| To: | "David S. Miller" <davem@xxxxxxxxxx> |
|---|---|
| Subject: | Re: zerocopy changes in 3c59x.c |
| From: | Andrew Morton <andrewm@xxxxxxxxxx> |
| Date: | Sat, 27 Jan 2001 00:27:32 +1100 |
| Cc: | Alexey Kuznetsov <kuznet@xxxxxxxxxxxxx>, "netdev@xxxxxxxxxxx" <netdev@xxxxxxxxxxx> |
| References: | <3A716447.CF6E8BB0@xxxxxxxxxx>, <3A714788.82C064BD@xxxxxxxxxx> <14961.24733.869800.77633@xxxxxxxxxxxxxxx> <3A716447.CF6E8BB0@xxxxxxxxxx> <14961.30181.671982.174763@xxxxxxxxxxxxxxx> |
| Sender: | owner-netdev@xxxxxxxxxxx |
"David S. Miller" wrote:
>
> Andrew Morton writes:
> > But why do this, rather than create new accounting fields? Let
> > me guess: short-term thing, intended to be removed, didn't want
> > to hack the userspace tools?
>
> Why add new fields when they are unnecessary? What is hurt
> by having these fields increment.
>
Hey, I was just asking! Dunno about others, but for me
the zc thing has basically come from nowhere - I'm still
coming up to speed about the design decisions which were
made, how it is implemented, etc.
/proc/net/dev says:
Inter-| Receive | Transmit
face |bytes packets errs drop fifo frame compressed multicast|bytes
packets errs drop fifo colls carrier compressed
lo: 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
eth0: 14274 104 0 0 0 0 54 0 5414
69 0 0 0 0 0 54
So this: ^^
tells us that 54 packets have been sent with h/w checksums and this:
^^
tells us that they were all multi-fragment.
I assume this is because the IP header is in a different frag?
Is there ever a situation in which these numbers will differ?
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: zerocopy changes in 3c59x.c, David S. Miller |
|---|---|
| Next by Date: | Re: zerocopy changes in 3c59x.c, David S. Miller |
| Previous by Thread: | Re: zerocopy changes in 3c59x.c, David S. Miller |
| Next by Thread: | Re: zerocopy changes in 3c59x.c, David S. Miller |
| Indexes: | [Date] [Thread] [Top] [All Lists] |