[RT-AC86U] I built cake!!!

[Testscript]

Cake has now been accepted into the mainline kernel in 4.19:
https://kernelnewbies.org/Linux_4.1...ence_with_the_CAKE_queue_management_algorithm

 

[Vexira]

Cake has now been accepted into the mainline kernel in 4.19:
https://kernelnewbies.org/Linux_4.1...ence_with_the_CAKE_queue_management_algorithm

I read that kernel version has support for ax wifi, makes me wonder if that's the version that the a 88u will run.

 

[RMerlin]

I read that kernel version has support for ax wifi, makes me wonder if that's the version that the a 88u will run.

RT-AX88U runs 4.1.51 (simply an updated version of the LTS version used by the RT-AC86U).

 

[Vexira]

would this worok on something like the ax 88u?

 

[Sinner]

would this worok on something like the ax 88u?

Hasnt merlin answered that for you a couple times already or was i dreaming. Think he said it would require something like a complete redesign of the tc command structure or something similar

 

[Vexira]

Hasnt merlin answered that for you a couple times already or was i dreaming. Think he said it would require something like a complete redesign of the tc command structure or something similar

I meant as in a direct port if it works on the 86 in theory the ax 88 which rinds on the same HND platform with a slightly newer kernel version number should be possible in theory but it is a massive amount of work, I was just thinking about getting an ax 88 and playing with it on my own out of curiosity that's all.

I dont see as to why the tc structure would have to be rebuilt unless it's to do with how cake itself, considering that the original poster got it working, but I understand that there are parts of it that would require some closed source code to be modified which is impossible, but my curiosity is In regards to the experiment that the op created, not some full blown ddwrt variants.

 

[Vexira]

Sorry to open an old topic but I'm curious if this is still being worked on, I'm down if a tester is needed.

I'm just also wondering what kind of differences in preformace may be yeilded compared to the FQ-CODEL that I'm currently using.

 

[Kingp1n]

Sorry to open an old topic but I'm curious if this is still being worked on, I'm down if a tester is needed.

I'm just also wondering what kind of differences in preformace may be yeilded compared to the FQ-CODEL that I'm currently using.

Anyone? Haha

 

[SomeWhereOverTheRainBow]

Anyone? Haha

I love forever topics..... Cake sounds interesting but for average use FQ-codel is sufficient considering most of this stuff does not have full Broadcom driver support anyways.

 

[Vexira]

Anyone? Haha

Missing driver level and the required kernel version for full support.

 

[Vexira]

I love forever topics..... Cake sounds interesting but for average use FQ-codel is sufficient considering most of this stuff does not have full Broadcom driver support anyways.

FQ-codel is missing driver level support i wish Broadcom would add it to the SDK so we could reap the benefits.

 

[RMerlin]

FQ-codel is missing driver level support i wish Broadcom would add it to the SDK so we could reap the benefits.

If I remember correctly, the missing driver support is mostly important for line speed support (i.e. for 1 Gbps connections).

 

[Vexira]

If I remember correctly, the missing driver support is mostly important for line speed support (i.e. for 1 Gbps connections).

So in other words it would benefit an asymmetrical 1 GB connection like I've see here in Australia with 1000/400 offered by some rare isps?

I guess by the time I get a 1Gb connection cake will be the standard implementation.

 

[RMerlin]

Asymmetric or not.

 

[Vexira]

Asymmetric or not.

Awesome, well I hope that in future it will happen and by that time I get a gigabit connection.

 

[Kingp1n]

Awesome, well I hope that in future it will happen and by that time I get a gigabit connection.

So in other words, no cake inclusion at this time for the RT-AC86U, correct? Bummer!

 

[Vexira]

So in other words, no cake inclusion at this time for the RT-AC86U, correct? Bummer!

Well unless Broadcom updates the kernel version in the SDK, and in that case trend micro would have to update Thier components.

 

[TheOldMan]

It's been a long week, but I finally got it working!

...Okay I haven't hit "apply" yet for fear of a nuclear explosion (and next to 100% asp illiteracy), but the queuing discipline itself works when manually applied. I poured over Eric's history to ensure that it has hooks in all the right places to act as a legit option alongside sfq and the codels.

Anyone want to work on this with me? Downsides: I'm a kernel/project hacker at best, am super messy, and suck at git. Upsides: I am completely willing to break my own hardware but no one else's, am just familiar enough with universal programming techniques to hack together whatever I need, and will not stop until I accomplish my goals.

Check out the screen shot!

Is this what you're talking about? https://www.bufferbloat.net/projects/codel/wiki/Cake/

 

[Frankflash]

FQ-codel is missing driver level support i wish Broadcom would add it to the SDK so we could reap the benefits.

I cant see that happening from useless broadcom i wish they would their are a lot more people who have fiber connections now

 

[Vexira]

I cant see that happening from useless broadcom i wish they would their are a lot more people who have fiber connections now

Well we would here in Australia but, politics, and Telstra, if I want it I'll have to fork out an absurd amount of money.

And I really wish I had a fiber connection would be better than the crappy copper line I have now.