Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Bill Fink
On Tue, 20 Nov 2007, Andrew Gallatin wrote: > David Miller wrote: > > From: Andrew Gallatin <[EMAIL PROTECTED]> > > Date: Tue, 20 Nov 2007 06:47:57 -0500 > > > >> David Miller wrote: > >> > From: Herbert Xu <[EMAIL PROTECTED]> > >> > Date: Tue, 20 Nov 2007 14:09:18 +0800 > >> > > >> >>

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread David Miller
From: Rick Jones <[EMAIL PROTECTED]> Date: Tue, 20 Nov 2007 11:45:54 -0800 > Sounds like one might as well go ahead and implement HP-UX/Solaris-like > ACK sending avoidance at the receiver and not bother with LRO-ACK on the > sender. > > In some experiements a while back I thought I saw that LR

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Rick Jones
David Miller wrote: From: Andrew Gallatin <[EMAIL PROTECTED]> Date: Tue, 23 Oct 2007 11:11:55 -0400 I've attached a patch which adds support to inet_lro for aggregating pure acks. I've applied this patch to net-2.6.25... but! This needs some serious thinking. What this patch ends up doing

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Evgeniy Polyakov
On Tue, Nov 20, 2007 at 10:08:31PM +0800, Herbert Xu ([EMAIL PROTECTED]) wrote: > Of course we still have the problem with the option in general > that Dave raised. That is this may cause the proliferation of > TCP receiver behaviour that may be undesirable. Yes, it results in bursts of traffic b

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Herbert Xu
On Tue, Nov 20, 2007 at 05:03:12PM +0300, Evgeniy Polyakov wrote: > > For software lro I agree, but this looks exactly like gso/tso case and > additional tweak for software gso. Having it per-system is fine, and I > believe no one should ever care that some distro will do bad/good things > with it

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Evgeniy Polyakov
On Tue, Nov 20, 2007 at 09:50:56PM +0800, Herbert Xu ([EMAIL PROTECTED]) wrote: > On Tue, Nov 20, 2007 at 04:35:09PM +0300, Evgeniy Polyakov wrote: > > > > On Tue, Nov 20, 2007 at 08:27:05AM -0500, Andrew Gallatin ([EMAIL > > PROTECTED]) wrote: > > > Hmm.. rather than a global tunable, what if it

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Herbert Xu
On Tue, Nov 20, 2007 at 04:35:09PM +0300, Evgeniy Polyakov wrote: > > On Tue, Nov 20, 2007 at 08:27:05AM -0500, Andrew Gallatin ([EMAIL PROTECTED]) > wrote: > > Hmm.. rather than a global tunable, what if it was a > > network driver managed tunable which toggled a flag in the > > lro_mgr features

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Evgeniy Polyakov
Hi. On Tue, Nov 20, 2007 at 08:27:05AM -0500, Andrew Gallatin ([EMAIL PROTECTED]) wrote: > Hmm.. rather than a global tunable, what if it was a > network driver managed tunable which toggled a flag in the > lro_mgr features? Would that be better? What about ethtool control to set LRO_simple and

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Andrew Gallatin
David Miller wrote: > From: Andrew Gallatin <[EMAIL PROTECTED]> > Date: Tue, 20 Nov 2007 06:47:57 -0500 > >> David Miller wrote: >> > From: Herbert Xu <[EMAIL PROTECTED]> >> > Date: Tue, 20 Nov 2007 14:09:18 +0800 >> > >> >> David Miller <[EMAIL PROTECTED]> wrote: >> >>> Fundamentally, I real

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread David Miller
From: Andrew Gallatin <[EMAIL PROTECTED]> Date: Tue, 20 Nov 2007 06:47:57 -0500 > David Miller wrote: > > From: Herbert Xu <[EMAIL PROTECTED]> > > Date: Tue, 20 Nov 2007 14:09:18 +0800 > > > >> David Miller <[EMAIL PROTECTED]> wrote: > >>> Fundamentally, I really don't like this change, it ba

Re: [PATCH] LRO ack aggregation

2007-11-20 Thread Andrew Gallatin
David Miller wrote: > From: Herbert Xu <[EMAIL PROTECTED]> > Date: Tue, 20 Nov 2007 14:09:18 +0800 > >> David Miller <[EMAIL PROTECTED]> wrote: >>> Fundamentally, I really don't like this change, it batches to the >>> point where it begins to erode the natural ACK clocking of TCP, and I >>> theref

Re: [PATCH] LRO ack aggregation

2007-11-19 Thread David Miller
From: Herbert Xu <[EMAIL PROTECTED]> Date: Tue, 20 Nov 2007 14:09:18 +0800 > David Miller <[EMAIL PROTECTED]> wrote: > > > > Fundamentally, I really don't like this change, it batches to the > > point where it begins to erode the natural ACK clocking of TCP, and I > > therefore am very likely to r

Re: [PATCH] LRO ack aggregation

2007-11-19 Thread Herbert Xu
David Miller <[EMAIL PROTECTED]> wrote: > > Fundamentally, I really don't like this change, it batches to the > point where it begins to erode the natural ACK clocking of TCP, and I > therefore am very likely to revert it before merging to Linus. Perhaps make it a tunable that defaults to off? Ch

Re: [PATCH] LRO ack aggregation

2007-11-19 Thread David Miller
From: Andrew Gallatin <[EMAIL PROTECTED]> Date: Tue, 23 Oct 2007 11:11:55 -0400 > I've attached a patch which adds support to inet_lro for aggregating > pure acks. I've applied this patch to net-2.6.25... but! This needs some serious thinking. What this patch ends up doing is creating big stret

[PATCH] LRO ack aggregation

2007-10-23 Thread Andrew Gallatin
Hi, We recently did some performance comparisons between the new inet_lro LRO support in the kernel, and our Myri10GE in-driver LRO. For receive, we found they were nearly identical. However, for transmit, we found that Myri10GE's LRO shows much lower CPU utilization. We traced the CPU utilizat