Re: [E1000-devel] e1000e giving message "Warning: detected ASPM enabled in EEPROM" on bootup

2009-03-05 Thread Chris Markle
Adding to this: > I compiled and tried the e1000e driver on my hardware with two PCI-E > NICs. The e1000e driver appears to be working but I noticed on bootup > that the following message was logged for one of the NICs: > > e1000e :03:00.0: Warning: detected ASPM enabled in EEPROM > > I see th

[E1000-devel] e1000e giving message "Warning: detected ASPM enabled in EEPROM" on bootup

2009-03-05 Thread Chris Markle
I compiled and tried the e1000e driver on my hardware with two PCI-E NICs. The e1000e driver appears to be working but I noticed on bootup that the following message was logged for one of the NICs: e1000e :03:00.0: Warning: detected ASPM enabled in EEPROM I see that ASPM is Active Power State

Re: [E1000-devel] 8.0.6 e1000.ko much much larger than older 7.0.33 version

2009-03-05 Thread Chris Leech
On Thu, 2009-03-05 at 17:31 -0800, Chris Markle wrote: > -rw-r--r-- 2 root root 2110478 Mar 5 16:35 e1000.ko > What concerned me and the reason for this inquiry is that the older > e1000 version had a MUCH smaller e1000.ko file, leading me to wonder > if I had done something wrong here. > [r...

Re: [E1000-devel] 8.0.6 e1000.ko much much larger than older 7.0.33 version

2009-03-05 Thread Tantilov, Emil S
"-k2" attached to the version means that the driver is included with the kernel. The standalone driver you downloaded from SF (8.0.6) is designed to work with variety of kernels which adds some compat code. It also may have functionality and/or device support not available in the kernel driver (

[E1000-devel] 8.0.6 e1000.ko much much larger than older 7.0.33 version

2009-03-05 Thread Chris Markle
I am a newby at e1000 building so please bear with me... I recently downloaded, built and installed the e1000 8.0.6 version on my i386 fc4 machine. When the dust cleared here's how things looked: [r...@dd42 ~]# ethtool -i eth0 driver: e1000 version: 8.0.6-NAPI firmware-version: N/A bus-info: :

Re: [E1000-devel] compiling the e1000 driver statically.

2009-03-05 Thread Tantilov, Emil S
Yeah - if none of the other solutions that were pointed out for setting ITR work for you ... Setting DEFAULT_ITR to 0 should do it. Thanks, Emil -Original Message- From: Ezra Taylor [mailto:ezra.tay...@gmail.com] Sent: Thursday, March 05, 2009 4:45 PM To: e1000-devel@lists.sourceforge.ne

Re: [E1000-devel] compiling the e1000 driver statically.

2009-03-05 Thread Chris Leech
On Thu, 2009-03-05 at 16:44 -0800, Ezra Taylor wrote: > Hello again: > Could you confirm that setting DEFAULT_ITR to 0 will > disable Interrupt Throttling. Do I have to set MAX_ITR and MIN_ITR to 0. > Also, are there any other files I have to edit? I won't bother you guys > a

Re: [E1000-devel] compiling the e1000 driver statically.

2009-03-05 Thread Ezra Taylor
Hello again: Could you confirm that setting DEFAULT_ITR to 0 will disable Interrupt Throttling. Do I have to set MAX_ITR and MIN_ITR to 0. Also, are there any other files I have to edit? I won't bother you guys anymore tonight. /* Interrupt Throttle Rate (interrupts/sec) *

Re: [E1000-devel] compiling the e1000 driver statically.

2009-03-05 Thread Ezra Taylor
Thank You Emil. I appreciate you help. On Thu, Mar 5, 2009 at 7:12 PM, Tantilov, Emil S wrote: > If your change is correct you will see a high number of interrupts/sec (for > example when measured with netperf using the RR tests or when transmitting > small packets). With interrupt throttling e

Re: [E1000-devel] compiling the e1000 driver statically.

2009-03-05 Thread Tantilov, Emil S
If your change is correct you will see a high number of interrupts/sec (for example when measured with netperf using the RR tests or when transmitting small packets). With interrupt throttling enabled your int/sec will be capped at either 8k or 20k int/sec (depending on the version of e1000 you'

[E1000-devel] compiling the e1000 driver statically.

2009-03-05 Thread Ezra Taylor
Hello all: First, I would like to apologize if this is the wrong place to ask this question. Here it goes. We are running Kernels with drivers we only need. The driver are compiled statically into the Linux Kernel. I was told recently to disable interrupt throttling to test its im

Re: [E1000-devel] issue in igb driver : Observing high value in rx_missed_errors & rx_no_buffer_count ( in Linux kernel 2.6)

2009-03-05 Thread Brandeburg, Jesse
you might be able to tune ITR using ethtool -C ethx rx-usecs N you may also want to increase or decrease the number of receive buffers using ethtool -G ethx rx N if the stack consumes too many cycles per packet it is normal for the driver to drop frames, that is what is architected to happen wi

Re: [E1000-devel] Kernel 2.6.27.10: e1000e hangs during boot at MTU change

2009-03-05 Thread Brandeburg, Jesse
Hi Carsten, any update? Carsten Aulbert wrote: > Hi Jesse, > > sorry for the late reply, too much other work to rank this question > high enough. > > I was able to work around the problem by adding netconsole module > loading to a dhclient hook. > > Brandeburg, Jesse schrieb: >>> >>> 0d:00.0 0