xdp: Remove map_to_flush and map swap detection

Now that all XDP maps that can be used with bpf_redirect_map() tracks
entries to be flushed in a global fashion, there is not need to track
that the map has changed and flush from xdp_do_generic_map()
anymore. All entries will be flushed in xdp_do_flush_map().

This means that the map_to_flush can be removed, and the corresponding
checks. Moving the flush logic to one place, xdp_do_flush_map(), give
a bulking behavior and performance boost.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-8-bjorn.topel@gmail.com

include/linux/filter.h

@@ -592,7 +592,6 @@ struct bpf_redirect_info {
 	u32 tgt_index;
 	void *tgt_value;
 	struct bpf_map *map;
-	struct bpf_map *map_to_flush;
 	u32 kern_flags;
 };
 

net/core/filter.c

@@ -3547,26 +3547,9 @@ static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 
 void xdp_do_flush_map(void)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
-	struct bpf_map *map = ri->map_to_flush;
-
-	ri->map_to_flush = NULL;
-	if (map) {
-		switch (map->map_type) {
-		case BPF_MAP_TYPE_DEVMAP:
-		case BPF_MAP_TYPE_DEVMAP_HASH:
-			__dev_map_flush();
-			break;
-		case BPF_MAP_TYPE_CPUMAP:
-			__cpu_map_flush();
-			break;
-		case BPF_MAP_TYPE_XSKMAP:
-			__xsk_map_flush();
-			break;
-		default:
-			break;
-		}
-	}
+	__dev_map_flush();
+	__cpu_map_flush();
+	__xsk_map_flush();
 }
 EXPORT_SYMBOL_GPL(xdp_do_flush_map);
 
@@ -3615,14 +3598,10 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	ri->tgt_value = NULL;
 	WRITE_ONCE(ri->map, NULL);
 
-	if (ri->map_to_flush && unlikely(ri->map_to_flush != map))
-		xdp_do_flush_map();
-
 	err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
 	if (unlikely(err))
 		goto err;
 
-	ri->map_to_flush = map;
 	_trace_xdp_redirect_map(dev, xdp_prog, fwd, map, index);
 	return 0;
 err: