The following reply was made to PR kern/36926; it has been noted by GNATS.

From: "Greg A. Woods" <woods@planix.com>
To: Pavel Cahyna <pavel@NetBSD.org>
Cc: NetBSD GNATS <gnats-bugs@NetBSD.org>,
	<kern-bug-people@NetBSD.org>, <gnats-admin@NetBSD.org>
Subject: Re: kern/36926: recent pullups to bge(4) on netbsd-4 seem to have broken hardware-assisted header checksums for "ASIC BCM5701 B5 (0x0105)"
Date: Tue, 02 Oct 2007 16:38:44 -0400

 --pgp-sign-Multipart_Tue_Oct__2_16:38:44_2007-1
 Content-Type: text/plain; charset=US-ASCII
 Content-Transfer-Encoding: quoted-printable
 
 At Mon, 1 Oct 2007 21:51:40 +0200, Pavel Cahyna wrote:
 Subject: Re: kern/36926: recent pullups to bge(4) on netbsd-4 seem to have =
 broken hardware-assisted header checksums for "ASIC BCM5701 B5 (0x0105)"
 >=20
 > does the latest fix to bge
 > (http://mail-index.netbsd.org/source-changes/2007/09/28/0008.html) fix
 > your problems?
 
 Yes, indeed it does!
 
 Thanks very much!
 
 I'm looking forward to the pullup to netbsd-4  :-)
 
 
 On the DELL PE2650 (where I first noticed the problem):
 
 	bge0 at pci1 dev 6 function 0: Broadcom BCM5701 Gigabit Ethernet
 	bge0: interrupting at ioapic1 pin 12 (irq 10)
 	bge0: ASIC BCM5701 B5 (0x0105), Ethernet address 00:06:5b:ed:e2:d1
 	brgphy0 at bge0 phy 1: BCM5701 1000BASE-T media interface, rev. 0
 
 $ /sbin/ifconfig bge0
 bge0: flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
         capabilities=3D3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx,U=
 DP4CSUM_Rx,UDP4CSUM_Tx>
         enabled=3D3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx,UDP4CS=
 UM_Rx,UDP4CSUM_Tx>
         address: 00:06:5b:ed:e2:d1
         media: Ethernet autoselect (100baseTX full-duplex,flowcontrol,rxpau=
 se,txpause)
         status: active
         inet 204.92.254.13 netmask 0xffffff00 broadcast 204.92.254.255
 
 
 On the IBM x336:
 
 	bge0 at pci6 dev 0 function 0: Broadcom BCM5721 Gigabit Ethernet
 	bge0: interrupting at ioapic0 pin 16 (irq 11)
 	bge0: pcie mode=3D0x105000
 	bge0: ASIC BCM5751 A1 (0x4101), Ethernet address 00:14:5e:33:69:ca
 	bge0: setting short Tx thresholds
 	brgphy0 at bge0 phy 1: BCM5750 1000BASE-T media interface, rev. 0
 
 $ /sbin/ifconfig bge0
 bge0: flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
         capabilities=3D3f80<TSO4,IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM=
 _Tx,UDP4CSUM_Rx,UDP4CSUM_Tx>
         enabled=3D3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx,UDP4CS=
 UM_Rx,UDP4CSUM_Tx>
         address: 00:14:5e:33:69:ca
         media: Ethernet autoselect (100baseTX full-duplex)
         status: active
         inet 204.92.254.7 netmask 0xffffff00 broadcast 204.92.254.255
 
 
 
 As an interesting side note, the Dell variant is significantly faster,
 at least at ping floods, even though the IBM is a significantly faster
 system otherwise:
 
 
 	ibm:
 
 # ping -q -c 10000 -n -f 204.92.254.9
 PING 204.92.254.9 (204.92.254.9): 56 data bytes
 
 ----204.92.254.9 PING Statistics----
 10000 packets transmitted, 10000 packets received, 0.0% packet loss
 round-trip min/avg/max/stddev =3D 0.241/0.243/0.501/0.003 ms
   4079.4 packets/sec sent,  4079.9 packets/sec received
 
 
 	dell:
 
 # ping -q -c 10000 -n -f 204.92.254.9
 PING 204.92.254.9 (204.92.254.9): 56 data bytes
 
 ----204.92.254.9 PING Statistics----
 10000 packets transmitted, 10000 packets received, 0.0% packet loss
 round-trip min/avg/max/stddev =3D 0.130/0.145/0.399/0.018 ms
   6689.9 packets/sec sent,  6690.5 packets/sec received
 
 
 (the .9 address is a DEC DE500-XA in my Alpha-PC264DP and there's an HP
 4000M switch in between them)
 
 
 --=20
 						Greg A. Woods
 
 H:+1 416 218-0098 W:+1 416 489-5852 x122 VE3TCP RoboHack <woods@robohack.ca>
 Planix, Inc. <woods@planix.com>       Secrets of the Weird <woods@weird.com>
 
 --pgp-sign-Multipart_Tue_Oct__2_16:38:44_2007-1
 Content-Type: application/pgp-signature
 Content-Transfer-Encoding: 7bit
 
 -----BEGIN PGP SIGNATURE-----
 Version: PGPfreeware 5.0i for non-commercial use
 MessageID: 7nISvPPsSiUfw67BtYCRGwr3aJsBm185
 
 iQA/AwUBRwKsVGZ9cbd4v/R/EQKAZgCghJ60JlPJUXjyMl+m2HBD564TdgkAoJIU
 TjZ2XUasnwBAO3i0oqmMC9Ow
 =Y5ma
 -----END PGP SIGNATURE-----
 
 --pgp-sign-Multipart_Tue_Oct__2_16:38:44_2007-1--