[PATCH 0/7] [DCCP]: Fixes and enhancements

[PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

Please find following a series of patches for DCCP.

These have been tested against torvalds/linux-2.6.git and davem/net-2.6.19.git

My opinion is that 1 and 2 can go straight into 2.6.18 as documentation
changes only - Dave - are you able to do as Arnaldo is very busy at present.

I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
issues that have been in the DCCP tree since inception and have caught a
number of people.

Number 6 is just shifting code around to tidy it up and introduces no change
in logic. You could argue for it to go in either 2.6.18 or 2.6.19!

Number 7 is implementing transmit buffering and is 2.6.19 material.
Andrea - this might be quite useful for you in CCID2 as well I believe.

These patches are all capable of being done independently except 3, 4, 5 which
are a group.

Also on http://wand.net.nz/~iam4/dccp/patches/ are the following further
patches which are not ready for merge but others might be interested in:
-DCCP-Probe ala TCP-Probe
-The starts of memory buffer limiting (this is not actually needed for number
7 as it is actually receive where problems occur which is an existing issue)
-My research code

-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

[PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

Please find following a series of patches for DCCP.

These have been tested against torvalds/linux-2.6.git and davem/net-2.6.19.git

My opinion is that 1 and 2 can go straight into 2.6.18 as documentation
changes only - Dave - are you able to do as Arnaldo is very busy at present.

I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
issues that have been in the DCCP tree since inception and have caught a
number of people.

Number 6 is just shifting code around to tidy it up and introduces no change
in logic. You could argue for it to go in either 2.6.18 or 2.6.19!

Number 7 is implementing transmit buffering and is 2.6.19 material.
Andrea - this might be quite useful for you in CCID2 as well I believe.

These patches are all capable of being done independently except 3, 4, 5 which
are a group.

Also on http://wand.net.nz/~iam4/dccp/patches/ are the following further
patches which are not ready for merge but others might be interested in:
-DCCP-Probe ala TCP-Probe
-The starts of memory buffer limiting (this is not actually needed for number
7 as it is actually receive where problems occur which is an existing issue)
-My research code

-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[David Miller]

From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Thu, 24 Aug 2006 15:16:33 +1200

> My opinion is that 1 and 2 can go straight into 2.6.18 as
> documentation changes only - Dave - are you able to do as Arnaldo is
> very busy at present.

I spent all of today on USAGI's IPSEC/MIPV6 patches and related
issues, so I'll look into this tomorrow.

Thanks Ian.

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[David Miller]

From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Thu, 24 Aug 2006 15:16:33 +1200

> My opinion is that 1 and 2 can go straight into 2.6.18 as
> documentation changes only - Dave - are you able to do as Arnaldo is
> very busy at present.

I spent all of today on USAGI's IPSEC/MIPV6 patches and related
issues, so I'll look into this tomorrow.

Thanks Ian.

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

> I spent all of today on USAGI's IPSEC/MIPV6 patches and related
> issues, so I'll look into this tomorrow.
>
> Thanks Ian.
>
Yes I saw that. Take your time as this is nowhere near as important!

Regards,

Ian
-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

> I spent all of today on USAGI's IPSEC/MIPV6 patches and related
> issues, so I'll look into this tomorrow.
>
> Thanks Ian.
>
Yes I saw that. Take your time as this is nowhere near as important!

Regards,

Ian
-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Arnaldo Carvalho de Melo]

On 8/24/06, Ian McDonald <ian.mcdonald@jandi.co.nz> wrote:
> > I spent all of today on USAGI's IPSEC/MIPV6 patches and related
> > issues, so I'll look into this tomorrow.
> >
> > Thanks Ian.
> >
> Yes I saw that. Take your time as this is nowhere near as important!

Sigh, I'm still busy indeed, gave a quick look at the series and up to
the 10th patch its OK at first sight, should improve the current
situation, thanks.

- Arnaldo

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Arnaldo Carvalho de Melo]

On 8/24/06, Ian McDonald <ian.mcdonald@jandi.co.nz> wrote:
> > I spent all of today on USAGI's IPSEC/MIPV6 patches and related
> > issues, so I'll look into this tomorrow.
> >
> > Thanks Ian.
> >
> Yes I saw that. Take your time as this is nowhere near as important!

Sigh, I'm still busy indeed, gave a quick look at the series and up to
the 10th patch its OK at first sight, should improve the current
situation, thanks.

- Arnaldo

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[David Miller]

From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Thu, 24 Aug 2006 15:16:33 +1200

> Please find following a series of patches for DCCP.
> 
> These have been tested against torvalds/linux-2.6.git and davem/net-2.6.19.git
> 
> My opinion is that 1 and 2 can go straight into 2.6.18 as documentation
> changes only - Dave - are you able to do as Arnaldo is very busy at present.
> 
> I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
> issues that have been in the DCCP tree since inception and have caught a
> number of people.

Ok, I'll toss 1-5 into 2.6.18

One thing I don't understand is this description from patch 5:

--------------------
This gives a theoretical speed of 71.9 Kbits/s. I measured across three
runs with this patch set and got 70.1 Kbits/s. Without this patchset the
average was 232 Kbits/s which means Linux can't be used for CCID3 research
properly.
--------------------

Decreasing the transfer rate is desirable?  I read this as saying
this "fix" drops the transfer rate down from 232Kb/sec to
70.1Kb/sec.  What's going on here?

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[David Miller]

From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Thu, 24 Aug 2006 15:16:33 +1200

> Please find following a series of patches for DCCP.
> 
> These have been tested against torvalds/linux-2.6.git and davem/net-2.6.19.git
> 
> My opinion is that 1 and 2 can go straight into 2.6.18 as documentation
> changes only - Dave - are you able to do as Arnaldo is very busy at present.
> 
> I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
> issues that have been in the DCCP tree since inception and have caught a
> number of people.

Ok, I'll toss 1-5 into 2.6.18

One thing I don't understand is this description from patch 5:

--------------------
This gives a theoretical speed of 71.9 Kbits/s. I measured across three
runs with this patch set and got 70.1 Kbits/s. Without this patchset the
average was 232 Kbits/s which means Linux can't be used for CCID3 research
properly.
--------------------

Decreasing the transfer rate is desirable?  I read this as saying
this "fix" drops the transfer rate down from 232Kb/sec to
70.1Kb/sec.  What's going on here?

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

On 8/27/06, David Miller <davem@davemloft.net> wrote:
> >
> > I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
> > issues that have been in the DCCP tree since inception and have caught a
> > number of people.
>
> Ok, I'll toss 1-5 into 2.6.18

Thanks for that. Are 6 and 7 going into 2.6.19 or do you want Arnaldo
to have a bit more of a look? 6 in particular is trivial.
>
> One thing I don't understand is this description from patch 5:
>
> --------------------
> This gives a theoretical speed of 71.9 Kbits/s. I measured across three
> runs with this patch set and got 70.1 Kbits/s. Without this patchset the
> average was 232 Kbits/s which means Linux can't be used for CCID3 research
> properly.
> --------------------
>
> Decreasing the transfer rate is desirable?  I read this as saying
> this "fix" drops the transfer rate down from 232Kb/sec to
> 70.1Kb/sec.  What's going on here?

DCCP CCID3 (RFC 4342) uses TFRC (RFC 3448) to calculate the desired
rate to send at based on feedback from the receiver. The reason for
this is that TFRC is not ACK/Window based to control rate and TFRC
calculates a rate so that the flow is "fair" when competing with TCP.
TFRC is designed to be smoother than TCP at dealing with loss - more
sine wave than saw tooth.

The calculation is based on the work Padhye et al did in this paper -
http://citeseer.ist.psu.edu/padhye98modeling.html

As it turns out this is based on TCP Reno at that time and modern TCP
variants are more efficient when dealing with loss as can be verified
through iperf but we should implement what the RFC says.

Basically the implementation in the DCCP code was buggy and was
transmitting too fast so I have made it conform to the RFC much
closer.

Ian
-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

On 8/27/06, David Miller <davem@davemloft.net> wrote:
> >
> > I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
> > issues that have been in the DCCP tree since inception and have caught a
> > number of people.
>
> Ok, I'll toss 1-5 into 2.6.18

Thanks for that. Are 6 and 7 going into 2.6.19 or do you want Arnaldo
to have a bit more of a look? 6 in particular is trivial.
>
> One thing I don't understand is this description from patch 5:
>
> --------------------
> This gives a theoretical speed of 71.9 Kbits/s. I measured across three
> runs with this patch set and got 70.1 Kbits/s. Without this patchset the
> average was 232 Kbits/s which means Linux can't be used for CCID3 research
> properly.
> --------------------
>
> Decreasing the transfer rate is desirable?  I read this as saying
> this "fix" drops the transfer rate down from 232Kb/sec to
> 70.1Kb/sec.  What's going on here?

DCCP CCID3 (RFC 4342) uses TFRC (RFC 3448) to calculate the desired
rate to send at based on feedback from the receiver. The reason for
this is that TFRC is not ACK/Window based to control rate and TFRC
calculates a rate so that the flow is "fair" when competing with TCP.
TFRC is designed to be smoother than TCP at dealing with loss - more
sine wave than saw tooth.

The calculation is based on the work Padhye et al did in this paper -
http://citeseer.ist.psu.edu/padhye98modeling.html

As it turns out this is based on TCP Reno at that time and modern TCP
variants are more efficient when dealing with loss as can be verified
through iperf but we should implement what the RFC says.

Basically the implementation in the DCCP code was buggy and was
transmitting too fast so I have made it conform to the RFC much
closer.

Ian
-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[David Miller]

From: "Ian McDonald" <ian.mcdonald@jandi.co.nz>
Date: Sun, 27 Aug 2006 15:46:58 +1200

> On 8/27/06, David Miller <davem@davemloft.net> wrote:
> > >
> > > I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
> > > issues that have been in the DCCP tree since inception and have caught a
> > > number of people.
> >
> > Ok, I'll toss 1-5 into 2.6.18
> 
> Thanks for that. Are 6 and 7 going into 2.6.19 or do you want Arnaldo
> to have a bit more of a look? 6 in particular is trivial.

Those patches are already in net-2.6.19

> Basically the implementation in the DCCP code was buggy and was
> transmitting too fast so I have made it conform to the RFC much
> closer.

Thanks for the explanation.

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[David Miller]

From: "Ian McDonald" <ian.mcdonald@jandi.co.nz>
Date: Sun, 27 Aug 2006 15:46:58 +1200

> On 8/27/06, David Miller <davem@davemloft.net> wrote:
> > >
> > > I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
> > > issues that have been in the DCCP tree since inception and have caught a
> > > number of people.
> >
> > Ok, I'll toss 1-5 into 2.6.18
> 
> Thanks for that. Are 6 and 7 going into 2.6.19 or do you want Arnaldo
> to have a bit more of a look? 6 in particular is trivial.

Those patches are already in net-2.6.19

> Basically the implementation in the DCCP code was buggy and was
> transmitting too fast so I have made it conform to the RFC much
> closer.

Thanks for the explanation.

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

> > On 8/27/06, David Miller <davem@davemloft.net> wrote:
> > > >
> > > > I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
> > > > issues that have been in the DCCP tree since inception and have caught a
> > > > number of people.
> > >
> > > Ok, I'll toss 1-5 into 2.6.18
> >
> > Thanks for that. Are 6 and 7 going into 2.6.19 or do you want Arnaldo
> > to have a bit more of a look? 6 in particular is trivial.
>
> Those patches are already in net-2.6.19
>
Yes I see that now. However I can't see #5 in net-2.6.git in your tree
or Linus' where 1-4 made it in...

Ian
-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

> > On 8/27/06, David Miller <davem@davemloft.net> wrote:
> > > >
> > > > I would love 3, 4, 5 to go into 2.6.18 as these resolve long standing CCID3
> > > > issues that have been in the DCCP tree since inception and have caught a
> > > > number of people.
> > >
> > > Ok, I'll toss 1-5 into 2.6.18
> >
> > Thanks for that. Are 6 and 7 going into 2.6.19 or do you want Arnaldo
> > to have a bit more of a look? 6 in particular is trivial.
>
> Those patches are already in net-2.6.19
>
Yes I see that now. However I can't see #5 in net-2.6.git in your tree
or Linus' where 1-4 made it in...

Ian
-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[David Miller]

From: "Ian McDonald" <ian.mcdonald@jandi.co.nz>
Date: Sun, 27 Aug 2006 16:57:17 +1200

> Yes I see that now. However I can't see #5 in net-2.6.git in your tree
> or Linus' where 1-4 made it in...

Resend it to me privately and I'll figure out what happened.

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[David Miller]

From: "Ian McDonald" <ian.mcdonald@jandi.co.nz>
Date: Sun, 27 Aug 2006 16:57:17 +1200

> Yes I see that now. However I can't see #5 in net-2.6.git in your tree
> or Linus' where 1-4 made it in...

Resend it to me privately and I'll figure out what happened.

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

On 8/27/06, David Miller <davem@davemloft.net> wrote:
> From: "Ian McDonald" <ian.mcdonald@jandi.co.nz>
> Date: Sun, 27 Aug 2006 16:57:17 +1200
>
> > Yes I see that now. However I can't see #5 in net-2.6.git in your tree
> > or Linus' where 1-4 made it in...
>
> Resend it to me privately and I'll figure out what happened.

Done.

-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand

Re: [PATCH 0/7] [DCCP]: Fixes and enhancements

[Ian McDonald]

On 8/27/06, David Miller <davem@davemloft.net> wrote:
> From: "Ian McDonald" <ian.mcdonald@jandi.co.nz>
> Date: Sun, 27 Aug 2006 16:57:17 +1200
>
> > Yes I see that now. However I can't see #5 in net-2.6.git in your tree
> > or Linus' where 1-4 made it in...
>
> Resend it to me privately and I'll figure out what happened.

Done.

-- 
Ian McDonald
Web: http://wand.net.nz/~iam4
Blog: http://imcdnzl.blogspot.com
WAND Network Research Group
Department of Computer Science
University of Waikato
New Zealand