| To: | shemminger@xxxxxxxx |
|---|---|
| Subject: | Re: [PATCH] PPP handling fragmented skbuff's |
| From: | "David S. Miller" <davem@xxxxxxxxxx> |
| Date: | Sat, 28 Jun 2003 00:21:29 -0700 (PDT) |
| Cc: | netdev@xxxxxxxxxxx |
| In-reply-to: | <20030627163524.347b2c8e.shemminger@xxxxxxxx> |
| References: | <20030627163524.347b2c8e.shemminger@xxxxxxxx> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
From: Stephen Hemminger <shemminger@xxxxxxxx> Date: Fri, 27 Jun 2003 16:35:24 -0700 Don't think this ever happens today, but if PPP ever gets a fragmented a skbuff and decides to copy it then bad things will happen. The following replaces the places that memcpy() with skb_copy_bits(). Please review carefully before applying, it builds and runs but can't really force these code path to occur under normal systems and devices. It looks ok. But I'll let this one sit over the weekend before applying so others can test it out. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH] Fix PPP async regression, David S. Miller |
|---|---|
| Next by Date: | Re: networking bugs and bugme.osdl.org, John Bradford |
| Previous by Thread: | [PATCH] PPP handling fragmented skbuff's, Stephen Hemminger |
| Next by Thread: | [PATCH] IPV6: Fixed M-Flag in last fragment, YOSHIFUJI Hideaki / 吉藤英明 |
| Indexes: | [Date] [Thread] [Top] [All Lists] |