On Sat 03-08-19 07:01:55, Tejun Heo wrote:

+void mem_cgroup_track_foreign_dirty_slowpath(struct page *page,
+					     struct bdi_writeback *wb)
+{
+	struct mem_cgroup *memcg = page->mem_cgroup;
+	struct memcg_cgwb_frn *frn;
+	u64 now = jiffies_64;
+	u64 oldest_at = now;
+	int oldest = -1;
+	int i;
+
+	/*
+	 * Pick the slot to use.  If there is already a slot for @wb, keep
+	 * using it.  If not replace the oldest one which isn't being
+	 * written out.
+	 */
+	for (i = 0; i < MEMCG_CGWB_FRN_CNT; i++) {
+		frn = &memcg->cgwb_frn[i];
+		if (frn->bdi_id == wb->bdi->id &&
+		    frn->memcg_id == wb->memcg_css->id)
+			break;
+		if (frn->at < oldest_at && atomic_read(&frn->done.cnt) == 1) {
+			oldest = i;
+			oldest_at = frn->at;
+		}
+	}
+
+	if (i < MEMCG_CGWB_FRN_CNT) {
+		unsigned long update_intv =
+			min_t(unsigned long, HZ,
+			      msecs_to_jiffies(dirty_expire_interval * 10) / 8);
+		/*
+		 * Re-using an existing one.  Let's update timestamp lazily
+		 * to avoid making the cacheline hot.
+		 */
+		if (frn->at < now - update_intv)
+			frn->at = now;
+	} else if (oldest >= 0) {
+		/* replace the oldest free one */
+		frn = &memcg->cgwb_frn[oldest];
+		frn->bdi_id = wb->bdi->id;
+		frn->memcg_id = wb->memcg_css->id;
+		frn->at = now;
+	}

I have to say I'm a bit nervous about the completely lockless handling
here. I understand that garbage in the cgwb_frn will just result in this
mechanism not working and possibly flushing wrong wb's but still it seems a
bit fragile. But I don't see any cheap way of synchronizing this so I guess
let's try how this will work in practice.

								Honza
--
Jan Kara [off-list ref]
SUSE Labs, CR