Re: [PATCH bpf v2] xdp: Fix handling of devmap in generic XDP

On Wed, Jun 13, 2018 at 8:40 PM, Toshiaki Makita
[off-list ref] wrote:

On 2018/06/14 11:56, Y Song wrote:
...

quoted

quoted

@@ -586,6 +589,15 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
        return 0;
 }

+struct sk_buff;
+
+static inline int dev_map_generic_redirect(struct bpf_dtab_netdev *dst,
+                                          struct sk_buff *skb,
+                                          struct bpf_prog *xdp_prog)
+{
+       return 0;

should you return an error code here?

My understanding is that this function never be called if
CONFIG_BPF_SYSCALL is not set so any value is OK.
I aligned it with other functions of devmap, specifically
dev_map_enqueue() which returns 0.

quoted

quoted

+}
+
 static inline
 struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
 {

diff --git a/include/linux/filter.h b/include/linux/filter.h
index 45fc0f5..8ddff1f 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h

@@ -19,6 +19,7 @@
 #include <linux/cryptohash.h>
 #include <linux/set_memory.h>
 #include <linux/kallsyms.h>
+#include <linux/if_vlan.h>

 #include <net/sch_generic.h>

@@ -786,6 +787,21 @@ static inline bool bpf_dump_raw_ok(void)
 struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off,
                                       const struct bpf_insn *patch, u32 len);

+static inline int __xdp_generic_ok_fwd_dev(struct sk_buff *skb,
+                                          struct net_device *fwd)

Previously this function is only used in filter.c and now it is only
used in devmap.c. Maybe this function should be in devmap.c
until it will be used cross different files?

This function is also called from xdp_do_generic_redirect() in
net/core/filter.c, so I can't move it to devmap.c.

Thanks for the explanation. Make sense.
Acked-by: Yonghong Song <redacted>