Merge tag 'bcachefs-2023-10-30' of https://evilpiepirate.org/git/bcachefs

Pull initial bcachefs updates from Kent Overstreet:
 "Here's the bcachefs filesystem pull request.

  One new patch since last week: the exportfs constants ended up
  conflicting with other filesystems that are also getting added to the
  global enum, so switched to new constants picked by Amir.

  The only new non fs/bcachefs/ patch is the objtool patch that adds
  bcachefs functions to the list of noreturns. The patch that exports
  osq_lock() has been dropped for now, per Ingo"

* tag 'bcachefs-2023-10-30' of https://evilpiepirate.org/git/bcachefs: (2781 commits)
  exportfs: Change bcachefs fid_type enum to avoid conflicts
  bcachefs: Refactor memcpy into direct assignment
  bcachefs: Fix drop_alloc_keys()
  bcachefs: snapshot_create_lock
  bcachefs: Fix snapshot skiplists during snapshot deletion
  bcachefs: bch2_sb_field_get() refactoring
  bcachefs: KEY_TYPE_error now counts towards i_sectors
  bcachefs: Fix handling of unknown bkey types
  bcachefs: Switch to unsafe_memcpy() in a few places
  bcachefs: Use struct_size()
  bcachefs: Correctly initialize new buckets on device resize
  bcachefs: Fix another smatch complaint
  bcachefs: Use strsep() in split_devs()
  bcachefs: Add iops fields to bch_member
  bcachefs: Rename bch_sb_field_members -> bch_sb_field_members_v1
  bcachefs: New superblock section members_v2
  bcachefs: Add new helper to retrieve bch_member from sb
  bcachefs: bucket_lock() is now a sleepable lock
  bcachefs: fix crc32c checksum merge byte order problem
  bcachefs: Fix bch2_inode_delete_keys()
  ...

MAINTAINERS

@@ -3482,6 +3482,14 @@ W:	http://bcache.evilpiepirate.org
 C:	irc://irc.oftc.net/bcache
 F:	drivers/md/bcache/
 
+BCACHEFS
+M:	Kent Overstreet <kent.overstreet@linux.dev>
+R:	Brian Foster <bfoster@redhat.com>
+L:	linux-bcachefs@vger.kernel.org
+S:	Supported
+C:	irc://irc.oftc.net/bcache
+F:	fs/bcachefs/
+
 BDISP ST MEDIA DRIVER
 M:	Fabien Dessenne <fabien.dessenne@foss.st.com>
 L:	linux-media@vger.kernel.org
@@ -5068,6 +5076,14 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers/core
 F:	Documentation/devicetree/bindings/timer/
 F:	drivers/clocksource/
 
+CLOSURES
+M:	Kent Overstreet <kent.overstreet@linux.dev>
+L:	linux-bcachefs@vger.kernel.org
+S:	Supported
+C:	irc://irc.oftc.net/bcache
+F:	include/linux/closure.h
+F:	lib/closure.c
+
 CMPC ACPI DRIVER
 M:	Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
 M:	Daniel Oliveira Nascimento <don@syst.com.br>
@@ -8748,6 +8764,13 @@ S:	Supported
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm.git
 F:	drivers/pmdomain/
 
+GENERIC RADIX TREE
+M:	Kent Overstreet <kent.overstreet@linux.dev>
+S:	Supported
+C:	irc://irc.oftc.net/bcache
+F:	include/linux/generic-radix-tree.h
+F:	lib/generic-radix-tree.c
+
 GENERIC RESISTIVE TOUCHSCREEN ADC DRIVER
 M:	Eugen Hristev <eugen.hristev@microchip.com>
 L:	linux-input@vger.kernel.org

drivers/md/bcache/Kconfig

@@ -4,6 +4,7 @@ config BCACHE
 	tristate "Block device as cache"
 	select BLOCK_HOLDER_DEPRECATED if SYSFS
 	select CRC64
+	select CLOSURES
 	help
 	Allows a block device to be used as cache for other devices; uses
 	a btree for indexing and the layout is optimized for SSDs.
@@ -19,15 +20,6 @@ config BCACHE_DEBUG
 	Enables extra debugging tools, allows expensive runtime checks to be
 	turned on.
 
-config BCACHE_CLOSURES_DEBUG
-	bool "Debug closures"
-	depends on BCACHE
-	select DEBUG_FS
-	help
-	Keeps all active closures in a linked list and provides a debugfs
-	interface to list them, which makes it possible to see asynchronous
-	operations that get stuck.
-
 config BCACHE_ASYNC_REGISTRATION
 	bool "Asynchronous device registration"
 	depends on BCACHE

drivers/md/bcache/Makefile

@@ -2,6 +2,6 @@
 
 obj-$(CONFIG_BCACHE)	+= bcache.o
 
-bcache-y		:= alloc.o bset.o btree.o closure.o debug.o extents.o\
-	io.o journal.o movinggc.o request.o stats.o super.o sysfs.o trace.o\
+bcache-y		:= alloc.o bset.o btree.o debug.o extents.o io.o\
+	journal.o movinggc.o request.o stats.o super.o sysfs.o trace.o\
 	util.o writeback.o features.o

drivers/md/bcache/bcache.h

@@ -179,6 +179,7 @@
 #define pr_fmt(fmt) "bcache: %s() " fmt, __func__
 
 #include <linux/bio.h>
+#include <linux/closure.h>
 #include <linux/kobject.h>
 #include <linux/list.h>
 #include <linux/mutex.h>
@@ -192,7 +193,6 @@
 #include "bcache_ondisk.h"
 #include "bset.h"
 #include "util.h"
-#include "closure.h"
 
 struct bucket {
 	atomic_t	pin;

drivers/md/bcache/super.c

@@ -2906,7 +2906,6 @@ static int __init bcache_init(void)
 		goto err;
 
 	bch_debug_init();
-	closure_debug_init();
 
 	bcache_is_reboot = false;
 

drivers/md/bcache/util.h

@@ -4,6 +4,7 @@
 #define _BCACHE_UTIL_H
 
 #include <linux/blkdev.h>
+#include <linux/closure.h>
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/sched/clock.h>
@@ -13,8 +14,6 @@
 #include <linux/workqueue.h>
 #include <linux/crc64.h>
 
-#include "closure.h"
-
 struct closure;
 
 #ifdef CONFIG_BCACHE_DEBUG

fs/Kconfig

@@ -48,6 +48,7 @@ source "fs/ocfs2/Kconfig"
 source "fs/btrfs/Kconfig"
 source "fs/nilfs2/Kconfig"
 source "fs/f2fs/Kconfig"
+source "fs/bcachefs/Kconfig"
 source "fs/zonefs/Kconfig"
 
 endif # BLOCK

fs/Makefile

@@ -123,6 +123,7 @@ obj-$(CONFIG_OCFS2_FS)		+= ocfs2/
 obj-$(CONFIG_BTRFS_FS)		+= btrfs/
 obj-$(CONFIG_GFS2_FS)           += gfs2/
 obj-$(CONFIG_F2FS_FS)		+= f2fs/
+obj-$(CONFIG_BCACHEFS_FS)	+= bcachefs/
 obj-$(CONFIG_CEPH_FS)		+= ceph/
 obj-$(CONFIG_PSTORE)		+= pstore/
 obj-$(CONFIG_EFIVAR_FS)		+= efivarfs/

fs/bcachefs/Kconfig

+
+config BCACHEFS_FS
+	tristate "bcachefs filesystem support (EXPERIMENTAL)"
+	depends on BLOCK
+	select EXPORTFS
+	select CLOSURES
+	select LIBCRC32C
+	select CRC64
+	select FS_POSIX_ACL
+	select LZ4_COMPRESS
+	select LZ4_DECOMPRESS
+	select LZ4HC_COMPRESS
+	select LZ4HC_DECOMPRESS
+	select ZLIB_DEFLATE
+	select ZLIB_INFLATE
+	select ZSTD_COMPRESS
+	select ZSTD_DECOMPRESS
+	select CRYPTO_SHA256
+	select CRYPTO_CHACHA20
+	select CRYPTO_POLY1305
+	select KEYS
+	select RAID6_PQ
+	select XOR_BLOCKS
+	select XXHASH
+	select SRCU
+	select SYMBOLIC_ERRNAME
+	select MEAN_AND_VARIANCE
+	help
+	The bcachefs filesystem - a modern, copy on write filesystem, with
+	support for multiple devices, compression, checksumming, etc.
+
+config BCACHEFS_QUOTA
+	bool "bcachefs quota support"
+	depends on BCACHEFS_FS
+	select QUOTACTL
+
+config BCACHEFS_POSIX_ACL
+	bool "bcachefs POSIX ACL support"
+	depends on BCACHEFS_FS
+	select FS_POSIX_ACL
+
+config BCACHEFS_DEBUG_TRANSACTIONS
+	bool "bcachefs runtime info"
+	depends on BCACHEFS_FS
+	default y
+	help
+	This makes the list of running btree transactions available in debugfs.
+
+	This is a highly useful debugging feature but does add a small amount of overhead.
+
+config BCACHEFS_DEBUG
+	bool "bcachefs debugging"
+	depends on BCACHEFS_FS
+	help
+	Enables many extra debugging checks and assertions.
+
+	The resulting code will be significantly slower than normal; you
+	probably shouldn't select this option unless you're a developer.
+
+config BCACHEFS_TESTS
+	bool "bcachefs unit and performance tests"
+	depends on BCACHEFS_FS
+	help
+	Include some unit and performance tests for the core btree code
+
+config BCACHEFS_LOCK_TIME_STATS
+       bool "bcachefs lock time statistics"
+       depends on BCACHEFS_FS
+       help
+       Expose statistics for how long we held a lock in debugfs
+
+config BCACHEFS_NO_LATENCY_ACCT
+	bool "disable latency accounting and time stats"
+	depends on BCACHEFS_FS
+	help
+	This disables device latency tracking and time stats, only for performance testing
+
+config MEAN_AND_VARIANCE_UNIT_TEST
+	tristate "mean_and_variance unit tests" if !KUNIT_ALL_TESTS
+	depends on KUNIT
+	select MEAN_AND_VARIANCE
+	default KUNIT_ALL_TESTS
+	help
+	  This option enables the kunit tests for mean_and_variance module.
+	  If unsure, say N.

fs/bcachefs/Makefile

+
+obj-$(CONFIG_BCACHEFS_FS)	+= bcachefs.o
+
+bcachefs-y		:=	\
+	acl.o			\
+	alloc_background.o	\
+	alloc_foreground.o	\
+	backpointers.o		\
+	bkey.o			\
+	bkey_methods.o		\
+	bkey_sort.o		\
+	bset.o			\
+	btree_cache.o		\
+	btree_gc.o		\
+	btree_io.o		\
+	btree_iter.o		\
+	btree_journal_iter.o	\
+	btree_key_cache.o	\
+	btree_locking.o		\
+	btree_trans_commit.o	\
+	btree_update.o		\
+	btree_update_interior.o	\
+	btree_write_buffer.o	\
+	buckets.o		\
+	buckets_waiting_for_journal.o	\
+	chardev.o		\
+	checksum.o		\
+	clock.o			\
+	compress.o		\
+	counters.o		\
+	debug.o			\
+	dirent.o		\
+	disk_groups.o		\
+	data_update.o		\
+	ec.o			\
+	errcode.o		\
+	error.o			\
+	extents.o		\
+	extent_update.o		\
+	fs.o			\
+	fs-common.o		\
+	fs-ioctl.o		\
+	fs-io.o			\
+	fs-io-buffered.o	\
+	fs-io-direct.o		\
+	fs-io-pagecache.o	\
+	fsck.o			\
+	inode.o			\
+	io_read.o		\
+	io_misc.o		\
+	io_write.o		\
+	journal.o		\
+	journal_io.o		\
+	journal_reclaim.o	\
+	journal_sb.o		\
+	journal_seq_blacklist.o	\
+	keylist.o		\
+	logged_ops.o		\
+	lru.o			\
+	mean_and_variance.o	\
+	migrate.o		\
+	move.o			\
+	movinggc.o		\
+	nocow_locking.o		\
+	opts.o			\
+	printbuf.o		\
+	quota.o			\
+	rebalance.o		\
+	recovery.o		\
+	reflink.o		\
+	replicas.o		\
+	sb-clean.o		\
+	sb-members.o		\
+	siphash.o		\
+	six.o			\
+	snapshot.o		\
+	subvolume.o		\
+	super.o			\
+	super-io.o		\
+	sysfs.o			\
+	tests.o			\
+	trace.o			\
+	two_state_shared_lock.o	\
+	util.o			\
+	varint.o		\
+	xattr.o
+
+obj-$(CONFIG_MEAN_AND_VARIANCE_UNIT_TEST)   += mean_and_variance_test.o

fs/bcachefs/acl.h

+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_ACL_H
+#define _BCACHEFS_ACL_H
+
+struct bch_inode_unpacked;
+struct bch_hash_info;
+struct bch_inode_info;
+struct posix_acl;
+
+#define BCH_ACL_VERSION	0x0001
+
+typedef struct {
+	__le16		e_tag;
+	__le16		e_perm;
+	__le32		e_id;
+} bch_acl_entry;
+
+typedef struct {
+	__le16		e_tag;
+	__le16		e_perm;
+} bch_acl_entry_short;
+
+typedef struct {
+	__le32		a_version;
+} bch_acl_header;
+
+void bch2_acl_to_text(struct printbuf *, const void *, size_t);
+
+#ifdef CONFIG_BCACHEFS_POSIX_ACL
+
+struct posix_acl *bch2_get_acl(struct mnt_idmap *, struct dentry *, int);
+
+int bch2_set_acl_trans(struct btree_trans *, subvol_inum,
+		       struct bch_inode_unpacked *,
+		       struct posix_acl *, int);
+int bch2_set_acl(struct mnt_idmap *, struct dentry *, struct posix_acl *, int);
+int bch2_acl_chmod(struct btree_trans *, subvol_inum,
+		   struct bch_inode_unpacked *,
+		   umode_t, struct posix_acl **);
+
+#else
+
+static inline int bch2_set_acl_trans(struct btree_trans *trans, subvol_inum inum,
+				     struct bch_inode_unpacked *inode_u,
+				     struct posix_acl *acl, int type)
+{
+	return 0;
+}
+
+static inline int bch2_acl_chmod(struct btree_trans *trans, subvol_inum inum,
+				 struct bch_inode_unpacked *inode,
+				 umode_t mode,
+				 struct posix_acl **new_acl)
+{
+	return 0;
+}
+
+#endif /* CONFIG_BCACHEFS_POSIX_ACL */
+
+#endif /* _BCACHEFS_ACL_H */

fs/bcachefs/alloc_background.h

+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_ALLOC_BACKGROUND_H
+#define _BCACHEFS_ALLOC_BACKGROUND_H
+
+#include "bcachefs.h"
+#include "alloc_types.h"
+#include "buckets.h"
+#include "debug.h"
+#include "super.h"
+
+enum bkey_invalid_flags;
+
+/* How out of date a pointer gen is allowed to be: */
+#define BUCKET_GC_GEN_MAX	96U
+
+static inline bool bch2_dev_bucket_exists(struct bch_fs *c, struct bpos pos)
+{
+	struct bch_dev *ca;
+
+	if (!bch2_dev_exists2(c, pos.inode))
+		return false;
+
+	ca = bch_dev_bkey_exists(c, pos.inode);
+	return pos.offset >= ca->mi.first_bucket &&
+		pos.offset < ca->mi.nbuckets;
+}
+
+static inline u64 bucket_to_u64(struct bpos bucket)
+{
+	return (bucket.inode << 48) | bucket.offset;
+}
+
+static inline struct bpos u64_to_bucket(u64 bucket)
+{
+	return POS(bucket >> 48, bucket & ~(~0ULL << 48));
+}
+
+static inline u8 alloc_gc_gen(struct bch_alloc_v4 a)
+{
+	return a.gen - a.oldest_gen;
+}
+
+static inline enum bch_data_type __alloc_data_type(u32 dirty_sectors,
+						   u32 cached_sectors,
+						   u32 stripe,
+						   struct bch_alloc_v4 a,
+						   enum bch_data_type data_type)
+{
+	if (stripe)
+		return data_type == BCH_DATA_parity ? data_type : BCH_DATA_stripe;
+	if (dirty_sectors)
+		return data_type;
+	if (cached_sectors)
+		return BCH_DATA_cached;
+	if (BCH_ALLOC_V4_NEED_DISCARD(&a))
+		return BCH_DATA_need_discard;
+	if (alloc_gc_gen(a) >= BUCKET_GC_GEN_MAX)
+		return BCH_DATA_need_gc_gens;
+	return BCH_DATA_free;
+}
+
+static inline enum bch_data_type alloc_data_type(struct bch_alloc_v4 a,
+						 enum bch_data_type data_type)
+{
+	return __alloc_data_type(a.dirty_sectors, a.cached_sectors,
+				 a.stripe, a, data_type);
+}
+
+static inline enum bch_data_type bucket_data_type(enum bch_data_type data_type)
+{
+	return data_type == BCH_DATA_stripe ? BCH_DATA_user : data_type;
+}
+
+static inline u64 alloc_lru_idx_read(struct bch_alloc_v4 a)
+{
+	return a.data_type == BCH_DATA_cached ? a.io_time[READ] : 0;
+}
+
+#define DATA_TYPES_MOVABLE		\
+	((1U << BCH_DATA_btree)|	\
+	 (1U << BCH_DATA_user)|		\
+	 (1U << BCH_DATA_stripe))
+
+static inline bool data_type_movable(enum bch_data_type type)
+{
+	return (1U << type) & DATA_TYPES_MOVABLE;
+}
+
+static inline u64 alloc_lru_idx_fragmentation(struct bch_alloc_v4 a,
+					      struct bch_dev *ca)
+{
+	if (!data_type_movable(a.data_type) ||
+	    a.dirty_sectors >= ca->mi.bucket_size)
+		return 0;
+
+	return div_u64((u64) a.dirty_sectors * (1ULL << 31), ca->mi.bucket_size);
+}
+
+static inline u64 alloc_freespace_genbits(struct bch_alloc_v4 a)
+{
+	return ((u64) alloc_gc_gen(a) >> 4) << 56;
+}
+
+static inline struct bpos alloc_freespace_pos(struct bpos pos, struct bch_alloc_v4 a)
+{
+	pos.offset |= alloc_freespace_genbits(a);
+	return pos;
+}
+
+static inline unsigned alloc_v4_u64s(const struct bch_alloc_v4 *a)
+{
+	unsigned ret = (BCH_ALLOC_V4_BACKPOINTERS_START(a) ?:
+			BCH_ALLOC_V4_U64s_V0) +
+		BCH_ALLOC_V4_NR_BACKPOINTERS(a) *
+		(sizeof(struct bch_backpointer) / sizeof(u64));
+
+	BUG_ON(ret > U8_MAX - BKEY_U64s);
+	return ret;
+}
+
+static inline void set_alloc_v4_u64s(struct bkey_i_alloc_v4 *a)
+{
+	set_bkey_val_u64s(&a->k, alloc_v4_u64s(&a->v));
+}
+
+struct bkey_i_alloc_v4 *
+bch2_trans_start_alloc_update(struct btree_trans *, struct btree_iter *, struct bpos);
+
+void __bch2_alloc_to_v4(struct bkey_s_c, struct bch_alloc_v4 *);
+
+static inline const struct bch_alloc_v4 *bch2_alloc_to_v4(struct bkey_s_c k, struct bch_alloc_v4 *convert)
+{
+	const struct bch_alloc_v4 *ret;
+
+	if (unlikely(k.k->type != KEY_TYPE_alloc_v4))
+		goto slowpath;
+
+	ret = bkey_s_c_to_alloc_v4(k).v;
+	if (BCH_ALLOC_V4_BACKPOINTERS_START(ret) != BCH_ALLOC_V4_U64s)
+		goto slowpath;
+
+	return ret;
+slowpath:
+	__bch2_alloc_to_v4(k, convert);
+	return convert;
+}
+
+struct bkey_i_alloc_v4 *bch2_alloc_to_v4_mut(struct btree_trans *, struct bkey_s_c);
+
+int bch2_bucket_io_time_reset(struct btree_trans *, unsigned, size_t, int);
+
+int bch2_alloc_v1_invalid(const struct bch_fs *, struct bkey_s_c,
+			  enum bkey_invalid_flags, struct printbuf *);
+int bch2_alloc_v2_invalid(const struct bch_fs *, struct bkey_s_c,
+			  enum bkey_invalid_flags, struct printbuf *);
+int bch2_alloc_v3_invalid(const struct bch_fs *, struct bkey_s_c,
+			  enum bkey_invalid_flags, struct printbuf *);
+int bch2_alloc_v4_invalid(const struct bch_fs *, struct bkey_s_c,
+			  enum bkey_invalid_flags, struct printbuf *);
+void bch2_alloc_v4_swab(struct bkey_s);
+void bch2_alloc_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c);
+
+#define bch2_bkey_ops_alloc ((struct bkey_ops) {	\
+	.key_invalid	= bch2_alloc_v1_invalid,	\
+	.val_to_text	= bch2_alloc_to_text,		\
+	.trans_trigger	= bch2_trans_mark_alloc,	\
+	.atomic_trigger	= bch2_mark_alloc,		\
+	.min_val_size	= 8,				\
+})
+
+#define bch2_bkey_ops_alloc_v2 ((struct bkey_ops) {	\
+	.key_invalid	= bch2_alloc_v2_invalid,	\
+	.val_to_text	= bch2_alloc_to_text,		\
+	.trans_trigger	= bch2_trans_mark_alloc,	\
+	.atomic_trigger	= bch2_mark_alloc,		\
+	.min_val_size	= 8,				\
+})
+
+#define bch2_bkey_ops_alloc_v3 ((struct bkey_ops) {	\
+	.key_invalid	= bch2_alloc_v3_invalid,	\
+	.val_to_text	= bch2_alloc_to_text,		\
+	.trans_trigger	= bch2_trans_mark_alloc,	\
+	.atomic_trigger	= bch2_mark_alloc,		\
+	.min_val_size	= 16,				\
+})
+
+#define bch2_bkey_ops_alloc_v4 ((struct bkey_ops) {	\
+	.key_invalid	= bch2_alloc_v4_invalid,	\
+	.val_to_text	= bch2_alloc_to_text,		\
+	.swab		= bch2_alloc_v4_swab,		\
+	.trans_trigger	= bch2_trans_mark_alloc,	\
+	.atomic_trigger	= bch2_mark_alloc,		\
+	.min_val_size	= 48,				\
+})
+
+int bch2_bucket_gens_invalid(const struct bch_fs *, struct bkey_s_c,
+			     enum bkey_invalid_flags, struct printbuf *);
+void bch2_bucket_gens_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c);
+
+#define bch2_bkey_ops_bucket_gens ((struct bkey_ops) {	\
+	.key_invalid	= bch2_bucket_gens_invalid,	\
+	.val_to_text	= bch2_bucket_gens_to_text,	\
+})
+
+int bch2_bucket_gens_init(struct bch_fs *);
+
+static inline bool bkey_is_alloc(const struct bkey *k)
+{
+	return  k->type == KEY_TYPE_alloc ||
+		k->type == KEY_TYPE_alloc_v2 ||
+		k->type == KEY_TYPE_alloc_v3;
+}
+
+int bch2_alloc_read(struct bch_fs *);
+
+int bch2_trans_mark_alloc(struct btree_trans *, enum btree_id, unsigned,
+			  struct bkey_s_c, struct bkey_i *, unsigned);
+int bch2_check_alloc_info(struct bch_fs *);
+int bch2_check_alloc_to_lru_refs(struct bch_fs *);
+void bch2_do_discards(struct bch_fs *);
+
+static inline u64 should_invalidate_buckets(struct bch_dev *ca,
+					    struct bch_dev_usage u)
+{
+	u64 want_free = ca->mi.nbuckets >> 7;
+	u64 free = max_t(s64, 0,
+			   u.d[BCH_DATA_free].buckets
+			 + u.d[BCH_DATA_need_discard].buckets
+			 - bch2_dev_buckets_reserved(ca, BCH_WATERMARK_stripe));
+
+	return clamp_t(s64, want_free - free, 0, u.d[BCH_DATA_cached].buckets);
+}
+
+void bch2_do_invalidates(struct bch_fs *);
+
+static inline struct bch_backpointer *alloc_v4_backpointers(struct bch_alloc_v4 *a)
+{
+	return (void *) ((u64 *) &a->v +
+			 (BCH_ALLOC_V4_BACKPOINTERS_START(a) ?:
+			  BCH_ALLOC_V4_U64s_V0));
+}
+
+static inline const struct bch_backpointer *alloc_v4_backpointers_c(const struct bch_alloc_v4 *a)
+{
+	return (void *) ((u64 *) &a->v + BCH_ALLOC_V4_BACKPOINTERS_START(a));
+}
+
+int bch2_dev_freespace_init(struct bch_fs *, struct bch_dev *, u64, u64);
+int bch2_fs_freespace_init(struct bch_fs *);
+
+void bch2_recalc_capacity(struct bch_fs *);
+
+void bch2_dev_allocator_remove(struct bch_fs *, struct bch_dev *);
+void bch2_dev_allocator_add(struct bch_fs *, struct bch_dev *);
+
+void bch2_fs_allocator_background_init(struct bch_fs *);
+
+#endif /* _BCACHEFS_ALLOC_BACKGROUND_H */

fs/bcachefs/alloc_types.h

+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_ALLOC_TYPES_H
+#define _BCACHEFS_ALLOC_TYPES_H
+
+#include <linux/mutex.h>
+#include <linux/spinlock.h>
+
+#include "clock_types.h"
+#include "fifo.h"
+
+struct bucket_alloc_state {
+	u64	buckets_seen;
+	u64	skipped_open;
+	u64	skipped_need_journal_commit;
+	u64	skipped_nocow;
+	u64	skipped_nouse;
+};
+
+#define BCH_WATERMARKS()		\
+	x(stripe)			\
+	x(normal)			\
+	x(copygc)			\
+	x(btree)			\
+	x(btree_copygc)			\
+	x(reclaim)
+
+enum bch_watermark {
+#define x(name)	BCH_WATERMARK_##name,
+	BCH_WATERMARKS()
+#undef x
+	BCH_WATERMARK_NR,
+};
+
+#define BCH_WATERMARK_BITS	3
+#define BCH_WATERMARK_MASK	~(~0U << BCH_WATERMARK_BITS)
+
+#define OPEN_BUCKETS_COUNT	1024
+
+#define WRITE_POINT_HASH_NR	32
+#define WRITE_POINT_MAX		32
+
+/*
+ * 0 is never a valid open_bucket_idx_t:
+ */
+typedef u16			open_bucket_idx_t;
+
+struct open_bucket {
+	spinlock_t		lock;
+	atomic_t		pin;
+	open_bucket_idx_t	freelist;
+	open_bucket_idx_t	hash;
+
+	/*
+	 * When an open bucket has an ec_stripe attached, this is the index of
+	 * the block in the stripe this open_bucket corresponds to:
+	 */
+	u8			ec_idx;
+	enum bch_data_type	data_type:6;
+	unsigned		valid:1;
+	unsigned		on_partial_list:1;
+
+	u8			dev;
+	u8			gen;
+	u32			sectors_free;
+	u64			bucket;
+	struct ec_stripe_new	*ec;
+};
+
+#define OPEN_BUCKET_LIST_MAX	15
+
+struct open_buckets {
+	open_bucket_idx_t	nr;
+	open_bucket_idx_t	v[OPEN_BUCKET_LIST_MAX];
+};
+
+struct dev_stripe_state {
+	u64			next_alloc[BCH_SB_MEMBERS_MAX];
+};
+
+#define WRITE_POINT_STATES()		\
+	x(stopped)			\
+	x(waiting_io)			\
+	x(waiting_work)			\
+	x(running)
+
+enum write_point_state {
+#define x(n)	WRITE_POINT_##n,
+	WRITE_POINT_STATES()
+#undef x
+	WRITE_POINT_STATE_NR
+};
+
+struct write_point {
+	struct {
+		struct hlist_node	node;
+		struct mutex		lock;
+		u64			last_used;
+		unsigned long		write_point;
+		enum bch_data_type	data_type;
+
+		/* calculated based on how many pointers we're actually going to use: */
+		unsigned		sectors_free;
+
+		struct open_buckets	ptrs;
+		struct dev_stripe_state	stripe;
+
+		u64			sectors_allocated;
+	} __aligned(SMP_CACHE_BYTES);
+
+	struct {
+		struct work_struct	index_update_work;
+
+		struct list_head	writes;
+		spinlock_t		writes_lock;
+
+		enum write_point_state	state;
+		u64			last_state_change;
+		u64			time[WRITE_POINT_STATE_NR];
+	} __aligned(SMP_CACHE_BYTES);
+};
+
+struct write_point_specifier {
+	unsigned long		v;
+};
+
+#endif /* _BCACHEFS_ALLOC_TYPES_H */

fs/bcachefs/bbpos.h

+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_BBPOS_H
+#define _BCACHEFS_BBPOS_H
+
+#include "bkey_methods.h"
+
+struct bbpos {
+	enum btree_id		btree;
+	struct bpos		pos;
+};
+
+static inline struct bbpos BBPOS(enum btree_id btree, struct bpos pos)
+{
+	return (struct bbpos) { btree, pos };
+}
+
+#define BBPOS_MIN	BBPOS(0, POS_MIN)
+#define BBPOS_MAX	BBPOS(BTREE_ID_NR - 1, POS_MAX)
+
+static inline int bbpos_cmp(struct bbpos l, struct bbpos r)
+{
+	return cmp_int(l.btree, r.btree) ?: bpos_cmp(l.pos, r.pos);
+}
+
+static inline struct bbpos bbpos_successor(struct bbpos pos)
+{
+	if (bpos_cmp(pos.pos, SPOS_MAX)) {
+		pos.pos = bpos_successor(pos.pos);
+		return pos;
+	}
+
+	if (pos.btree != BTREE_ID_NR) {
+		pos.btree++;
+		pos.pos = POS_MIN;
+		return pos;
+	}
+
+	BUG();
+}
+
+static inline void bch2_bbpos_to_text(struct printbuf *out, struct bbpos pos)
+{
+	prt_str(out, bch2_btree_ids[pos.btree]);
+	prt_char(out, ':');
+	bch2_bpos_to_text(out, pos.pos);
+}
+
+#endif /* _BCACHEFS_BBPOS_H */

fs/bcachefs/btree_write_buffer.h

+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BCACHEFS_BTREE_WRITE_BUFFER_H
+#define _BCACHEFS_BTREE_WRITE_BUFFER_H
+
+int __bch2_btree_write_buffer_flush(struct btree_trans *, unsigned, bool);
+int bch2_btree_write_buffer_flush_sync(struct btree_trans *);
+int bch2_btree_write_buffer_flush(struct btree_trans *);
+
+int bch2_btree_insert_keys_write_buffer(struct btree_trans *);
+
+void bch2_fs_btree_write_buffer_exit(struct bch_fs *);
+int bch2_fs_btree_write_buffer_init(struct bch_fs *);
+
+#endif /* _BCACHEFS_BTREE_WRITE_BUFFER_H */

fs/bcachefs/buckets_waiting_for_journal.h

+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _BUCKETS_WAITING_FOR_JOURNAL_H
+#define _BUCKETS_WAITING_FOR_JOURNAL_H
+
+#include "buckets_waiting_for_journal_types.h"
+
+bool bch2_bucket_needs_journal_commit(struct buckets_waiting_for_journal *,
+				      u64, unsigned, u64);
+int bch2_set_bucket_needs_journal_commit(struct buckets_waiting_for_journal *,
+					 u64, unsigned, u64, u64);
+
+void bch2_fs_buckets_waiting_for_journal_exit(struct bch_fs *);
+int bch2_fs_buckets_waiting_for_journal_init(struct bch_fs *);
+
+#endif /* _BUCKETS_WAITING_FOR_JOURNAL_H */