Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts the # of mbufs is before
the
code that adds the tcp/ip header
On 19 Aug 2015, at 16:00, Rick Macklem rmack...@uoguelph.ca wrote:
Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts
Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts the # of mbufs is before
the
code that adds the tcp/ip header
Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:51:44AM +0200, Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts
Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts the # of mbufs is before
the
code that adds the tcp/ip header
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=201428
Mark Linimon lini...@freebsd.org changed:
What|Removed |Added
Keywords||patch
Hi All,
FreeBSD 10.2-STABLE
2*CPU Intel E5-2643v3 with HyperThreading enabled
Intel XL710 network adapter
I updated the ixl driver to version 1.4.0 from download.intel.com
Every ixl interface create 24 queues (6 cores *2 HT *2 CPUs) but
utilizes only 16-17 of them. Where is the reason of such
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=202484
Mark Linimon lini...@freebsd.org changed:
What|Removed |Added
Assignee|freebsd-b...@freebsd.org
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=202351
Mark Linimon lini...@freebsd.org changed:
What|Removed |Added
Assignee|freebsd-b...@freebsd.org
On Tue, Aug 18, 2015 at 04:20:07PM -0700, David Wolfskill wrote:
I was minding my own business in a staff meeting this afternoon, and my
laptop rebooted; seems it got a panic. I've copied the core.txt.0 file
to http://www.catwhisker.org/~david/FreeBSD/stable_10/, along with a
verbose
The IXLV_MAX_QUEUES value is for the VF driver; the standard driver should
be able to allocate and properly use up to 64 queues.
That said, you're only getting rx traffic on the first 16 queues, so that
looks like a bug in the driver. I'll take a look at it.
- Eric
On Wed, Aug 19, 2015 at 11:00
On 08/19/15 at 05:43P, Evgeny Khorokhorin wrote:
Hi All,
FreeBSD 10.2-STABLE
2*CPU Intel E5-2643v3 with HyperThreading enabled
Intel XL710 network adapter
I updated the ixl driver to version 1.4.0 from download.intel.com
Every ixl interface create 24 queues (6 cores *2 HT *2 CPUs) but
hi,
you'll have to do some debugging. it looks like it's some kind of odd
race - line 461 is _ieee80211_crypto_delkey(); line 105 is
cipher_detach() and it blows up there.
Try wlandebug +crypto during your next boot and let's see what it
logs for the key.
If you can 'print *key' in kgdb on the
On Wed, Aug 19, 2015 at 12:25:38PM -0700, Adrian Chadd wrote:
... But we definitely ahe enough to put into a PR..
...
Bug 202494 - Panic [page fault] in _ieee80211_crypto_delkey()
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=202494
Peace,
david
--
David H. Wolfskill
Eric,
I updated this driver in kernel, not as module. And I removed
#include opt_rss.h
from if_ixl.c and ixl_txrx.c:
#ifndef IXL_STANDALONE_BUILD
#include opt_inet.h
#include opt_inet6.h
#include opt_rss.h
#endif
because RSS for is only in HEAD
Could I break smth by doing this?
Best regards,
Yeah; it should be able to do up to 64 queues for the PF's. It's possible
for the NVM to limit the RSS table size and entry width, but that seems
unlikely.
- Eric
On Wed, Aug 19, 2015 at 12:41 PM Adrian Chadd adrian.ch...@gmail.com
wrote:
no, it's not the RSS option - it's the RSS
Does it do RSS distribution into 16 queues?
-a
On 19 August 2015 at 11:17, Eric Joyner e...@freebsd.org wrote:
The IXLV_MAX_QUEUES value is for the VF driver; the standard driver should
be able to allocate and properly use up to 64 queues.
That said, you're only getting rx traffic on the
no, it's not the RSS option - it's the RSS configuration in the NIC
for steering traffic into different queues based on header contents.
The RSS kernel option includes the framework that ties it all together
into the network stack - if you don't use it (which is the default),
the NICs are free to
Since FreeBSD is a general-purpose operating system, a fresh install with
default options will certainly not meet all the requirements. However, from a
quick read of the RFC it looks like it would be straightforward to configure a
FreeBSD box to meet the requirements.
For simple routing, the
Daniel Braniss wrote:
On 19 Aug 2015, at 16:00, Rick Macklem rmack...@uoguelph.ca wrote:
Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now
On Wed, Aug 19, 2015 at 09:00:35AM -0400, Rick Macklem wrote:
Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts the # of mbufs is before the
code that adds the tcp/ip header mbuf.
In my opinion, this should be fixed by setting if_hw_tsomaxsegcount to whatever
the driver provides - 1. It is not the driver's
On Tue, Aug 18, 2015 at 06:04:25PM -0400, Rick Macklem wrote:
Hans Petter Selasky wrote:
On 08/18/15 14:53, Rick Macklem wrote:
If this is just a test machine, maybe you could test with these lines (at
about #880)
in sys/netinet/tcp_output.c commented out? (It looks to me like this
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts the # of mbufs is before the
code that adds the tcp/ip header mbuf.
In my opinion, this should be
On Wed, Aug 19, 2015 at 09:51:44AM +0200, Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts the # of mbufs is before
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts the # of mbufs is before the
code that adds the tcp/ip header mbuf.
In my opinion, this should be fixed by setting if_hw_tsomaxsegcount
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick Macklem wrote:
Ouch! Yes, I now see that the code that counts the # of mbufs is before the
code that adds the tcp/ip header mbuf.
In my opinion, this should be
On Wed, Aug 19, 2015 at 08:13:59AM -0400, Rick Macklem wrote:
Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:51:44AM +0200, Hans Petter Selasky wrote:
On 08/19/15 09:42, Yonghyeon PYUN wrote:
On Wed, Aug 19, 2015 at 09:00:52AM +0200, Hans Petter Selasky wrote:
On 08/18/15 23:54, Rick
28 matches
Mail list logo
