2.6.16.13 e1000 reports incorrect PCI-X bus speed?

netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed

* 2.6.16.13 e1000 reports incorrect PCI-X bus speed?
@ 2006-05-12 21:15 Ben Greear
  2006-05-12 21:33 ` Auke Kok
  2006-05-12 21:41 ` Rick Jones
  0 siblings, 2 replies; 5+ messages in thread
From: Ben Greear @ 2006-05-12 21:15 UTC (permalink / raw)
  To: NetDev

I'm running a slightly modified 2.6.16.13 kernel on FC5-amd64.  The motherboard
is SuperMicro H8SSL dual-core AMD system.  According to super-micro web site,
the PCI-X slot is 133Mhz.  I'm using a 4-port pro/1000 NIC.

dmesg shows a listing of 120Mhz:

Intel(R) PRO/1000 Network Driver - version 6.3.9-k4-NAPI
Copyright (c) 1999-2005 Intel Corporation.
GSI 18 sharing vector 0xB9 and IRQ 18
ACPI: PCI Interrupt 0000:03:04.0[A] -> GSI 20 (level, low) -> IRQ 18
e1000: 0000:03:04.0: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bc
e1000: eth2: e1000_probe: Intel(R) PRO/1000 Network Connection
GSI 19 sharing vector 0xC1 and IRQ 19
ACPI: PCI Interrupt 0000:03:04.1[B] -> GSI 21 (level, low) -> IRQ 19
e1000: 0000:03:04.1: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bd
e1000: eth3: e1000_probe: Intel(R) PRO/1000 Network Connection
GSI 20 sharing vector 0xC9 and IRQ 20
ACPI: PCI Interrupt 0000:03:06.0[A] -> GSI 22 (level, low) -> IRQ 20
e1000: 0000:03:06.0: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:be
e1000: eth4: e1000_probe: Intel(R) PRO/1000 Network Connection
GSI 21 sharing vector 0xD1 and IRQ 21
ACPI: PCI Interrupt 0000:03:06.1[B] -> GSI 23 (level, low) -> IRQ 21
e1000: 0000:03:06.1: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bf
e1000: eth5: e1000_probe: Intel(R) PRO/1000 Network Connection

ethtool -d eth3 shows 66Mhz:
....
0x00008: STATUS (Device status register) 0x00003B87
       Duplex:                            full
       Link up:                           link config
       TBI mode:                          disabled
       Link speed:                        1000Mb/s
       Bus type:                          PCI-X
       Bus speed:                         66MHz
       Bus width:                         64-bit

Any idea why I see the slower bus speed reported?

More info available as needed.

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: 2.6.16.13 e1000 reports incorrect PCI-X bus speed?
  2006-05-12 21:15 2.6.16.13 e1000 reports incorrect PCI-X bus speed? Ben Greear
@ 2006-05-12 21:33 ` Auke Kok
  2006-05-12 22:26   ` Ben Greear
  2006-05-12 21:41 ` Rick Jones
  1 sibling, 1 reply; 5+ messages in thread
From: Auke Kok @ 2006-05-12 21:33 UTC (permalink / raw)
  To: netdev

Ben Greear wrote:
> I'm running a slightly modified 2.6.16.13 kernel on FC5-amd64.  The 
> motherboard
> is SuperMicro H8SSL dual-core AMD system.  According to super-micro web 
> site,
> the PCI-X slot is 133Mhz.  I'm using a 4-port pro/1000 NIC.
> 
> dmesg shows a listing of 120Mhz:
> 
> Intel(R) PRO/1000 Network Driver - version 6.3.9-k4-NAPI
> Copyright (c) 1999-2005 Intel Corporation.
> GSI 18 sharing vector 0xB9 and IRQ 18
> ACPI: PCI Interrupt 0000:03:04.0[A] -> GSI 20 (level, low) -> IRQ 18
> e1000: 0000:03:04.0: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bc
> e1000: eth2: e1000_probe: Intel(R) PRO/1000 Network Connection
> GSI 19 sharing vector 0xC1 and IRQ 19
> ACPI: PCI Interrupt 0000:03:04.1[B] -> GSI 21 (level, low) -> IRQ 19
> e1000: 0000:03:04.1: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bd
> e1000: eth3: e1000_probe: Intel(R) PRO/1000 Network Connection
> GSI 20 sharing vector 0xC9 and IRQ 20
> ACPI: PCI Interrupt 0000:03:06.0[A] -> GSI 22 (level, low) -> IRQ 20
> e1000: 0000:03:06.0: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:be
> e1000: eth4: e1000_probe: Intel(R) PRO/1000 Network Connection
> GSI 21 sharing vector 0xD1 and IRQ 21
> ACPI: PCI Interrupt 0000:03:06.1[B] -> GSI 23 (level, low) -> IRQ 21
> e1000: 0000:03:06.1: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bf
> e1000: eth5: e1000_probe: Intel(R) PRO/1000 Network Connection
> 
> ethtool -d eth3 shows 66Mhz:
> ....
> 0x00008: STATUS (Device status register) 0x00003B87
>       Duplex:                            full
>       Link up:                           link config
>       TBI mode:                          disabled
>       Link speed:                        1000Mb/s
>       Bus type:                          PCI-X
>       Bus speed:                         66MHz
>       Bus width:                         64-bit
> 
> Any idea why I see the slower bus speed reported?

ethtool currently doesn't know about some bus speeds, and will report a wrong
speed. Here's what ethtool currently uses: (ethtool-3/e1000.c)

327                 (reg & E1000_STATUS_PCIX_MODE) ?
328                         ((reg & E1000_STATUS_PCIX_SPEED_133) ? "133MHz" :
329                         (reg & E1000_STATUS_PCIX_SPEED_100) ? "100MHz" :
330                         "66MHz")>              :

obviously 120 is missing, but PCI-E speeds are also missing (2500gbps). The
output of the e1000 module is correct.

Auke


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: 2.6.16.13 e1000 reports incorrect PCI-X bus speed?
  2006-05-12 21:33 ` Auke Kok
@ 2006-05-12 22:26   ` Ben Greear
  2006-05-14  0:54     ` Chris Leech
  0 siblings, 1 reply; 5+ messages in thread
From: Ben Greear @ 2006-05-12 22:26 UTC (permalink / raw)
  To: Auke Kok; +Cc: netdev

Auke Kok wrote:

> obviously 120 is missing, but PCI-E speeds are also missing (2500gbps). The
> output of the e1000 module is correct.

Any idea why 120Mhz is used instead of 133?  It doesn't
seem to matter in my performance tests, but I am curious...

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: 2.6.16.13 e1000 reports incorrect PCI-X bus speed?
  2006-05-12 22:26   ` Ben Greear
@ 2006-05-14  0:54     ` Chris Leech
  0 siblings, 0 replies; 5+ messages in thread
From: Chris Leech @ 2006-05-14  0:54 UTC (permalink / raw)
  To: Ben Greear; +Cc: Auke Kok, netdev

> Any idea why 120Mhz is used instead of 133?  It doesn't
> seem to matter in my performance tests, but I am curious...

I think Rick is right, the bus between the bridge on the card and the
e1000s is running at 120Mhz.

- Chris

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: 2.6.16.13 e1000 reports incorrect PCI-X bus speed?
  2006-05-12 21:15 2.6.16.13 e1000 reports incorrect PCI-X bus speed? Ben Greear
  2006-05-12 21:33 ` Auke Kok
@ 2006-05-12 21:41 ` Rick Jones
  1 sibling, 0 replies; 5+ messages in thread
From: Rick Jones @ 2006-05-12 21:41 UTC (permalink / raw)
  To: Ben Greear; +Cc: NetDev

Ben Greear wrote:
> I'm running a slightly modified 2.6.16.13 kernel on FC5-amd64.  The 
> motherboard
> is SuperMicro H8SSL dual-core AMD system.  According to super-micro web 
> site,
> the PCI-X slot is 133Mhz.  I'm using a 4-port pro/1000 NIC.
> 
> dmesg shows a listing of 120Mhz:

IIRC there is a bridge chip on those four-port cards.  Perhaps the 
"north" side of the bridge is at 133 MHz, but the south side (since it 
is more than one load) is only 120?

> Intel(R) PRO/1000 Network Driver - version 6.3.9-k4-NAPI
> Copyright (c) 1999-2005 Intel Corporation.
> GSI 18 sharing vector 0xB9 and IRQ 18
> ACPI: PCI Interrupt 0000:03:04.0[A] -> GSI 20 (level, low) -> IRQ 18
> e1000: 0000:03:04.0: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bc
> e1000: eth2: e1000_probe: Intel(R) PRO/1000 Network Connection
> GSI 19 sharing vector 0xC1 and IRQ 19
> ACPI: PCI Interrupt 0000:03:04.1[B] -> GSI 21 (level, low) -> IRQ 19
> e1000: 0000:03:04.1: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bd
> e1000: eth3: e1000_probe: Intel(R) PRO/1000 Network Connection
> GSI 20 sharing vector 0xC9 and IRQ 20
> ACPI: PCI Interrupt 0000:03:06.0[A] -> GSI 22 (level, low) -> IRQ 20
> e1000: 0000:03:06.0: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:be
> e1000: eth4: e1000_probe: Intel(R) PRO/1000 Network Connection
> GSI 21 sharing vector 0xD1 and IRQ 21
> ACPI: PCI Interrupt 0000:03:06.1[B] -> GSI 23 (level, low) -> IRQ 21
> e1000: 0000:03:06.1: e1000_probe: (PCI-X:120MHz:64-bit) 00:04:23:ad:41:bf
> e1000: eth5: e1000_probe: Intel(R) PRO/1000 Network Connection
> 
> ethtool -d eth3 shows 66Mhz:
> ....
> 0x00008: STATUS (Device status register) 0x00003B87
>       Duplex:                            full
>       Link up:                           link config
>       TBI mode:                          disabled
>       Link speed:                        1000Mb/s
>       Bus type:                          PCI-X
>       Bus speed:                         66MHz
>       Bus width:                         64-bit
> 
> Any idea why I see the slower bus speed reported?

That one though escapes me.

rick jones

> 
> More info available as needed.
> 
> Thanks,
> Ben
> 


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2006-05-14  0:54 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-05-12 21:15 2.6.16.13 e1000 reports incorrect PCI-X bus speed? Ben Greear
2006-05-12 21:33 ` Auke Kok
2006-05-12 22:26   ` Ben Greear
2006-05-14  0:54     ` Chris Leech
2006-05-12 21:41 ` Rick Jones

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).