Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread One Thousand Gnomes
> These results for Toeplitz are not plausible. Given random input you > cannot expect any hash function to produce such uniform results. I > suspect either your input data is biased or how your applying the hash > is. > > When I run 64 random IPv4 3-tuples through Toeplitz and Jenkins I get >

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Eric Dumazet
On Wed, 2016-01-13 at 23:10 +, Haiyang Zhang wrote: > I have done a comparison of the Toeplitz v.s. Jenkins Hash algorithms, > and found that the Toeplitz provides much better distribution of the > connections into send-indirection-table entries. See the data below -- > showing how many

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Tom Herbert
> I have done a comparison of the Toeplitz v.s. Jenkins Hash algorithms, > and found that the Toeplitz provides much better distribution of the > connections into send-indirection-table entries. See the data below -- > showing how many TCP connections are distributed into each of the > sixteen

RE: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Haiyang Zhang
uk.ukuu.org.uk>; David Miller <da...@davemloft.net>; > vkuzn...@redhat.com; net...@vger.kernel.org; KY Srinivasan > <k...@microsoft.com>; de...@linuxdriverproject.org; linux- > ker...@vger.kernel.org > Subject: Re: [PATCH net-next] hv_netvsc: don't make assumptions on > struct flow_ke

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Tom Herbert
rguk.ukuu.org.uk> >> Cc: Tom Herbert <t...@herbertland.com>; Haiyang Zhang >> <haiya...@microsoft.com>; David Miller <da...@davemloft.net>; >> vkuzn...@redhat.com; net...@vger.kernel.org; KY Srinivasan >> <k...@microsoft.com>; de...@linuxdriverproject.or

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Eric Dumazet
On Thu, 2016-01-14 at 17:53 +, One Thousand Gnomes wrote: > > These results for Toeplitz are not plausible. Given random input you > > cannot expect any hash function to produce such uniform results. I > > suspect either your input data is biased or how your applying the hash > > is. > > > >

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Tom Herbert
nd.com>; Haiyang Zhang >> >> <haiya...@microsoft.com>; David Miller <da...@davemloft.net>; >> >> vkuzn...@redhat.com; net...@vger.kernel.org; KY Srinivasan >> >> <k...@microsoft.com>; de...@linuxdriverproject.org; linux- >> >> ker...@vge

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Tom Herbert
; One Thousand Gnomes >> >> <gno...@lxorguk.ukuu.org.uk>; David Miller <da...@davemloft.net>; >> >> vkuzn...@redhat.com; net...@vger.kernel.org; KY Srinivasan >> >> <k...@microsoft.com>; de...@linuxdriverproject.org; linux- >> >> ker...@vge

RE: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Haiyang Zhang
uk.ukuu.org.uk>; David Miller <da...@davemloft.net>; > vkuzn...@redhat.com; net...@vger.kernel.org; KY Srinivasan > <k...@microsoft.com>; de...@linuxdriverproject.org; linux- > ker...@vger.kernel.org > Subject: Re: [PATCH net-next] hv_netvsc: don't make assumptions on > struct flow_k

RE: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Haiyang Zhang
@microsoft.com>; David Miller <da...@davemloft.net>; > vkuzn...@redhat.com; net...@vger.kernel.org; KY Srinivasan > <k...@microsoft.com>; de...@linuxdriverproject.org; linux- > ker...@vger.kernel.org > Subject: Re: [PATCH net-next] hv_netvsc: don't make assumptions on > struct flow_k

RE: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Haiyang Zhang
uk.ukuu.org.uk>; David Miller <da...@davemloft.net>; > vkuzn...@redhat.com; net...@vger.kernel.org; KY Srinivasan > <k...@microsoft.com>; de...@linuxdriverproject.org; linux- > ker...@vger.kernel.org > Subject: Re: [PATCH net-next] hv_netvsc: don't make assumptions on > struct flow_k

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread David Miller
From: Tom Herbert Date: Thu, 14 Jan 2016 13:44:24 -0800 > The fact that we can negatively affect the output of Toeplitz so > predictably is actually a liability and not a benefit. This sort of > thing can be the basis of a DOS attack and is why we kicked out XOR > hash in

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-14 Thread Eric Dumazet
On Thu, 2016-01-14 at 20:23 +, Haiyang Zhang wrote: > > For non-random inputs, I used the port selection of iperf that increases > the port number by 2 for each connection. Only send-port numbers are > different, other values are the same. I also tested some other fixed > increment,

RE: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-13 Thread Haiyang Zhang
riverproject.org; linux- > ker...@vger.kernel.org; eric.duma...@gmail.com > Subject: Re: [PATCH net-next] hv_netvsc: don't make assumptions on > struct flow_keys layout > > From: Vitaly Kuznetsov <vkuzn...@redhat.com> > Date: Thu, 7 Jan 2016 10:33:09 +0100 > > >

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-13 Thread David Miller
From: Haiyang Zhang Date: Wed, 13 Jan 2016 23:10:57 + > I have done a comparison of the Toeplitz v.s. Jenkins Hash algorithms, > and found that the Toeplitz provides much better distribution of the > connections into send-indirection-table entries. This fails to

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-10 Thread David Miller
From: Vitaly Kuznetsov Date: Thu, 7 Jan 2016 10:33:09 +0100 > Recent changes to 'struct flow_keys' (e.g commit d34af823ff40 ("net: Add > VLAN ID to flow_keys")) introduced a performance regression in netvsc > driver. Is problem is, however, not the above mentioned commit

RE: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-08 Thread KY Srinivasan
t > <eric.duma...@gmail.com> > Cc: Tom Herbert <t...@herbertland.com>; net...@vger.kernel.org; > Haiyang Zhang <haiya...@microsoft.com>; de...@linuxdriverproject.org; > linux-ker...@vger.kernel.org; David Miller <da...@davemloft.net> > Subject: Re: [PATCH

RE: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-08 Thread Haiyang Zhang
@vger.kernel.org; KY > Srinivasan <k...@microsoft.com>; Haiyang Zhang <haiya...@microsoft.com>; > de...@linuxdriverproject.org; linux-ker...@vger.kernel.org; David Miller > <da...@davemloft.net> > Subject: Re: [PATCH net-next] hv_netvsc: don't make assumptions on > stru

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-08 Thread Tom Herbert
On Thu, Jan 7, 2016 at 4:52 AM, Eric Dumazet wrote: > On Thu, 2016-01-07 at 10:33 +0100, Vitaly Kuznetsov wrote: >> Recent changes to 'struct flow_keys' (e.g commit d34af823ff40 ("net: Add >> VLAN ID to flow_keys")) introduced a performance regression in netvsc >> driver.

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-07 Thread John Fastabend
On 16-01-07 05:28 AM, Vitaly Kuznetsov wrote: > Eric Dumazet writes: > >> On Thu, 2016-01-07 at 10:33 +0100, Vitaly Kuznetsov wrote: >>> Recent changes to 'struct flow_keys' (e.g commit d34af823ff40 ("net: Add >>> VLAN ID to flow_keys")) introduced a performance

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-07 Thread Eric Dumazet
On Thu, 2016-01-07 at 10:33 +0100, Vitaly Kuznetsov wrote: > Recent changes to 'struct flow_keys' (e.g commit d34af823ff40 ("net: Add > VLAN ID to flow_keys")) introduced a performance regression in netvsc > driver. Is problem is, however, not the above mentioned commit but the > fact that

RE: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-07 Thread KY Srinivasan
Herbert <t...@herbertland.com>; net...@vger.kernel.org; KY > Srinivasan <k...@microsoft.com>; Haiyang Zhang <haiya...@microsoft.com>; > de...@linuxdriverproject.org; linux-ker...@vger.kernel.org; David Miller > <da...@davemloft.net> > Subject: Re: [PATCH

Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct flow_keys layout

2016-01-07 Thread John Fastabend
rnel.org; David Miller >> <da...@davemloft.net> >> Subject: Re: [PATCH net-next] hv_netvsc: don't make assumptions on struct >> flow_keys layout >> >> On 16-01-07 05:28 AM, Vitaly Kuznetsov wrote: >>> Eric Dumazet <eric.duma...@gmail.com> writes: >&