misc/62324: Onboard Broadcom BCM5705 controller not initialized properly

Will Saxon saxonww at ufl.edu
Tue Feb 3 16:00:46 PST 2004


>Number:         62324
>Category:       misc
>Synopsis:       Onboard Broadcom BCM5705 controller not initialized properly
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-bugs
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Tue Feb 03 16:00:36 PST 2004
>Closed-Date:
>Last-Modified:
>Originator:     Will Saxon
>Release:        5.2
>Organization:
>Environment:
Not currently booted into FreeBSD on this machine. This is an amd64 machine and the problem exists in 5.2-RELEASE in both amd64 and i386 builds.
>Description:
During module insertion or using GENERIC with the compiled in bge driver, with or without ACPI enabled:

bge0: <Broadcom BCM5705 Gigabit Ethernet, ASIC rev. 0x3003> mem 0xfb100000-0xfb10ffff irq 5 at device 11.0 on pci0

bge0: firmware handshake timed out

bge0: RX CPU self-diagnostics failed!

bge0: chip initialization failed

device_probe_and_attach: bge0 attach returned 6


This problem also seems to occur with the linux tg3 driver, although the linux bcm5700 driver does work.

This chip is integrated on an MSI K8T Master2-FAR dual Opteron board. 
>How-To-Repeat:
     Happens every time you attempt to initialize the adapter
>Fix:
      I read in the mail archives about this possibly being fixed by altering the DELAY(n) statements in /usr/src/sys/dev/bge/if_bge.c. I went through that file and multiplied each value by 10 and then by 1000, but that did not work.
>Release-Note:
>Audit-Trail:
>Unformatted:


More information about the freebsd-bugs mailing list