| To: | Casey Carter <Casey@xxxxxxxxxx> |
|---|---|
| Subject: | Re: [PATCH] Make SO_BINDTODEVICE bind very tightly |
| From: | "David S. Miller" <davem@xxxxxxxxxx> |
| Date: | Wed, 8 Oct 2003 04:49:48 -0700 |
| Cc: | netdev@xxxxxxxxxxx |
| In-reply-to: | <3F833B39.4020606@Carter.net> |
| References: | <3F833B39.4020606@Carter.net> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
On Tue, 07 Oct 2003 17:16:25 -0500 Casey Carter <Casey@xxxxxxxxxx> wrote: > In the presence of both sockets that are and sockets that are not bound > to a specific interface, those bound to the interface should always be > favored for packet delivery over those not so bound. No, the fix made in the original posting of this thread is the correct one. One may not have some sockets with SO_BINDTODEVICE and some with not bound at the same time to the same identity. Therefore the fix should be a bind() time. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | RE: e1000 close (NAPI), Robert Olsson |
|---|---|
| Next by Date: | Re: e1000 close (NAPI), David S. Miller |
| Previous by Thread: | [PATCH] Make SO_BINDTODEVICE bind very tightly, Casey Carter |
| Next by Thread: | [PATCH] dvb-net -- allocate network device structures, Stephen Hemminger |
| Indexes: | [Date] [Thread] [Top] [All Lists] |