On Fri, Dec 16, 2005 at 03:05:13PM -0800, David S. Miller ([EMAIL PROTECTED])
wrote:
> From: Evgeniy Polyakov <[EMAIL PROTECTED]>
> Date: Sat, 17 Dec 2005 02:00:34 +0300
>
> > The most expensive is ioread according to oprofile, which can not be
> > changed.
>
> Sometimes clever device driver pro
From: Evgeniy Polyakov <[EMAIL PROTECTED]>
Date: Sat, 17 Dec 2005 02:00:34 +0300
> The most expensive is ioread according to oprofile, which can not be
> changed.
Sometimes clever device driver programmers can eliminate some
ioread operations :-)
> The second one is memory allocation and then co
On Fri, Dec 16, 2005 at 09:40:48AM -0800, Stephen Hemminger ([EMAIL PROTECTED])
wrote:
> On Fri, 16 Dec 2005 11:39:52 +0300
> Evgeniy Polyakov <[EMAIL PROTECTED]> wrote:
>
> > On Thu, Dec 15, 2005 at 12:55:48PM -0800, Stephen Hemminger ([EMAIL
> > PROTECTED]) wrote:
> > > Given the pretty extens
On Fri, 16 Dec 2005 11:39:52 +0300
Evgeniy Polyakov <[EMAIL PROTECTED]> wrote:
> On Thu, Dec 15, 2005 at 12:55:48PM -0800, Stephen Hemminger ([EMAIL
> PROTECTED]) wrote:
> > Given the pretty extensive modifications necessary, do you have any
> > useful benchmark data to show that this is a win. P
Here is initial support for size truncation.
It has some problems with unconnected socket closing.
diff --git a/include/net/zerocopy.h b/include/net/zerocopy.h
--- a/include/net/zerocopy.h
+++ b/include/net/zerocopy.h
@@ -152,6 +152,7 @@ struct zsock
u32 zc_seq_first;
> CPU usage graph attached.
> 'zc' - zero-copy transfer.
> 'recv/write' - recv()/write() transfer.
It is really attached now.
--
Evgeniy Polyakov
cpu_usage.png
Description: PNG image
On Thu, Dec 15, 2005 at 12:55:48PM -0800, Stephen Hemminger ([EMAIL PROTECTED])
wrote:
> Given the pretty extensive modifications necessary, do you have any
> useful benchmark data to show that this is a win. Previous zero
> copy (like page flipping) has not be merged because it did not
> prove to
Given the pretty extensive modifications necessary, do you have any
useful benchmark data to show that this is a win. Previous zero
copy (like page flipping) has not be merged because it did not
prove to be a net gain.
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body
