 	btrfs_sysfs_remove_mounted(fs_info);
-	return error;
+	return ret;
 }
 
 static ssize_t qgroup_enabled_show(struct kobject *qgroups_kobj,