Re: [RFC] sctp/tcp: Question -- ICMPv4 length check (not) redundant?
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: 2008-07-27 04:48:30
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: 2008-07-27 04:48:30
David Miller [off-list ref] wrote:
What if the lower protocol doesn't have a header that it at least 8 bytes? :-)
When such a protocol enters our tree then we can undo this patch :) Seriously apart from raw sockets it isn't possible to generate anything with less than 8 bytes in the IP payload. As our raw error handler is separate anyway all we have to do is move the common check below the raw handler and then we can proceed to remove the protocol-specific checks as Gerrit proposed. The stastic is just an oversight AFAICS from looking at git history. Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} [off-list ref] Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt