Tomato for ARM routers

Discussion in 'Tomato Firmware' started by kthaddock, Feb 28, 2014.

  1. Connor McCaffrey

    Connor McCaffrey Networkin' Nut Member

    yeah but judging by times test im guessing they also have worse 5GHz performance due to some kind of other small change, the rt-ac68p didnt perform aswell as the rt-ac68u, even though that makes no sense...
  2. jsmiddleton4

    jsmiddleton4 Network Guru Member

    As Mr. Doh indicated buying it locally gives you a chance to look at the box label. Which I did for another I just purchased to make sure it was manufactured recently and the B1 version.

    It is also 1ghz.

    I wouldn't purchase it on Amazon because I'd do so blindly. Fry's matches Amazon so I was able to get the second one at the 150 price. Both I've purchased were for 150. Its it an outstanding value at 150.....

    "And would anyone know if the USB 3.0 speed increases with the AC68P are also there in those B1 AC68U...."

    I have no way to compare.

    As I understand it a 68U Revision C1 is coming in 2016 but its still the same chip that's in the 68U now just faster.

    The one I have is the same CPU as in the 68P not a faster 68U cpu.

    Also I am not saying for sure across the board any newer made B1's have the same cpu as the 68P's. I have no way of knowing such.
  3. redsandvb

    redsandvb Network Guru Member

    Thanks for that.

    There's a C1 coming too? Really interesting.

    So just checked BestBuy's page, the 68P is on sale for $161 at the moment. That's what I'll probably end up with. :)
  4. MrDoh

    MrDoh Addicted to LI Member

    Good choice *smile*!
    jsmiddleton4 likes this.
  5. jsmiddleton4

    jsmiddleton4 Network Guru Member

    68P for 161.

    I'd probably have done than than play with revision numbers and manufacturing dates at Fry's Electronics.
  6. AndreDVJ

    AndreDVJ LI Guru Member

    So I figured out what was wrong. The developer from that Sabai repo overlooked few things. So far my tests shown I managed to fix USB LED's and their behavior on R7000. The LED's turn on/off as I plug/unplug devices as expected and:

    1) For USB 3.0 port, you need to take in consideration you may connect either a USB 2.0 or 3.0 device. They are addressed differently.
    2) I listed the addresses and how the devices was listed while plugging & unplugging USB 2.0 and 3.0 devices across both ports.

    One limitation I found is if you plug/unplug a device too quickly, the LED's will stop working for that port. I don't know how to resolve this, and this is yet another hack to fix something. I tested with two USB 2.0 and one 3.0 flash drive.

    I am attaching a patch to this post. The file name in question is located at release/src-rt-6.x.4708/router/rc/usb.c , so you may test of you are brave enough.

    I run Jacky's Advanced Tomato, so if you want the firmware I am currently running to test yourself (though I recommend you compile your own Tomato) and you feel concerned about USB LED's, leave a message here.

    If anyone has an idea or found an issue somewhere and fixed, feel free to share. I wasn't very bright with variable names, though I gave a description of few things.

    @shibby20 can you test and let me know if this is good enough to be merged upstream?

    EDIT: I missed an address in case you disable USB 3.0 driver, the USB 2.0 port light wouldn't work. Not only that, if you turn on/off USB 3.0 without rebooting, the addresses would change. Good thing is that I did not have to test lots of cases, one address and updating a test was enough. Probably that's it, I found all the four possible addresses.

    Attached Files:

    Last edited: Dec 26, 2015
    IcyTexx and Elfew like this.
  7. pedro311

    pedro311 Networkin' Nut Member

    Any side effects?
    //EDIT: I mean, what can be broken by applying this patch?
    Last edited: Apr 11, 2018
  8. kille72

    kille72 LI Guru Member

    @tvlz, I'm also curious about your patch, can you please answer @pedro311's question?
  9. tvlz

    tvlz LI Guru Member

    What those iptables rules do is to remove the changes made by @KyleS in the fix ipt not saving marks patch
    and return to the way it was, as I explained here.

    There is no real downside, maybe a tiny speed reduction (test anyone ) due to having to process the QOS rules for every packet, the gain is you get the rule# back.
    kille72 likes this.
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice