eeupdate is not a public domain tool and is only available under NDA. Please
contact your factory representative for technical support.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From:
There is no publically accessible git repository for the out-of-tree driver.
There is only the README, so it's more useful to do diffs between the versions.
The other useful document is the Specification Update for the networking part
you are using. It's available publically on intel.com. Just
Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Jeff Westfahl [mailto:jeff.westf...@ni.com]
Sent: Tuesday, May 20, 2014 12:34 PM
To: Fujinaka, Todd
Cc: Jeff Westfahl; e1000-devel@lists.sourceforge.net
Subject: RE: [E1000-devel] i210 register TXPBSIZE is only reset
So the offsets are off because you are tunneling and it looks like you're
coming from the same port and address.
To answer some of your other questions:
RSS is defined by Microsoft and you'll have to take that up with them.
You've now been given two options to balance your traffic: vmdq (not
RSS hashing is pretty simple. Do you have the same source and destination
address and port? If so, you get one hash. I'm not sure what your packets look
like, I've been told by those who know more about PPoE that you'll need to use
VMDq.
I know there are other ways for traffic to be routed to
I think it's going to take me a bit of time to look into this, but I don' t
think you can reset it with a static value. Let me go poke at some other people
and see what they think.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
Can you send lspci with more verbosity? Perhaps run as root?
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Ben Greear [mailto:gree...@candelatech.com]
Sent: Thursday, May 08, 2014
-
From: Fujinaka, Todd [mailto:todd.fujin...@intel.com]
Sent: Thursday, May 08, 2014 12:56 PM
To: Ben Greear; e1000-devel list
Subject: Re: [E1000-devel] Hardware unit hang in 3.14.3+ kernel, e1000e driver.
Can you send lspci with more verbosity? Perhaps run as root?
Todd Fujinaka
Software
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
From: shiv prakash Agarwal [mailto:chhotu.s...@gmail.com]
Sent: Sunday, May 04, 2014 11:55 PM
To: Duyck, Alexander H; linux-arm-ker...@lists.infradead.org
Cc: Fujinaka, Todd; e1000-devel
The filtering is enabled because you don't have any VLANs configured. In igb we
assume you have the 802.1q module installed and that enabled VLAN 0.
I think our ixgbe driver enables VLAN 0 by default (or so I've been told) and
we could do that for igb as well if you think that's necessary.
I'm afraid that the issue you're seeing is with the PCIe in your ARM system.
You're going to have to clear that up before you can get anything working on
the PCIe bus such as an Ethernet controller. I would suggest contacting an ARM
mailing list.
Todd Fujinaka
Software Application Engineer
The latest driver is igb-5.2.5. Please try that.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Grzegorz Kuczyński [mailto:gk180...@interia.pl]
Sent: Monday, April 28, 2014 10:29 AM
I would suggest that you ask your FAE this question. It's unclear exactly what
you're asking for and the FAE should be able to provide a wider range of
documentation than what we can provide publically.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
I think you're barking up the wrong tree here.
First, make sure both sides aren't forcing speed or duplex. They should both be
set to 1G auto. If one side is forced to a particular speed or to a particular
duplex, that just changes the advertised capabilities and the other side will
be
Also, I'm not sure what you mean by 0xBABA. That would be an interesting
coincidence if the checksum was 0xBABA, but the checksum should be recalculated
when the iNVM is changed. There could be a chance that whoever programmed the
part was using an older broken version of the programming tool.
To: Fujinaka, Todd
Cc: Ronciak, John; e1000-devel@lists.sourceforge.net
Subject: Re: [E1000-devel] igb: v5.2.5 fails doing NVM checksum check on i211
On Mon, Apr 14, 2014 at 07:51:16PM +, Fujinaka, Todd wrote:
Also, I'm not sure what you mean by 0xBABA. That would be an interesting
coincidence
We also need to know what part you're using. lspci | grep Ethernet should
narrow that down.
Thanks.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Borislav Petkov
, April 02, 2014 7:04 PM
To: David Miller; Fujinaka, Todd; Skidmore, Donald C;
linux-ker...@vger.kernel.org; e1000-devel@lists.sourceforge.net;
net...@vger.kernel.org
Cc: Hiroshi Shimamoto; Hiroshi Baba
Subject: RE: [E1000-devel] fail to add 64 VLANs or more when SR-IOV is enabled
-Original
I think we know what's going on and we're looking into this.
Thanks.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Ben Greear [mailto:gree...@candelatech.com]
Sent: Monday, March
Ben,
Can you tell us your exact hardware just in case that's required for this issue
to reproduce?
Thanks.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Fujinaka, Todd
) 712-4565
-Original Message-
From: Fujinaka, Todd
Sent: Thursday, April 03, 2014 1:21 PM
To: Fujinaka, Todd; Ben Greear; e1000-devel list
Subject: RE: [E1000-devel] irq, watchdog errors with 82574L, TSO, and small TCP
MSS.
Ben,
Can you tell us your exact hardware just in case that's
I think we need more details. What part and what driver?
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: nirmoy das [mailto:nirmoy.ai...@gmail.com]
Sent: Wednesday, April 02, 2014 2:51
I think you're running into the filter limit for VFs.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Madoka Komatsubara [mailto:m-komatsub...@ab.jp.nec.com]
Sent: Wednesday, April
I think some of these questions would be best answered under NDA. Can you
contact your vendor and go through that channel?
Thanks.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: x
: x aus [mailto:fos...@gmail.com]
Sent: Wednesday, March 26, 2014 12:38 PM
To: Fujinaka, Todd
Cc: e1000-devel@lists.sourceforge.net
Subject: Re: [E1000-devel] 82576/igb custom driver
We had NDA in place, other than the MACSEC item(#6), which item do you think
should better go to Intel too
There should be a doc file with the flashless update that you should follow.
What is the version of eeupdate you are running? And is there a reason you're
running an older version of igb? Can you try the newest version on
e1000.sourceforge.net? igb-5.1.2 changed some of the features for
-
From: Lauper Alexander [mailto:alexander.lau...@comlab.ch]
Sent: Thursday, March 06, 2014 9:10 AM
To: Fujinaka, Todd; e1000-devel@lists.sourceforge.net
Subject: RE: igb-5.1.2 compilation fails
Hi Todd
Any news concerning this issue?
Kind regards,
Alex
-Ursprüngliche Nachricht-
Von
The PXE spec does not support IPv6. The only way you might be able to get this
going is with uEFI, and we don't do the uEFI drivers. You'll have to contact
the uEFI folks.
Thanks.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
: Wednesday, February 26, 2014 7:55 AM
To: Fujinaka, Todd
Cc: e1000-devel@lists.sourceforge.net
Subject: Re: [E1000-devel] burn ipxe to card
2014-02-26 19:50 GMT+04:00 Fujinaka, Todd todd.fujin...@intel.com:
The PXE spec does not support IPv6. The only way you might be able to get
this going
You have to fix these in the reverse order. First, compile the LATEST driver,
which is 3.19.1, not 3.18.7. It's available on e1000.sourceforge.net. That
should resolve the build issue.
Second, use the module parameter allow_unsupported_sfp=1,1 when you load the
driver, and this may work. We're
: Fujinaka, Todd [mailto:todd.fujin...@intel.com]
Sent: Thursday, December 05, 2013 11:01 AM
To: Julien Houles; e1000-devel@lists.sourceforge.net
Subject: Re: [E1000-devel] Problem with SYSTIMH register readout
We're going to try to reproduce this locally. We may or may not have the same
motherboard so
[mailto:noname.e...@yahoo.com]
Sent: Tuesday, February 18, 2014 1:00 AM
To: Duyck, Alexander H; Skidmore, Donald C; Fujinaka, Todd;
e1000-devel@lists.sourceforge.net
Subject: Re: [E1000-devel] Access to cards serial numbers
Thank you all
The following is the output of lspci -xxx.
0c:00.0 Ethernet
The only serial number available is the MAC address, AFAIK.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Nomad Esst [mailto:noname.e...@yahoo.com]
Sent: Wednesday, February 12,
If you're doing this for school, I say go for it. If you're doing this for
performance, I'm not sure how useful the effort will be. We should be able to
handle hundreds of thousands of MSI-X interrupts per second, as they're
messages on the PCIe bus rather than a separate asynchronous signal.
Where did you get the parts? That's who you should contact.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Lauper Alexander [mailto:alexander.lau...@comlab.ch]
Sent: Thursday,
Is the Debian kernel the same as the upstream kernel? We don't support Debian
directly, be we can look into why this is not compiling on the upstream kernel.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
I think I've figured out the problem. Unfortunately, I was remotely testing it
tonight and my system is no longer on the network. I'll have to get back to you
tomorrow.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Vijay Agrawal [mailto:vagra...@ixiacom.com]
Sent: Tuesday, January 28, 2014 10:07 PM
To: Fujinaka, Todd; e1000-de...@lists.sf.net
Subject: RE: [E1000-devel] ixgbe-3.17.3 :VLAN tagged packets does not work when
using PF-PACKET socket
Hi Todd,
Just wanted
/cores.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
From: Sharva Pathak [mailto:pat...@princeton.edu]
Sent: Wednesday, January 22, 2014 2:09 PM
To: Fujinaka, Todd
Cc: e1000-devel@lists.sourceforge.net
Subject: Re: [E1000
Looks like you're using custom hardware from Doremi Labs. There's a lot of
questions we need to ask about the hardware setup as well as the software. Can
you file a bug on sourceforge? Also, have you tried any other kernels with
success?
Thanks.
Todd Fujinaka
Software Application Engineer
We've found in preliminary testing that the PCIe bus (1 lane of PCIe Gen 1) was
the real bottleneck for us.
Enable support in this case means you get to do a lot of modifications to the
driver for, perhaps, not much gain. I would suggest it would be quicker to find
a newer part with multiqueue
To: Fujinaka, Todd
Cc: e1000-devel@lists.sourceforge.net
Subject: Re: [E1000-devel] [PATCH] e1000e: Permit operation without
non-volatile memory
Disabling the NVM might not work with all of your parts. I can only test with
what I have available.
It's definitely working with the 82574 here. And if it's
Can you tell us which motherboard you're using? This is the weekend for me so
I'm just giving you a quick (little-research) answer, but most recent
motherboards I've seen require 3 DIMMs per CPU or performance will be impacted.
Todd Fujinaka
Software Application Engineer
Networking Division
(ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Fujinaka, Todd [mailto:todd.fujin...@intel.com]
Sent: Tuesday, September 17, 2013 8:54 AM
To: ??; e1000-de...@lists.sf.net
Subject: Re: [E1000-devel] About NIC 82574L
There is only one queue enabled
If rx_dma_resources show the count from QPRDC, which is in section 8.2.3.23.75
of the datasheet. The latest datasheet is available on intel.com by searching
for 82599 datasheet.
Total number of receive packets dropped for the queue. Packets can be dropped
for the following reasons:
1. Rx queue
There is only one queue enabled because we found in initial testing that we
weren't seeing any performance benefit from the second queue. The single lane
of PCIe at 2.5GT/s was our main bottleneck at that time.
You can certainly enable a second queue if you write your own driver. You have
to
Is this on the e1000 driver? Can you tell us which part you're using as well?
How are you configuring the link? Are you bringing the link down with ifdown?
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
I don't see anything in particular in your logs (I was only looking for a
couple of things in the lspci) and if the system is working for you now I'll
consider this problem solved. Please let us know if you have any further
questions.
Thanks.
Todd Fujinaka
Software Application Engineer
I just noticed there were questions at the bottom of the logs.
The power savings and CPU scaling need to be balanced with performance. The
time between packets is long enough for the CPU to go to sleep, even at 10G. I
can't make any suggestions without knowing your setup, but for full
I will look into this. Thanks for the report.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Sergey Egorov [mailto:egorovh...@gmail.com]
Sent: Friday, August 16, 2013 3:21 AM
To:
Can you send us your lspci -vvv (run as root)? You may want to open a bug at
sourceforge and attach the logs there. We would also like to see the full dmesg
and /var/log/messages.
Also, send us ethtool -S. Ideally we would like to see it before and after the
test.
Can you describe your
I don't believe mcelog has anything to do with the networking driver. You may
need to contact supp...@intel.com.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: ratheesh kannoth
I'm afraid you're going to have to ask the provider of the kernel and kernel
module for support. We don't support 82574 in the e1000 driver and haven't
since the e1000e split in 2008. There have been a lot of updates to the driver
since that time.
I would also suggest checking the
The latest kernel should turn off ASPM as well, but you should be able to check
by looking at lspci -vvv. I think LnkCtl should say ASPM disabled.
Sorry for top-posting.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
Nothing jumped out at me when I checked, but here's the documentation check I
did to see if there was something obvious. First, download the specification
update for your part (go to intel.com and search for 82571EB spec update).
Then, check the release date for your driver (I usually just
The simple answer is that you are running a stress test and there is some
finite processing being done separating traffic into queues which is showing
some performance impact on your setup. At that rate you could also be running
into limitations of the PCIe bus and the added latency of going
Just to reply to the list, this happened to be due to a bug in ethtool (present
in Ubuntu 12.04). The checksums aren't recalculated for all four ports on the
i350 due to this bug. Michael had to download the latest ethtool and that fixed
things (or so I hope!)
Todd Fujinaka
Software
The latest ixgbe driver from e1000.sourceforge.net has the set_irq_affinity.sh
script included. I would certainly suggest you use a newer driver because we're
constantly improving the driver for performance and fixing bugs we've found.
The version you have is a couple of years old.
As far as
)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Ethan Zhao [mailto:ethan.ker...@gmail.com]
Sent: Wednesday, November 28, 2012 7:10 PM
To: Fujinaka, Todd
Cc: Joe Jin; Ben Hutchings; Mary Mcgrath; net...@vger.kernel.org;
e1000-de...@lists.sf.net; linux-ker
-
From: Joe Jin [mailto:joe@oracle.com]
Sent: Wednesday, November 28, 2012 12:31 AM
To: Ben Hutchings
Cc: Fujinaka, Todd; Mary Mcgrath; net...@vger.kernel.org;
e1000-de...@lists.sf.net; linux-ker...@vger.kernel.org; linux-pci
Subject: Re: [E1000-devel] 82571EB: Detected Hardware Unit Hang
Sent: Monday, November 26, 2012 8:00 PM
To: Fujinaka, Todd
Cc: Dave, Tushar N; net...@vger.kernel.org; e1000-de...@lists.sf.net;
linux-ker...@vger.kernel.org; Mary Mcgrath
Subject: Re: [E1000-devel] 82571EB: Detected Hardware Unit Hang
On 11/27/12 00:23, Fujinaka, Todd wrote:
If you look
Those are completely different parts with completely different architectures
(and completely different problems). All the fixes I know for your onboard part
(the 82579LM if I'm reading things correctly) are from the motherboard group so
you should check to see that your BIOS is up-to-date.
, 2012 10:11 AM
To: Fujinaka, Todd; Mary Mcgrath
Cc: Joe Jin; net...@vger.kernel.org; e1000-de...@lists.sf.net;
linux-ker...@vger.kernel.org; linux-pci
Subject: RE: [E1000-devel] 82571EB: Detected Hardware Unit Hang
On Tue, 2012-11-27 at 17:32 +, Fujinaka, Todd wrote:
Forgive me if I'm being
On Tue, 20 Nov 2012, Joe Jin wrote:
On 11/20/12 16:59, Dave, Tushar N wrote:
Have you power off the system completely after modifying eeprom? If not
please do so.
Hi Tushar,
Seems not works for me, would you please help to check what is wrong of my
operations?
...
# lspci -s
The latest driver igb-3.2.9 certainly has support and a grep shows:
#define E1000_DEV_ID_I350_COPPER 0x1521
#define E1000_DEV_ID_I350_FIBER 0x1522
#define E1000_DEV_ID_I350_SERDES 0x1523
#define E1000_DEV_ID_I350_SGMII 0x1524
#define
I am currently aware of ET2 drivers for Linux, FreeBSD, and Windows for x86,
x86_64, and ia64 hardware. The ET card, which I'm fairly sure you don't have
(the sticker on the card will say ET2 on the card in one place on the ET2 card)
may have drivers for your hardware as it is a year older.
If you need Intel Support (and not Intel support, there is a difference) you
should email supp...@intel.com and not the sourceforge list. While we are
official support (the verb) the organization that provides Intel Support
(factory returns, etc) is not on this list.
Todd Fujinaka
Technical
...@lsi.com]
Sent: Thursday, February 17, 2011 7:50 AM
To: Fujinaka, Todd; Brandeburg, Jesse
Cc: e1000-de...@lists.sf.net; Ronciak, John
Subject: RE: [E1000-devel] Intel E1G44ET PCIe card
Well, I do not know who is who. All I know is that we bought a card that does
not work.
We have wasted so much time
You'll need a newer driver. Go to http://e1000.sourceforge.net and get the
latest igb driver.
Todd Fujinaka
Technical Marketing Engineer
LAN Access Division (LAD)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Etemadi, Mohammad Reza (Reza)
Actually, I'm the keeper of the patches. I believe we only give them
out through the FAEs, but I can find out if you'd like. Please let me know.
Todd Fujinaka
Technical Marketing Engineer
LAN Access Division (LAD)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original
On 09/07/2010 04:16 PM, Chris Friesen wrote:
I'm investigating an issue where under load we eventually suddenly stop
receiving packets. We're sending out packets, I can see the reply being
sent with tcpdump on other hosts, but we never receive the reply. The
driver doesn't complain about
Access Division (LAD)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Chris Friesen [mailto:chris.frie...@genband.com]
Sent: Monday, September 13, 2010 3:20 PM
To: Fujinaka, Todd
Cc: e1000-devel@lists.sourceforge.net
Subject: Re: [E1000-devel] strangeness
301 - 372 of 372 matches
Mail list logo