Split the global taskqueue used to process all UFS trim completions,

into per-mount taskqueue with the private taskqueue processing thread.
This allows to drain the taskqueue on unmount, to ensure that all
TRIMs are finished before mount structures are freed.

But just draining the taskqueue where TRIM biodone geom-up completions
are processed is not enough, since ffs_blkfree(), called by the task,
might result in more writes.  Count inflight delayed blkfree's and
pause() unmount until the counter drains as well.

Reported by:	Nick Evans <nevans@talkpoint.com>
Tested by:	Nick Evans <nevans@talkpoint.com>, pho
Sponsored by:	The FreeBSD Foundation
MFC after:	2 weeks

3 files changed, 19 insertions, 3 deletions

diff --git a/sys/ufs/ffs/ffs_alloc.c b/sys/ufs/ffs/ffs_alloc.c
index 19c00de6e1ba..799efe3fa94b 100644
--- a/sys/ufs/ffs/ffs_alloc.c
+++ b/sys/ufs/ffs/ffs_alloc.c
@@ -2270,8 +2270,6 @@ ffs_blkfree_cg(ump, fs, devvp, bno, size, inum, dephd)
 	bdwrite(bp);
 }
 
-TASKQUEUE_DEFINE_THREAD(ffs_trim);
-
 struct ffs_blkfree_trim_params {
 	struct task task;
 	struct ufsmount *ump;
@@ -2294,6 +2292,7 @@ ffs_blkfree_trim_task(ctx, pending)
 	ffs_blkfree_cg(tp->ump, tp->ump->um_fs, tp->devvp, tp->bno, tp->size,
 	    tp->inum, tp->pdephd);
 	vn_finished_secondary_write(UFSTOVFS(tp->ump));
+	atomic_add_int(&tp->ump->um_trim_inflight, -1);
 	free(tp, M_TEMP);
 }
 
@@ -2306,7 +2305,7 @@ ffs_blkfree_trim_completed(bip)
 	tp = bip->bio_caller2;
 	g_destroy_bio(bip);
 	TASK_INIT(&tp->task, 0, ffs_blkfree_trim_task, tp);
-	taskqueue_enqueue(taskqueue_ffs_trim, &tp->task);
+	taskqueue_enqueue(tp->ump->um_trim_tq, &tp->task);
 }
 
 void
@@ -2350,6 +2349,7 @@ ffs_blkfree(ump, fs, devvp, bno, size, inum, vtype, dephd)
 	 * reordering, TRIM might be issued after we reuse the block
 	 * and write some new data into it.
 	 */
+	atomic_add_int(&ump->um_trim_inflight, 1);
 	tp = malloc(sizeof(struct ffs_blkfree_trim_params), M_TEMP, M_WAITOK);
 	tp->ump = ump;
 	tp->devvp = devvp;
diff --git a/sys/ufs/ffs/ffs_vfsops.c b/sys/ufs/ffs/ffs_vfsops.c
index a82ef61bc5b9..b89dc332c287 100644
--- a/sys/ufs/ffs/ffs_vfsops.c
+++ b/sys/ufs/ffs/ffs_vfsops.c
@@ -42,6 +42,7 @@ __FBSDID("$FreeBSD$");
 #include <sys/namei.h>
 #include <sys/priv.h>
 #include <sys/proc.h>
+#include <sys/taskqueue.h>
 #include <sys/kernel.h>
 #include <sys/vnode.h>
 #include <sys/mount.h>
@@ -1005,6 +1006,12 @@ ffs_mountfs(devvp, mp, td)
 			    mp->mnt_stat.f_mntonname);
 			ump->um_candelete = 0;
 		}
+		if (ump->um_candelete) {
+			ump->um_trim_tq = taskqueue_create("trim", M_WAITOK,
+			    taskqueue_thread_enqueue, &ump->um_trim_tq);
+			taskqueue_start_threads(&ump->um_trim_tq, 1, PVFS,
+			    "%s trim", mp->mnt_stat.f_mntonname);
+		}
 	}
 
 	ump->um_mountp = mp;
@@ -1260,6 +1267,12 @@ ffs_unmount(mp, mntflags)
 	}
 	if (susp)
 		vfs_write_resume(mp, VR_START_WRITE);
+	if (ump->um_trim_tq != NULL) {
+		while (ump->um_trim_inflight != 0)
+			pause("ufsutr", hz);
+		taskqueue_drain_all(ump->um_trim_tq);
+		taskqueue_free(ump->um_trim_tq);
+	}
 	DROP_GIANT();
 	g_topology_lock();
 	if (ump->um_fsckpid > 0) {
diff --git a/sys/ufs/ufs/ufsmount.h b/sys/ufs/ufs/ufsmount.h
index 6b357c1b6cc9..838c1e3b53f8 100644
--- a/sys/ufs/ufs/ufsmount.h
+++ b/sys/ufs/ufs/ufsmount.h
@@ -50,6 +50,7 @@ MALLOC_DECLARE(M_UFSMNT);
 struct buf;
 struct inode;
 struct nameidata;
+struct taskqueue;
 struct timeval;
 struct ucred;
 struct uio;
@@ -85,6 +86,8 @@ struct ufsmount {
 	int64_t	um_savedmaxfilesize;		/* XXX - limit maxfilesize */
 	int	um_candelete;			/* devvp supports TRIM */
 	int	um_writesuspended;		/* suspension in progress */
+	u_int	um_trim_inflight;
+	struct taskqueue *um_trim_tq;
 	int	(*um_balloc)(struct vnode *, off_t, int, struct ucred *,
 		    int, struct buf **);
 	int	(*um_blkatoff)(struct vnode *, off_t, char **, struct buf **);