Re: Hardware checksums on bge(4) interfaces

[Chris Ross <cross+netbsd%distal.com@localhost>]

On Sep 14, 2010, at 6:25 PM, Brian Buhrow wrote:

>       Hello.  Looking at the Linux tg3 driver from the 2.6-36 kernel, it
> appears that the problem is a driver bug in NetBSD.  Only the 5700B series
> of chips appear to suffer from the hardware checksumming problem.  And,
> yes, it affects udp packets as well.
>       If I cook up a patch, are any of you in a position to test?

  My primary development machine is running current, and soon will be reloaded 
with a -current from last night (that I'm in the midst of deploying).  It has:

bge0 at pci1 dev 6 function 0: Broadcom BCM5703X Gigabit Ethernet
bge0: interrupting at ioapic1 pin 12
bge0: ASIC BCM5702/5703 A2 (0x1002), Ethernet address 00:0f:1f:6e:57:94
brgphy0 at bge0 phy 1: BCM5703 1000BASE-T media interface, rev. 2
brgphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 
1000baseT-FDX, auto
bge1 at pci1 dev 8 function 0: Broadcom BCM5703X Gigabit Ethernet
bge1: interrupting at ioapic1 pin 13
bge1: ASIC BCM5702/5703 A2 (0x1002), Ethernet address 00:0f:1f:6e:57:95
brgphy1 at bge1 phy 1: BCM5703 1000BASE-T media interface, rev. 2
brgphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 
1000baseT-FDX, auto

  And can test whatever you like.  I also have a DL380G3 with it running 
netbsd-5 and using the bge's in an agr, but it looks like they're the same part 
anyway:

bge0 at pci4 dev 1 function 0: Broadcom BCM5703X Gigabit Ethernet
bge0: interrupting at ioapic1 pin 13
bge0: ASIC BCM5703 A2 (0x1002), Ethernet address 00:0e:7f:22:3b:1b
brgphy0 at bge0 phy 1: BCM5703 1000BASE-T media interface, rev. 2
brgphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 
1000baseT-FDX, auto
bge1 at pci4 dev 2 function 0: Broadcom BCM5703X Gigabit Ethernet
bge1: interrupting at ioapic1 pin 15
bge1: ASIC BCM5703 A2 (0x1002), Ethernet address 00:0e:7f:22:3b:1a
brgphy1 at bge1 phy 1: BCM5703 1000BASE-T media interface, rev. 2
brgphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 
1000baseT-FDX, auto


  Happy to test things on my 5703 A2 (0x1002) if you have a patch to test.  I 
admit I'm not sure how to demonstrate the presence or lack of the problem, 
however.  I'd need some guidance on that front.  I have only just started 
looking into enabling any of the checksumming features of these interfaces.

                        - Chris