>Number:         11301
>Category:       kern
>Synopsis:       rtk driver spins after receiving a packet
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Oct 23 17:40:00 PDT 2000
>Closed-Date:
>Last-Modified:
>Originator:     Chuck Silvers
>Release:        -current as of today (2000/10/23)
>Organization:
myself
>Environment:
NetBSD spiffy.chuq.com 1.5H NetBSD 1.5H (GENERIC) #4: Mon Oct 23 08:42:00 PDT 2000     chs@spathi.chuq.com:/home/chs/netbsd/src/sys/arch/i386/compile/GENERIC i386


>Description:

the rtk driver spins in rtk_intr() after receiving a packet until
it receives another packet.  this was determined by setting breakpoints
in ddb and otherwise messing around.

another odd thing that probably related:  I occasionally get this message
on the console:

rtk0: discarding oversize frame (len=65535)


boot messages for this card are:

rtk0 at pci0 dev 13 function 0: RealTek 8139 10/100BaseTX
rtk0: interrupting at irq 15
rtk0: Ethernet address 00:48:54:80:1d:fc
ukphy1 at rtk0 phy 7: Generic IEEE 802.3u media interface
ukphy1: OUI 0x000000, model 0x0000, rev. 0
ukphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto


the numbers on the second of the "ukphy1" lines being all zeroes
kinda looks suspicious, but I know diddly-squat about modern
ethernet hardware so maybe that's normal.


>How-To-Repeat:

try to use the rtk driver.  notice your machine hang, off and on.
having another machine ping the machine with the rtk card produces
an amusing start-and-stop behaviour.


>Fix:

dunno.
>Release-Note:
>Audit-Trail:
>Unformatted: