Updated zerocopy patch up on kernel.org

Updated zerocopy patch up on kernel.org

[David S. Miller]

Nothing interesting or new, just merges up with the latest 2.4.1-pre1
patch from Linus.

ftp.kernel.org:/pub/linux/kernel/people/davem/zerocopy-2.4.1p1-1.diff.gz

I haven't had any reports from anyone, which must mean that it is
working perfectly fine and adds no new bugs, testers are thus in
nirvana and thus have nothing to report.  :-)

As much as I would like to believe this, I know that there must be
some (however minor) bug in there.  So please make the effort to
report bugs if you do spot them.  Again, a reminder to test bugs also
against vanilla 2.4.1pre1 and report the bug against that if the bug
appears there too. This way, I know what bugs are specific to the
zerocopy stuff and which are not.

Thanks.

Later,
David S. Miller
davem@redhat.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: Updated zerocopy patch up on kernel.org

[David S. Miller]

   Date: Tue, 09 Jan 2001 20:23:17 -0500
   From: Jeff Garzik <jgarzik@mandrakesoft.com>

   Is there any value to supporting fragments in a driver which
   doesn't do hardware checksumming?  IIRC Alexey had a patch to do
   such for Tulip, but I don't see it in the above patchset.

I'm actually considering making the SG w/o hwcsum situation illegal.

Basically, with SG only, you can run into some problems.
We don't prevent anyone from making modifications to the
paged data, we just grab references to the pages.

This works perfectly fine when the card does the checksumming,
the card DMAs a snapshot of the data into it's internal buffers,
checksums that local snapshot of the data, and the checksum is fine.

If, on the other hand, we're doing this in software, we can send out
packets with bad checksums.  The next retransmit could fix it up, but
this is a horrible scheme quality of implementation wise.

Later,
David S. Miller
davem@redhat.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: Updated zerocopy patch up on kernel.org

[Jeff Garzik]

"David S. Miller" wrote:
> 
> Nothing interesting or new, just merges up with the latest 2.4.1-pre1
> patch from Linus.
> 
> ftp.kernel.org:/pub/linux/kernel/people/davem/zerocopy-2.4.1p1-1.diff.gz
> 
> I haven't had any reports from anyone, which must mean that it is
> working perfectly fine and adds no new bugs, testers are thus in
> nirvana and thus have nothing to report.  :-)

Is there any value to supporting fragments in a driver which doesn't do
hardware checksumming?  IIRC Alexey had a patch to do such for Tulip,
but I don't see it in the above patchset.

	Jeff


-- 
Jeff Garzik       | "You see, in this world there's two kinds of
Building 1024     |  people, my friend: Those with loaded guns
MandrakeSoft      |  and those who dig. You dig."  --Blondie
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: Updated zerocopy patch up on kernel.org

[David S. Miller]

   Date: Thu, 11 Jan 2001 11:41:30 +0100 (CET)
   From: Ingo Molnar <mingo@elte.hu>

   On Tue, 9 Jan 2001, David S. Miller wrote:

   > I'm actually considering making the SG w/o hwcsum situation illegal.

   i believe it might still make some limited sense for normal sendmsg()
   and higher MTUs (or 8k NFS) - we could copy & checksum stuff into the
   ->tcp_page if SG is possible and thus the SG capability improves the VM.
   (because we can allocate at PAGE_SIZE granularity.)

Basically what your advocating for is to take advantage of SG-only
devices when we have full control of the page contents.

Sure this would work.

But honestly the real gain from SG-only devices would be (as you know)
the memory usage savings when sending a single static file object to
several thousand clients.

Later,
David S. Miller
davem@redhat.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: Updated zerocopy patch up on kernel.org

[Ingo Molnar]

On Tue, 9 Jan 2001, David S. Miller wrote:

> Nothing interesting or new, just merges up with the latest 2.4.1-pre1
> patch from Linus.
>
> ftp.kernel.org:/pub/linux/kernel/people/davem/zerocopy-2.4.1p1-1.diff.gz
>
> I haven't had any reports from anyone, which must mean that it is
> working perfectly fine and adds no new bugs, testers are thus in
> nirvana and thus have nothing to report.  :-)

(works like a charm here.)

	Ingo

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: Updated zerocopy patch up on kernel.org

[Ingo Molnar]

On Tue, 9 Jan 2001, David S. Miller wrote:

>    Is there any value to supporting fragments in a driver which
>    doesn't do hardware checksumming?  IIRC Alexey had a patch to do
>    such for Tulip, but I don't see it in the above patchset.
>
> I'm actually considering making the SG w/o hwcsum situation illegal.

i believe it might still make some limited sense for normal sendmsg()
and higher MTUs (or 8k NFS) - we could copy & checksum stuff into the
->tcp_page if SG is possible and thus the SG capability improves the VM.
(because we can allocate at PAGE_SIZE granularity.)

	Ingo

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: Updated zerocopy patch up on kernel.org

[Andrew Morton]

Ingo Molnar wrote:
> 
> On Tue, 9 Jan 2001, David S. Miller wrote:
> 
> > Nothing interesting or new, just merges up with the latest 2.4.1-pre1
> > patch from Linus.
> >
> > ftp.kernel.org:/pub/linux/kernel/people/davem/zerocopy-2.4.1p1-1.diff.gz
> >
> > I haven't had any reports from anyone, which must mean that it is
> > working perfectly fine and adds no new bugs, testers are thus in
> > nirvana and thus have nothing to report.  :-)
> 
> (works like a charm here.)

For the record...

I've been running it since release on 2.4.0-UP/x86.  The
NIC is a 3c905B so we're doing scater/gather and hw
checksumming.  It does a lot of NFS client work
against a Netapp server.  rsize=wsize=8192.

I'm not using sendfile().

IOW: me too.

-
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: Updated zerocopy patch up on kernel.org

[Anton Blanchard]

 
> > Nothing interesting or new, just merges up with the latest 2.4.1-pre1
> > patch from Linus.
> >
> > ftp.kernel.org:/pub/linux/kernel/people/davem/zerocopy-2.4.1p1-1.diff.gz
> >
> > I haven't had any reports from anyone, which must mean that it is
> > working perfectly fine and adds no new bugs, testers are thus in
> > nirvana and thus have nothing to report.  :-)
> 
> (works like a charm here.)

Likewise here running a sendfile hacked samba :)

Anton
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/