Merge tag 'bcachefs-2024-03-13' of https://evilpiepirate.org/git/bcachefs

Pull bcachefs updates from Kent Overstreet:

 - Subvolume children btree; this is needed for providing a userspace
   interface for walking subvolumes, which will come later

 - Lots of improvements to directory structure checking

 - Improved journal pipelining, significantly improving performance on
   high iodepth write workloads

 - Discard path improvements: the discard path is more efficient, and no
   longer flushes the journal unnecessarily

 - Buffered write path can now avoid taking the inode lock

 - new mm helper: memalloc_flags_{save|restore}

 - mempool now does kvmalloc mempools

* tag 'bcachefs-2024-03-13' of https://evilpiepirate.org/git/bcachefs: (128 commits)
  bcachefs: time_stats: shrink time_stat_buffer for better alignment
  bcachefs: time_stats: split stats-with-quantiles into a separate structure
  bcachefs: mean_and_variance: put struct mean_and_variance_weighted on a diet
  bcachefs: time_stats: add larger units
  bcachefs: pull out time_stats.[ch]
  bcachefs: reconstruct_alloc cleanup
  bcachefs: fix bch_folio_sector padding
  bcachefs: Fix btree key cache coherency during replay
  bcachefs: Always flush write buffer in delete_dead_inodes()
  bcachefs: Fix order of gc_done passes
  bcachefs: fix deletion of indirect extents in btree_gc
  bcachefs: Prefer struct_size over open coded arithmetic
  bcachefs: Kill unused flags argument to btree_split()
  bcachefs: Check for writing superblocks with nonsense member seq fields
  bcachefs: fix bch2_journal_buf_to_text()
  lib/generic-radix-tree.c: Make nodes more reasonably sized
  bcachefs: copy_(to|from)_user_errcode()
  bcachefs: Split out bkey_types.h
  bcachefs: fix lost journal buf wakeup due to improved pipelining
  bcachefs: intercept mountoption value for bool type
  ...

1 files changed, 27 insertions, 10 deletions

diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c
index d7c81beac14a..562561a9a510 100644
--- a/fs/bcachefs/btree_cache.c
+++ b/fs/bcachefs/btree_cache.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include "bcachefs.h"
+#include "bbpos.h"
 #include "bkey_buf.h"
 #include "btree_cache.h"
 #include "btree_io.h"
@@ -60,7 +61,7 @@ static void btree_node_data_free(struct bch_fs *c, struct btree *b)
 
 	clear_btree_node_just_written(b);
 
-	kvpfree(b->data, btree_buf_bytes(b));
+	kvfree(b->data);
 	b->data = NULL;
 #ifdef __KERNEL__
 	kvfree(b->aux_data);
@@ -94,7 +95,7 @@ static int btree_node_data_alloc(struct bch_fs *c, struct btree *b, gfp_t gfp)
 {
 	BUG_ON(b->data || b->aux_data);
 
-	b->data = kvpmalloc(btree_buf_bytes(b), gfp);
+	b->data = kvmalloc(btree_buf_bytes(b), gfp);
 	if (!b->data)
 		return -BCH_ERR_ENOMEM_btree_node_mem_alloc;
 #ifdef __KERNEL__
@@ -107,7 +108,7 @@ static int btree_node_data_alloc(struct bch_fs *c, struct btree *b, gfp_t gfp)
 		b->aux_data = NULL;
 #endif
 	if (!b->aux_data) {
-		kvpfree(b->data, btree_buf_bytes(b));
+		kvfree(b->data);
 		b->data = NULL;
 		return -BCH_ERR_ENOMEM_btree_node_mem_alloc;
 	}
@@ -208,6 +209,18 @@ static int __btree_node_reclaim(struct bch_fs *c, struct btree *b, bool flush)
 	int ret = 0;
 
 	lockdep_assert_held(&bc->lock);
+
+	struct bbpos pos = BBPOS(b->c.btree_id, b->key.k.p);
+
+	u64 mask = b->c.level
+		? bc->pinned_nodes_interior_mask
+		: bc->pinned_nodes_leaf_mask;
+
+	if ((mask & BIT_ULL(b->c.btree_id)) &&
+	    bbpos_cmp(bc->pinned_nodes_start, pos) < 0 &&
+	    bbpos_cmp(bc->pinned_nodes_end, pos) >= 0)
+		return -BCH_ERR_ENOMEM_btree_node_reclaim;
+
 wait_on_io:
 	if (b->flags & ((1U << BTREE_NODE_dirty)|
 			(1U << BTREE_NODE_read_in_flight)|
@@ -408,7 +421,7 @@ void bch2_fs_btree_cache_exit(struct bch_fs *c)
 	if (c->verify_data)
 		list_move(&c->verify_data->list, &bc->live);
 
-	kvpfree(c->verify_ondisk, c->opts.btree_node_size);
+	kvfree(c->verify_ondisk);
 
 	for (i = 0; i < btree_id_nr_alive(c); i++) {
 		struct btree_root *r = bch2_btree_id_root(c, i);
@@ -711,6 +724,9 @@ static noinline struct btree *bch2_btree_node_fill(struct btree_trans *trans,
 	b = bch2_btree_node_mem_alloc(trans, level != 0);
 
 	if (bch2_err_matches(PTR_ERR_OR_ZERO(b), ENOMEM)) {
+		if (!path)
+			return b;
+
 		trans->memory_allocation_failure = true;
 		trace_and_count(c, trans_restart_memory_allocation_failure, trans, _THIS_IP_, path);
 		return ERR_PTR(btree_trans_restart(trans, BCH_ERR_transaction_restart_fill_mem_alloc_fail));
@@ -760,8 +776,9 @@ static noinline struct btree *bch2_btree_node_fill(struct btree_trans *trans,
 	}
 
 	if (!six_relock_type(&b->c.lock, lock_type, seq)) {
-		if (path)
-			trace_and_count(c, trans_restart_relock_after_fill, trans, _THIS_IP_, path);
+		BUG_ON(!path);
+
+		trace_and_count(c, trans_restart_relock_after_fill, trans, _THIS_IP_, path);
 		return ERR_PTR(btree_trans_restart(trans, BCH_ERR_transaction_restart_relock_after_fill));
 	}
 
@@ -901,7 +918,7 @@ retry:
 
 	if (unlikely(btree_node_read_error(b))) {
 		six_unlock_type(&b->c.lock, lock_type);
-		return ERR_PTR(-EIO);
+		return ERR_PTR(-BCH_ERR_btree_node_read_error);
 	}
 
 	EBUG_ON(b->c.btree_id != path->btree_id);
@@ -992,7 +1009,7 @@ struct btree *bch2_btree_node_get(struct btree_trans *trans, struct btree_path *
 
 	if (unlikely(btree_node_read_error(b))) {
 		six_unlock_type(&b->c.lock, lock_type);
-		return ERR_PTR(-EIO);
+		return ERR_PTR(-BCH_ERR_btree_node_read_error);
 	}
 
 	EBUG_ON(b->c.btree_id != path->btree_id);
@@ -1075,7 +1092,7 @@ lock_node:
 
 	if (unlikely(btree_node_read_error(b))) {
 		six_unlock_read(&b->c.lock);
-		b = ERR_PTR(-EIO);
+		b = ERR_PTR(-BCH_ERR_btree_node_read_error);
 		goto out;
 	}
 
@@ -1096,7 +1113,7 @@ int bch2_btree_node_prefetch(struct btree_trans *trans,
 	struct btree_cache *bc = &c->btree_cache;
 	struct btree *b;
 
-	BUG_ON(trans && !btree_node_locked(path, level + 1));
+	BUG_ON(path && !btree_node_locked(path, level + 1));
 	BUG_ON(level >= BTREE_MAX_DEPTH);
 
 	b = btree_cache_find(bc, k);