* [Drbd-dev] Re: [DRBD-cvs] r1600 - branches/drbd-0.7/drbd
[not found] <20041019091152.50E053BE6D@garcon.linbit.com>
@ 2004-10-19 9:30 ` Philipp Reisner
2004-10-19 10:03 ` Lars Ellenberg
0 siblings, 1 reply; 3+ messages in thread
From: Philipp Reisner @ 2004-10-19 9:30 UTC (permalink / raw)
To: drbd-dev
On Tuesday 19 October 2004 11:11, svn@svn.drbd.org wrote:
> Author: lars
> Date: 2004-10-19 11:11:49 +0200 (Tue, 19 Oct 2004)
> New Revision: 1600
>
> Modified:
> branches/drbd-0.7/drbd/drbd_int.h
> branches/drbd-0.7/drbd/drbd_main.c
> branches/drbd-0.7/drbd/drbd_receiver.c
> branches/drbd-0.7/drbd/drbd_req.c
> branches/drbd-0.7/drbd/drbd_worker.c
> Log:
> get rid of the HERE macro
> provide (experimental) code for decreasing latency on 2.6 kernels by
> pushing the tcp stack
>
Hi Lars,
Do you really think it is an good idea to put that tcp-stack pushing
into drbd-0.7 ??
- Do you know that it improves performance ?
Or is it just an assumption.
- Do we really want to do this with the stable code
I think we should have a least one or two reports from the list,
that
a) it does not break anything
b) it improves performance.
-philipp
--
: Dipl-Ing Philipp Reisner Tel +43-1-8178292-50 :
: LINBIT Information Technologies GmbH Fax +43-1-8178292-82 :
: Schönbrunnerstr 244, 1120 Vienna, Austria http://www.linbit.com :
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [Drbd-dev] Re: [DRBD-cvs] r1600 - branches/drbd-0.7/drbd
2004-10-19 9:30 ` [Drbd-dev] Re: [DRBD-cvs] r1600 - branches/drbd-0.7/drbd Philipp Reisner
@ 2004-10-19 10:03 ` Lars Ellenberg
2004-10-19 10:40 ` Philipp Reisner
0 siblings, 1 reply; 3+ messages in thread
From: Lars Ellenberg @ 2004-10-19 10:03 UTC (permalink / raw)
To: drbd-dev
/ 2004-10-19 11:30:24 +0200
\ Philipp Reisner:
> > get rid of the HERE macro
> > provide (experimental) code for decreasing latency on 2.6 kernels by
> > pushing the tcp stack
> >
>
> Hi Lars,
>
> Do you really think it is an good idea to put that tcp-stack pushing
> into drbd-0.7 ??
I only put it in drbd07 to ease propagation with svnp.
and so it does not get lost.
currently I go over all those small patches I sent somewhen and never
got any reaction about, and just remove them if they meanwhile are no
longer usefull, or apply them if I think they could do something good.
> - Do you know that it improves performance ?
> Or is it just an assumption.
I do not know.
> - Do we really want to do this with the stable code
> I think we should have a least one or two reports from the list,
> that
> a) it does not break anything
> b) it improves performance.
it is #if 0'ed in drbd 0.7
I think we should #if 1 it in trunk,
benchmark it, and drop it if it does not do any good.
lge
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [Drbd-dev] Re: [DRBD-cvs] r1600 - branches/drbd-0.7/drbd
2004-10-19 10:03 ` Lars Ellenberg
@ 2004-10-19 10:40 ` Philipp Reisner
0 siblings, 0 replies; 3+ messages in thread
From: Philipp Reisner @ 2004-10-19 10:40 UTC (permalink / raw)
To: drbd-dev
[...]
> > Do you really think it is an good idea to put that tcp-stack pushing
> > into drbd-0.7 ??
[...]
> it is #if 0'ed in drbd 0.7
> I think we should #if 1 it in trunk,
> benchmark it, and drop it if it does not do any good.
>
Ok, I did not saw the #if 0 in drbd0.7
What's about posting it to the user list, maybe there is
someone out there who wants to benchmark it...
-Phil
--
: Dipl-Ing Philipp Reisner Tel +43-1-8178292-50 :
: LINBIT Information Technologies GmbH Fax +43-1-8178292-82 :
: Schönbrunnerstr 244, 1120 Vienna, Austria http://www.linbit.com :
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2004-10-19 10:38 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20041019091152.50E053BE6D@garcon.linbit.com>
2004-10-19 9:30 ` [Drbd-dev] Re: [DRBD-cvs] r1600 - branches/drbd-0.7/drbd Philipp Reisner
2004-10-19 10:03 ` Lars Ellenberg
2004-10-19 10:40 ` Philipp Reisner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox