Re: bat0 not capable of MTU > 1500

Re: bat0 not capable of MTU > 1500

[Rich Chase]

I am trying to create an unimpeded path MTU of 1500 on my OpenWRT router (Belkin RT-3200). I built the firmware from git Master, with kernel version 6.1, and batman-adv version 2023.2.
I have a batmesh hard-if device mastering an 802.1S mesh, which are both set to MTU=2304.

Using Luci (OpenWRT web gui for uci), I set the MTU of bat0 to 1504 to accommodate .1q vlans on bat0 -- bat0.5, bat0.110, etc.

I see in the system log:

    kern.err kernel: [751585.375855] bat0: mtu greater than device maximum


Also, if I do:

    #> ip link set bat0 mtu 1504


I get:

    ip: SIOCSIFMTU: Invalid argument


and if I test:

    #> ip link show bat0

...it says that my mtu is 1500 still.

For my life I can't see anything I'm doing wrong. Why can't bat0
have MTU greater than 1500?

I don't have any other devices mastered by bat0, other than my .1s wifi mesh link.

Throwing myself at your mercy. Help?

Re: bat0 not capable of MTU > 1500

[Sven Eckelmann]

[-- Attachment #1: Type: text/plain, Size: 425 bytes --]

On Tuesday, 12 December 2023 20:30:24 CET Rich Chase wrote:
> For my life I can't see anything I'm doing wrong. Why can't bat0
> have MTU greater than 1500?
> 
> I don't have any other devices mastered by bat0, other than my .1s wifi mesh link.
> 
> Throwing myself at your mercy. Help?

https://lore.kernel.org/all/20230727-jumbo-mtu-v1-3-036ce1a6e33c@narfation.org/
https://www.open-mesh.org/issues/365

Kind regards,
	Sven

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: bat0 not capable of MTU > 1500

[rchase47]

Sven,

Amazing. I applied your patch (applied cleanly to 2023.3), recompiled my firmware (I had to modify the package to use 2023.3, and to exclude outdated patches -- your jumbo-frames patch is the only one I applied), and so far it's working great I can see that bat0 is able to take an MTU of 1504, which is what I wanted. 

I didn't try to violate the MTU "space" by reducing my hard-if MTU, and I don't use NC or the other suspect cases you suggested needed testing, but I'm willing to give back and try some additional configurations, if you can give me a list of test cases (and what you want to see).

Thank you so much for maintaining this project, and for your prompt and helpful reply. I'd love to help close this issue, as it looks to me like you likely have created a good patch. I naively don't see any likely reasons why it won't always work. The logic seems right to me.

Respectfully,
Rich

Re: bat0 not capable of MTU > 1500

[Rich Chase]

Sven,
I have been using jumbo frames for almost nine months now with no ill affects that I have noticed. I applied the patch you provided. Seems to be fine, functionally. 

I haven't done pcap or anything, so caveat emptor. Please let me know if you would like me to perform any specific forensic analysis or tests. Otherwise, this is just a sincere thank you!!!
Cheers,
Rich

Sent from Proton Mail Android


-------- Original Message --------
On 12/12/23 11:53 AM, Sven Eckelmann <sven@narfation.org> wrote:

>  On Tuesday, 12 December 2023 20:30:24 CET Rich Chase wrote:
>  > For my life I can't see anything I'm doing wrong. Why can't bat0
>  > have MTU greater than 1500?
>  >
>  > I don't have any other devices mastered by bat0, other than my .1s wifi mesh link.
>  >
>  > Throwing myself at your mercy. Help?
>  
>  https://lore.kernel.org/all/20230727-jumbo-mtu-v1-3-036ce1a6e33c@narfation.org/
>  https://www.open-mesh.org/issues/365
>  
>  Kind regards,
>  	Sven