2009/12/8 Nicolas Droux <[email protected]>:
>
>
> Chavdar Ivanov wrote:
>>
>> Hi,
>>
>> 'man rge' states, among the rest, .....
>>     The  values  returned  by  the  driver  in the DL_INFO_ACK primitive
>> in
>>       response to the DL_INFO_REQ are as follows:
>>
>>           o      Maximum SDU (with jumbo frame) is 7000.
>>
>>           o      Minimum SDU is 0.
>> .....
>>
>> The hardware apparently supports it; I am using Jumbo frames on three
>> NetBSD machines with the same network adapter; it only warns that the max is
>> 7500, not 9000 as expected:
>>
>>  $ ifconfig re0
>> re0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 7500
>>
>>  capabilities=3f80<TSO4,IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx,UDP4CSUM_Rx,UDP4CSUM_Tx>
>>        enabled=0
>>        address: 00:14:c1:4c:3a:f4
>>        media: Ethernet autoselect (1000baseT full-duplex)
>>        status: active
>>        inet 10.1.32.9 netmask 0xff000000 broadcast 10.255.255.255
>>        inet6 fe80::214:c1ff:fe4c:3af4%re0 prefixlen 64 scopeid 0x1
>>
>> On OpenSolaris I get:  # dladm show-linkprop -p mtu rge0
>> LINK         PROPERTY        PERM VALUE          DEFAULT        POSSIBLE
>> rge0         mtu             r-   1500           1500           1500
>>
>> Is there any particular reason Jumbo frames are not supported on this
>> adapter?
>
> rge doesn't implement the driver properties interfaces and the MTU is not
> going to be manageable through dladm for these NICs. You have to enable
> jumbo frames by setting the "default_mtu" property in /kernel/drv/rge.conf.

Thanks, I'll try it tomorrow (don't want to do it over the VPN from
home right now...)

>
> I filed 6807010 (rge driver should support GLDv3 properties) back in
> February but the driver team has obviously not yet addressed the issue.
>
> Nicolas.
>
>>
>>

(just read PSARC/2009/638; cool).

Chavdar Ivanov

>



-- 
----
Pablo Picasso  - "Computers are useless. They can only give you
answers." - http://www.brainyquote.com/quotes/authors/p/pablo_picasso.html
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to