On Tue, Aug 16, 2005 at 09:21:24AM -0700, Arthur Kepner wrote: > > So, your patch would accept up to sysctl_ipfrag_max_dist "disorder" > between calls to ip_frag_too_far() (for a particluar fragment queue). > (I'm using "disorder" loosely, but I think it's clear enough what is > meant?) With my change it would accept up to sysctl_ipfrag_max_dist > "disorder" over the lifetime of a fragment queue. > > The latter is the behavior that I intended. But your original may be > better, especially if there are many fragment queues with a common > source.
I see. I chose that behaviour because 1) We're not trying to come up with the perfect solution since there really isn't one. So my objective was for a solution that worked well enough with NFS/UDP. 2) Given that requirement this was the behaviour that was least strict while still being relatively simple to code. So my preference for this isn't really based on particularly strong reasons. However, as I couldn't see any strong reason for it to be more strict either I'd rather stay with the one since it's slightly simpler to code. Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <[EMAIL PROTECTED]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
