Merge tag 'bcachefs-2024-06-28' of https://evilpiepirate.org/git/bcachefs

Pull bcachefs fixes from Kent Overstreet:
 "Simple stuff:

   - NULL ptr/err ptr deref fixes

   - fix for getting wedged on shutdown after journal error

   - fix missing recalc_capacity() call, capacity now changes correctly
     after a device goes read only

     however: our capacity calculation still doesn't take into account
     when we have mixed ro/rw devices and the ro devices have data on
     them, that's going to be a more involved fix to separate accounting
     for "capacity used on ro devices" and "capacity used on rw devices"

   - boring syzbot stuff

  Slightly more involved:

   - discard, invalidate workers are now per device

     this has the effect of simplifying how we take device refs in these
     paths, and the device ref cleanup fixes a longstanding race between
     the device removal path and the discard path

   - fixes for how the debugfs code takes refs on btree_trans objects we
     have debugfs code that prints in use btree_trans objects.

     It uses closure_get() on trans->ref, which is mainly for the cycle
     detector, but the debugfs code was using it on a closure that may
     have hit 0, which is not allowed; for performance reasons we cannot
     avoid having not-in-use transactions on the global list.

     Introduce some new primitives to fix this and make the
     synchronization here a whole lot saner"

* tag 'bcachefs-2024-06-28' of https://evilpiepirate.org/git/bcachefs:
  bcachefs: Fix kmalloc bug in __snapshot_t_mut
  bcachefs: Discard, invalidate workers are now per device
  bcachefs: Fix shift-out-of-bounds in bch2_blacklist_entries_gc
  bcachefs: slab-use-after-free Read in bch2_sb_errors_from_cpu
  bcachefs: Add missing bch2_journal_do_writes() call
  bcachefs: Fix null ptr deref in journal_pins_to_text()
  bcachefs: Add missing recalc_capacity() call
  bcachefs: Fix btree_trans list ordering
  bcachefs: Fix race between trans_put() and btree_transactions_read()
  closures: closure_get_not_zero(), closure_return_sync()
  bcachefs: Make btree_deadlock_to_text() clearer
  bcachefs: fix seqmutex_relock()
  bcachefs: Fix freeing of error pointers

1 files changed, 46 insertions, 6 deletions

diff --git a/lib/closure.c b/lib/closure.c
index 2e1ee9fdec08..c971216d9d77 100644
--- a/lib/closure.c
+++ b/lib/closure.c
@@ -13,7 +13,7 @@
 #include <linux/seq_file.h>
 #include <linux/sched/debug.h>
 
-static inline void closure_put_after_sub(struct closure *cl, int flags)
+static inline void closure_put_after_sub_checks(int flags)
 {
 	int r = flags & CLOSURE_REMAINING_MASK;
 
@@ -22,12 +22,17 @@ static inline void closure_put_after_sub(struct closure *cl, int flags)
 		 flags & CLOSURE_GUARD_MASK, (unsigned) __fls(r)))
 		r &= ~CLOSURE_GUARD_MASK;
 
-	if (!r) {
-		smp_acquire__after_ctrl_dep();
+	WARN(!r && (flags & ~CLOSURE_DESTRUCTOR),
+	     "closure ref hit 0 with incorrect flags set: %x (%u)",
+	     flags & ~CLOSURE_DESTRUCTOR, (unsigned) __fls(flags));
+}
+
+static inline void closure_put_after_sub(struct closure *cl, int flags)
+{
+	closure_put_after_sub_checks(flags);
 
-		WARN(flags & ~CLOSURE_DESTRUCTOR,
-		     "closure ref hit 0 with incorrect flags set: %x (%u)",
-		     flags & ~CLOSURE_DESTRUCTOR, (unsigned) __fls(flags));
+	if (!(flags & CLOSURE_REMAINING_MASK)) {
+		smp_acquire__after_ctrl_dep();
 
 		cl->closure_get_happened = false;
 
@@ -145,6 +150,41 @@ void __sched __closure_sync(struct closure *cl)
 }
 EXPORT_SYMBOL(__closure_sync);
 
+/*
+ * closure_return_sync - finish running a closure, synchronously (i.e. waiting
+ * for outstanding get()s to finish) and returning once closure refcount is 0.
+ *
+ * Unlike closure_sync() this doesn't reinit the ref to 1; subsequent
+ * closure_get_not_zero() calls waill fail.
+ */
+void __sched closure_return_sync(struct closure *cl)
+{
+	struct closure_syncer s = { .task = current };
+
+	cl->s = &s;
+	set_closure_fn(cl, closure_sync_fn, NULL);
+
+	unsigned flags = atomic_sub_return_release(1 + CLOSURE_RUNNING - CLOSURE_DESTRUCTOR,
+						   &cl->remaining);
+
+	closure_put_after_sub_checks(flags);
+
+	if (unlikely(flags & CLOSURE_REMAINING_MASK)) {
+		while (1) {
+			set_current_state(TASK_UNINTERRUPTIBLE);
+			if (s.done)
+				break;
+			schedule();
+		}
+
+		__set_current_state(TASK_RUNNING);
+	}
+
+	if (cl->parent)
+		closure_put(cl->parent);
+}
+EXPORT_SYMBOL(closure_return_sync);
+
 int __sched __closure_sync_timeout(struct closure *cl, unsigned long timeout)
 {
 	struct closure_syncer s = { .task = current };