Re: Calculation of inflight data

2013-05-04 Thread Andre Oppermann
On 03.05.2013 09:28, Aris Angelo wrote: Hi, I am trying to implement an extension to the FreeBSD TCP stack. In order to do that, I have a question regarding the calculation of the pipe variable, the amount of data that the sender calculates as being inflight. I am puzzled for the case when no

IGMP with no matching rules

2013-05-04 Thread Jason Hellenthal
Hey Everyone, Has anyone seen IGMP traffic hit there pflog interface even if there are no rules matching that tell it to log ? Anyone that has a pointer to eliminate the logging of the IGMP traffic would be extremely helpful. This has been fairly frustrating up to this point trying to either

Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Richard Sharpe
Hi folks, I understand better why I am seeing EINVAL intermittently when sending data from Samba via SMB2. The ixgbe driver, for TSO reasons, limits the amount of data that can be DMA'd to 65535 bytes. It returns EINVAL for any mbuf chain larger than that. The SO_SNDBUF for that socket is set

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Adrian Chadd
.. and please file a PR. I'm sure Jack will love this kind of feedback. :) Thanks for doing this debugging! I'm glad to see others getting dirty in the network stack. Adrian On 4 May 2013 06:52, Richard Sharpe realrichardsha...@gmail.com wrote: Hi folks, I understand better why I am

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Richard Sharpe
On Sat, May 4, 2013 at 10:39 AM, Adrian Chadd adr...@freebsd.org wrote: On 4 May 2013 06:52, Richard Sharpe realrichardsha...@gmail.com wrote: Hi folks, I understand better why I am seeing EINVAL intermittently when sending data from Samba via SMB2. The ixgbe driver, for TSO reasons, limits

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Jack Vogel
If you don't use TSO you will hurt your TX performance significantly from the tests that I've run. What exactly is the device you are using, I don't have the source in front of me now, but I'm almost sure that the limit is not 64K but 256K, or are you using some ancient version of the driver?

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Jack Vogel
Yes, I checked: #define IXGBE_TSO_SIZE 262140 So, the driver is not limiting you to 64K assuming you are using a version of recent vintage. Jack On Sat, May 4, 2013 at 1:41 PM, Jack Vogel jfvo...@gmail.com wrote: If you don't use TSO you will hurt your TX performance significantly from

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Richard Sharpe
On Sat, May 4, 2013 at 1:41 PM, Jack Vogel jfvo...@gmail.com wrote: If you don't use TSO you will hurt your TX performance significantly from the tests that I've run. What exactly is the device you are using, I don't have the source in front of me now, but I'm almost sure that the limit is not

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Andre Oppermann
On 04.05.2013 22:47, Jack Vogel wrote: Yes, I checked: #define IXGBE_TSO_SIZE 262140 So, the driver is not limiting you to 64K assuming you are using a version of recent vintage. The stack won't generate TCP and IP packets larger than 64K. However the ethernet header gets prepended to it

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Jack Vogel
Hmmm, so its the stack, can that be easily increased Andre? Regards, Jack On Sat, May 4, 2013 at 2:00 PM, Andre Oppermann an...@freebsd.org wrote: On 04.05.2013 22:47, Jack Vogel wrote: Yes, I checked: #define IXGBE_TSO_SIZE 262140 So, the driver is not limiting you to 64K assuming

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Jack Vogel
Ahh, Twinville, new hardware :) The version at the tip is 2.5.8 and I am working on version 2.5.12 internally that I hope to commit next week... so your version is a bit old :) I would do some testing on newer code. Jack On Sat, May 4, 2013 at 1:54 PM, Richard Sharpe

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Richard Sharpe
On Sat, May 4, 2013 at 2:18 PM, Jack Vogel jfvo...@gmail.com wrote: Ahh, Twinville, new hardware :) The version at the tip is 2.5.8 and I am working on version 2.5.12 internally that I hope to commit next week... so your version is a bit old :) I would do some testing on newer code. I would

Re: kern/178288: [fxp] fxp network driver broken in 7.4

2013-05-04 Thread linimon
Old Synopsis: fxp network driver broken in 7.4 New Synopsis: [fxp] fxp network driver broken in 7.4 Responsible-Changed-From-To: freebsd-i386-freebsd-net Responsible-Changed-By: linimon Responsible-Changed-When: Sat May 4 21:30:24 UTC 2013 Responsible-Changed-Why: Over to maintainer(s).

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Eric van Gyzen
On 05/04/2013 04:19 PM, Richard Sharpe wrote: On Sat, May 4, 2013 at 2:18 PM, Jack Vogel jfvo...@gmail.com wrote: Ahh, Twinville, new hardware :) The version at the tip is 2.5.8 and I am working on version 2.5.12 internally that I hope to commit next week... so your version is a bit old :) I

Re: Is there any way to limit the amount of data in an mbuf chain submitted to a driver?

2013-05-04 Thread Richard Sharpe
On Sat, May 4, 2013 at 4:59 PM, Eric van Gyzen e...@vangyzen.net wrote: On 05/04/2013 04:19 PM, Richard Sharpe wrote: On Sat, May 4, 2013 at 2:18 PM, Jack Vogel jfvo...@gmail.com wrote: Ahh, Twinville, new hardware :) The version at the tip is 2.5.8 and I am working on version 2.5.12