Modify

Opened 7 years ago

Closed 4 years ago

Last modified 3 years ago

#6580 closed task (fixed)

ASUS RT-N16 Support

Reported by: Coool Owned by: hauke
Priority: low Milestone: Features Paradise
Component: kernel Version: Trunk
Keywords: ASUS RT-N16, ASUS, RT-N16 Cc:

Description

I would like to see ASUS RT-N16 support. I can be a tester.

Attachments (5)

asus.diff (36.7 KB) - added by bernhardloos@… 7 years ago.
booting kernel
n16_1.diff (1.5 KB) - added by rtz2 6 years ago.
fixes an illegal backplane access in the cache managment code
2.6.37-bootlog.txt (10.5 KB) - added by GKashperko 6 years ago.
bootlog
cfi_cmdset_0002-add-buffer-write-cmd-timeout.patch (977 bytes) - added by gkashperko 5 years ago.
2011-11-02_TARGET_brcm4716_patches.tar.gz (158.2 KB) - added by anonymous 5 years ago.

Download all attachments as: .zip

Change History (473)

comment:1 Changed 7 years ago by anonymous

Would be nice. I have not seen any info if bcm4718 is supported by b43, and I don't know if the proprietary Broadcom driver works with newer 2.6 kernels. Currently the 2.6 builds available (dd-wrt, Tomato, stock) use kernel versions around 2.6.22-24 and the proprietary Broadcom driver. OpenWrt trunk currently uses 2.6.3x kernels.

wiki:
http://wiki.openwrt.org/inbox/asusrt-n16

Relevant discussions:
https://forum.openwrt.org/viewtopic.php?id=22567
https://forum.openwrt.org/viewtopic.php?id=22016
https://forum.openwrt.org/viewtopic.php?id=23330 <-- reports some progress (almost bootable?)

(Coool: I know you saw those threads, but maybe others didn't)

comment:2 Changed 7 years ago by anonymous

It seems that it has a (relevant part of lspci -vnn)
00:01.0 Network controller [0280]: Broadcom Corporation BCM43XG [14e4:4329] (rev 01)
If I am correct, this chipset (BCM43XG) is supported by b43 (perhaps w/o n)

comment:3 follow-up: Changed 7 years ago by Tony Butler <spudz76@…>

I added some bits to the wiki page for this model the other day. I don't have the hardware to do anything really useful, but I tossed in a few things gleaned from GPL and DD-WRT sources which target this model, I suspect the non-boot issue is a matter of understanding the boot process this model uses (which appears to be something different than the usual Broadcom stuff).

comment:4 in reply to: ↑ 3 Changed 7 years ago by anonymous

Replying to Tony Butler <spudz76@…>:

I suspect the non-boot issue is a matter of understanding the boot process this model uses (which appears to be something different than the usual Broadcom stuff).

Others also suggest that the cause could be the 128MB memory access problem. There is a discussion about this in the first thread (no solution yet)

comment:5 follow-up: Changed 7 years ago by Nistor Andrei <coder.tux@…>

I've tried a 2.4 build... it failed just as the other ones. Does this mean it's not a memory problem?
From what I understood on the forum, the 2.4 tree has the 128MB patch.

comment:6 in reply to: ↑ 5 ; follow-up: Changed 7 years ago by anonymous

Replying to Nistor Andrei <coder.tux@…>:

I've tried a 2.4 build... it failed just as the other ones. Does this mean it's not a memory problem?
From what I understood on the forum, the 2.4 tree has the 128MB patch.

What you say makes sense to me. When trying the 2.4 build, did you try the advised NVRAM settings? (although the factory settings should be fine, since the stock 2.6 firmware works out of the box)

(Still, the others on the thread say, that there are still problems with the 128Mb support in the 2.6 tree (not many people use 128Mb modded Wl-500gP routers, so it is not easy to tell).)
Anyways the boot loader (or whatever stuff that is not working) could be adjusted based on the Tomato or DD-WRT firmwares. Only we need some who knows what to look for...

comment:7 in reply to: ↑ 6 Changed 7 years ago by Nistor Andrei <coder.tux@…>

Replying to anonymous:

What you say makes sense to me. When trying the 2.4 build, did you try the advised NVRAM settings? (although the factory settings should be fine, since the stock 2.6 firmware works out of the box)

(Still, the others on the thread say, that there are still problems with the 128Mb support in the 2.6 tree (not many people use 128Mb modded Wl-500gP routers, so it is not easy to tell).)
Anyways the boot loader (or whatever stuff that is not working) could be adjusted based on the Tomato or DD-WRT firmwares. Only we need some who knows what to look for...

I was using the stock NVRAM values... Right now I'm bugging rtz to take a look at the vendor's firmware, maybe he can find out why it's not booting... he suspects it's some kind of alignment issue in the trx file...

comment:8 follow-up: Changed 7 years ago by Nistor Andrei <coder.tux@…>

i tried to boot the kernel through tftp from the CFE, and it hangs... does anyone know a way to make it a little more verbose?

comment:9 follow-up: Changed 7 years ago by gregd72002

Could someone update us on the progress please?

comment:10 in reply to: ↑ 9 ; follow-up: Changed 7 years ago by gregd72002

Apparently, the issue with 128MB ram has been solved in:
https://dev.openwrt.org/ticket/3177

But as some testing, it did not solve the problem:
https://forum.openwrt.org/viewtopic.php?pid=104299#p104299

Anyone can confirm this?

comment:11 in reply to: ↑ 10 ; follow-up: Changed 7 years ago by tmg

Replying to gregd72002:
[...]

Anyone can confirm this?

still doesn't work for me, boot fails at

Closing network.
Starting program at 0x80001000

comment:12 in reply to: ↑ 11 Changed 7 years ago by tmd

err, rather

Entry at 0x80001000
Starting program at 0x80001000

comment:13 Changed 7 years ago by gregd72002

Some testing (conducted by 2 independent testers) pointed out that the reason for kernel dying while initialising is ssb initialisation (ssb scanning to be precise). This is just a suggestion as further testing has to be done.

Please note that this bug has nothing to do with TRX header and asus ADDVER. The bootloader loads the kernel without any fancy headers. Kernel gets initialised by discovering CPU, memory etc.. but ssb scanning seems to fail.

comment:14 in reply to: ↑ 8 Changed 7 years ago by gregd72002

Replying to Nistor Andrei <coder.tux@…>:

i tried to boot the kernel through tftp from the CFE, and it hangs... does anyone know a way to make it a little more verbose?

Look for setting up early console, in particular early printk. The trunk has a patch that disables early printk (patch id is somewhere around 900). Remove the patch, recompile, flash, plug serial cable and debug.

comment:15 Changed 7 years ago by Nistor Andrei <coder.tux@…>

rtz (from IRC) built a kernel with early printk enabled, but it didn't output any more info than the stock kernel.
Don't know how he built it, but I'll give it a shot again sometime at the beginning of april when I'll have more spare time.
I also tried the kernel form linux-mips with the same results...
About the ssb initialisation, I agree. I've looked at the asus sources, and the only thing that's radically different is just that. I don't have the knowledge to dig deeper into it, unfortunately.

comment:16 follow-up: Changed 7 years ago by anonymous

Hi,

Any update on this matter?

(I was wondering, that (as far as I kow) the newest working kernel being used is 2.6.23/24 in DD-wrt (which is newer than the stock/tomato 2.6.22). Maybe that would be a better starting point.)

comment:17 in reply to: ↑ 16 Changed 7 years ago by frosch6669

The next step should be discovering the dd-wrt kernel config/source/patches to find out if there is any difference to the openwrt-kernel in the ssb-scanning/ssb-initialisation.
I do not eaven have an idea of how to start on this discovering due to missing knowlege of kernel-structure.
Does anyone know something about ssb-scanning/ssb-initialisation?

Replying to anonymous:

Hi,

Any update on this matter?

(I was wondering, that (as far as I kow) the newest working kernel being used is 2.6.23/24 in DD-wrt (which is newer than the stock/tomato 2.6.22). Maybe that would be a better starting point.)

comment:18 follow-up: Changed 7 years ago by bernhardloos@…

The main problem is, that the SSB in the BRCM4716 used for this router is very different from the version used otherwise.
I did manage to produce a working kernel, but almost nothing works, not even wired network, so this isn' exactly ready for prime-time.

Changed 7 years ago by bernhardloos@…

booting kernel

comment:19 in reply to: ↑ 18 Changed 7 years ago by anonymous

Replying to bernhardloos@…:

The main problem is, that the SSB in the BRCM4716 used for this router is very different from the version used otherwise.
I did manage to produce a working kernel, but almost nothing works, not even wired network, so this isn' exactly ready for prime-time.

Maybe this is a stupid question, but: is there a working gigabit ethernet driver in trunk? Stock firmware seems to include its own driver:

eth0: Broadcom BCM47XX 10/100/1000 Mbps Ethernet Controller 5.10.56.27

(Broadcom wireless driver is also 5.10.56.27)

((Also -march=74kc -mtune=74kc should be ok?))

comment:20 follow-up: Changed 7 years ago by gregd72002

I've done some testing and can confirm that it is booting correctly.
Please see the link for 'dmesg' and '/proc/interrupts':
http://openwrt.pastebin.com/knCq6hjn

Note the two things:

  • "Disabling SSB PCI-E host due to code brokeness"
  • there are no interrupts assigned to usb/ethernet/wireless

I believe that correcting the above would solve most of the problems.

Another problem might be missing drivers. However, USB should work out of box (but it does not). Once USB is done we can look into drivers for ethernet and wireless.

Thanks,

comment:21 in reply to: ↑ 20 ; follow-up: Changed 6 years ago by anonymous

Replying to gregd72002:

I've done some testing and can confirm that it is booting correctly.
Please see the link for 'dmesg' and '/proc/interrupts':
http://openwrt.pastebin.com/knCq6hjn

Note the two things:

  • "Disabling SSB PCI-E host due to code brokeness"
  • there are no interrupts assigned to usb/ethernet/wireless

I believe that correcting the above would solve most of the problems.

Another problem might be missing drivers. However, USB should work out of box (but it does not). Once USB is done we can look into drivers for ethernet and wireless.

Thanks,

Hi,
Thanks for the great work!
Is there any progress? Is it necessary to have PCIe working? I ask this, since the PCIe core is only included in the bcm4718. The other products of this series (the bcm4716 and the bcm4717) doesn't have PCIe, so maybe it is not essential to have the basic functions (USB, ethernet and wireless) working.
I took a look, at the ssb init codes of the two sources (Broadcom vs trunk). To me they look very different, it will take somebody with far more skills than me to make something out of that.

comment:22 in reply to: ↑ 21 ; follow-up: Changed 6 years ago by gregd72002

Replying to anonymous:

Hi,
Thanks for the great work!
Is there any progress? Is it necessary to have PCIe working? I ask this, since the PCIe core is only included in the bcm4718. The other products of this series (the bcm4716 and the bcm4717) doesn't have PCIe, so maybe it is not essential to have the basic functions (USB, ethernet and wireless) working.
I took a look, at the ssb init codes of the two sources (Broadcom vs trunk). To me they look very different, it will take somebody with far more skills than me to make something out of that.

Hi,

You are right, PCIe is not of any concern anymore. Just the drivers are missing. Indeed, the Broadcom code is very different from trunk and therefore it takes so much time. It looks like BCM4716/16/18 is a very new platform. But we have to do it. The broadcom code is very much influenced by kernel 2.4. In here we aim to provide to utilize fully kernel 2.6.
There is no other way than simply play (debug) and try with this stuff. USB and ethernet drivers are the things that should be looked at now.

comment:23 in reply to: ↑ 22 ; follow-up: Changed 6 years ago by anonymous

Replying to gregd72002:

Hi,

You are right, PCIe is not of any concern anymore. Just the drivers are missing. Indeed, the Broadcom code is very different from trunk and therefore it takes so much time. It looks like BCM4716/16/18 is a very new platform. But we have to do it. The broadcom code is very much influenced by kernel 2.4. In here we aim to provide to utilize fully kernel 2.6.
There is no other way than simply play (debug) and try with this stuff. USB and ethernet drivers are the things that should be looked at now.

Hi,

I am not a skilled programmer, but I would like to help, if there is some task, that I could do (with the drivers etc). I have an RT-N16 (currently running dd-wrt), but I have no access to the serial console.

comment:24 Changed 6 years ago by rtz2

  • Owner changed from developers to rtz2
  • Status changed from new to accepted
  • Version changed from Kamikaze 8.09 to Trunk

Changed 6 years ago by rtz2

fixes an illegal backplane access in the cache managment code

comment:25 in reply to: ↑ 23 Changed 6 years ago by rtz2

Replying to anonymous:

Hi,

I am not a skilled programmer, but I would like to help, if there is some task, that I could do (with the drivers etc). I have an RT-N16 (currently running dd-wrt), but I have no access to the serial console.

USB should work, there are only device ids missing in th driver.
The biggest problem is the wired network driver, I didn't even start to investigate it yet. So, feel free to take a look at it.
You also need the other patch I attached here.

comment:26 follow-up: Changed 6 years ago by anonymous

Hello, is there any progress to get the RT-N16 running with openwrt ?
I need it in combination with the 2.4 kernel.

comment:27 in reply to: ↑ 26 Changed 6 years ago by gregd72002

Replying to anonymous:

Hello, is there any progress to get the RT-N16 running with openwrt ?
I need it in combination with the 2.4 kernel.

I doubt it will ever be supported on 2.4 kernel... The kernel boots, but the drivers need to be ported.

comment:28 Changed 6 years ago by toegangscontrole@…

Probably the RT-N16 will be running with Kamikaze 8.09. How long do you think this will take to manage ? I just bought a RT-N16 running dd-wrt right now. There are a lot of people waiting for this generation to work on openwrt.

comment:29 follow-up: Changed 6 years ago by rob@…

I've applied both patches attached in this thread to an SVN trunk build but still can't get passed the "Entry at 0x80001000

Starting program at 0x80001000" boot issue." I've got a serial console connected and am not afraid to get down and dirty with the coding, I just don't know where to start to try and help out here.

Can somebody at least contribute the patches that exist right now to SVN so that fresh checkouts will at least boot on this device? That would probably make it at a lot easier to get more devels working on addressing the other problems such as missing drivers... though I fear if this turns out to be anything like all of the old Broadcomm stuff we may have to setup a separate build target for this device that will be based around 2.6.22 (or somewhere around it) so that the existing binary drivers can be used for now...

comment:30 in reply to: ↑ 29 Changed 6 years ago by anonymous

Replying to rob@…:

I've applied both patches attached in this thread to an SVN trunk build but still can't get passed the "Entry at 0x80001000

Starting program at 0x80001000" boot issue." I've got a serial console connected and am not afraid to get down and dirty with the coding, I just don't know where to start to try and help out here.

Can somebody at least contribute the patches that exist right now to SVN so that fresh checkouts will at least boot on this device? That would probably make it at a lot easier to get more devels working on addressing the other problems such as missing drivers... though I fear if this turns out to be anything like all of the old Broadcomm stuff we may have to setup a separate build target for this device that will be based around 2.6.22 (or somewhere around it) so that the existing binary drivers can be used for now...

Some time ago I tried the patches on some recent trunk and they work well for me. The device boots correctly. When testing, please do not forget to apply both the patches that are provided!

comment:31 Changed 6 years ago by michael

No success... I hope there's someone who can tell us all how to get patches work under backfire or trunk. Ther's a lot of rt-n16 owners just waiting booting kernel... I really don't have skills to make kernel patches but at least I can fight with drivers and so on with ttl rs232 converter.

Please...
This is my kind little petition to get that board to work.

comment:32 follow-up: Changed 6 years ago by rtz2

The patch here can't go into trunk as it is, because it kills brcm4710 devices.
Fixing this is rather hard, I would need a dump of the registers of the first core.
There are also a number of other problems, that are difficult to fix.

Parts of this code is already in trunk, so the patch probably won't apply, try r21268 or earilier.

There is also this patch: https://dev.openwrt.org/browser/trunk/target/linux/brcm47xx/patches-2.6.32/900-disable_early_printk.patch
Remove it and then enable early printk in kernel_menuconfig.

comment:33 in reply to: ↑ 32 ; follow-up: Changed 6 years ago by rob@…

Replying to rtz2:

The patch here can't go into trunk as it is, because it kills brcm4710 devices.
Fixing this is rather hard, I would need a dump of the registers of the first core.

Anyone have any ideas what devices actually run brcm4710? I've got WL-500gD, 500gp and 500gpV2 devices kicking around but I don't think any of them actually use 4710... I could be wrong though.

There are also a number of other problems, that are difficult to fix.

Okay so I don't intend to be negative but what is the point in the current patches then if there's no chance they'll ever be useful to the main trunk? Perhaps we need to create a new target that you can select from the build menu that will incorporate these patches, and leave them out of the main Brcm47xx target?

Parts of this code is already in trunk, so the patch probably won't apply, try r21268 or earilier.

That makes sense, as I did notice a few skipped chunks when I applied them.

There is also this patch: https://dev.openwrt.org/browser/trunk/target/linux/brcm47xx/patches-2.6.32/900-disable_early_printk.patch
Remove it and then enable early printk in kernel_menuconfig.

Did that but forgot that I had to enable early printk in the kernel menuconfig, so will try again.

comment:34 Changed 6 years ago by michael

Parts of this code is already in trunk, so the patch probably won't apply, try r21268 or earilier.

That makes sense, as I did notice a few skipped chunks when I applied them.

Sounds clever. I'll try that too.

Okay so I don't intend to be negative but what is the point in the current patches then if there's no chance they'll ever be useful to the main trunk? Perhaps we need to create a new target that you can select from the build menu that will incorporate these patches, and leave them out of the main Brcm47xx target?

I have same kind of feelings too. At least that's much better solution than no kernel or kernel patches at all. One and only target would be great idea but I suppose we all can live without it if it's hard/impossible to make work nicely...

Anyway rt-n16 board is something we really need to get working desperatively because wl500gp and wl500gp2 are sold out and both are end of life boards.

comment:35 in reply to: ↑ 33 Changed 6 years ago by rtz2

Replying to rob@…:

Replying to rtz2:

The patch here can't go into trunk as it is, because it kills brcm4710 devices.
Fixing this is rather hard, I would need a dump of the registers of the first core.

Anyone have any ideas what devices actually run brcm4710? I've got WL-500gD, 500gp and 500gpV2 devices kicking around but I don't think any of them actually use 4710... I could be wrong though.

There are also a number of other problems, that are difficult to fix.

Okay so I don't intend to be negative but what is the point in the current patches then if there's no chance they'll ever be useful to the main trunk? Perhaps we need to create a new target that you can select from the build menu that will incorporate these patches, and leave them out of the main Brcm47xx target?

I didn't say that, I said there are a number of problems with them.
Most of those problems will also be there with a new target.
I'm working on it, but I have a lot of other stuff on my plate, so it's pretty slow going.

comment:36 Changed 6 years ago by rob@…

Okay I finally got the thing to boot. The patches listed here do the trick, but as pointed out can't be applied directly to SVN beyond 21268 since a lot of other code has changed around and the file structures are different (I manually sat down this morning and spent an hour inserting the code from the patches into the correct places in the kernel files).

As pointed out the system now boots, but drivers are non-existent to functionally it's quite useless at the moment. I noticed the kernel does try to load all of the USB modules, but they don't seem to actually do anything as any USB device I connect doesn't even generate an entry in the kernel log so I'm guessing there's still something to be done there. Hats off to rtz2 for even getting this far as I have no idea how all that patch code got figured out but cross-referencing with both the Asus GPL sources and DD-WRT sources is a crazy mess as neither are as clearly laid out as the OpenWRT stuff and both are built around older kernel versions...

I'm tempted to try and see if I can shoe-horn the patch stuff into 8.09 since that's still based on Kernel 2.6.22 which is probably more likely to work with the existing Broadcom binary drivers for ethernet and wireless... not a great long term plan but I think it makes the most sense right now for a short term solution to get things running (since I've recently come to the conclusion that DD-WRT is a coding disaster and is not an option at all).

comment:37 follow-up: Changed 6 years ago by rtz2

I finally managed to get the patch in a proper shape and fix (hopefully) the problems with it.

https://lists.openwrt.org/pipermail/openwrt-devel/2010-May/007049.html

The USB driver won't work, because the broadcom has a different vendor id on ai busses, so the driver won't bind to the core. Even if he does, the thing will crash, but I already have an idea why.
No clue about the ethernet driver, I don't even know, which one this thing uses.

comment:38 in reply to: ↑ 37 ; follow-up: Changed 6 years ago by anonymous

Replying to rtz2:

I finally managed to get the patch in a proper shape and fix (hopefully) the problems with it.

https://lists.openwrt.org/pipermail/openwrt-devel/2010-May/007049.html

Dang, see if I'd waited a few hours I could have saved myself all of that manual patching! Thanks for the awesome work though the next time I have to rebuild the kernel I'll try applying this patch directly...

The USB driver won't work, because the broadcom has a different vendor id on ai busses, so the driver won't bind to the core. Even if he does, the thing will crash, but I already have an idea why.

Okay cool good to know you've got a plan (it's beyond me!)

No clue about the ethernet driver, I don't even know, which one this thing uses.

The device I have infront of me says BCM53115. A quick google looks like there are just some ID's missing as somebody else got this going for another target:

http://www.mail-archive.com/openwrt-devel@lists.openwrt.org/msg03763.html

Maybe that'll help us out?

comment:39 in reply to: ↑ 38 ; follow-up: Changed 6 years ago by rtz2

Replying to anonymous:

Dang, see if I'd waited a few hours I could have saved myself all of that manual patching! Thanks for the awesome work though the next time I have to rebuild the kernel I'll try applying this patch directly...

Sorry, I didn't get it done earlier.

The device I have infront of me says BCM53115. A quick google looks like there are just some ID's missing as somebody else got this going for another target:

http://www.mail-archive.com/openwrt-devel@lists.openwrt.org/msg03763.html

Maybe that'll help us out?

That's only the switch. It may come in handy later on, but it should work at least to some degree out of the box.

comment:40 in reply to: ↑ 39 ; follow-up: Changed 6 years ago by anonymous

Replying to rtz2:

Sorry, I didn't get it done earlier.

lol no need to apologize, I understand that awesomeness takes time!

That's only the switch. It may come in handy later on, but it should work at least to some degree out of the box.

lol oh yea I feel like a goof! Switch connects to Ethernet interface on CPU, thus driver ethernet is needed... duh... And yea since it's a hardware switch it forwards packets around between ports on it's own, no problem, just no way to actually get to the device itself from those ports since driver is missing, got it...

Okay I'll keep digging... not sure if I'm helping or causing more headaches but I'm trying :-)

comment:41 in reply to: ↑ 40 Changed 6 years ago by anonymous

After racking through tonnes of source code and information I'm really not getting very far with the ethernet. All information seems to point towards it being the same as on the 4705 (since they're part of the same "big family"/generation of chips), but whether that's true or not is beyond me. I've been digging through the ASUS GPL code, but since it was all written for a kernel that didn't integrate SSB I'm pretty lost in comparing the two.

The patches add a definition for SSB_DEV_ETHERNET_GBIT2, which seems to line up with the BCM47XX_GMAC_ID defined in the ASUS CFE code (figured the CFE would be a bit easier to dig through than the whole kernel since it provides a working ethernet interface anyway); the CFE code referencing that seems to just load the same (I'm assuming) proprietary 47xx ethernet driver...

comment:42 Changed 6 years ago by anonymous

Hi all,
Any news?

comment:43 Changed 6 years ago by Coool

Look like none.

comment:44 Changed 6 years ago by anonymous

Looks like its time to look elsewere for some linux support for these pippies.

comment:45 Changed 6 years ago by anonymous

Hi all,
Could you please tell us some updates on the progress of this project, or on the factors that could speed up the progress?
I could also imagine a solution like in the Lyx project (sponsoring features), if it is in compliance with the directives / policy of OpenWrt.

comment:46 Changed 6 years ago by codertux

I'd be interested in donating to see my RT-N16 supported in OpenWRT.

comment:47 Changed 6 years ago by rob@…

I've been keeping track on the forums and things for the various aspects of this (as well as trying to do my own driver porting).

As far as I can tell here's where things stand:
rtz2 Said as of last posting on this bug tracker that the USB driver needed some tweaking (it was missing the ID for the chipset, and was also causing some faults), but he had some plans as to how to fix it... no word on whether that's been successful or not.

tathagata over in this thread has been working on the wired/wireless driver ports from Tomato/DDWRT (technically he's working on the WNR3500L but since it's the same core chipset the drivers apply here as well). Last I heard he was having a PCI bus problem that he raised in this thread and it looks like he's making progress there.

If we can get reports from those developers directly there might be more that can be done to help out...

comment:48 follow-up: Changed 6 years ago by anonymous

Does r21809

Changeset [21809] by nbd
Add a new kernel-version independent broadcom binary driver for brcm47xx …

add support for the wireless of RT-N16 (brcm4718 / BCM43XG [14e4:4329])?

comment:49 in reply to: ↑ 48 Changed 6 years ago by rob@…

Replying to anonymous:

Does r21809

Changeset [21809] by nbd
Add a new kernel-version independent broadcom binary driver for brcm47xx …

add support for the wireless of RT-N16 (brcm4718 / BCM43XG [14e4:4329])?

Good eye on that... it looks like it's a fairly direct port of the broadcom driver that ships with the GPL for the RT-N16. I'll try grabbing it and rolling it into my current build to see if it does anything; I have a feeling this may still require some tweaking in the main kernel to get running, but let's try and see (fingers crossed).

comment:50 Changed 6 years ago by rob@…

Just incase anyone is trying this/working on things, it should be noted that rtz2's diff file has a bug. The section that reads:

+ case 0x4322:
+- if (cc->pmu.rev == 2) {
+- chipco_write32(cc, SSB_CHIPCO_PLLCTL_ADDR, 0x0000000A);
+- chipco_write32(cc, SSB_CHIPCO_PLLCTL_DATA, 0x380005C0);
+- }
++ if (cc->pmu.rev == 2)
++ ssb_chipco_pll_write(0xa, 0x380005C0);
+ break;

The ssb_chipco_pll_write function should be called using cc as the first argument, so:
++ if (cc->pmu.rev == 2)
++ ssb_chipco_pll_write(cc, 0xa, 0x380005C0);

I'm compiling against it now, but that was the only error that I got initially.

comment:51 follow-up: Changed 6 years ago by anonymous

This ticket now exists for about 5 months. Is there any hope we will have openwrt running this year on this device ?

comment:52 in reply to: ↑ 51 Changed 6 years ago by anonymous

Replying to anonymous:

This ticket now exists for about 5 months. Is there any hope we will have openwrt running this year on this device ?

Welcome to the world of open source... you get what you pay for. There's a lot more support now than there was when the ticket was opened; all that's needed now is to get the proprietary device drivers sorted out, which is no easy task but there's progress being made. Best source for updates is this thread:
https://forum.openwrt.org/viewtopic.php?id=23330&p=4

Since WNR3500L is essentially the same device as RT-N16. For the record, btw, the main reason it takes this long to try to get support in OpenWRT vs Tomato/DDWRT/etc is because we are using the latest Kernel and modern files... where as Tomato/DD-Wrt literally just take the source code GPL'd by Asus/whoever, change the build options, and call it their own firmware; so they don't have to actually make drivers or anything. Not taking anything away from them, but if Asus was a car maker, their source code would be a crumpled up ratty old blue print that they "gave out" for their car; DD-Wrt/Tomato is the equivalent of taking that blueprint, adding some bigger wheels and maybe an extra speaker for the radio, and building the same car. OpenWrt is like taking that blueprint, understanding how it works, thinking about what makes sense and what doesn't, and taking the best features and adding them to an existing solidly thought out design. So takes longer/is more work, but is much better in the end.

comment:53 Changed 6 years ago by anonymous

Anybody out there???

comment:54 follow-up: Changed 6 years ago by anonymous

Please see ticket #7905 -- does it help?

comment:55 in reply to: ↑ 54 Changed 6 years ago by anonymous

Replying to anonymous:

Please see ticket #7905 -- does it help?

Nice but I doubt that will be of any use to us here as it's the drivers for their desktop/laptop chipsets mainly. Remember that Broadcomm does have "sortof open source" drivers already available for 2.6.x kernels, but they're written for an older version that what OpenWRT currently implements, and require some major modifications in order to get rolling along with... trying to make those changes and at least get the drivers to compile is where everyone's at right now...

comment:56 follow-up: Changed 6 years ago by anonymous

No hope at all? Should I throw my RT-N16 away?

comment:57 in reply to: ↑ 56 ; follow-up: Changed 6 years ago by anonymous

No hope at all? Should I throw my RT-N16 away?

If you want to _use_ your RT-N16, not _play_ with it, just flash dd-wrt or tomatoUsb and start using your router.

comment:58 in reply to: ↑ 57 Changed 6 years ago by anonymous

Replying to anonymous:

No hope at all? Should I throw my RT-N16 away?

If you want to _use_ your RT-N16, not _play_ with it, just flash dd-wrt or tomatoUsb and start using your router.

If they would implement overlaying the root file system (to have a "writeable" root partition, instead of the ugly configuration strings in NVRAM) and straightforward placement of configuration files, than I would use either of them. But they wouldn't be DD-wrt/TomatoUSB anymore would they?
Until then I stick with my wl-500gPv1, which is almost completely supported (with the Broadcom wireless driver, since B43 still isn't working well, if I heard correctly). And have a relatively expensive paperweight next to it.

comment:59 Changed 6 years ago by anonymous

I see, you just wanna play with configs, not to perfrom one-time setup and use your box as router. It is understandable.

Then, throw your RT-N16 away and buy another box (but this time read forums first!), or use your outdated wl500g, even if it still doesn't working well.

comment:60 Changed 6 years ago by KanjiMonster

There is also always the possibility to give the device to a dev, this normally speeds up the process of adding support for a device or platform.

comment:61 Changed 6 years ago by anonymous

OFF:
I have to say sorry, I did not mean to sound arrogant or ignorant. I just wanted to say that Openwrt is in my opinion superior in some aspects to the other mentioned firmwares.
\OFF:
As I see in the other threads, at least two people are working on the port of the missing ethernet driver (the same way as the proprietary Broadcom wireless driver was ported - somehow without using the kernel SSB layer???). Unfortunately posts are getting scarcer in that thread.

comment:62 Changed 6 years ago by tick

the rt-n16 can use Tomato DualWAN and dd-wrt firmware. when openwrt support? expecting!!!!

comment:63 Changed 6 years ago by meabru@…

What's the BroadComm open-source announcement going to do here? Is there a plan now that we can get our hands on the driver code? Is OpenWrt going to have it as a build requirement, we link-in and voila?

Just wondering how this goes down.

-B

comment:64 Changed 6 years ago by jow

  • Milestone changed from Kamikaze to Kamikaze Features Paradise
  • Priority changed from normal to low

comment:65 Changed 6 years ago by anonymous

Thanks for slowin' down this otherwise superfast process. So now it is declared, that the RT-N16 (which is/was one of the best routers otherwise) will not get any support from the Openwrt team.

comment:66 Changed 6 years ago by jow

rtz2 is mia, no other developer has such a device and there where also no donations as far as I can tell so how do you expect the OpenWrt team to develop a port without any hardware and specs? And no, it is no help that DD-Wrt and Tomato work, they use proprietary components which are no option for OpenWrt.

comment:67 follow-up: Changed 6 years ago by anonymous

So, what should we do to help?

comment:68 in reply to: ↑ 67 ; follow-up: Changed 6 years ago by KanjiMonster

Replying to anonymous:

So, what should we do to help?

Get a RT-N16 into the hands of a developer.

comment:69 in reply to: ↑ 68 ; follow-up: Changed 6 years ago by anonymous

Replying to KanjiMonster:

Replying to anonymous:

So, what should we do to help?

Get a RT-N16 into the hands of a developer.

I checked on ebay: a new RT-N16 costs about a hundred bucks. I own one RT-N16, I would like to see Openwrt support, but I can not afford to buy another one.
However I am willing to donate 20$.
4 more, and we are good to go.
Buy the way, how do you donate to Openwrt (for a specific goal?)?

comment:70 in reply to: ↑ 69 ; follow-up: Changed 6 years ago by KanjiMonster

Replying to anonymous:

Replying to KanjiMonster:

Replying to anonymous:

So, what should we do to help?

Get a RT-N16 into the hands of a developer.

I checked on ebay: a new RT-N16 costs about a hundred bucks. I own one RT-N16, I would like to see Openwrt support, but I can not afford to buy another one.
However I am willing to donate 20$.
4 more, and we are good to go.

Great!

Buy the way, how do you donate to Openwrt (for a specific goal?)?

Good question. For generally donating, see SupportDonate, but I don't see anything on how to donate for a specific goal.

This should probably best coordinated by someone who knows to which developer this device then should go (probably not to nbd, as he already has his hands full with Atheros stuff ;).

comment:71 in reply to: ↑ 70 Changed 6 years ago by KanjiMonster

I just talked with glp, send him an email (see the SupportDonate page for his email address) when you did the donation.

comment:72 Changed 6 years ago by jow

I ordered one now (privately) but it might take some time until I can work on it.

comment:73 Changed 6 years ago by anonymous

I tried to donate through SPI, but they ask an insane amount of personal data (including phone number, address etc), which I am unwilling to provide. On the other hand they are not interested in details like what specific reason I am donating for, providing a comment and alike. Isn't there another way?

comment:74 Changed 6 years ago by anonymous

I would really like to see support for RT-N16, I donated $20 bucks for support of this device, it looks like we asking for mission impossible to the developers.

comment:75 Changed 6 years ago by moj0@…

Clearly most developers are just plain jealous they're wasting their time on useless routers when suddenly this WRT-54GL-like awesome-O-router rt-n16 arrives and all the ddwrt 'n00bs' own it and so they shun it like it's a NASA sparepart so hard to port Openwrt.

comment:76 Changed 6 years ago by Kasper Schoonman

Dear OpenWRT development,

We own approx 10 of these AccessPoints, we are currently struggling with dd-WRT (or better said Broadcom's NAS application).
We really would like to use OpenWRT which we use on the older Accesspoints (Linksys WRT54GL's), HostAPd works much better than NAS, but the WRT54GL has not enough resources to handle our needs (School network with 200+ wireless clients using Roaming Profiles on a Windows Domain). The Linksys AP's have high loadaverages with 5 - 10 clients, while our Asus RT-N16 AP's handle 20+ clients without any issue.

What can I do to get this device supported? Should I donate through the website? Should I donate a Asus RT-N16?

Kasper

comment:77 Changed 6 years ago by anonymous

Hey devs, these are on sale at Newegg and Buy.com for $89.99 with free shipping right now. Not too shabby for a device with 2x the flash and ram of most of the other devices available, and USB...

(Can you tell I'm trying to encourage support? :) )

comment:78 Changed 6 years ago by anonymous

I have a nearly new RT-N16 that I accidentally bricked while dicking with the firmware (can’t even perform the usual early-boot tftp) that I would be happy to mail to any OpenWRT developer that would use it in his porting efforts. Any problems are purely software related, so it should be JTAG recoverable.

I just need to know how to direct this to the right (legitimate) OpenWRT developer. Any help?

comment:79 Changed 6 years ago by anonymous

BTW, I am in the US. So someplace I can ship to easily and without exorbitant postage would be preferable.

comment:80 Changed 6 years ago by anonymous

https://dev.openwrt.org/wiki/SupportDonate

You could mention that cshore has volunteered to work on it (I'm in Canada so that shouldn't be too bad for shipping).

comment:81 follow-up: Changed 6 years ago by anonymous

I was recommended the RT-N16 and a copy of Gargoyle, I wish I'd researched it further -- finding out that OpenWRT (the base for Gargoyle) is still incompatible with the RT-N16 is disappointing.

Any developer want mine? It may have a processor second to none, but without Gargoyle, it's inert plastic junk.

comment:82 in reply to: ↑ 81 Changed 6 years ago by KanjiMonster

Replying to anonymous:

Any developer want mine? It may have a processor second to none, but without Gargoyle, it's inert plastic junk.

Depends on where you are, and what you want for it ;P.

But seriously, work has been recently restarted to get the new broadcoms supported, so if you still have a bit patience, you might have something that runs OpenWrt (but probably without wifi for now) in a few weeks/months.

comment:83 Changed 6 years ago by anonymous

Sorry, what i need to do? to configure openwrt to build it for my rt-n16 ? and it's possible ?

comment:84 Changed 6 years ago by anonymous

work stopped ? sources are awable on asus web site !!!! what time to wait ?

comment:85 Changed 6 years ago by anonymous

Some one answered ??? how time we need waiting ? and some peaple had donated a lot of money to develop one.....

comment:86 Changed 6 years ago by anonymous

please look at... asus.com

the GPL code now are available!!!!!!!!!!!!!!

comment:87 Changed 6 years ago by anonymous

Is it possible to use dd-wrt 2.6.24 kernel in openwrt? I think, it's a good temporary solution.

comment:88 Changed 6 years ago by adrien@…

Will donate 200$US for RT-N16 support.

comment:89 follow-up: Changed 6 years ago by gregd72002@…

I was playing around with the router together with 'rtz2', this was almost a year ago (please see the first posts). Unfortunately, I run out of time and have left the project. It looks the progress has stopped as well. Not sure about 'rtz2', I do not seem to find him on IRC anymore.
Anyway, if someone in person could take on this project I'm happy to make a personal donation as well..
As far as I am concerned, it should be straight forward to get openwrt boot on this router (rtz2 and me had managed to successfully boot it). The reminding issues were/are:

  • ethernet driver
  • usb driver
  • wireless driver

Anyone in OpenWRT community up to get the money? Note that WNR3500L uses (almost) the same components, therefore there might be some money to get from WNR3500L owners as well!

Thanks,
Gregory
gregd72002(_at_)gmail.com

comment:90 in reply to: ↑ 89 ; follow-up: Changed 6 years ago by anonymous

Replying to gregd72002@…:

As far as I am concerned, it should be straight forward to get openwrt boot on this router (rtz2 and me had managed to successfully boot it). The reminding issues were/are:

  • ethernet driver
  • usb driver
  • wireless driver

usb works. ethernet is in progress, soon gonna finish.

comment:91 in reply to: ↑ 90 Changed 6 years ago by GKashperko

Replying to anonymous:

usb works. ethernet is in progress, soon gonna finish.

Finally ethernet works too. Neew few days to play around with it and to finish few things. After that will post sources at openwrt developer lists.
Patches to get box up with usb working already there.

comment:92 Changed 6 years ago by anonymous

As far as I understood the source for the Broadcom-stuff is not published yet and probably wont be. Will this affect or limit the use of this device or will it be fully supported? (As you may have noticed I have no clue about routers...)

comment:93 follow-up: Changed 6 years ago by lauri@…

How can I help out with testing/development? I have the hardware and I am not afraid of bricking :)

comment:94 in reply to: ↑ 93 Changed 6 years ago by anonymous

Replying to lauri@…:

How can I help out with testing/development? I have the hardware and I am not afraid of bricking :)

Is it possible to brick it at all? ;) I do not think so :D as long as you have serial cable, you should be fine with testing openwrt in any possible way ;)

Changed 6 years ago by GKashperko

bootlog

comment:95 Changed 6 years ago by GKashperko

Posted 3 prerequisite patches few days ago (broadcom-diag, base-files/netconfig and bugfix for kmod-switch). No response yet. Ready to start posting kernel patches as soon as those 3 will be commited (hope they will be as changes are obvious).

comment:96 follow-up: Changed 6 years ago by 8p8c

Ready to test here a.s.a.p

comment:97 in reply to: ↑ 96 ; follow-ups: Changed 6 years ago by GKashperko

Replying to 8p8c:

Ready to test here a.s.a.p

Started with posting AI-bus support stuff at linux-wireless. Hopefully we will get AI-based socks' support with kernel some day. ATM making separate target for bcm4716/Openwrt as not sure if it would be merged with bcm47xx one. Might will be done by tomorrow.
And still waiting for _any_ feedback with broadcom-diag, base-files/netconfig and kmod-switch.

comment:98 in reply to: ↑ 97 Changed 6 years ago by lauri@…

Replying to GKashperko:

And still waiting for _any_ feedback with broadcom-diag, base-files/netconfig and kmod-switch.

A little generic how-to would help out with testing :)

comment:99 Changed 6 years ago by hauke

  • Owner changed from rtz2 to hauke
  • Status changed from accepted to assigned

comment:100 Changed 6 years ago by anonymous

Toolchain installed on Ubuntu 10.10 - as per OpenWRT instructions - make menuconfig settings and specific file replacements needed.

comment:101 Changed 6 years ago by gregd72002@…

I guess it might be useful to read this if anyone is interested in testing this stuff.
There are clear instructions how to get OpenWRT on Asus RT-N16 running.
https://lists.openwrt.org/pipermail/openwrt-devel/2011-February/009634.html

Also, please do report on any issues!

comment:102 follow-up: Changed 6 years ago by lauri@…

I tried out the prepared image, seems to be working nicely altough:

a) No packages for this architecture, it should be brcm47xx compatible right?
b) No WiFi, what exactly is missing?

comment:103 in reply to: ↑ 102 Changed 6 years ago by gregd72002@…

Replying to lauri@…:

I tried out the prepared image, seems to be working nicely altough:

a) No packages for this architecture, it should be brcm47xx compatible right?
b) No WiFi, what exactly is missing?


If you build the image yourself and select "broadcom-wl" you should get wifi working... that's what ppl on mailing list say... give it a try and report back.

comment:104 Changed 6 years ago by 8p8c

Up and running here on N16 - Congratulations and a big thank you. Going to try building the image next as running pre-compiled at the moment - still testing.

comment:105 in reply to: ↑ 97 ; follow-up: Changed 6 years ago by anonymous

Replying to GKashperko:

Replying to 8p8c:

Ready to test here a.s.a.p

Started with posting AI-bus support stuff at linux-wireless. Hopefully we will get AI-based socks' support with kernel some day. ATM making separate target for bcm4716/Openwrt as not sure if it would be merged with bcm47xx one. Might will be done by tomorrow.
And still waiting for _any_ feedback with broadcom-diag, base-files/netconfig and kmod-switch.

Thank you it's working here from few hours. I built from trunk with your latest patches.
No network issue, I get wireless working too. Dnsmasq seems not working, something wrong in configuration on dhcp-range on trunk.

comment:106 Changed 6 years ago by anonymous

I am new to this, so would like to ask what kind of openwrt version can be used with this patch?
And if it is possible to describe steps, so that new people could also try out OpenWrt with Asus RT-N16 :)

comment:108 Changed 6 years ago by hcontrol

Hello I try to get the patches applied (rev 25536) but i get following errors:

patching file package/broadcom-diag/src/diag.c
Hunk #1 FAILED at 59.
Hunk #2 FAILED at 74.
Hunk #3 FAILED at 363.
Hunk #4 FAILED at 502.
Hunk #5 FAILED at 950.
Hunk #6 FAILED at 999.
6 out of 6 hunks FAILED -- saving rejects to file package/broadcom-diag/src/diag.c.rej
patching file package/switch/src/switch-robo.c
Hunk #1 FAILED at 301.
Hunk #2 FAILED at 331.
Hunk #3 FAILED at 344.
Hunk #4 FAILED at 361.
Hunk #5 FAILED at 610.
5 out of 5 hunks FAILED -- saving rejects to file package/switch/src/switch-robo.c.rej

patching file package/broadcom-diag/Makefile
Hunk #1 FAILED at 15.
1 out of 1 hunk FAILED -- saving rejects to file package/broadcom-diag/Makefile.rej
patching file package/broadcom-wl/Makefile
Hunk #1 FAILED at 26.
1 out of 1 hunk FAILED -- saving rejects to file package/broadcom-wl/Makefile.rej
patching file package/broadcom-wl/patches/005-broadcom2-w11.patch
patching file package/kernel/modules/netdevices.mk
Hunk #1 FAILED at 516.
1 out of 1 hunk FAILED -- saving rejects to file package/kernel/modules/netdevices.mk.rej
patching file package/kernel/modules/other.mk
Hunk #1 FAILED at 697.
1 out of 1 hunk FAILED -- saving rejects to file package/kernel/modules/other.mk.rej
patching file package/mac80211/Makefile
Hunk #1 FAILED at 35.
1 out of 1 hunk FAILED -- saving rejects to file package/mac80211/Makefile.rej
patching file package/nvram/Makefile
Hunk #1 FAILED at 19.
1 out of 1 hunk FAILED -- saving rejects to file package/nvram/Makefile.rej
patching file package/switch/Makefile
Hunk #1 FAILED at 15.
Hunk #2 FAILED at 33.
2 out of 2 hunks FAILED -- saving rejects to file package/switch/Makefile.rej

Can you please tell me how to get it working ?

comment:109 follow-up: Changed 6 years ago by hcontrol

Up and running now !! I just made a mistake by putting the patches in the wrong
directory.

I changed the opkg.conf to be able to use existing packages..

src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay
arch all 1
arch mipsel 200
arch brcm47xx 10

comment:110 Changed 6 years ago by anonymous

Only that does not run yet is the wireless...

wifi enable gives:

wl0(mac80211): Interface type not supported
wl0(mac80211): Interface type not supported

What can i do to solve ?

comment:111 in reply to: ↑ 105 ; follow-up: Changed 6 years ago by anonymous

No network issue, I get wireless working too. Dnsmasq seems not working, something wrong in configuration on dhcp-range on trunk.

Dnsmasq is just missing /usr/lib/common.awk. Copy it there and it will work fine.

comment:112 follow-up: Changed 6 years ago by anonymous

Can you please post your .config file to get wireless working ? I mean the one used for compiling

comment:113 in reply to: ↑ 112 ; follow-up: Changed 6 years ago by anonymous

Replying to anonymous:

Can you please post your .config file to get wireless working ? I mean the one used for compiling

I have just used the patches from the earlier post:
https://lists.openwrt.org/pipermail/openwrt-devel/2011-February/009634.html

and then applied the wl patch from a later post in that thread:
https://lists.openwrt.org/pipermail/openwrt-devel/2011-February/009634.html

I was using the (then) latest Openwrt revision, which was r25522. The patches applied without a problem.

The router is running fine, wireless is working well (only tested 54g, no idea about n)

This is a huge step, I would really like to thank the developers for this!

comment:114 in reply to: ↑ 113 Changed 6 years ago by anonymous

comment:115 Changed 6 years ago by tones111

These patches are a great step toward getting 4716 based routers supported! I've applied the 3 non wireless patches and now the kernel starts booting on WRT160N V3. What's the expected timeline before these get commited to mainline? Thanks!!!

comment:116 in reply to: ↑ 111 Changed 6 years ago by anonymous

Replying to anonymous:

No network issue, I get wireless working too. Dnsmasq seems not working, something wrong in configuration on dhcp-range on trunk.

Dnsmasq is just missing /usr/lib/common.awk. Copy it there and it will work fine.

All been working fine since one week.

comment:117 Changed 6 years ago by moj0@…

Could someone please compile an image with LUCI working out of the box?

comment:118 follow-up: Changed 6 years ago by yrleu

George's this post details the steps to build OpenWrt for ASUS RT-N16. It's a MUST-READ.
http://permalink.gmane.org/gmane.comp.embedded.openwrt.devel/7922

comment:119 in reply to: ↑ 118 Changed 6 years ago by anonymous

Replying to yrleu:

George's this post details the steps to build OpenWrt for ASUS RT-N16. It's a MUST-READ.
http://permalink.gmane.org/gmane.comp.embedded.openwrt.devel/7922

I would advise to rather use the patch set in this thread, since the one you are pointing to is based on the code released by Asus (kernel 2.6.22), which contains proprietary parts that are not compatible with newer kernels (like the one used by Openwrt -- 2.6.37), and is therefore unmaintainable. The patches included in this thread include newly written code needed to support the brcm4716 based (like SSB AI) hardware in a manner that is compatibe with the current structure of the kernels.

comment:120 follow-up: Changed 6 years ago by yrleu

The patches described in this thread were released by George. But only in http://permalink.gmane.org/gmane.comp.embedded.openwrt.devel/7922 could I find detailed steps to build the image.

Although George suggested kernel 2.6.22 in the post, I have used the patches with 2.6.37 and have built a bootable image.

comment:121 in reply to: ↑ 120 ; follow-up: Changed 6 years ago by anonymous

Replying to yrleu:

The patches described in this thread were released by George. But only in http://permalink.gmane.org/gmane.comp.embedded.openwrt.devel/7922 could I find detailed steps to build the image.

Although George suggested kernel 2.6.22 in the post, I have used the patches with 2.6.37 and have built a bootable image.

George's post says:
"Attached is the patch against current (last tried it with 24667) Openwrt trunk to get it run on RT-N16 with
2.6.22.19 kernel."

Which suggests, that the patches do not actually apply to the 2.6.37 kernel, but rather they bring back 2.6.22 support and Openwrt is built with a 2.6.22 kernel instead. You can check the version of the kernel by running "uname-a" on your router.

comment:122 in reply to: ↑ 121 Changed 6 years ago by GKashperko

Replying to anonymous:

Replying to yrleu:

The patches described in this thread were released by George. But only in http://permalink.gmane.org/gmane.comp.embedded.openwrt.devel/7922 could I find detailed steps to build the image.

Although George suggested kernel 2.6.22 in the post, I have used the patches with 2.6.37 and have built a bootable image.

George's post says:
"Attached is the patch against current (last tried it with 24667) Openwrt trunk to get it run on RT-N16 with
2.6.22.19 kernel."

Which suggests, that the patches do not actually apply to the 2.6.37 kernel, but rather they bring back 2.6.22 support and Openwrt is built with a 2.6.22 kernel instead. You can check the version of the kernel by running "uname-a" on your router.

2.6.22 patches you reference are 2+ month old ones I did based on ASUS GPL package.
They in no way will let you compile 2.6.22+ kernel (2.6.37 as well ofcourse) unless you patch them hardly.

To get modern kernels' support use another ssb+ai-based patchworks posted lately in openwrt developer mail lists.

comment:123 follow-up: Changed 6 years ago by anonymous

Hello everyone,I already tried the latest firmware but i can't download packages.the error says wrong architecture.

Thanks in advance

comment:124 in reply to: ↑ 123 Changed 6 years ago by anonymous

Replying to anonymous:

Hello everyone,I already tried the latest firmware but i can't download packages.the error says wrong architecture.

Thanks in advance

What do you mean by latest firmware? Have you built it yourself? Witch patches did you use? If you used the patches reported by gregd72002, than a separate target is created, therefore the existing packages will not show up. You can either built the packages yourself or use the hack in hcontrol's post from a few days ago.

comment:125 Changed 6 years ago by anonymous

Good Day!

Can anyone send me the step by step procedure to compile including where to download the trunk

Thanks in advance

comment:126 follow-up: Changed 6 years ago by Olo

Hi all,

succesfully flashed image. When I do:
# opkg update
Downloading http://downloads.openwrt.org/snapshots/trunk/brcm4716/packages/Packages.gz.
wget: server returned error: HTTP/1.1 404 Not Found
Collected errors:

my question is where from to get compatible packages (I would like to run ipv6 dibbler) or how to create right one (mean working one).

best regards && thanks for effort!

comment:127 in reply to: ↑ 126 Changed 6 years ago by anonymous

Replying to Olo:

Hi all,

succesfully flashed image. When I do:
# opkg update
Downloading http://downloads.openwrt.org/snapshots/trunk/brcm4716/packages/Packages.gz.
wget: server returned error: HTTP/1.1 404 Not Found
Collected errors:

my question is where from to get compatible packages (I would like to run ipv6 dibbler) or how to create right one (mean working one).

best regards && thanks for effort!

Please read the previuos posts in this thread, before asking. This question was already answered, you can use a hack described by hcontrol (make opkg use the brcm47xx packages, since they should be compatible). If you wish to create them yourself, go ahead compile trunk and your packages after applying the patches (which can also be found in this thread).

comment:128 Changed 6 years ago by anonymous

Everything seems to work out nicely, except for one: Transmission seems to trigger a reboot after a few minutes. Because of that, I do not have any logs. At one occasion however (when the router did not reboot because of Transmission, but wan stopped working -- could be a different bug) I found ssb_dma_rx related error messages.
I built my kernel with the 2011-02-13T23-17 patches (selected mips32r2 (there is a prompt asking to choose between r1 and r2, when building with make V=99) and used march=74kc)

comment:129 follow-up: Changed 6 years ago by anonymous

which trunk version did you get for compiling ? 25469 for me works excellent, I have no issues so far. If you want also ipv6 you can use my image: https://e-lista.pl/OpenWrt/

comment:130 in reply to: ↑ 129 ; follow-up: Changed 6 years ago by anonymous

Replying to anonymous:

which trunk version did you get for compiling ? 25469 for me works excellent, I have no issues so far. If you want also ipv6 you can use my image: https://e-lista.pl/OpenWrt/

Thanks. It was already present in my first build (around 25522). I only have this issue when using Transmission. Did you test your build with transmission?

comment:131 Changed 6 years ago by anonymous

If you mean Transmission - torrent client - no, instead I use torrentflux. As I stated before - no issues.

comment:132 Changed 6 years ago by maurer

did someone tried .11N transfers? what throughput/speeds?

comment:133 follow-up: Changed 6 years ago by anonymous

was to try the pre-built images from George... anyone know what is the root password for it?

comment:134 in reply to: ↑ 133 Changed 6 years ago by anonymous

Replying to anonymous:

was to try the pre-built images from George... anyone know what is the root password for it?

nvm... telnet works :)

comment:135 in reply to: ↑ 130 ; follow-up: Changed 6 years ago by anonymous

Replying to anonymous:

Replying to anonymous:

which trunk version did you get for compiling ? 25469 for me works excellent, I have no issues so far. If you want also ipv6 you can use my image: https://e-lista.pl/OpenWrt/

Thanks. It was already present in my first build (around 25522). I only have this issue when using Transmission. Did you test your build with transmission?

I have tested it with ctorrent (running on the router), and it seems to crash as well, if there is (high) traffic. (There is one comment stating that TorrentFlux is ok, however I doubt that Torrentflux runs on the router)

comment:136 in reply to: ↑ 135 Changed 6 years ago by gregd72002@…

Replying to anonymous:

Replying to anonymous:

Replying to anonymous:

which trunk version did you get for compiling ? 25469 for me works excellent, I have no issues so far. If you want also ipv6 you can use my image: https://e-lista.pl/OpenWrt/

Thanks. It was already present in my first build (around 25522). I only have this issue when using Transmission. Did you test your build with transmission?

I have tested it with ctorrent (running on the router), and it seems to crash as well, if there is (high) traffic. (There is one comment stating that TorrentFlux is ok, however I doubt that Torrentflux runs on the router)

It happened once to me as well... it looks like it crashes under heavy traffic... this does not seem to be related to number of simultaneous connections as it crashed when using "Onlive" service in my case...

comment:137 follow-up: Changed 6 years ago by jow

Torrents crashing routers are hardly a new phenomena, it happens on all platforms, with all versions. I suggest to take the torrent discussion to the forum as it is a bit out of the scope of this model support ticket.

comment:138 in reply to: ↑ 137 Changed 6 years ago by anonymous

Replying to jow:

Torrents crashing routers are hardly a new phenomena, it happens on all platforms, with all versions. I suggest to take the torrent discussion to the forum as it is a bit out of the scope of this model support ticket.

I only mentioned this matter, since the same config does not crash on the wl-500gPv1 (and the only difference between the two targets is the bcm4716 patches) I suspect, that the support patches introduce some kind of instability, which only shows itself under heavy load (did not try yet with the fixed gcc-linaro; anyways the crash happens consistently in 2-5 minutes, which is not that usual on all platforms).

comment:139 Changed 6 years ago by anonymous

I have a problem with compiling my distro when i give make -j3 V=99 first after some compilation i get a question:

  • CPU selection

*
CPU type

  1. MIPS32 Release 1 (CPU_MIPS32_R1)
  1. MIPS32 Release 2 (CPU_MIPS32_R2) (NEW)

choice[1-2]:
And no matter what I choose iget this error:
SHELL= /home/paczki/OPENWRT/RT-N16/trunk/staging_dir/host/bin/flock /home/paczki/OPENWRT/RT-N16/trunk/tmp/.root-copy.flock -c 'cp -fpR /home/paczki/OPENWRT/RT-N16/trunk/staging_dir/target-mipsel_uClibc-0.9.32/root-brcm4716/tmp-kmod-fuse/. /home/paczki/OPENWRT/RT-N16/trunk/staging_dir/target-mipsel_uClibc-0.9.32/root-brcm4716/'
rm -rf /home/paczki/OPENWRT/RT-N16/trunk/staging_dir/target-mipsel_uClibc-0.9.32/root-brcm4716/tmp-kmod-fuse
touch /home/paczki/OPENWRT/RT-N16/trunk/staging_dir/target-mipsel_uClibc-0.9.32/root-brcm4716/stamp/.kmod-fuse_installed
make[3]: Leaving directory `/home/paczki/OPENWRT/RT-N16/trunk/package/fuse'
make[2]: Leaving directory `/home/paczki/OPENWRT/RT-N16/trunk'
make[1]: * home/paczki/OPENWRT/RT-N16/trunk/staging_dir/target-mipsel_uClibc-0.9.32/stamp/.package_compile Error 2
make[1]: Leaving directory `/home/paczki/OPENWRT/RT-N16/trunk'
make:
* [world] Error 2

Can anyone help me ?

comment:140 Changed 6 years ago by anonymous

The bcm4716 is MIPS32r2 and some of the patches seem to add support for r2, so I have selected 2, and it seems to work fine. (This question will pop up a second time near the end of compilation as well)

comment:141 Changed 6 years ago by dylek

I can see that today was uploaded new packages for brcm47xx and opkg doesn't work :/
I changed opkg.conf as posted by hcontrol and it was working until today. Is there any
idea chow to change the opkg.conf to start it working again ?

comment:142 Changed 6 years ago by anonymous

I was making update yesterday and every think works fine instead of kmod_tun :/
Can anybody share with me this module compiled for kernel 2.6.37 ? Without this
module i can not make openvpn connections.

Thanks a lot.

comment:143 Changed 6 years ago by gregd72002@…

I spent some days testing and it looks like there are some stability issues...
First of all trunk r16126 is the latest release George's patches will work. Trunk r16127 introduces some changes into ssb...

Apart from that was running r16126 on the router without major issues, however got the log produces the following (but the router continuous to work):

CPU 0 Unable to handle kernel paging request at virtual address 00000000, epc == 80009ee0, ra == 800ed940
Oops[#1]:
Cpu 0
$ 0   : 00000000 1100bc01 00000000 87a65e18
$ 4   : 8758c25c 81094b60 80269058 00000088
$ 8   : 00010000 00000007 00000007 00000008
$12   : 2b03ebd8 00200000 00000000 00000003
$16   : 8758c25c 8758c260 878e88f8 00000000
$20   : 86b58e40 00200200 00100100 80271488
$24   : 004650b8 2adf475c                  
$28   : 87a64000 87a65e08 003fffff 800ed940
Hi    : 00000000
Lo    : 00000000
epc   : 80009ee0 0x80009ee0
    Tainted: P           
ra    : 800ed940 0x800ed940
Status: 1100bc03    KERNEL EXL IE 
Cause : 0080000c
BadVA : 00000000
PrId  : 00019740 (MIPS 74Kc)
Modules linked in: usb_storage snd_usb_audio snd_usbmidi_lib evdev uhci_hcd ohci_hcd ledtrig_usbdev xt_LED ledtrig_morse ipt_SET ipt_set ip_set_setlist ip_set_portmap ip_set_nethash ip_set_macipmap ip_set_iptreemap ip_set_iptree ip_set_ipportnethash ip_set_ipportiphash ip_set_ipporthash ip_set_ipmap ip_set_iphash ip_set nf_nat_tftp nf_conntrack_tftp nf_nat_snmp_basic nf_nat_sip nf_conntrack_sip nf_nat_rtsp nf_conntrack_rtsp nf_nat_pptp nf_conntrack_pptp nf_nat_h323 nf_conntrack_h323 nf_nat_proto_gre nf_conntrack_proto_gre nf_nat_amanda nf_conntrack_amanda nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp xt_policy xt_esp ipt_ah ipt_REDIRECT ipt_NETMAP ipt_MASQUERADE iptable_nat nf_nat xt_recent xt_helper xt_conntrack xt_connmark xt_connbytes xt_NOTRACK iptable_raw xt_state nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack ehci_hcd sd_mod pppoe pppox pppoatm ipt_REJECT xt_TCPMSS ipt_LOG xt_comment xt_multiport xt_mac xt_limit iptable_mangle iptable_filter ip_tables xt_tcpudp x_tables nfs ip_gre gre l2tp_ip l2tp_eth l2tp_netlink l2tp_core ipcomp6 xfrm6_tunnel xfrm6_mode_tunnel xfrm6_mode_transport xfrm6_mode_beet esp6 ah6 ipcomp xfrm4_tunnel xfrm4_mode_tunnel xfrm4_mode_transport xfrm4_mode_beet esp4 ah4 ipip ext2 ppp_mppe tunnel6 tunnel4 ext3 jbd tun snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd_rawmidi snd_seq_device snd_hwdep snd_page_alloc snd soundcore ppp_async ppp_generic slhc loop xfrm_user xfrm_ipcomp af_key vfat fat ntfs lockd sunrpc isofs ext4 jbd2 wl(P) br2684 atm nls_utf8 usbcore ts_fsm ts_bm ts_kmp switch_robo switch_core scsi_mod nls_base crc16 crc_ccitt input_core bgmac sha1_generic crypto_null md5 hmac des_generic cbc arc4 aes_generic ecb cryptomgr crypto_hash crypto_blkcipher aead crypto_algapi ipv6 diag [last unloaded: switch_core]
Process transmission-da (pid: 1807, threadinfo=87a64000, task=878e88f8, tls=2b046970)
Stack : 81094b60 8758c32c 802a0000 00000000 8758c260 00000000 00100100 800597b8
        87a65e70 80098e20 8758c25c 8758c32c 00000200 00000000 86b58e40 00200200
        00100100 800ed940 000001ff 000001ff 86b58e40 00000200 81094b60 80271488
        81094b60 80064350 00000000 00000000 00000000 00000000 81088758 81095e38
        000017e4 00000000 00000000 00000200 00000000 00000200 8758c32c 00000200
        ...
Call Trace:[<800597b8>] 0x800597b8
[<80098e20>] 0x80098e20
[<800ed940>] 0x800ed940
[<80064350>] 0x80064350
[<800646e8>] 0x800646e8
[<800932a4>] 0x800932a4
[<8005dbe4>] 0x8005dbe4
[<80003630>] 0x80003630
[<800c45f8>] 0x800c45f8


Code: ac830008  afb10010  afa20014 <ac430000> afb20018  3c16802a  26c23760  8c420008  3c030001 
Unhandled kernel unaligned access[#2]:
Cpu 0
$ 0   : 00000000 1100bc01 80000000 00000009
$ 4   : 00000000 0fc1a000 84052000 0fc1a000
$ 8   : 00010000 00000026 00000007 00000008
$12   : 2b8c8a40 00200000 00000000 00000003
$16   : 0fc1a000 84052000 00000200 0fc1b000
$20   : 873fff40 8758c02c 87a11400 80271488
$24   : 004650b8 2b67e75c                  
$28   : 84d64000 84d65dc8 003fffff 800eef24
Hi    : 0000001a
Lo    : 00000000
epc   : 800edecc 0x800edecc
    Tainted: P      D    
ra    : 800eef24 0x800eef24
Status: 1100bc03    KERNEL EXL IE 
Cause : 80800010
BadVA : 0000001d
PrId  : 00019740 (MIPS 74Kc)
Modules linked in: usb_storage snd_usb_audio snd_usbmidi_lib evdev uhci_hcd ohci_hcd ledtrig_usbdev xt_LED ledtrig_morse ipt_SET ipt_set ip_set_setlist ip_set_portmap ip_set_nethash ip_set_macipmap ip_set_iptreemap ip_set_iptree ip_set_ipportnethash ip_set_ipportiphash ip_set_ipporthash ip_set_ipmap ip_set_iphash ip_set nf_nat_tftp nf_conntrack_tftp nf_nat_snmp_basic nf_nat_sip nf_conntrack_sip nf_nat_rtsp nf_conntrack_rtsp nf_nat_pptp nf_conntrack_pptp nf_nat_h323 nf_conntrack_h323 nf_nat_proto_gre nf_conntrack_proto_gre nf_nat_amanda nf_conntrack_amanda nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp xt_policy xt_esp ipt_ah ipt_REDIRECT ipt_NETMAP ipt_MASQUERADE iptable_nat nf_nat xt_recent xt_helper xt_conntrack xt_connmark xt_connbytes xt_NOTRACK iptable_raw xt_state nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack ehci_hcd sd_mod pppoe pppox pppoatm ipt_REJECT xt_TCPMSS ipt_LOG xt_comment xt_multiport xt_mac xt_limit iptable_mangle iptable_filter ip_tables xt_tcpudp x_tables nfs ip_gre gre l2tp_ip l2tp_eth l2tp_netlink l2tp_core ipcomp6 xfrm6_tunnel xfrm6_mode_tunnel xfrm6_mode_transport xfrm6_mode_beet esp6 ah6 ipcomp xfrm4_tunnel xfrm4_mode_tunnel xfrm4_mode_transport xfrm4_mode_beet esp4 ah4 ipip ext2 ppp_mppe tunnel6 tunnel4 ext3 jbd tun snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd_rawmidi snd_seq_device snd_hwdep snd_page_alloc snd soundcore ppp_async ppp_generic slhc loop xfrm_user xfrm_ipcomp af_key vfat fat ntfs lockd sunrpc isofs ext4 jbd2 wl(P) br2684 atm nls_utf8 usbcore ts_fsm ts_bm ts_kmp switch_robo switch_core scsi_mod nls_base crc16 crc_ccitt input_core bgmac sha1_generic crypto_null md5 hmac des_generic cbc arc4 aes_generic ecb cryptomgr crypto_hash crypto_blkcipher aead crypto_algapi ipv6 diag [last unloaded: switch_core]
Process transmission-da (pid: 3354, threadinfo=84d64000, task=878e9178, tls=2b8d0970)
Stack : 00000001 00000000 003fffff 80271488 00000002 800612dc 00000000 81080a40
        8758c0fc 00000200 00000000 873fff40 00200200 00100100 80271488 800ed3ac
        81080a40 8758c0fc 802a0000 00000000 00001000 00200200 00100100 800597b8
        81080a40 800ed8fc 86b5ae20 2b8c8dcc 0000fc1a 0000467f 81080a40 8758c0fc
        8758c02c 800ed950 000001ff 000001ff 873fff40 00000200 81080a40 80271488
        ...
Call Trace:[<800612dc>] 0x800612dc
[<800ed3ac>] 0x800ed3ac
[<800597b8>] 0x800597b8
[<800ed8fc>] 0x800ed8fc
[<800ed950>] 0x800ed950
[<80064350>] 0x80064350
[<800646e8>] 0x800646e8
[<800932a4>] 0x800932a4
[<8005dbe4>] 0x8005dbe4
[<80003630>] 0x80003630
[<868a5f3e>] 0x868a5f3e


Code: 10c00005  00000000  8c630008 <5460ffef> 8c660014  00801021  03e00008  00000000  27bdffc8 

Any idea how to fix this?

comment:144 Changed 6 years ago by gregd72002@…

Concerning my previous post, I meant r26126 (not r16126) is working and r26127 (not r16126) introduces ssb changes.

comment:145 Changed 6 years ago by krey@…

when this patches goes to trunk? now a hard to compile completely custom openwrt with this patches. Build often falls.

comment:146 Changed 5 years ago by Krey@…

i get this:

root@OpenWrt:/# wifi enable
wl0: Invalid argument
wl0: Invalid argument
root@OpenWrt:/# wl radio on
wl: wl driver adapter not found

comment:147 Changed 5 years ago by yrleu

The Wi-Fi link of my DIR-627 with Linux Kernel 2.6.37 andbroadcom-wl 5.10.56.27.3 is quite unstable.

Therefore, I have to try to downgrade to Linux 2.6.22 and use the proprietary wireless-related .o files (Broadcom 5.10.147.0) provided by ASUS. :(

comment:148 follow-up: Changed 5 years ago by gregd72002@…

Have been running r26126 on my rt n16 for over a month now. Haven't had any issues at all. No reboots or hangs. Wireless connectivity (though just at 130Mbps) is much more stable than on ddwrt...
Running usb-audio, mpd, usb hdd, asterisk, timecapsule and other stuff - no issues at all! Great stuff!

The only thing we need is to get this integrate this into the openwrt trunk!!


comment:149 Changed 5 years ago by hauke

There is a complete redesign done of the the bus. The last part of the discussion can be found here: https://lkml.org/lkml/2011/4/14/156 . George is still working on this. I do not want to create a new target for these devices as they could share the code with the old Broadcom target. As long as the old devices do not work when integrating support for the bcm4716 chips I will not merge this. The brcm47xx target should contain two subtargets on with support for older mips cpus and one with optimizations and only for newer mips cpus. They should share the source code and should only differ in compiler options.

I would like to see a patch which adds support for these new SoCs to the existing brcm47xx target so that the image will boot on old and new devices.

The Ethernet driver in this patch is not licensed under the GPL as it was created from non GPL code from Braodcom.

comment:150 Changed 5 years ago by keknek

How we can get this work? End users are not intrested or even skilled how those patches should be or not to be.

There's a lot of people who need rt-n16 under openwrt. So I trully hope you can make some solution that end users can run their rt-n16 boards under openwrt without 'alien' skills...

I see that some people have done very hard work to get rt-n16 run under openwrt. So this is my petition that their work should be visible...

If it's not possible I hope there's some person who can make some unofficial version of openwrt or detailed instructions how we can run rt-16 under openwrt. Please...

comment:151 follow-up: Changed 5 years ago by anonymous

I cannot find any working links to George's patches anymore (links in list are returning 404). Can You please repost working links?

comment:152 in reply to: ↑ 151 Changed 5 years ago by GKashperko

Replying to anonymous:

I cannot find any working links to George's patches anymore (links in list are returning 404). Can You please repost working links?

Site was down for maintenance. Original links are valid again.

comment:153 in reply to: ↑ 148 Changed 5 years ago by anonymous

Hey Greg? Could you please post the (post patch) image that you made for the rest of us?
thanks!

Replying to gregd72002@…:

Have been running r26126 on my rt n16 for over a month now. Haven't had any issues at all. No reboots or hangs. Wireless connectivity (though just at 130Mbps) is much more stable than on ddwrt...
Running usb-audio, mpd, usb hdd, asterisk, timecapsule and other stuff - no issues at all! Great stuff!

The only thing we need is to get this integrate this into the openwrt trunk!!


comment:154 follow-up: Changed 5 years ago by anonymous

It seems I got my RT-N16 working with r26126 and George's patches 000 to 002 using the default configuration (going to see why my configuration doesn't want to boot and since I don't have CFE it's gona take a while). The only problem seems to be, that power led is not working (always turned off).

comment:155 in reply to: ↑ 154 ; follow-up: Changed 5 years ago by anonymous

Replying to anonymous:

It seems I got my RT-N16 working with r26126 and George's patches 000 to 002 using the default configuration (going to see why my configuration doesn't want to boot and since I don't have CFE it's gona take a while). The only problem seems to be, that power led is not working (always turned off).

Nevermind that power led proble, it magically started working.

comment:156 in reply to: ↑ 155 Changed 5 years ago by gregd72002@…

Replying to anonymous:

Nevermind that power led proble, it magically started working.

The power led switches on only after all init scripts are finished executing. In some cases (especially the very first boot) it might take a few minutes.

comment:157 Changed 5 years ago by ncruz

Hi,

Can anyone post a link to a compiled firmware? Thanks!

comment:158 follow-up: Changed 5 years ago by anonymous

I've been following this ticket from afar as I am not skilled enough to aid in the devolopment. How close is the rt-n16 to have built in support in the main openwrt builds? Should the wiki be updated from WIP to supported?

comment:159 in reply to: ↑ 158 Changed 5 years ago by anonymous

Replying to anonymous:

I've been following this ticket from afar as I am not skilled enough to aid in the devolopment. How close is the rt-n16 to have built in support in the main openwrt builds? Should the wiki be updated from WIP to supported?

No, OpenWRT trunk or releases does not support Asus RT-N16. You can use George's patches from https://lists.openwrt.org/pipermail/openwrt-devel/2011-February/009653.html to patch r26126 and then configure and compile OpenWRT manually. These patches cannot be added to the OpenWRT trunk because this code uses Broadcom's proprietary components.

comment:160 Changed 5 years ago by anonymous

Great thanks for the reply. Would you be able to give me and anyone else wondering a hint as to how I'd go about applying these patches to these source? I've compiled software on linux before but never modified it before doing so with patches.

comment:161 Changed 5 years ago by shinton@…

I too have wondered about how to go about doing this, is it just as easy as putting the files in the folder that contains all the .patch files?

comment:162 Changed 5 years ago by anonymous

Great work !! I think now we all need clear step-by-step manual how to get openwrt compiled with patches.

comment:163 Changed 5 years ago by joeshint@…

Has anyone looked at integrating these patches into debwrt? Debwrt is debian linux running on an openwrt base allowing for a complete linux experience on our routers. http://dev.debwrt.net/ As far as I can tell it would be the same patching process as for openwrt to get debwrt to work with the rt-n16. I agree with the previous posts, if someone could point us in the right direction and describe how to patch openwrt it would be much appreciated. Great work everybody!

comment:164 in reply to: ↑ 109 Changed 5 years ago by shinton@…

I too have this same problem, which folder did you put the patches in to get it to work?

Replying to hcontrol:

Up and running now !! I just made a mistake by putting the patches in the wrong
directory.

I changed the opkg.conf to be able to use existing packages..

src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay
arch all 1
arch mipsel 200
arch brcm47xx 10

comment:165 Changed 5 years ago by shinton@…

People it's actually quite easy I realized, 1. download openwrt trunk source. 2. cd into trunk source folder. 3. download patches into this folder 4. execute this command patch -p1 < file.patch where file.patch is where you would put the name of the 3 patches one at a time in order.

comment:166 Changed 5 years ago by anonymous

"1. download openwrt trunk source"
Latest?

Need I chmod like George?

comment:167 Changed 5 years ago by shinton@…

I used r26126 which can be downloaded here:
https://dev.openwrt.org/changeset/26126/trunk?old_path=%2F&old=26126&format=zip
Not too sure about the latest but you could try, might need to modify some of the patches to work. And you shouldn't need to chmod anything...

comment:168 Changed 5 years ago by cromo

I just finished compiling and but got a little confused about which of the binary should I use to upgrade the router. There's no file having n16 in its name in bin/brcm4716 directory, even though I can see e.g. openwrt-e3000_v1-squashfs.bin for which his patch also provides support. openwrt-brcm4716-squashfs.trx is my guess but still I'd like to confirm that anyway.

comment:169 Changed 5 years ago by shinton@…

When you tun make menuconfig you must select BCM94716 and then under target profile select your device (RT-N16). These option will only come up if you patch the source files FIRST. Please read the README found here: http://www.znau.edu.ua/temp/asus-rt-n16/openwrt-bcm4716-README.txt
and follow its instructions at the bottom exactly. Hope this helps!

comment:170 Changed 5 years ago by cromo

@shinton, I did patch the source and select target accordingly, I am only asking which of the bin files I am supposed to flash.

comment:171 follow-up: Changed 5 years ago by shinton@…

N16 comes up with a -none in the filename.

comment:172 in reply to: ↑ 171 Changed 5 years ago by anonymous

Replying to shinton@…:

N16 comes up with a -none in the filename.

I've tried the openwrt-brcm4716-squashfs.trx compiled after 3 patches were applied. It boots after tftp flash but I can't seem to login to it via the IP specified in the .config file.

I can't seem to find any .trx files that has N16/-None in it... did a locate as well it's none existence in my VM environment. I tried the same file from the author's website and it works well, I can telnet into tty fine. Am I missing something? the file size appears to be the same.

comment:173 Changed 5 years ago by cromo

@anonymous - I was having same problem, I just found a solution. I forgot to do the chmod +x on netconfig as it is mentioned in George's readme file. Don't forget do the 'make clean' or deleting the 'base package' manually.

Now I have other problem - I can't get the WAN to work. I only can see eth0, eth0.0 and eth0.1 in ifconfig, and dmesg says:
roboswitch: Probing device eth0: No such device
roboswitch: Probing device eth1: No such device
roboswitch: Probing device eth2: No such device
roboswitch: Probing device eth3: No such device

Any idea why is it like this? I am using the 26126 revision.

comment:174 Changed 5 years ago by cromo

@shinton, anonymous
There is no file with "-none" built here at all, I flashed the bin/brcm4716/openwrt-brcm4716-squashfs.trx and it works, despite my WAN problem as described above.

Also, George's site (http://www.znau.edu.ua/temp/asus-rt-n16/2011-02-13T23-17/) also has a precompiled binary for n16 available for download and it has the same filename, i.e. openwrt-brcm4716-squashfs.trx.

comment:175 follow-up: Changed 5 years ago by shinton@…

Ya sorry, realized just now that the -none flag in the filename is actually from the rt-n16 debwrt build that I'm working on. If anyone is interested I will share these files if I have success.

comment:176 in reply to: ↑ 175 Changed 5 years ago by anonymous

@cromo, thanks I think that's what I missed; will give that a try again. I also have the similar errors and the only interface that came up (using the author's stock image) is eth0. I noticed when applying the patch to r26126 there is a "switch-robo.c.rej". However, I've checked briefly and it appears to have been patched previously and there were only cosmetic changes that are different... I guess that means it's only good to use it as a wireless bridge as opposed to a router? :)

comment:177 Changed 5 years ago by anonymous

Update - I got it up and running with webif GUI. I ended up check out tag r25618 and applied all 3 patches with no errors (no much difference really but I found this is r25619 has the change from George). The problem is indeed the permission with netconfig. Cheers!

comment:178 Changed 5 years ago by cromo

@anonymous, is WAN working fine with r25618? Do you get any roboswitch errors with this revision?

comment:179 Changed 5 years ago by anonymous

OK, I got it working with r25618, eth0.1 has picked up the dhcp IP from my cable modem. Thanks!

comment:180 Changed 5 years ago by bdkoepke@…

@shinton

I would be extremely interested in trying your debwrt builds. I am also currently working on getting openwrt r25618 in debwrt.

comment:181 Changed 5 years ago by shinton@…

@bdkoepke

So far I've been unable to get any of the switch ports to work. It would seem that it boots ok as you can hear the HDD making some noises but that's just a guess and I won't fully know what's going on until my serial cable comes in. So far my approach has been to modify the openwrt.mk file to automatically apply the patches and do the chmod +x from the readme as well as edit the .config file created when running make menuconfig from the top directory and changing the appropriate values to brcm4716 as this is what make openwrt/menuconfig uses to populate its menu. I suspect it is a problem with improper permissions as many people here have faced but please do let me know if you make any headway.

comment:182 Changed 5 years ago by bdkoepke@…

@shinton

Thanks for the info, I chose to compile the kernel separately by specifying it download r25618 manually. i.e. make openwrt/prepare (after editing openwrt.mk) && make openwrt/menuconfig && make. I'm just finishing up the debian compile.

I don't want to flash it unless I know I can recover it. Where did you order a serial cable for debugging? I'm new to the whole jtag thing. (Not to kernel development, I just haven't done much MIPS work. I have done a significant amount with ARM-based devices though...)

Thanks in advance. I might flash it in the meantime anyway, I have a spare router. If all goes well I will let you know.

comment:183 Changed 5 years ago by shinton@…

@bdkoepke

You won't brick your rt-n16 with a bad flash, it has a built in recovery mode that has worked for me everytime. (see here: http://infodepot.wikia.com/wiki/Asus_RT-N16#Recovery_Mode ) The serial cable is a nokia ca-42 and for convenience I also ordered a 4 pin audio cable which I will use as a plug to connect up to the headers without soldering. These are the two I bought here: http://cgi.ebay.ca/ws/eBayISAPI.dll?ViewItem&item=280671264438 http://cgi.ebay.ca/ws/eBayISAPI.dll?ViewItem&item=370116779876

comment:184 Changed 5 years ago by anonymous

@shinton

Thank you, I will try flashing this soon, and I will report back.

comment:185 Changed 5 years ago by bdkoepke@…

@shinton

Looks like I'm in the same boat as you. (The drive spins up, but I can't get an address or ping the ip with a static address). I may have done the procedure incorrectly, the instructions are a little bit unclear. They specify that you have to build an openwrt kernel, which is fine, but then they talk about a debwrt kernel? What is the difference? You can't boot two kernels, so I don't get it. I think they're referencing the same one...

Shinton, if you don't mind would you email me @ bdkoepke at gmail dot com since we are working on the same thing, and then any *BREAKTHROUGHS* we can post here? I just don't want to be spamming everyone in this ticket. If not that's fine, I just don't want to drive anyone away.

comment:186 Changed 5 years ago by anonymous

Update - got webif installed but getting this error when configuring wifi:

wl0(mac80211): Interface type not supported

When I try from the command line "/sbin/wifi" I get the same error... dmesg says:

wl0: Broadcom BCM4329 802.11 Wireless Controller 5.10.56.27

comment:187 Changed 5 years ago by jow

"wl0(mac80211)" sounds completely wrong. wl0 usually indicates proprietary broadcom.
Where did you get your wifi config from? I suggest to try "rm /etc/config/wireless; wifi detect > /etc/config/wireless"

comment:188 Changed 5 years ago by anonymous

Thanks, I'll give that I try. I think what I need to use is the "nas" package? :)

comment:189 Changed 5 years ago by cromo

@anonynous
I got the wifi working with following settings, that were added to /etc/config/wireless by luci and they work fine. Except maybe for 802.11n that I can't get to work, but according to openwrt manuals it's supposed to be impossible. Which is weird as the same broadcom-wl driver is used by DDWRT where I could easly get the 300Mbps bit rate.

config 'wifi-device' 'wl0'

option 'type' 'broadcom'
option 'disabled' '0'
option 'channel' '6'
option 'txpower' '16'
option 'hwmode' '11bg'
option 'frameburst' '1'

comment:190 Changed 5 years ago by cromo

I also have a problem with my usb pendrive. It gets detected (according to dmesg) and relevant /dev/sda[1] nodes are created (and are readable with cat/less), but "mount -t ext2 /dev/sda1 /mnt/share" fails with "mount: mounting /dev/sda1 on /mnt/share failed: No such device". Any ideas?

comment:191 Changed 5 years ago by cromo

OK, never mind, I thought that ext2 is supported out-of-the-box, but /proc/filesystems says it supports ext3 only, so I reformatted the pendrive and it works fine now.

comment:192 Changed 5 years ago by anonymous

@cromo no 802.11n is perfectly fine with me, I don't plan to use N in 2.4ghz and so B/G is good enough for me.
I tried to put in the config and did a wifi command but am not sure if I'm doing it right or not. Another question, the "luci" mentioned above is it program or person you are referring to? Thanks again.

comment:193 Changed 5 years ago by cromo

I, on the other hand, would make good use of 802.11n.

luci is a web frontend for OpenWrt:
http://wiki.openwrt.org/doc/howto/luci.essentials
http://luci.subsignal.org/ (site currently down)

comment:194 Changed 5 years ago by dylek

Hi, I have one litle request could some one put link to this post for example in
wiki: http://wiki.openwrt.org/inbox/asusrt-n16/wip, it is really hard to find this solutions.

comment:195 Changed 5 years ago by anonymous

@dylek There already is a link, look closer

comment:196 follow-up: Changed 5 years ago by cromo

When running transmission and downloading, ethernet driver fails after some time and relevant interfaces disappear from the ifconfig list. r25618. Here's the stack trace from dmesg:

BUG: scheduling while atomic: transmission-da/1526/0x10000100
Modules linked in: usb_storage uhci_hcd nf_nat_tftp nf_conntrack_tftp nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp ipt_MASQUERADE iptable_nat nf_nat xt_NOTRACK iptable_raw xt_state nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack ehci_hcd sd_mod pppoe pppox ipt_REJECT xt_TCPMSS ipt_LOG xt_comment xt_multiport xt_mac xt_limit iptable_mangle iptable_filter ip_tables xt_tcpudp x_tables ext3 jbd ppp_async ppp_generic slhc wl(P) busbcore switch_robo switch_core scsi_mod nls_base crc_ccitt bgmac diag [last unloaded: switch_core]
Cpu 0
$ 0 : 00000000 1100bc00 1100bc01 00000001
$ 4 : 8750c050 0000001b 8750c050 00000000
$ 8 : 8750c000 ffffffff 0000000d 87803988
$12 : 87803990 ffffffff 00000001 87fe2c60
$16 : 8750c050 1100bc01 000000d0 87801660
$20 : 86c2be08 86c2be18 86c2be10 00200200
$24 : 00000000 8011cb28
$28 : 86c2a000 86c2bd50 00100100 80091b4c
Hi : 00000000
Lo : 00000000
epc : 80091b78 0x80091b78

Tainted: P

ra : 80091b4c 0x80091b4c
Status: 1100bc03 KERNEL EXL IE
Cause : 00801000
PrId : 00019740 (MIPS 74Kc)
BUG: scheduling while atomic: klogd/291/0x10000100
Modules linked in: usb_storage uhci_hcd nf_nat_tftp nf_conntrack_tftp nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp ipt_MASQUERADE iptable_nat nf_nat xt_NOTRACK iptable_raw xt_state nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack ehci_hcd sd_mod pppoe pppox ipt_REJECT xt_TCPMSS ipt_LOG xt_comment xt_multiport xt_mac xt_limit iptable_mangle iptable_filter ip_tables xt_tcpudp x_tables ext3 jbd ppp_async ppp_generic slhc wl(P) usbcore switch_robo switch_core scsi_mod nls_base crc_ccitt bgmac diag [last unloaded: switch_core]
Cpu 0
$ 0 : 00000000 1100bc01 00002105 00000068
$ 4 : 00002104 87aa9ef0 87aa9efc 1100bc01
$ 8 : 87aa9fe0 0000bc00 00000000 86fa4000
$12 : 00000000 ffffffff 00000001 00000000
$16 : 00000000 00000014 00001000 802c0000
$20 : 00000000 0047b7d8 80280000 80280000
$24 : 00000000 8016a494
$28 : 87aa8000 87aa9ee0 802c0000 80023c9c
Hi : 0000008c
Lo : a898af00
epc : 80023c84 0x80023c84

Tainted: P

ra : 80023c9c 0x80023c9c
Status: 1100bc03 KERNEL EXL IE
Cause : 00802000
PrId : 00019740 (MIPS 74Kc)
BUG: scheduling while atomic: klogd/291/0x10000100
Modules linked in: usb_storage uhci_hcd nf_nat_tftp nf_conntrack_tftp nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp ipt_MASQUERADE iptable_nat nf_nat xt_NOTRACK iptable_raw xt_state nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack ehci_hcd sd_mod pppoe pppox ipt_REJECT xt_TCPMSS ipt_LOG xt_comment xt_multiport xt_mac xt_limit iptable_mangle iptable_filter ip_tables xt_tcpudp x_tables ext3 jbd ppp_async ppp_generic slhc wl(P) usbcore switch_robo switch_core scsi_mod nls_base crc_ccitt bgmac diag [last unloaded: switch_core]
Cpu 0
$ 0 : 00000000 1100bc01 00002105 00000068
$ 4 : 00002104 87aa9ef0 87aa9efc 1100bc01
$ 8 : 87aa9fe0 0000bc00 00000000 86fa4000
$12 : 00000000 ffffffff 00000001 00000000
$16 : 00000000 00000014 00001000 802c0000
$20 : 00000000 0047b7d8 80280000 80280000
$24 : 00000000 8016a494
$28 : 87aa8000 87aa9ee0 802c0000 80023c9c
Hi : 0000008c
Lo : a898af00
epc : 80023c84 0x80023c84

Tainted: P

ra : 80023c9c 0x80023c9c
Status: 1100bc03 KERNEL EXL IE
Cause : 00802000
PrId : 00019740 (MIPS 74Kc)
ssb-dma: ssb_dma_rxfill: BUG: drivers/ssb/ssb_dma.c:989
ssb-dma: ssb_dma_rxfill: BUG: drivers/ssb/ssb_dma.c:989
ssb-dma: ssb_dma_rxfill: BUG: drivers/ssb/ssb_dma.c:989
ssb-dma: ssb_dma_rxfill: BUG: drivers/ssb/ssb_dma.c:989
ssb-dma: ssb_dma_rxfill: BUG: drivers/ssb/ssb_dma.c:989
... and so on.

comment:197 in reply to: ↑ 196 ; follow-up: Changed 5 years ago by GKashperko

Replying to cromo:

When running transmission and downloading, ethernet driver fails after some time and relevant interfaces disappear from the ifconfig list. r25618. Here's the stack trace from dmesg:

ssb-dma: ssb_dma_rxfill: BUG: drivers/ssb/ssb_dma.c:989
... and so on.

Could you please contact me on icq #12962491 or by mail (george <at> znau.edu.ua), think I could fix that.

comment:198 follow-up: Changed 5 years ago by 2steps@…

Do I still have to apply the patches or did the changes go to the repo? Checking out the latest from repository, the patches did result in errors...

comment:199 in reply to: ↑ 198 Changed 5 years ago by anonymous

Replying to 2steps@…:

Do I still have to apply the patches or did the changes go to the repo? Checking out the latest from repository, the patches did result in errors...

Of course they do.

comment:200 in reply to: ↑ 197 ; follow-up: Changed 5 years ago by chand

Replying to GKashperko:

Replying to cromo:

When running transmission and downloading, ethernet driver fails after some time and relevant interfaces disappear from the ifconfig list. r25618. Here's the stack trace from dmesg:

ssb-dma: ssb_dma_rxfill: BUG: drivers/ssb/ssb_dma.c:989
... and so on.

Could you please contact me on icq #12962491 or by mail (george <at> znau.edu.ua), think I could fix that.

Hello,

I used the r25469 two months without trouble on a RT-N16. Since I switch to r25618 I've got this error or just a freeze. Could you provide a fix ?

Thanks.

comment:201 in reply to: ↑ 200 ; follow-up: Changed 5 years ago by anonymous

Replying to chand:
Are you sure this has only appeared after updating from r25469? If so, it will probably help a lot to determine the cause of the problem.

comment:202 in reply to: ↑ 201 Changed 5 years ago by chand

Replying to anonymous:

Replying to chand:
Are you sure this has only appeared after updating from r25469? If so, it will probably help a lot to determine the cause of the problem.

Not sure if the bug already appear on r25469, I remember on or two freeze but no logs in last months.

comment:203 Changed 5 years ago by GKashperko

Updated patchset for recent trunks (.trx image is built with r26913)
http://www.znau.edu.ua/temp/asus-rt-n16/2011-05-17T04-28/

comment:204 Changed 5 years ago by anonymous

Thank you George!

comment:205 Changed 5 years ago by anonymous

Thanks George! Very close to having debwrt running as a result of your patches :)

comment:206 follow-up: Changed 5 years ago by windozer

Please could somebody compile one with luci working out of the box? Or probably direct me to a build that has one.

comment:207 in reply to: ↑ 206 Changed 5 years ago by chand

Replying to windozer:

Please could somebody compile one with luci working out of the box? Or probably direct me to a build that has one.

I built one with r26913 then from trunk with luci enabled (also from luci trunk), just after login in luci my rt-n16 reboot. No problem if I don't try to login. Something wrong with Luci from trunk perhaps.

comment:208 Changed 5 years ago by cromo

chand, this is actually quite a different bug with wireless interface that luci attempts to access on a status page. George is investigating that at the moment.

comment:209 follow-up: Changed 5 years ago by GKashperko

There are problems with wl if building trunk with 2011-05-17 patchset. Currently i'm looking into it. If you suffer from wl crashing (most likely you are) then remove package/broadcom-wl/patches/006-generic-dma-api.patch and rebuild the image.

comment:210 in reply to: ↑ 209 Changed 5 years ago by GKashperko

Replying to GKashperko:

There are problems with wl if building trunk with 2011-05-17 patchset. Currently i'm looking into it. If you suffer from wl crashing (most likely you are) then remove package/broadcom-wl/patches/006-generic-dma-api.patch and rebuild the image.

Updated patchset with wl crash fixed
www.znau.edu.ua/temp/asus-rt-n16/2011-05-18T00-05

comment:211 Changed 5 years ago by anonymous

For those interested, a thread for rt-n16 debwrt support using these patches has been started here

As stated in the linked thread, the current issue faced in debwrt is the inability to connect to the device after flashing an image. The approach so far has been to try and use the steps outlined in the openwrt brcm4716 readme by George as ckised as possible in their build environment. Sorry if this is too off-topic

comment:212 follow-up: Changed 5 years ago by anonymous

George,

when applying the final patch (002) on build 26966 I get the following error:

"The next patch would create the file package/broadcom-wl/patches/006-generic-dma-api.patch,
which already exists! Skipping patch.
1 out of 1 hunk ignored"

Any idea how to fix this?

comment:213 in reply to: ↑ 212 Changed 5 years ago by anonymous

Replying to anonymous:

when applying the final patch (002) on build 26966 I get the following error:

"The next patch would create the file package/broadcom-wl/patches/006-generic-dma-api.patch,
which already exists! Skipping patch.
1 out of 1 hunk ignored"

Any idea how to fix this?

006-generic-dma-api.patch merged with trunk since #26949, just ignore this message.

comment:214 Changed 5 years ago by shinton@…

The RT-N16 is now *Supported for testing purposes on Debwrt. See the following link:

http://dev.debwrt.net/wiki/AsusRT-N16

I say *Supported as it is still very fresh and untested
There is a forum there where any future support discussion should continue.

Try at your own risk.

comment:215 follow-up: Changed 5 years ago by anonymous

I have a problem with networking on build r27014 and r26970 (the box boots well, power led is turning on but no connection to it). No problems with r26913. Any ideas what's the problem with last releases? What release is to be used for compilation?

comment:216 in reply to: ↑ 215 ; follow-up: Changed 5 years ago by GKashperko

Replying to anonymous:

I have a problem with networking on build r27014 and r26970 (the box boots well, power led is turning on but no connection to it). No problems with r26913. Any ideas what's the problem with last releases? What release is to be used for compilation?

My current development tree is based on r26949, so far it builds and works just fine.
Does target/linux/brcm4716/base-files/etc/init.d/netconfig have +x attribute set ?

comment:217 in reply to: ↑ 216 Changed 5 years ago by anonymous

Replying to GKashperko:

Replying to anonymous:

I have a problem with networking on build r27014 and r26970 (the box boots well, power led is turning on but no connection to it). No problems with r26913. Any ideas what's the problem with last releases? What release is to be used for compilation?

My current development tree is based on r26949, so far it builds and works just fine.
Does target/linux/brcm4716/base-files/etc/init.d/netconfig have +x attribute set ?

Yes I remember to make it executable. I've tried to include ip6tables options to config. Is there some issues with IPv6 support? Thanks for your work George!

comment:218 Changed 5 years ago by anonymous

I too have issues with IPv6. IPv4 with luci works fine tho

comment:219 follow-up: Changed 5 years ago by OpenSys

Hello,

I decided the compile openwrt for rt-n16 but, openwrt trunk give me:
"ERROR: please fix target/linux/brcm4716/Makefile"

I compare the Makefile with others targets, and seams ok.
the revisions is r26913 and the last one. but say th same and of course that the target, don't show up.
What is the problem?

OpenSys

comment:220 in reply to: ↑ 219 Changed 5 years ago by anonymous

Replying to OpenSys:

Hello,

I decided the compile openwrt for rt-n16 but, openwrt trunk give me:
"ERROR: please fix target/linux/brcm4716/Makefile"

I compare the Makefile with others targets, and seams ok.
the revisions is r26913 and the last one. but say th same and of course that the target, don't show up.
What is the problem?

OpenSys

Found it :)

required to copy the config brcm4716 kernel to openwrt kernel src dir.

comment:221 Changed 5 years ago by anonymous

Anyone kind enough to post the latest Working .trx image for this? I know there have been some updates/patches but the build on George's site is dated 05/17 and it seems he has some patches after that.

Thanks in advance.

comment:222 Changed 5 years ago by Krey@…

How set 40Mhz bandwith in config/wireless? htmode not work for broadcom.

comment:223 follow-up: Changed 5 years ago by cromo@…

George, is there any reason why http://wireless.kernel.org/en/users/Drivers/brcm80211
driver wouldn't work for wifi?

comment:224 Changed 5 years ago by cromo@…

OK, I think I answered the question myself - not yet available for non x86/64 architectures.

comment:225 follow-up: Changed 5 years ago by cromo@…

Can anyone with a decent internet connection and external USB drive can check whether he also experiences router rebooting soon after transmission reaches download transfer of about 1MB/s? I keep struggling to find a reason of reboots but without serial console I can't tell much. I know this has been reported already in this ticket and advised to be treaded as a common transmission pitfall when running on router, but this one seems kind of specific, as if I limit the download speed to about 200Kb/s, I no longer experience reboots, but WAN dies instead after some time (~15 minutes) and SSB errors appear in dmesg (as already reported by me and others in this topic)
So once again, if anyone could please help testing and hopefully getting some useful kernel oops dump, I'd be very grateful.

comment:226 in reply to: ↑ 225 Changed 5 years ago by GKashperko

Replying to cromo@…:

Can anyone with a decent internet connection and external USB drive can check whether he also experiences router rebooting soon after transmission reaches download transfer of about 1MB/s? I keep struggling to find a reason of reboots but without serial console I can't tell much. I know this has been reported already in this ticket and advised to be treaded as a common transmission pitfall when running on router, but this one seems kind of specific, as if I limit the download speed to about 200Kb/s, I no longer experience reboots, but WAN dies instead after some time (~15 minutes) and SSB errors appear in dmesg (as already reported by me and others in this topic)
So once again, if anyone could please help testing and hopefully getting some useful kernel oops dump, I'd be very grateful.

I tried to track out kernel oops caused by transmission under high loads but after recent broadcom-wl fixes I don't anymore have any ssb-related messages when oops happens. It still can be caused by some ssb/ai-related bug ofcourse but atm I don't have any evidence for that.

comment:227 Changed 5 years ago by OpenSys

Seams that wifi driver was falling, something like ifconfig wl0 up = kernel panic and instant reboot.

comment:228 follow-up: Changed 5 years ago by cromo@…

George, are these broadcom-wl fixes you mention included in your recent patches or SVN trunk? Yesterday I compiled HEAD with your patches from 18th may and the router still reboots. I will try again today with make distclean instead of make clean, maybe that will help.

comment:229 in reply to: ↑ 228 ; follow-ups: Changed 5 years ago by gkashperko

Replying to cromo@…:

George, are these broadcom-wl fixes you mention included in your recent patches or SVN trunk? Yesterday I compiled HEAD with your patches from 18th may and the router still reboots. I will try again today with make distclean instead of make clean, maybe that will help.

There were two wl patches I mention - both merged with trunk as of r26949.
Updated patchset reflecting these changes is at http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-06T16-59/
.trx image built from scratch (r27114 changeset) boots just fine (available at the ftp above). Have no wl crashes with default wifi setup. If you manage to get stock .trx crash please let me know.

comment:230 in reply to: ↑ 229 ; follow-up: Changed 5 years ago by anonymous

Replying to gkashperko:

Replying to cromo@…:

George, are these broadcom-wl fixes you mention included in your recent patches or SVN trunk? Yesterday I compiled HEAD with your patches from 18th may and the router still reboots. I will try again today with make distclean instead of make clean, maybe that will help.

There were two wl patches I mention - both merged with trunk as of r26949.
Updated patchset reflecting these changes is at http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-06T16-59/
.trx image built from scratch (r27114 changeset) boots just fine (available at the ftp above). Have no wl crashes with default wifi setup. If you manage to get stock .trx crash please let me know.

George, I've pulled the .trx from the URL above (-59) and wifi config fails with 'wl0: Invalid argument' x 2. Wired ethernet seems to work ok. This is my /etc/config/wireless

config 'wifi-device' 'wl0'

option 'type' 'broadcom'
option 'channel' '11'

config 'wifi-iface'

option 'device' 'wl0'
option 'network' 'lan'
option 'mode' 'ap'
option 'ssid' 'xxxxxxxxxxxx'
option 'encryption' 'psk2'
option 'key' 'xxxxxxxxxxxxxxx'

(SSID and key removed).

/etc/config/network

#### VLAN configuration
config switch eth0

option enable 1

config switch_vlan eth0_0

option device "eth0"
option vlan 0
option ports "1 2 3 4 8*"

config switch_vlan eth0_1

option device "eth0"
option vlan 1
option ports "0 8"

#### Loopback configuration
config interface loopback

option ifname "lo"
option proto static
option ipaddr 127.0.0.1
option netmask 255.0.0.0

#### LAN configuration
config interface lan

option type bridge
option ifname "eth0.0"
option proto static
option ipaddr 192.168.1.5
option netmask 255.255.255.0

#### WAN configuration
config interface wan

option ifname "eth0.1"
option proto dhcp


config interface 6rd

option proto 6to4
option adv_subnet 1 # Selects the advertised /64 prefix, default 1 if not specified

Also, using the opkg.conf hack above, I can install packages but not kmods as the kernel revision on trunk/brcm47xx seems to be different. This kinda blocks ipv6.

/etc/opkg.conf
cat /etc/opkg.conf
src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay

packages:
6scripts - 0.1-6
6to4 - 3-1
base-files - 71-r27114
busybox - 1.18.4-1
dnsmasq - 2.57-2
dropbear - 0.53.1-3
hotplug2 - 1.0-beta-4
iptables - 1.4.10-2
iptables-mod-conntrack - 1.4.10-2
iptables-mod-nat - 1.4.10-2
kernel - 2.6.38.2-1
kmod-bgmac - 2.6.38.2-1
kmod-brcm-wl - 2.6.38.2+5.10.56.27.3-1
kmod-crc-ccitt - 2.6.38.2-1
kmod-crc16 - 2.6.38.2-1
kmod-diag - 2.6.38.2-10
kmod-fs-ext4 - 2.6.38.2-1
kmod-fs-mbcache - 2.6.38.2-1
kmod-ipt-conntrack - 2.6.38.2-1
kmod-ipt-core - 2.6.38.2-1
kmod-ipt-nat - 2.6.38.2-1
kmod-ipt-nathelper - 2.6.38.2-1
kmod-nls-base - 2.6.38.2-1
kmod-ppp - 2.6.38.2-1
kmod-pppoe - 2.6.38.2-1
kmod-scsi-core - 2.6.38.2-1
kmod-switch - 2.6.38.2-4
kmod-usb-core - 2.6.38.2-1
kmod-usb-storage - 2.6.38.2-1
kmod-usb-uhci - 2.6.38.2-1
kmod-usb2 - 2.6.38.2-1
libc - 0.9.32-71
libgcc - linaro-71
libip4tc - 1.4.10-2
libuci - 2011-03-27.2-1
libxtables - 1.4.10-2
mtd - 15
nas - 5.10.56.27.3-1
nvram - 9
opkg - 618-1
ppp - 2.4.4-17
ppp-mod-pppoe - 2.4.4-17
uci - 2011-03-27.2-1
wireless-tools - 29-4
wl - 5.10.56.27.3-1
wlc - 5.10.56.27.3-1

Any suggestions?

comment:231 in reply to: ↑ 230 Changed 5 years ago by anonymous

Replying to anonymous:

George, I've pulled the .trx from the URL above (-59) and wifi config fails with 'wl0: Invalid argument' x 2. Wired ethernet seems to work ok.

I have these two messages if issue /sbin/wifi on my test box as well but still can access n16 ap from my notebook. My guess these are caused by wl config utility trying to ioctl something wrong or unsupported.

Also, using the opkg.conf hack above, I can install packages but not kmods as the kernel revision on trunk/brcm47xx seems to be different. This kinda blocks ipv6.

The only I can help you with here - I can build and publish image with extra commonly-used packages or just packages alone.

comment:232 Changed 5 years ago by linvinus

gkashperko, thanks for patches https://dev.openwrt.org/changeset/26949
now wifi, and router it self working stable!
no more kernel oops

comment:233 Changed 5 years ago by anonymous

George - thanks for your quick response. I can confirm that despite these errors wifi does work, but I had to do some manual work with wl. I believe this may be down to the existing configuration being preserved? So my OpenWRT config was slightly different to my Tomato-USB config and something wasn't getting re-initialised correctly? This is just speculation but the wifi did start at some point during my fiddling. Possible areas of investigation are - setting SSID, setting encryption mode? I did find one report which said you have to turn the radio off before setting the SSID? Anyway, somehow I coaxed it to life and it has been stable for over 12 hours now. The IP6 stuff - I decided to pull the source, patched it ok with your patches and I'm currently building the world (r27114) including the IPV6 support - I'll let you know if my custom build is ok.

comment:234 Changed 5 years ago by dylek

Hi i have some problem while i was compiling r27124, can anyone help me ?

fread failure or file "/home/paczki/OPENWRT/RT-N16/trunk.r27124/build_dir/linux-brcm4716/fs_mark" too large
make[5]: * [install] Error 1
make[5]: Leaving directory `/home/paczki/OPENWRT/RT-N16/trunk.r27124/target/linux/brcm4716/image'
make[4]:
* [install] Error 2
make[4]: Leaving directory `/home/paczki/OPENWRT/RT-N16/trunk.r27124/target/linux/brcm4716'
make[3]: * [install] Error 2
make[3]: Leaving directory `/home/paczki/OPENWRT/RT-N16/trunk.r27124/target/linux'
make[2]:
* [target/linux/install] Error 2
make[2]: Leaving directory `/home/paczki/OPENWRT/RT-N16/trunk.r27124'
make[1]: * home/paczki/OPENWRT/RT-N16/trunk.r27124/staging_dir/target-mipsel_uClibc-0.9.32[[BR]/stamp/.target_install] Error 2
make[1]: Leaving directory `/home/paczki/OPENWRT/RT-N16/trunk.r27124'

comment:235 follow-up: Changed 5 years ago by anonymous

GKashperko
I've installed latest openwrt software, it has 2.6.38.2 integrated. Problem is I want to use modules such as Tee, but the kmod packages that I can obtain from the brcm47xx folder will not go along with your kernel.
So please may you upload the .config file you used to compile your kernel, so I can compile kernel and have packages of my own or, can you upload packages aswell, which I and other people can use to perfect the device :)
Spasibo!

comment:236 in reply to: ↑ 235 Changed 5 years ago by gkashperko

Replying to anonymous:

GKashperko
I've installed latest openwrt software, it has 2.6.38.2 integrated. Problem is I want to use modules such as Tee, but the kmod packages that I can obtain from the brcm47xx folder will not go along with your kernel.
So please may you upload the .config file you used to compile your kernel, so I can compile kernel and have packages of my own or, can you upload packages aswell, which I and other people can use to perfect the device :)
Spasibo!

There is default Openwrt config in each patchset release - see brcm4716_defconfig in trunk root.

comment:237 Changed 5 years ago by anonymous

I have built a large number of packages for brcm4716 using r27114 with the -59 RT-N16 patchset, does anyone have a preferred location for me to upload them?

ls bin/brcm4716/packages/ | wc -l
2307

comment:238 follow-up: Changed 5 years ago by linvinus

GKashperko, i'm trying to build openwrt with linux-2.6.38.2 kernel (with patches from http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-06T16-59/), but i can't build broadcom-wl

openwrt_2.6.38/openwrt/build_dir/linux-brcm4716/broadcom-wl-5.10.56.27.3/driver/bcmsrom.c:2140: error: implicit declaration of function ‘ssb_get_fallback_sprom’

with kernel 2.6.37 was patch 977-ssb_export_fallback_sprom.patch which export this function.
with 2.6.38 it gone.

any suggestions?

comment:239 in reply to: ↑ 238 Changed 5 years ago by anonymous

Replying to linvinus:

GKashperko, i'm trying to build openwrt with linux-2.6.38.2 kernel (with patches from http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-06T16-59/), but i can't build broadcom-wl

openwrt_2.6.38/openwrt/build_dir/linux-brcm4716/broadcom-wl-5.10.56.27.3/driver/bcmsrom.c:2140: error: implicit declaration of function ‘ssb_get_fallback_sprom’

with kernel 2.6.37 was patch 977-ssb_export_fallback_sprom.patch which export this function.
with 2.6.38 it gone.

any suggestions?

i'm sorry, my fault, it works!

comment:240 follow-up: Changed 5 years ago by gregd72002

Ok, I tried the latest patches along with r27143 and the router behaves strangely... it boots the first time after firmware flash (I got everything working after the first boot)... but does not boot second time and it get into awaiting TFTP upload straight away.

I did test it two times (I flashed two times the firmware and took dmesg after each of them):

Connected to 192.168.1.1.
Escape character is '^]'.
 === IMPORTANT ============================
  Use 'passwd' to set your login password
  this will disable telnet and enable SSH
 ------------------------------------------


BusyBox v1.18.4 (2011-06-08 19:46:32 BST) built-in shell (ash)
Enter 'help' for a list of built-in commands.

  _______                     ________        __
 |       |.-----.-----.-----.|  |  |  |.----.|  |_
 |   -   ||  _  |  -__|     ||  |  |  ||   _||   _|
 |_______||   __|_____|__|__||________||__|  |____|
          |__| W I R E L E S S   F R E E D O M
 ATTITUDE ADJUSTMENT (bleeding edge, r27143) ----------
  * 1/4 oz Vodka      Pour all ingredients into mixing
  * 1/4 oz Gin        tin with ice, strain into glass.
  * 1/4 oz Amaretto
  * 1/4 oz Triple sec
  * 1/4 oz Peach schnapps
  * 1/4 oz Sour mix
  * 1 splash Cranberry juice
 -----------------------------------------------------
root@OpenWrt:/# dmesg
Linux version 2.6.38.2 (kiwi@debian) (gcc version 4.5.4 20110505 (prerelease) (Linaro GCC 4.5-2011.05-0) ) #2 Wed Jun 8 20:29:27 BST 2011
assume 128MB RAM
CPU revision is: 00019740 (MIPS 74Kc)
ssb: Core 0 found: ChipCommon (cc 0x800, rev 0x1F, vendor 0x04BF)
ssb: Core 1 found: IEEE 802.11 (cc 0x812, rev 0x11, vendor 0x04BF)
ssb: Core 2 found: Gigabit MAC (cc 0x82D, rev 0x00, vendor 0x04BF)
ssb: Core 3 found: MIPS 74K (cc 0x82C, rev 0x01, vendor 0x04A7)
ssb: Core 4 found: USB 2.0 Host (cc 0x819, rev 0x04, vendor 0x04BF)
ssb: Core 5 found: PCI-E (cc 0x820, rev 0x0E, vendor 0x04BF)
ssb: Core 6 found: DDR1/2 memory controller (cc 0x82E, rev 0x01, vendor 0x04BF)
ssb: Core 7 found: Internal Memory (cc 0x80E, rev 0x07, vendor 0x04BF)
ssb: Core 8 found: I2S (cc 0x834, rev 0x00, vendor 0x04BF)
ssb: Found END of erom after 9 cores
ssb: chipcommon status is 0x800
ssb: Found rev 5 PMU (capabilities 0x11242C05)
ssb: ssb_pmu_resources_init: Resource count 7
ssb: Initializing MIPS core...
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: set_irq: core 0x0812, irq 3 => 3
ssb: set_irq: core 0x082d, irq 4 => 4
ssb: set_irq: core 0x0819, irq 5 => 5
ssb: after irq reconfiguration
ssb: core 0x0800, irq : 2(S)* 3  4  5  6  D  I 
ssb: core 0x0812, irq : 2(S)  3* 4  5  6  D  I 
ssb: core 0x082d, irq : 2(S)  3  4* 5  6  D  I 
ssb: core 0x082c, irq : 2(S)  3  4  5  6  D  I*
ssb: core 0x0819, irq : 2(S)  3  4  5* 6  D  I 
ssb: core 0x0820, irq : 2(S)  3  4  5  6* D  I 
ssb: core 0x082e, irq : 2(S)* 3  4  5  6  D  I 
ssb: core 0x080e, irq : 2(S)  3  4  5  6  D  I*
ssb: core 0x0834, irq : 2(S)* 3  4  5  6  D  I 
ssb: Sonics Silicon Backplane (AI type) found at address 0x18000000
Serial init done.
Determined physical RAM map:
 memory: 08000000 @ 00000000 (usable)
Initrd not found or empty - disabling initrd
Zone PFN ranges:
  Normal   0x00000000 -> 0x00008000
Movable zone start PFN for each node
early_node_map[1] active PFN ranges
    0: 0x00000000 -> 0x00008000
On node 0 totalpages: 32768
free_area_init_node: node 0, pgdat 80249950, node_mem_map 81000000
  Normal zone: 256 pages used for memmap
  Normal zone: 0 pages reserved
  Normal zone: 32512 pages, LIFO batch:7
pcpu-alloc: s0 r0 d32768 u32768 alloc=1*32768
pcpu-alloc: [0] 0 
Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 32512
Kernel command line:  root=/dev/mtdblock2 rootfstype=squashfs,jffs2 noinitrd console=ttyS0,115200
PID hash table entries: 512 (order: -1, 2048 bytes)
Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
Primary instruction cache 32kB, VIPT, 4-way, linesize 32 bytes.
Primary data cache 32kB, 4-way, VIPT, cache aliases, linesize 32 bytes
Writing ErrCtl register=00000000
Readback ErrCtl register=00000000
Memory: 127248k/131072k available (2021k kernel code, 3824k reserved, 323k data, 156k init, 0k highmem)
NR_IRQS:128
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m1=2, fc=960, clock=480
console [ttyS0] enabled
Calibrating delay loop... 239.10 BogoMIPS (lpj=478208)
pid_max: default: 32768 minimum: 301
Mount-cache hash table entries: 512
NET: Registered protocol family 16
bio: create slab <bio-0> at 0
Switching to clocksource MIPS
Switched to NOHz mode on CPU #0
Disabling SSB PCI-E host due to code brokeness
NET: Registered protocol family 2
IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
TCP established hash table entries: 4096 (order: 3, 32768 bytes)
TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
TCP: Hash tables configured (established 4096 bind 4096)
TCP reno registered
UDP hash table entries: 256 (order: 0, 4096 bytes)
UDP-Lite hash table entries: 256 (order: 0, 4096 bytes)
NET: Registered protocol family 1
PCI: CLS 0 bytes, default 32
squashfs: version 4.0 (2009/01/31) Phillip Lougher
JFFS2 version 2.2 (NAND) (SUMMARY) (LZMA) (RTIME) (CMODE_PRIORITY) (c) 2001-2006 Red Hat, Inc.
msgmni has been set to 248
io scheduler noop registered
io scheduler deadline registered (default)
Serial: 8250/16550 driver, 2 ports, IRQ sharing enabled
serial8250: ttyS0 at MMIO 0xb8000300 (irq = 2) is a U6_16550A
serial8250.0: ttyS0 at MMIO 0xb8000300 (irq = 2) is a U6_16550A
flash init: 0x1c000000 0x02000000
Physically mapped flash: Found 1 x16 devices at 0x0 in 8-bit bank. Manufacturer ID 0x0000c2 Chip ID 0x00007e
Amd/Fujitsu Extended Query Table at 0x0040
  Amd/Fujitsu Extended Query version 1.3.
number of CFI chips: 1
Flash device: 0x2000000 at 0x1fc00000
bootloader size: 262144
Looking for dual image
TRX offset : 0
Updating TRX offsets and length:
old trx = [0x0000001c, 0x00000968, 0x000cfc00], len=0x00291000 crc32=0xf42394d2
new trx = [0x0000001c, 0x00000968, 0x000cfc00], len=0x000cfc00 crc32=0xeec3bfc8
Done
Creating 4 MTD partitions on "Physically mapped flash":
0x000000000000-0x000000040000 : "cfe"
0x000000040000-0x000001fe0000 : "linux"
0x00000010fc00-0x000001fe0000 : "rootfs"
mtd: partition "rootfs" must either start or end on erase block boundary or be smaller than an erase block -- forcing read-only
mtd: partition "rootfs" set to be root filesystem
mtd: partition "rootfs_data" created automatically, ofs=2C0000, len=1D20000 
0x0000002c0000-0x000001fe0000 : "rootfs_data"
0x000001fe0000-0x000002000000 : "nvram"
BCM47xx Watchdog Timer enabled (30 seconds, nowayout)
TCP westwood registered
NET: Registered protocol family 17
802.1Q VLAN Support v1.8 Ben Greear <greearb@candelatech.com>
All bugs added by David S. Miller <davem@redhat.com>
VFS: Mounted root (squashfs filesystem) readonly on device 31:2.
Freeing unused kernel memory: 156k freed
diag: Detected 'ASUS RT-N16'
roboswitch: Probing device eth0: No such device
roboswitch: Probing device eth1: No such device
roboswitch: Probing device eth2: No such device
roboswitch: Probing device eth3: No such device
bgmac ssb0:1: eth0: Broadcom SSB GMAC ethernet driver (ext. Robo Switch) e0:cb:4e:43:9e:94
SCSI subsystem initialized
roboswitch: Probing device eth0: found a 53115! It's a 5350.
usbcore: registered new interface driver usbfs
usbcore: registered new interface driver hub
usbcore: registered new device driver usb
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
wl: module license 'unspecified' taints kernel.
Disabling lock debugging due to kernel taint
wl0: Broadcom BCM4329 802.11 Wireless Controller 5.10.56.27
PPP generic driver version 2.4.2
device eth0.0 entered promiscuous mode
device eth0 entered promiscuous mode
ip_tables: (C) 2000-2006 Netfilter Core Team
br-lan: port 1(eth0.0) entering forwarding state
br-lan: port 1(eth0.0) entering forwarding state
NET: Registered protocol family 24
ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m1=2, fc=960, clock=480
USB20H mdio control register : 0x8000000c
ehci_hcd ssb0:2: SSB EHCI Controller
ehci_hcd ssb0:2: new USB bus registered, assigned bus number 1
ehci_hcd ssb0:2: irq 5, io mem 0x18004000
ehci_hcd ssb0:2: USB 0.0 started, EHCI 1.00
hub 1-0:1.0: USB hub found
hub 1-0:1.0: 2 ports detected
nf_conntrack version 0.5.0 (1990 buckets, 7960 max)
uhci_hcd: USB Universal Host Controller Interface driver
Initializing USB Mass Storage driver...
usbcore: registered new interface driver usb-storage
USB Mass Storage support registered.
jffs2_scan_eraseblock(): End of filesystem marker found at 0x0
jffs2_build_filesystem(): unlocking the mtd device... done.
jffs2_build_filesystem(): erasing all blocks after the end marker... done.
JFFS2 notice: (1120) jffs2_build_xattr_subsystem: complete building xattr subsystem, 0 of xdatum (0 unchecked, 0 orphan) and 0 of xref (0 dead, 0 orphan) found.
Node CRC 717097b2 != calculated CRC a8450937 for node at 01ce00f0
Node CRC ffc14c6e != calculated CRC 0f4b6123 for node at 01ce0188
root@OpenWrt:/# reboot
root@OpenWrt:/# Connection closed by foreign host.



== DEAD END AFTER REBOOT - HAD TO FLASH THE FIRMWARE AGAIN ==


-----
Connected to 192.168.1.1.
Escape character is '^]'.
 === IMPORTANT ============================
  Use 'passwd' to set your login password
  this will disable telnet and enable SSH
 ------------------------------------------


BusyBox v1.18.4 (2011-06-08 19:46:32 BST) built-in shell (ash)
Enter 'help' for a list of built-in commands.

  _______                     ________        __
 |       |.-----.-----.-----.|  |  |  |.----.|  |_
 |   -   ||  _  |  -__|     ||  |  |  ||   _||   _|
 |_______||   __|_____|__|__||________||__|  |____|
          |__| W I R E L E S S   F R E E D O M
 ATTITUDE ADJUSTMENT (bleeding edge, r27143) ----------
  * 1/4 oz Vodka      Pour all ingredients into mixing
  * 1/4 oz Gin        tin with ice, strain into glass.
  * 1/4 oz Amaretto
  * 1/4 oz Triple sec
  * 1/4 oz Peach schnapps
  * 1/4 oz Sour mix
  * 1 splash Cranberry juice
 -----------------------------------------------------
root@OpenWrt:/# dmesg
Linux version 2.6.38.2 (kiwi@debian) (gcc version 4.5.4 20110505 (prerelease) (Linaro GCC 4.5-2011.05-0) ) #2 Wed Jun 8 20:29:27 BST 2011
assume 128MB RAM
CPU revision is: 00019740 (MIPS 74Kc)
ssb: Core 0 found: ChipCommon (cc 0x800, rev 0x1F, vendor 0x04BF)
ssb: Core 1 found: IEEE 802.11 (cc 0x812, rev 0x11, vendor 0x04BF)
ssb: Core 2 found: Gigabit MAC (cc 0x82D, rev 0x00, vendor 0x04BF)
ssb: Core 3 found: MIPS 74K (cc 0x82C, rev 0x01, vendor 0x04A7)
ssb: Core 4 found: USB 2.0 Host (cc 0x819, rev 0x04, vendor 0x04BF)
ssb: Core 5 found: PCI-E (cc 0x820, rev 0x0E, vendor 0x04BF)
ssb: Core 6 found: DDR1/2 memory controller (cc 0x82E, rev 0x01, vendor 0x04BF)
ssb: Core 7 found: Internal Memory (cc 0x80E, rev 0x07, vendor 0x04BF)
ssb: Core 8 found: I2S (cc 0x834, rev 0x00, vendor 0x04BF)
ssb: Found END of erom after 9 cores
ssb: chipcommon status is 0x800
ssb: Found rev 5 PMU (capabilities 0x11242C05)
ssb: ssb_pmu_resources_init: Resource count 7
ssb: Initializing MIPS core...
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: set_irq: core 0x0812, irq 3 => 3
ssb: set_irq: core 0x082d, irq 4 => 4
ssb: set_irq: core 0x0819, irq 5 => 5
ssb: after irq reconfiguration
ssb: core 0x0800, irq : 2(S)* 3  4  5  6  D  I 
ssb: core 0x0812, irq : 2(S)  3* 4  5  6  D  I 
ssb: core 0x082d, irq : 2(S)  3  4* 5  6  D  I 
ssb: core 0x082c, irq : 2(S)  3  4  5  6  D  I*
ssb: core 0x0819, irq : 2(S)  3  4  5* 6  D  I 
ssb: core 0x0820, irq : 2(S)  3  4  5  6* D  I 
ssb: core 0x082e, irq : 2(S)* 3  4  5  6  D  I 
ssb: core 0x080e, irq : 2(S)  3  4  5  6  D  I*
ssb: core 0x0834, irq : 2(S)* 3  4  5  6  D  I 
ssb: Sonics Silicon Backplane (AI type) found at address 0x18000000
Serial init done.
Determined physical RAM map:
 memory: 08000000 @ 00000000 (usable)
Initrd not found or empty - disabling initrd
Zone PFN ranges:
  Normal   0x00000000 -> 0x00008000
Movable zone start PFN for each node
early_node_map[1] active PFN ranges
    0: 0x00000000 -> 0x00008000
On node 0 totalpages: 32768
free_area_init_node: node 0, pgdat 80249950, node_mem_map 81000000
  Normal zone: 256 pages used for memmap
  Normal zone: 0 pages reserved
  Normal zone: 32512 pages, LIFO batch:7
pcpu-alloc: s0 r0 d32768 u32768 alloc=1*32768
pcpu-alloc: [0] 0 
Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 32512
Kernel command line:  root=/dev/mtdblock2 rootfstype=squashfs,jffs2 noinitrd console=ttyS0,115200
PID hash table entries: 512 (order: -1, 2048 bytes)
Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
Primary instruction cache 32kB, VIPT, 4-way, linesize 32 bytes.
Primary data cache 32kB, 4-way, VIPT, cache aliases, linesize 32 bytes
Writing ErrCtl register=00000000
Readback ErrCtl register=00000000
Memory: 127248k/131072k available (2021k kernel code, 3824k reserved, 323k data, 156k init, 0k highmem)
NR_IRQS:128
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m1=2, fc=960, clock=480
console [ttyS0] enabled
Calibrating delay loop... 239.10 BogoMIPS (lpj=478208)
pid_max: default: 32768 minimum: 301
Mount-cache hash table entries: 512
NET: Registered protocol family 16
bio: create slab <bio-0> at 0
Switching to clocksource MIPS
Switched to NOHz mode on CPU #0
Disabling SSB PCI-E host due to code brokeness
NET: Registered protocol family 2
IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
TCP established hash table entries: 4096 (order: 3, 32768 bytes)
TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
TCP: Hash tables configured (established 4096 bind 4096)
TCP reno registered
UDP hash table entries: 256 (order: 0, 4096 bytes)
UDP-Lite hash table entries: 256 (order: 0, 4096 bytes)
NET: Registered protocol family 1
PCI: CLS 0 bytes, default 32
squashfs: version 4.0 (2009/01/31) Phillip Lougher
JFFS2 version 2.2 (NAND) (SUMMARY) (LZMA) (RTIME) (CMODE_PRIORITY) (c) 2001-2006 Red Hat, Inc.
msgmni has been set to 248
io scheduler noop registered
io scheduler deadline registered (default)
Serial: 8250/16550 driver, 2 ports, IRQ sharing enabled
serial8250: ttyS0 at MMIO 0xb8000300 (irq = 2) is a U6_16550A
serial8250.0: ttyS0 at MMIO 0xb8000300 (irq = 2) is a U6_16550A
flash init: 0x1c000000 0x02000000
Physically mapped flash: Found 1 x16 devices at 0x0 in 8-bit bank. Manufacturer ID 0x0000c2 Chip ID 0x00007e
Amd/Fujitsu Extended Query Table at 0x0040
  Amd/Fujitsu Extended Query version 1.3.
number of CFI chips: 1
Flash device: 0x2000000 at 0x1fc00000
bootloader size: 262144
Looking for dual image
TRX offset : 0
Updating TRX offsets and length:
old trx = [0x0000001c, 0x00000968, 0x000cfc00], len=0x00291000 crc32=0xf42394d2
new trx = [0x0000001c, 0x00000968, 0x000cfc00], len=0x000cfc00 crc32=0xeec3bfc8
Done
Creating 4 MTD partitions on "Physically mapped flash":
0x000000000000-0x000000040000 : "cfe"
0x000000040000-0x000001fe0000 : "linux"
0x00000010fc00-0x000001fe0000 : "rootfs"
mtd: partition "rootfs" must either start or end on erase block boundary or be smaller than an erase block -- forcing read-only
mtd: partition "rootfs" set to be root filesystem
mtd: partition "rootfs_data" created automatically, ofs=2C0000, len=1D20000 
0x0000002c0000-0x000001fe0000 : "rootfs_data"
0x000001fe0000-0x000002000000 : "nvram"
BCM47xx Watchdog Timer enabled (30 seconds, nowayout)
TCP westwood registered
NET: Registered protocol family 17
802.1Q VLAN Support v1.8 Ben Greear <greearb@candelatech.com>
All bugs added by David S. Miller <davem@redhat.com>
VFS: Mounted root (squashfs filesystem) readonly on device 31:2.
Freeing unused kernel memory: 156k freed
diag: Detected 'ASUS RT-N16'
roboswitch: Probing device eth0: No such device
roboswitch: Probing device eth1: No such device
roboswitch: Probing device eth2: No such device
roboswitch: Probing device eth3: No such device
bgmac ssb0:1: eth0: Broadcom SSB GMAC ethernet driver (ext. Robo Switch) e0:cb:4e:43:9e:94
SCSI subsystem initialized
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
roboswitch: Probing device eth0: found a 53115! It's a 5350.
usbcore: registered new interface driver usbfs
usbcore: registered new interface driver hub
usbcore: registered new device driver usb
device eth0.0 entered promiscuous mode
device eth0 entered promiscuous mode
br-lan: port 1(eth0.0) entering forwarding state
br-lan: port 1(eth0.0) entering forwarding state
wl: module license 'unspecified' taints kernel.
Disabling lock debugging due to kernel taint
wl0: Broadcom BCM4329 802.11 Wireless Controller 5.10.56.27
PPP generic driver version 2.4.2
ip_tables: (C) 2000-2006 Netfilter Core Team
NET: Registered protocol family 24
ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m1=2, fc=960, clock=480
USB20H mdio control register : 0x8000000c
ehci_hcd ssb0:2: SSB EHCI Controller
ehci_hcd ssb0:2: new USB bus registered, assigned bus number 1
ehci_hcd ssb0:2: irq 5, io mem 0x18004000
ehci_hcd ssb0:2: USB 0.0 started, EHCI 1.00
hub 1-0:1.0: USB hub found
hub 1-0:1.0: 2 ports detected
nf_conntrack version 0.5.0 (1990 buckets, 7960 max)
uhci_hcd: USB Universal Host Controller Interface driver
Initializing USB Mass Storage driver...
usbcore: registered new interface driver usb-storage
USB Mass Storage support registered.
jffs2_scan_eraseblock(): End of filesystem marker found at 0x0
jffs2_build_filesystem(): unlocking the mtd device... done.
jffs2_build_filesystem(): erasing all blocks after the end marker... done.
JFFS2 notice: (1092) jffs2_build_xattr_subsystem: complete building xattr subsystem, 0 of xdatum (0 unchecked, 0 orphan) and 0 of xref (0 dead, 0 orphan) found.
root@OpenWrt:/# reboot

SAME STORY AGAIN - ROUTER DOES NOT INITIATE AFTER REBOOT

comment:241 Changed 5 years ago by shinton@…

@gregd72002 I am experiencing the exact same issue, only started happening within the past week or so. Some update to the openwrt svn must be causing this. George any ideas??

comment:242 in reply to: ↑ 240 ; follow-up: Changed 5 years ago by anonymous

Replying to gregd72002:

Ok, I tried the latest patches along with r27143 and the router behaves strangely... it boots the first time after firmware flash (I got everything working after the first boot)... but does not boot second time and it get into awaiting TFTP upload straight away.

Replying to shinton@…:

@gregd72002 I am experiencing the exact same issue, only started happening within the past week or so. Some update to the openwrt svn must be causing this. George any ideas??

This happens since r27118, r27117 works as intended. Will try to find out why tomorrow

comment:243 in reply to: ↑ 242 ; follow-up: Changed 5 years ago by gkashperko

Replying to anonymous:

Replying to gregd72002:

Ok, I tried the latest patches along with r27143 and the router behaves strangely... it boots the first time after firmware flash (I got everything working after the first boot)... but does not boot second time and it get into awaiting TFTP upload straight away.

Replying to shinton@…:

@gregd72002 I am experiencing the exact same issue, only started happening within the past week or so. Some update to the openwrt svn must be causing this. George any ideas??

This happens since r27118, r27117 works as intended. Will try to find out why tomorrow

Seems compiling with linaro-4.5-2011.05 leads to some code behaving wrong which in turn leads to mtd corruption. Image built with gcc-4.5.2 works as intended.

comment:244 in reply to: ↑ 243 Changed 5 years ago by gkashperko

Replying to gkashperko:

Seems compiling with linaro-4.5-2011.05 leads to some code behaving wrong which in turn leads to mtd corruption. Image built with gcc-4.5.2 works as intended.

Image built with linaro-4.5-2011.04 seems to work fine too.

comment:245 follow-up: Changed 5 years ago by jonimus

I was able to use the patches provided here and some other config changes to get it semi booting on my WRT320N, but the kernel failing the do_map_probe to find the mtd blocks and thus cannot mount the rootfs. Any suggestions on where to look at fixing that?

comment:246 in reply to: ↑ 245 ; follow-up: Changed 5 years ago by gkashperko

Replying to jonimus:

I was able to use the patches provided here and some other config changes to get it semi booting on my WRT320N, but the kernel failing the do_map_probe to find the mtd blocks and thus cannot mount the rootfs. Any suggestions on where to look at fixing that?

Could it be related to https://dev.openwrt.org/ticket/7246 by any chance ?

comment:247 in reply to: ↑ 246 Changed 5 years ago by jonimus

Replying to gkashperko:

Replying to jonimus:

I was able to use the patches provided here and some other config changes to get it semi booting on my WRT320N, but the kernel failing the do_map_probe to find the mtd blocks and thus cannot mount the rootfs. Any suggestions on where to look at fixing that?

Could it be related to https://dev.openwrt.org/ticket/7246 by any chance ?

That is exactly it, thanks I will try that.

comment:248 Changed 5 years ago by anonymous

@George

Do the latest patches here: http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-15T01-24/ fix the recovery mode on reboot issue? The reason I ask is that I haven't seen anything posted since then yet and am not sure if you are still testing or what the case may be. Thanks!

comment:249 follow-up: Changed 5 years ago by anonymous

Sorry I gave the wrong link, I meant here: http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-13T02-46/

ps the trac system thinks that this post is spam so I am just gonna type gibberish below so that it accepts this comment. aggjdgjsgjsdjgsdjgjaskgjaaksgka dgdkgjak kagkja kjadkgj ksjd kjaskjkajksgs ajg jka gjk jkgdjkgk jksjd gkjsd kjs jdkj kjsdgjksjakjsfkjaskj j sgjkgskj kja j jakj ajk kj sgjsdjgsdjgjaskgjaaksgka dgdkgjak kagkja kjadkgj ksjd kjaskjkajksgs ajg jka gjk jkgdjkgk jksjd gkjsd kjs jdkj kjsdgjksjakjsfkjaskj j sgjkgskj kja j jakj ajk kj sgjsdjgsdjgjaskgjaaksgka dgdkgjak kagkja kjadkgj ksjd kjaskjkajksgs ajg jka gjk jkgdjkgk jksjd gkjsd kjs jdkj kjsdgjksjakjsfkjaskj j sgjkgskj kja j jakj ajk kj

comment:250 in reply to: ↑ 249 ; follow-up: Changed 5 years ago by anonymous

Replying to anonymous:

Sorry I gave the wrong link, I meant here: http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-13T02-46/

These introduces small fix for BUG support. Also if you use brcm_defconfig supplied with patches as your configuration base then gcc-4.5.2 will be selected by default.

Recovery mode on reboot seems is caused by gcc-4.5.4 (the linaro-4.5-2011.05 is based on) optimisations of unsigned char comparisons leading to misscompiling cfi_cmdset_0002.c which in turn corrupt the mtd.

To address this compile trunk with gcc-4.5.2

comment:251 in reply to: ↑ 250 Changed 5 years ago by gkashperko

Replying to anonymous:

Replying to anonymous:

Sorry I gave the wrong link, I meant here: http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-13T02-46/

These introduces small fix for BUG support. Also if you use brcm_defconfig supplied with patches as your configuration base then gcc-4.5.2 will be selected by default.

Recovery mode on reboot seems is caused by gcc-4.5.4 (the linaro-4.5-2011.05 is based on) optimisations of unsigned char comparisons leading to misscompiling cfi_cmdset_0002.c which in turn corrupt the mtd.

To address this compile trunk with gcc-4.5.2

Good news - I've finally tracked the bug out. My original suspect on linaro misscompiling things was plain wrong. Clearly source of the bug is mine :)
Will release fixed patches this evening after done sorting few things out.

comment:252 Changed 5 years ago by gkashperko

Fighting with N16 entering emergency mode right after firstboot I've run into troubles. While I've found and fixed one nasty bug of mine I still suffer from the box entering emergency recover.
The bug appeared after Openwrt moved to linaro-4.5-2011.05. Reverting patch described in http://www.mail-archive.com/gcc-patches@gcc.gnu.org/msg00784.html makes image work but I doubght that mtd corruption is caused by this gcc change. Whole difference between image working or failing after firstboot lies in 1ne assembly instruction in the whole kernel, in the cfi_cmdset_0002.c
After analysing cfi_cmdset_0002.c assembly source around the optimisation performed by the gcc with or without patch mentioned I don't see any reason for code missexecution with given input values I have for N16. Seems the whole difference between working/failing image lay in additional 32-bit instruction in cfi_cmdset_0002.o as substituting unoptimised andi $2,$2,0x00ff with nop still keep the image from failing.
Atm I'm reading through the code involved around drivers/mtd but actually I'm off the ideas how to fix this, where is the source for the bug and why the difference in single nop in cfi_cmdset_0002.o have such an impact.
Would be much appreciated for any help with this.

comment:253 follow-up: Changed 5 years ago by iiordanovv@…

This could be totally off the mark, but could it be that this $2,$2,0x00ff gets overwritten by a buffer overflow during execution with something else instead?

comment:254 in reply to: ↑ 253 Changed 5 years ago by anonymous

Replying to iiordanovv@…:

This could be totally off the mark, but could it be that this $2,$2,0x00ff gets overwritten by a buffer overflow during execution with something else instead?

No. The weird thing is, that $2,$2,0x00ff doesn't matters at all. Declaring
int foo(void)
{

sleep(5);

}
EXPORT_SYMBOL_GPL(foo);
within cfi_cmdset_0002.c or just making some message text longer makes image working:
printk(KERN_WARNING "Failed to allocate memory for MTD deviceXXXX\n");
instead of
printk(KERN_WARNING "Failed to allocate memory for MTD device\n");
still make the trick.

comment:255 Changed 5 years ago by gkashperko

Cleaned out everything ssb-related. The only I left from ssb setup code:

*(u32 *)KSEG1ADDR(0x18003028) = 1 << 31;
cp0_compare_irq = 7;

And box still enters emergency recover.

comment:256 Changed 5 years ago by gkashperko

Bug was caused by cfi_cmdset_0002 starting transmit data too early, before chip was done with processing Write Buffer Load command.
Added patch to the attachments as it is not ssb/ai specific.
Also the issue could be related to #9252
Fixed patchset is at http://www.znau.edu.ua/temp/asus-rt-n16/2011-06-18T05-55

comment:257 Changed 5 years ago by anonymous

SUCCESS! r27220 with 2011-06-19 patches. Two minor issues, neither related to George's patches - perl has issues under Ubuntu 11.04 host due to missing "-lm -lrt" flags in libs (can be resolved with a hack to Makefile.SH) and enabling something IPV6 related in the OpenWRT config switches ether bridge to [m]odule in linux config, this kills the router boot. So you need to be very careful what menuconfig options are changed and check the linux .config to ensure that CONFIG_BRIDGE=y not [m] before deploying. Wifi works with the two known error messages and the router seems stable.

comment:258 Changed 5 years ago by anonymous

Ah ha! Kernel bump to 2.6.38+ has introduced a bad dependency between ethernet bridge and IPV6. Fix is in trunk r27237+

comment:259 Changed 5 years ago by shinton@…

For those experiencing an error while compiling openwrt using 2.6.38 when applying the "822-usb_serial_option_add_more_devices.patch" I've updated the patch to correct this. I also created a patch to fix the patch for those wanting a quick and easy fix. See links below:

patch: http://dl.dropbox.com/u/29682150/822-usb_serial_option_add_more_devices.patch
patch to fix the patch: http://dl.dropbox.com/u/29682150/003-openwrt4716-TARGET_brcm4716-usb_serial_fix.patch

Not sure if this is brcm4716 specific, might be an idea to commit this to trunk though.

comment:260 Changed 5 years ago by shinton@…

I've realized that the above issue might be DebWrt specific, please disregard as this most likely doesn't affect you.

comment:261 Changed 5 years ago by jow

Status update:

As of r27760, the brcm47xx target (when switched to Linux 3.0 final) will boot on the RT-N16 using the newly added BCMA (AMBA) bus support. The proprietary broadcom-wl driver works as well, the generic mtd read back fix has been incorperated.

There is *no* ethernet support yet as the legality of the bgmac driver is not yet confirmed - the driver was built from code declared as trade secret - this is currently discussed on the LKML. See http://lkml.org/lkml/2011/5/7/175 .

There is no diag support yet, this one will follow soon.

comment:262 Changed 5 years ago by slydog@…

Does this yield any advantage over using George's patches?

comment:263 Changed 5 years ago by jow

Currently? No.

comment:264 in reply to: ↑ 229 ; follow-up: Changed 5 years ago by cromo

Replying to gkashperko:

.trx image built from scratch (r27114 changeset) boots just fine (available at the ftp above). Have no wl crashes with default wifi setup. If you manage to get stock .trx crash please let me know.

Hi George, sorry for a (very) late reply. I have just tried your stock r27724 and although I thought at first that everything was fine, it turnes out this bug is still there. However now it seems to me that it takes some more time between starting the DL and a reboot - it used to be up about couple of secs, now it's couple of minutes as far as my observation goes.

comment:265 in reply to: ↑ 264 Changed 5 years ago by cromo

And as previously, if I limit the DL speed, instead of reboot it will kill the ethernet connection and flood dmesg with that (52413 port is used by transmission):

TCP: Possible SYN flooding on port 52413. Sending cookies.
TCP: Possible SYN flooding on port 52413. Sending cookies.
TCP: Possible SYN flooding on port 52413. Sending cookies.
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
TCP: Possible SYN flooding on port 52413. Sending cookies.
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
TCP: Possible SYN flooding on port 52413. Sending cookies.
TCP: Possible SYN flooding on port 52413. Sending cookies.
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
net_ratelimit: 2 callbacks suppressed
TCP: Possible SYN flooding on port 52413. Sending cookies.
TCP: Possible SYN flooding on port 52413. Sending cookies.
TCP: Possible SYN flooding on port 52413. Sending cookies.
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
ssb: ssb_pmu5_clock: p1=1, p2=1, ndiv=48(0x30), m3=8, fc=960, clock=120
br-lan: port 1(eth0.0) entering forwarding state
br-lan: port 1(eth0.0) entering forwarding state

comment:266 Changed 5 years ago by anonymous

@jow/George Any update on BGMAC support for latest developments related to linux version 3.0?

comment:267 in reply to: ↑ 229 Changed 5 years ago by cromo <cromo@…>

Replying to gkashperko:

.trx image built from scratch (r27114 changeset) boots just fine (available at the ftp above). Have no wl crashes with default wifi setup. If you manage to get stock .trx crash please let me know.

I have now switched to debwrt and transmission is stable there - no errors at all in dmesg.
Yay! \o/

comment:268 follow-up: Changed 5 years ago by cromo <cromo@…>

Ignore what I just said - the bug is still there. Damn...

comment:269 Changed 5 years ago by anonymous

Latest patchset fails to build on current build as follows:

Applying patch platform/031-ssb-add-callback-for-sprom.patch
patching file arch/mips/bcm63xx/boards/board_bcm963xx.c
patching file drivers/ssb/pci.c
Hunk #1 FAILED at 662.
Hunk #2 FAILED at 707.
2 out of 2 hunks FAILED -- rejects in file drivers/ssb/pci.c
patching file drivers/ssb/sprom.c
Hunk #1 FAILED at 17.
Hunk #2 FAILED at 145.
Hunk #3 FAILED at 163.
3 out of 3 hunks FAILED -- rejects in file drivers/ssb/sprom.c
patching file drivers/ssb/ssb_private.h
Hunk #1 FAILED at 171.
1 out of 1 hunk FAILED -- rejects in file drivers/ssb/ssb_private.h
patching file include/linux/ssb/ssb.h
Hunk #1 FAILED at 404.
1 out of 1 hunk FAILED -- rejects in file include/linux/ssb/ssb.h
Patch platform/031-ssb-add-callback-for-sprom.patch does not apply (enforce with -f)
make[2]: * home/stephen/openwrt/build_dir/linux-brcm4716/linux-2.6.38.8/.quilt_checked Error 1
make[2]: Leaving directory `/home/stephen/openwrt/target/linux/brcm4716'
make[1]:
* [menuconfig] Error 2
make[1]: Leaving directory `/home/stephen/openwrt/target/linux'
make: * [kernel_menuconfig] Error 2

Is there a fix??

comment:270 follow-up: Changed 5 years ago by anonymous

@George/anyone... Could you kindly let me know the latest revision of openwrt when the patchset worked?

comment:271 in reply to: ↑ 270 Changed 5 years ago by anonymous

Replying to anonymous:

@George/anyone... Could you kindly let me know the latest revision of openwrt when the patchset worked?

Updated set for 3.0.1 kernel
http://www.znau.edu.ua/temp/asus-rt-n16/2011-08-08T22-51/

comment:272 Changed 5 years ago by linstarr24@…

Tried a buid last night, got this error:

cp: cannot stat `/home/linstarr/openwrt/build_dir/linux-brcm4716/linux-3.0.1/fs/mbcache.ko': No such file or directory
make[3]: * home/linstarr/openwrt/bin/packages/kmod-fs-ext4_3.0.1-1_brcm4716.ipk Error 1

comment:273 Changed 5 years ago by Jon Hermansen <jon.hermansen@…>

George,

Thank you so much for continuing to provide updates to these patches -- I pulled the latest build from your link.

Will we eventually see support for the RT-N16 in mainline OpenWrt?

comment:274 Changed 5 years ago by anonymous

I needed to use this patch:

https://patchwork.kernel.org/patch/1001022/

to build r27949 (checked out from trunk earlier today).

comment:275 in reply to: ↑ 268 Changed 5 years ago by cromo <cromo@…>

Replying to cromo <cromo@…>:

Ignore what I just said - the bug is still there. Damn...

I got my issue finally solved and it turnes out it was not related to firmware at all.
I described my findings here: http://www.debwrt.net/forum/hardware-group3/other-devices-forum6/support-for-rt-n16-thread4.15/#postid-280.

Hope it prevents other people from going the same path I did (i.e. blaming the firmware).

Thanks for everything!

comment:276 in reply to: ↑ 223 ; follow-up: Changed 5 years ago by linstar@…

Replying to cromo@…:

George, is there any reason why http://wireless.kernel.org/en/users/Drivers/brcm80211
driver wouldn't work for wifi?

@Cromo

I emailed the guys behind brcm80211 and this was the response:

"The brcmfmac driver in theory support mips platforms, both big and little
endian. However, there is basically no testing for those platforms at this
point. We're looking at expanding our test coverage, but don't have that in
place yet."

Maybe you could be a tester?

comment:277 in reply to: ↑ 276 Changed 5 years ago by <cromo@…>

Replying to linstar@…:

Maybe you could be a tester?

I would be happy to, but from what I saw at this driver description is that it does not support SSB based chips ("Does not support older PCI/PCIe chips with SSB backplane"), and AFIK that's the bus BCM4329 in RT-N16 uses?

comment:278 Changed 5 years ago by Dawid Wrobel <cromo@…>

Also it does not support AP mode yet, so testing it for usage with router doesn't make much sense at the moment.

comment:279 Changed 5 years ago by linstar@…

@Cromo
I think only the softmac driver doesn't support ssb yet whereas the fullmac one does (at least that's the way I read it. Anyways just thought I would put it out there.

comment:280 Changed 5 years ago by linstar@…

Actually I misread, you are correct Cromo.

comment:281 follow-up: Changed 5 years ago by azdps

i have up and running on my rt-n16 build 2011-08-09T22-58. everything seems to be working well. the only thing that i can see is that the wireless light on the router doesn't work when wireless is enabled or disabled. great job. this is the first time i've ever used openwrt. george i'm glad you are compiling builds for us to use otherwise i don't think i would have openwrt up and running on my router right now.

comment:282 Changed 5 years ago by anonymous

i've been trying to get dmz setup. i modified the /etc/config/network file but i lose internet connectivity. can someone modify the text below to allow me to put a wireless client with a static ip of 192.168.1.105 into dmz. thank you.

#### VLAN configuration 
config switch eth0
	option enable   1

config switch_vlan eth0_0
	option device   "eth0"
	option vlan     0
	option ports    "1 2 3 4 8*"

config switch_vlan eth0_1
	option device   "eth0"
	option vlan     1
	option ports    "0 8"

#### Loopback configuration
config interface loopback
	option ifname	"lo"
	option proto	static
	option ipaddr	127.0.0.1
	option netmask	255.0.0.0


#### LAN configuration
config interface lan
	option type 	bridge
	option ifname	"eth0.0"
	option proto	static
	option ipaddr	192.168.1.1
	option netmask	255.255.255.0


#### WAN configuration
config interface	wan
	option ifname	"eth0.1"
	option proto	dhcp

comment:283 Changed 5 years ago by thorin

If someone can give me clear instructions on how to get p910nd print server to successfully work on my RT-N16 using build 2011-08-09T22-58 I will pay you $25.00 via Paypal for you time. The USB printer that I'm using is HP P2015. Let me know thank you.

my contact email is arizonadps@…

thank you

comment:284 in reply to: ↑ 281 ; follow-ups: Changed 5 years ago by shinton@…

Replying to azdps:

i have up and running on my rt-n16 build 2011-08-09T22-58. everything seems to be working well. the only thing that i can see is that the wireless light on the router doesn't work when wireless is enabled or disabled. great job. this is the first time i've ever used openwrt. george i'm glad you are compiling builds for us to use otherwise i don't think i would have openwrt up and running on my router right now.

Same problem here, power/wlan led's do not work. Anyone have a fix?

comment:285 in reply to: ↑ 284 Changed 5 years ago by azdps

Replying to shinton@…:

Same problem here, power/wlan led's do not work. Anyone have a fix?

Just to clarify regarding the lights on my RT-N16. The power, lan and wan lights works. The Wireless light does not work.

comment:286 in reply to: ↑ 284 ; follow-up: Changed 5 years ago by gkashperko

Replying to shinton@…:

Same problem here, power/wlan led's do not work. Anyone have a fix?

Power led works for me. As for wlan it seems to be controlled with broadcom-wl, by that part which is supplied precompiled in .o file. Try echo "1" >/proc/diag/led/wlan while wl.ko and wl_glue.ko are unloaded - you should see wlan led turned on. As soon as you insmod both wl and wl_glue it won't work anymore.

comment:287 Changed 5 years ago by voltagex@…

Very interested in using some of the work from this model to get OpenWRT working on the Belkin N1 Vision.

comment:288 Changed 5 years ago by gsedej

Hi! I have RT-N16, openwrt trunk and using George's patches. I have problem with using ffmpeg. It compiles ok (with added codecs) but it segfaults each time when it should start processing (it shows all compiling options and versions but then it says Segmentation fault). I had to manually copy libavfilter.so and libavfilter.so.2 because it's somehow excluded from package. Any idea?
And also can't use my webcam because there is no "videodev.h" so I can't compile anything connected with v4l.

comment:289 follow-up: Changed 5 years ago by anonymous

Please tell me anyone how install openwrt step by step. I want to install in my Asus RT n16, but i don't no how do it. Thanks!

comment:290 in reply to: ↑ 289 Changed 5 years ago by anonymous

Replying to anonymous:

Please tell me anyone how install openwrt step by step. I want to install in my Asus RT n16, but i don't no how do it. Thanks!

For linux:

  1. Turn off the router, disable your ethernet connection:

sudo ifconfig eth0 down

  1. Turn on the router while holding down the 'restore' button on the back. You should be rewarded with a flashing power LED.
  1. Enable your network:

sudo ifconfig eth0 up 192.168.1.2

  1. Check you can ping your router.

ping 192.168.1.1

  1. Upload the firmware:

cd <place where it is downloaded>
ls *.trx
tftp 192.168.1.1
binary
put <the trx file>

  1. Then wait for the prompt to be returned, and wait for a further minute. Finally turn off the router, set your ethernet back to auto, wait a minute, turn on the router and you should have a tomato install!

comment:291 Changed 5 years ago by anonymous

Thanks for answer! I will try to install this. I hope not brick my router.

comment:292 follow-up: Changed 5 years ago by anonymous

For Windows operating sistem can anyone tell me what are the steps to install openwrt in my Asus rt n16...
Please give me the exact address where i can download the latest firmware openwrt for RT N16. Thank you very much for your help!

comment:293 in reply to: ↑ 286 ; follow-up: Changed 5 years ago by anonymous

Replying to gkashperko:

Replying to shinton@…:

Same problem here, power/wlan led's do not work. Anyone have a fix?

Power led works for me. As for wlan it seems to be controlled with broadcom-wl, by that part which is supplied precompiled in .o file. Try echo "1" >/proc/diag/led/wlan while wl.ko and wl_glue.ko are unloaded - you should see wlan led turned on. As soon as you insmod both wl and wl_glue it won't work anymore.

shinton i noticed you submitted a ticket for this problem (https://dev.openwrt.org/ticket/10032). i was wondering if you have a coding solution to the problem so I can just modify the diag.c file myself to get my wlan light working properly.

comment:294 in reply to: ↑ 293 Changed 5 years ago by shinton@…

Replying to anonymous:

shinton i noticed you submitted a ticket for this problem (https://dev.openwrt.org/ticket/10032). i was wondering if you have a coding solution to the problem so I can just modify the diag.c file myself to get my wlan light working properly.

Nope no solution unfortunately, when the wl.ko module loads up it seems to take over control of the led and then you are unable to control it from /proc/diag/led/wlan. I'm pretty sure that the broadcom-wl package would need to be modified in order to get the wlan activity light working.

comment:295 in reply to: ↑ 292 Changed 5 years ago by anonymous

Replying to anonymous:

For Windows operating sistem can anyone tell me what are the steps to install openwrt in my Asus rt n16...
Please give me the exact address where i can download the latest firmware openwrt for RT N16. Thank you very much for your help!

I just compiled the latest openwrt version. Contains the same stuff as Georges except for a few modifications. For instance opkg.conf doesn't need to be modified now for opkg to update or install packages.

you can download the latest working RT-N16 firmware from the following address;

http://www.datafilehost.com/download-bda4e8c0.html

comment:296 follow-up: Changed 5 years ago by Andrei Nistor <coder.tux@…>

I've also just compiled my first openwrt image for the rt-n16, per george's instructions. I've also built a few packages that were unusable for me from the brcm47xx target (mostly kernel modules, but some other stuff too). I can add other packages to this repository if time permits, and I plan to rebuild it with every new patchset George releases.

You can get the image at http://voinici.ceata.org/~codertux/openwrt/
You need to edit /etc/opkg.conf and add the following line:

src snapshots http://voinici.ceata.org/~codertux/openwrt/packages

A question for the previous poster, how did you change the packaged opkg.conf?

comment:297 in reply to: ↑ 296 Changed 5 years ago by anonymous

Replying to Andrei Nistor <coder.tux@…>:

I've also just compiled my first openwrt image for the rt-n16, per george's instructions. I've also built a few packages that were unusable for me from the brcm47xx target (mostly kernel modules, but some other stuff too). I can add other packages to this repository if time permits, and I plan to rebuild it with every new patchset George releases.

You can get the image at http://voinici.ceata.org/~codertux/openwrt/
You need to edit /etc/opkg.conf and add the following line:

src snapshots http://voinici.ceata.org/~codertux/openwrt/packages

A question for the previous poster, how did you change the packaged opkg.conf?

before you issue the make command, locate the opkg.conf in the trunk directory. make your changes and save the file then issue the make command. the changes you made to opkg.conf will be in the firmware.

comment:298 Changed 5 years ago by Aubrey

Fantastic work. I'm really impressed by how snappy it is once I got LuCI set up. Newbie question though: Should most of the backfire brcm47xx packages on the OpenWRT site work now if I add them to my sources? I suspect the answer is yes, but haven't tested yet.
Super grateful, many thanks!

comment:299 follow-up: Changed 5 years ago by Andrei Nistor <coder.tux@…>

Most of the packages for brch47xx should work

comment:300 in reply to: ↑ 299 ; follow-up: Changed 5 years ago by anonymous

Thanks alot for the great work guys.
I have a question : is DDNS and Asterisk working on the latest build for asus RT-n16 and if yest where to download it?

thanks again.

comment:301 Changed 5 years ago by Aubrey

see above post.

comment:302 in reply to: ↑ 300 ; follow-up: Changed 5 years ago by anonymous

Replying to anonymous:

Thanks alot for the great work guys.
I have a question : is DDNS and Asterisk working on the latest build for asus RT-n16 and if yest where to download it?

thanks again.

see above post. http://downloads.openwrt.org/backfire/10.03/brcm47xx/packages/

comment:303 in reply to: ↑ 302 ; follow-up: Changed 5 years ago by anonymous

Replying to anonymous:

Replying to anonymous:

Thanks alot for the great work guys.
I have a question : is DDNS and Asterisk working on the latest build for asus RT-n16 and if yest where to download it?

thanks again.

see above post. http://downloads.openwrt.org/backfire/10.03/brcm47xx/packages/

I am aware about backfire packages, the question was which firmware build is working on the ASus rt-n16 and is DDNS and asterisk working fine in this build?

thanks.

comment:304 in reply to: ↑ 303 ; follow-up: Changed 5 years ago by anonymous

I am aware about backfire packages, the question was which firmware build is working on the ASus rt-n16 and is DDNS and asterisk working fine in this build?

The Bleeding Edge version (trunk) works on the RT-N16. Backfire probably would work fine if the patches that George made where modified to work with the Backfire source code. You can download the packages that work with Bleeding Edge (trunk) from the following location:

http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/

You can also download the Bleeding Edge firmware I compiled a few days ago from:

http://www.datafilehost.com/download-bda4e8c0.html

comment:305 in reply to: ↑ 304 Changed 5 years ago by anonymous

Replying to anonymous:

I am aware about backfire packages, the question was which firmware build is working on the ASus rt-n16 and is DDNS and asterisk working fine in this build?

The Bleeding Edge version (trunk) works on the RT-N16. Backfire probably would work fine if the patches that George made where modified to work with the Backfire source code. You can download the packages that work with Bleeding Edge (trunk) from the following location:

http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/

You can also download the Bleeding Edge firmware I compiled a few days ago from:

http://www.datafilehost.com/download-bda4e8c0.html

Thanks, I will give it a try and report back.

comment:306 Changed 5 years ago by phaoost <phaoost@…>

http://www.datafilehost.com/download-bda4e8c0.html worked for me, but it seems I can't use 802.11n on the router, only 802.11b/g/g+turbo are available. is there anything I can do with it?

comment:307 follow-up: Changed 5 years ago by phaoost <phaoost@…>

another issue - it seems I cannot update kernel:

root@OpenWrt:~# opkg list-installed|grep kernel
kernel - 3.0.3-1
root@OpenWrt:~# uname -a
Linux OpenWrt 3.0.1 #4 Sun Sep 4 11:54:51 MST 2011 mips GNU/Linux

now I can't load modules, since they are 3.0.3

comment:308 in reply to: ↑ 307 ; follow-up: Changed 5 years ago by anonymous

Replying to phaoost <phaoost@…>:

another issue - it seems I cannot update kernel:
now I can't load modules, since they are 3.0.3

from my understanding you can't just upgrade the kernel. you would need to install a compiled firmware that has a newer kernel compiled in. you still should be able to install the modules. make sure your opkg.conf file contains the following line:

http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/

comment:309 in reply to: ↑ 308 ; follow-up: Changed 5 years ago by phaoost <phaoost@…>

Replying to anonymous:

from my understanding you can't just upgrade the kernel. you would need to install a compiled firmware that has a newer kernel compiled in. you still should be able to install the modules. make sure your opkg.conf file contains the following line:

http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/

I see modules are from 3.0.3 in repository. tun.ko fails to load for 3.0.1 since it's actially for 3.0.3

comment:310 in reply to: ↑ 309 Changed 5 years ago by azdps

I see modules are from 3.0.3 in repository. tun.ko fails to load for 3.0.1 since it's actially for 3.0.3

if you are having 3.0.3 problems look at Andrei Nistor's post a few posts up. he has a link to firmware and 3.0.1 packages. he listed the following link which you will need to put in your opkg.conf file:

http://voinici.ceata.org/~codertux/openwrt/packages/

comment:311 Changed 5 years ago by Andrei Nistor <coder.tux@…>

I'll try to update to 3.0.3 next weekend (tied up with exams atm). If you want more packages when I rebuild, let me know.

comment:312 Changed 5 years ago by OpenSys

Seams the not all modules was build in " http://voinici.ceata.org/~codertux/openwrt/packages/" gets: "FAT-fs (sda1): codepage cp437 not found"

Andrei Nistor, just all kernel modules that usually was build in openwrt.

is possible to get the wifi led on by gpio dev ?

comment:313 Changed 5 years ago by Andrei Nistor <coder.tux@…>

I'll have to build the kmod-nls-* packages to fix that error. You can try kmod-nls-base, maybe it'll fix it.

About the wifi led, I have no idea how to fix it.

comment:314 Changed 5 years ago by phaoost <phaoost@…>

Is there any possibility to activate 802.11n ?

comment:315 follow-up: Changed 5 years ago by gkashperko

Street magic to get wireless led working with broadcom-wl:

nvram unset sb/1/leddc
nvram set wl0gpio7=2
nvram commit

comment:316 in reply to: ↑ 315 ; follow-ups: Changed 5 years ago by Andrei Nistor <coder.tux@…>

Replying to gkashperko:

Street magic to get wireless led working with broadcom-wl:

nvram unset sb/1/leddc
nvram set wl0gpio7=2
nvram commit

Doesn't make any difference on my router...

comment:317 in reply to: ↑ 316 Changed 5 years ago by OpenSys

Replying to Andrei Nistor <coder.tux@…>:

Replying to gkashperko:

Street magic to get wireless led working with broadcom-wl:

nvram unset sb/1/leddc
nvram set wl0gpio7=2
nvram commit

Doesn't make any difference on my router...

In my, is the same the wifi led don't work.

comment:318 Changed 5 years ago by OpenSys

New build was made:

check in: www.znau.edu.ua/temp/asus-rt-n16/2011-09-19T15-44/

was news he have ?

comment:319 Changed 5 years ago by anonymous

How install web interface? Because i instaled trx file, but i not have grafic interface. I have ping to 192.168.1.1... Nothing else appear in screen! Thanks for help!

comment:320 Changed 5 years ago by phaoost <phaoost@…>

you need to telnet 192.168.1.1

comment:321 in reply to: ↑ 316 Changed 5 years ago by gkashperko

Replying to Andrei Nistor <coder.tux@…>:

Replying to gkashperko:

Street magic to get wireless led working with broadcom-wl:

nvram unset sb/1/leddc
nvram set wl0gpio7=2
nvram commit

Doesn't make any difference on my router...


wl0gpio7 nvram value controls led behaviour. Defined behaviours are there in wlioctl.h, see WL_LED_XXXX defines (look through broadcom-wl sources). I suggested value 2 (WL_LED_ACTIVITY) which make wlan led off, but blinking on activity. Default value of 8 (WL_LED_WI2) keep wlan led on, flashing on activity. Not in mood to play around with others.

leddc nvram value controls led powersave. Calculated as following:
leddc=(n << 16 | (100 - n))
0 < n < 100
The higher you set n the brighter led will be. Doing nvram unset sb/1/leddc resets this to default value of 10%.

I guess those who report wifi led still not working just don't have the box uncovered as I do and can't see the led directly - beleive me, it's blinking :)
Here is recipe to get it much much much brighter:
nvram set wl0gpio7=8
nvram set sb/1/leddc=0x5A000A
nvram commit

This will make wlan led 90% of max brightness, turned on when enabled, off when down, blinking on activity.

When done with nvram changes reboot or reload wl.ko

comment:322 Changed 5 years ago by Andrei Nistor <coder.tux@…>

I've set mine to 90% brightness, always on, blinking on activity (the other setting worked too, but wasn't really visible). Thanks!

comment:323 Changed 5 years ago by jon.hermansen@…

Tried to boot the snapshot bcm47xx build from tonight. I don't have serial access (yet), but as far as I can tell the router never becomes pingable or hands out DHCP addresses.

Has anyone got it working, and if so, when did you download it?

comment:324 Changed 5 years ago by azdps


I compiled the latest trunk version on 09/23/2011 which includes linux kernel 3.04. I would like to especially thank gkashperko for providing the information to get the RT-N16 up and running with OpenWrt.

Changes that I made before compiling:

  • updated opkg.conf (now points to right package download location)
  • updated banner (changed the way the banner looks)
  • updated profile (added clear command to clear screen at telnet/ssh login)
  • updated nvram (to properly enable wireless led)

you can download it at:

http://www.datafilehost.com/download-c2fe4d94.html


comment:325 follow-up: Changed 5 years ago by azdps

* IGNORE MY LAST POST. HAS INVALID DOWNLOAD LINK *

I compiled the latest trunk version on 09/23/2011 which includes linux kernel 3.04. I would like to especially thank gkashperko for providing the information to get the RT-N16 up and running with OpenWrt.

Changes that I made before compiling:

  • updated opkg.conf (now points to right package download location)
  • updated banner (changed the way the banner looks)
  • updated profile (added clear command to clear screen at telnet/ssh login)
  • updated nvram (to properly enable wireless led)

you can download it at:

http://www.datafilehost.com/download-9d857f90.html


comment:326 in reply to: ↑ 325 ; follow-up: Changed 5 years ago by anonymous

Replying to azdps:

I compiled the latest trunk version on 09/23/2011 which includes linux kernel 3.04. I would like to especially thank gkashperko for providing the information to get the RT-N16 up and running with OpenWrt.

Thanks for pretty stable image :)

Could you please point out exact steps to reproduce the image? I tried compiling it myself but resulting image does not load ssb kernel module in time so wireless module is not loaded. Upon booting diag utility detects my rt-n16 as WL500G Premium instead.

Lauri

comment:327 in reply to: ↑ 326 ; follow-up: Changed 5 years ago by anonymous

Could you please point out exact steps to reproduce the image? I tried compiling it myself but resulting image does not load ssb kernel module in time so wireless module is not loaded. Upon booting diag utility detects my rt-n16 as WL500G Premium instead.

I think the problem you are running into is in the menuconfig settings. Once you issue the command make menuconfig, you need to change a few settings. Change the following:

Target System from Broadcom BCM947XX/953xx to Broadcom BCM94716. Then the Target Profile will automatically change to ASUS RT-N16. Thats probably the snag you were running into.

comment:328 in reply to: ↑ 327 ; follow-up: Changed 5 years ago by anonymous

I think the problem you are running into is in the menuconfig settings. Once you issue the command make menuconfig, you need to change a few settings. Change the following:

Target System from Broadcom BCM947XX/953xx to Broadcom BCM94716. Then the Target Profile will automatically change to ASUS RT-N16. Thats probably the snag you were running into.

Is trunk supposed to have that target system or you used some patches? Is BCM94716 incompatible with other BCM947xx chips or we can eventually see them merged together?

Lauri

comment:329 in reply to: ↑ 328 Changed 5 years ago by azdps


Is trunk supposed to have that target system or you used some patches?


the trunk needs to be patched with 3 patches made by gkashperko. read his instructions page located at -

http://www.znau.edu.ua/temp/asus-rt-n16/openwrt-bcm4716-README.txt


download the patches located at the below location. go into location which has the newest date to download the newest patches -


http://www.znau.edu.ua/temp/asus-rt-n16/


>Is BCM94716 incompatible with other BCM947xx chips or we can eventually see them merged together?


not sure. i will let a real expert answer that question.

comment:330 Changed 5 years ago by Opensys

I was experiencing bad serial terminal enter key typing.
The enter key don't work... only write a strange chars in the same position over and over. All other keys that i type write ok. this problem happens in last trunk, all old trunks works fine.
Also i try the last trunk, and the problems persists.
I tested vt100, linux with same results.
Any idea ?

comment:331 follow-up: Changed 5 years ago by anonymous

@azdps
Thanks for your image. I experienced that compiling one can be quite annoying. Since I try to use my asus with a sound card, wiki says that a kernel with build-in alsa is required. If you plan to build another version could you or others include it, please?

comment:332 in reply to: ↑ 331 Changed 5 years ago by azdps

I experienced that compiling one can be quite annoying. Since I try to use my asus with a sound card, wiki says that a kernel with build-in alsa is required. If you plan to build another version could you or others include it, please?

I attempted to build a version with sound support and experienced compiling errors. I just don't have the time to work through the errors to figure it out. Hopefully someone else can assist you.

comment:333 follow-up: Changed 5 years ago by jon.hermansen@…

Did anyone notice the tg3 module fails to build with the latest patches? I'm building against trunk, compiling all (or as many as I can) modules along with the kernel so that I can set up an IPsec VPN.

  CC [M]  drivers/net/tg3.o
drivers/net/tg3.c: In function 'tg3_get_device_address':
drivers/net/tg3.c:14357:4: error: implicit declaration of function 'ssb_gige_get_macaddr'
make[7]: *** [drivers/net/tg3.o] Error 1
make[6]: *** [drivers/net] Error 2
make[5]: *** [drivers] Error 2
make[5]: Leaving directory `/home/jhermansen/Sandbox/openwrt/openwrt-trunk/build_dir/linux-brcm4716/linux-3.0.4'
make[4]: *** [/home/jhermansen/Sandbox/openwrt/openwrt-trunk/build_dir/linux-brcm4716/linux-3.0.4/.modules] Error 2
make[4]: Leaving directory `/home/jhermansen/Sandbox/openwrt/openwrt-trunk/target/linux/brcm4716'
make[3]: *** [compile] Error 2
make[3]: Leaving directory `/home/jhermansen/Sandbox/openwrt/openwrt-trunk/target/linux'
make[2]: *** [target/linux/compile] Error 2
make[2]: Leaving directory `/home/jhermansen/Sandbox/openwrt/openwrt-trunk'
make[1]: *** [/home/jhermansen/Sandbox/openwrt/openwrt-trunk/staging_dir/target-mipsel_uClibc-0.9.32/stamp/.target_compile] Error 2
make[1]: Leaving directory `/home/jhermansen/Sandbox/openwrt/openwrt-trunk'
make: *** [world] Error 2

comment:334 in reply to: ↑ 333 ; follow-up: Changed 5 years ago by gkashperko

Replying to jon.hermansen@…:

Did anyone notice the tg3 module fails to build with the latest patches? I'm building against trunk, compiling all (or as many as I can) modules along with the kernel so that I can set up an IPsec VPN.

AFAIK there are no brcm4716'ths paired with Tigon3. In any case if you compile for N16 you can skip it for sure.

comment:335 in reply to: ↑ 334 Changed 5 years ago by gkashperko

Replying to gkashperko:

Replying to jon.hermansen@…:

Did anyone notice the tg3 module fails to build with the latest patches? I'm building against trunk, compiling all (or as many as I can) modules along with the kernel so that I can set up an IPsec VPN.

AFAIK there are no brcm4716'ths paired with Tigon3. In any case if you compile for N16 you can skip it for sure.

Btw if you still want to get it compiled turn "Device Drivers->Sonics Silicon Backplane->SSB Broadcom Gigabit Ethernet driver" on in kernel menuconfig (check if CONFIG_SSB_DRIVER_GIGE is defined in your kernel .config).

comment:336 follow-up: Changed 5 years ago by OpenSys

get a kernel panic with restart, any idea ?

CPU 0 Unable to handle kernel paging request at virtual address 00100104, epc == 801865d4, ra == 80186570
Oops#1:
Cpu 0
$ 0 : 00000000 00000000 00200200 00100100
$ 4 : 802a61e0 00000024 0f01fc00 0000061e
$ 8 : 00000002 80110988 00100100 87803410
$12 : 00000001 87ae2800 00000800 00000800
$16 : 87b29cec 00000040 000000ac 80270000
$20 : 802c0000 80270000 00e43379 80270000
$24 : 00000000 80170e48
$28 : 80262000 80263d60 80270000 80186570
Hi : 00000035
Lo : 871f2c24
epc : 801865d4 0x801865d4

Tainted: P

ra : 80186570 0x80186570
Status: 1100bc02 KERNEL EXL
Cause : 0080000c
BadVA : 00100104
PrId : 00019740 (MIPS 74Kc)
Modules linked in: usb_storage uhci_hcd ohci_hcd ip6t_REJECT ip6t_LOG ip6t_rt ip6t_hbh ip6t_mh ip6t_ipv6header ip6t_frag ip6t_eui64 ip6t_ah ip6table_raw ip6_queue ip6table_mangle ip6table_filter ip6_tables nf_conntrack_ipv6 nf_defrag_ipv6 ipt_ULOG nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp xt_HL xt_hl ipt_ECN xt_CLASSIFY xt_time xt_tcpmss xt_statistic xt_mark xt_length ipt_ecn xt_DSCP xt_dscp xt_string xt_layer7 ipt_MASQUERADE iptable_nat nf_nat xt_recent xt_helper xt_connmark xt_connbytes xt_conntrack xt_NOTRACK iptable_raw xt_state nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack sata_nv ehci_hcd sd_mod pppoe pppox ipt_REJECT xt_TCPMSS ipt_LOG xt_comment xt_multiport xt_mac xt_limit iptable_mangle iptable_filter ip_tables xt_tcpudp x_tables ip_gre gre ifb ip6_tunnel tunnel6 ppp_async ppp_generic slhc ext4 jbd2 mbcache wl(P) wl_glue libata usbcore ts_fsm ts_bm ts_kmp switch_robo switch_core scsi_mod nls_base crc16 crc_ccitt bgmac ipv6 diag [last unloaded: switch_core]
Process swapper (pid: 0, threadinfo=80262000, task=80265ce0, tls=00000000)
Stack : 00000004 00000001 00000001 00000000 802b308c 00000001 00000003 00000100

0000000a 802b0000 802b0000 80267868 80267e50 8001ff38 80290000 80296a95
80260000 00000002 00000001 80400000 00000000 80292868 80290000 80296a95
80260000 00000002 00000001 80400000 ffffffff 80020140 1fb5ad00 0000ead9
00000800 80292868 00001000 80020320 80260000 00000002 00000001 80400000
...

Call Trace:[<8001ff38>] 0x8001ff38
[<80290000>] 0x80290000
[<80292868>] 0x80292868
[<80290000>] 0x80290000
[<80020140>] 0x80020140
[<80292868>] 0x80292868
[<80020320>] 0x80020320
[<800020bc>] 0x800020bc
[<80004fac>] 0x80004fac
[<80005180>] 0x80005180
[<80292868>] 0x80292868
[<80290000>] 0x80290000
[<8001588c>] 0x8001588c
[<80006ae8>] 0x80006ae8
[<800051a0>] 0x800051a0
[<802758d4>] 0x802758d4
[<80292868>] 0x80292868
[<8027521c>] 0x8027521c

Code: 08061980 26624ba0 8e030000 <ac620004> ac430000 27c34ba0 8c62000c ac70000c 3c038027
Kernel panic - not syncing: Fatal exception in interrupt

comment:337 in reply to: ↑ 336 ; follow-up: Changed 5 years ago by gkashperko

Replying to OpenSys:

get a kernel panic with restart, any idea ?

CONFIG_KALLSYMS ?

comment:338 in reply to: ↑ 337 ; follow-up: Changed 5 years ago by OpenSys

Replying to gkashperko:

Replying to OpenSys:

get a kernel panic with restart, any idea ?

CONFIG_KALLSYMS ?

Hello gkashperko,

i don't see this option in .config kernel 3.0.4.
The option enables module debug correct ?
can i put it manually ?

comment:339 in reply to: ↑ 338 ; follow-up: Changed 5 years ago by gkashperko

Replying to OpenSys:

i don't see this option in .config kernel 3.0.4.
The option enables module debug correct ?
can i put it manually ?

It adds symbolic crash information and symbolic stack backtraces.

make kernel_menuconfig

General setup -> Configure standard kernel features (expert users) -> Load all symbols for debugging/ksymoops

comment:340 in reply to: ↑ 339 Changed 5 years ago by OpenSys

Replying to gkashperko:

Replying to OpenSys:

i don't see this option in .config kernel 3.0.4.
The option enables module debug correct ?
can i put it manually ?

It adds symbolic crash information and symbolic stack backtraces.

make kernel_menuconfig

General setup -> Configure standard kernel features (expert users) -> Load all symbols for debugging/ksymoops

Thank You,

That's it, next time that happen i will get the method() that make me the kernel oops.

comment:341 follow-up: Changed 5 years ago by maurer

i don't know how to add the rt-n16 patches to openwrt.
can you guide/help me with a few commands?
Thanks

comment:342 in reply to: ↑ 341 Changed 5 years ago by anonymous

Replying to maurer:

i don't know how to add the rt-n16 patches to openwrt.
can you guide/help me with a few commands?
Thanks

  1. download openwrt trunk source.
  2. cd into trunk source folder.
  3. download patches into this folder
  4. execute this command

patch -p1 < file.patch
where file.patch is where you would put the name of the 3 patches one at a time in order.

comment:343 Changed 5 years ago by anonymous

I compiled the source from trunk rev 28400 which includes linux kernel 3.0.4 but downgraded to 3.0.3 for compatibility with most trunk ipk.

Fixed:

  • opkg.conf (to points to trunk package)
  • nvram (to enable wireless led)

Includes:

  • kmod-usb-core
  • kmod-usb2
  • kmod-usb-storage-extras
  • kmod-fs-ext4
  • kmod-rtl8187
  • block-mount
  • blkid
  • swap-utils
  • usbutils

Download link: http://www.mediafire.com/?x68zzi5z1179oha

Thank to all.

comment:344 Changed 5 years ago by OpenSys

i was experienced in some reboots a freeze problem in: Calibrating delay loop...

The line should be "Calibrating delay loop... 239.10 BogoMIPS (lpj=478208)"

seams some problem to find the correct cpu BogoMIPS.
I using trunk r28406

Any clue to fix that ?

comment:345 in reply to: ↑ description Changed 5 years ago by anonymous

Replying to Coool:

I would like to see ASUS RT-N16 support. I can be a tester.

comment:346 follow-up: Changed 5 years ago by squadra@…

Can someone please tell me what works and what doesnt work in azdps´s image?

comment:347 in reply to: ↑ 346 ; follow-up: Changed 5 years ago by anonymous

Replying to squadra@…:

Can someone please tell me what works and what doesnt work in azdps´s image?

It's pretty much as basic as you can get. If you want to add additional functionality such as bandwidth monitoring, vpn etc you will need to install packages to suit your needs.

What works


usb
wireless
ethernet
firewall

comment:348 in reply to: ↑ 347 Changed 5 years ago by anonymous

Replying to anonymous:

Replying to squadra@…:

Can someone please tell me what works and what doesnt work in azdps´s image?

It's pretty much as basic as you can get. If you want to add additional functionality such as bandwidth monitoring, vpn etc you will need to install packages to suit your needs.

What works


usb
wireless
ethernet
firewall

thanks! and this is stable so far?

comment:349 follow-up: Changed 5 years ago by anonymous

I have installed the image from http://www.mediafire.com/?x68zzi5z1179oha on my router and it works great with one exception. Any kernel modules installed by opkg trigger the following error message in the dmesg output (example below for ipv6). Any ideas on how to resolve this issue?

ipv6: version magic '3.0.3 mod_unload MIPS32_R1 32BIT ' should be '3.0.3 mod_unload MIPS32_R2 32BIT '

comment:350 in reply to: ↑ 349 ; follow-up: Changed 5 years ago by anonymous

Replying to anonymous:

I have installed the image from http://www.mediafire.com/?x68zzi5z1179oha on my router and it works great with one exception. Any kernel modules installed by opkg trigger the following error message in the dmesg output (example below for ipv6). Any ideas on how to resolve this issue?

ipv6: version magic '3.0.3 mod_unload MIPS32_R1 32BIT ' should be '3.0.3 mod_unload MIPS32_R2 32BIT '

All kernel modules, libraries and packages in thunk are compiled for MIPS32_R1 32BIT. So, all kernel modules working only for MIPS32_R1 32BIT except some libraries and packages will work both MIPS32_R1 and MIPS32_R2.

If you compile kernel modules for rt-n16 yourself then the issue will solve.
Please note the Openwrt are not native support for rt-n16 router.
Thank to gkashperko who providing the patch to get the RT-N16 up and running with OpenWrt.

comment:351 in reply to: ↑ 350 Changed 5 years ago by anonymous

Replying to anonymous:

Replying to anonymous:

I have installed the image from http://www.mediafire.com/?x68zzi5z1179oha on my router and it works great with one exception. Any kernel modules installed by opkg trigger the following error message in the dmesg output (example below for ipv6). Any ideas on how to resolve this issue?

ipv6: version magic '3.0.3 mod_unload MIPS32_R1 32BIT ' should be '3.0.3 mod_unload MIPS32_R2 32BIT '

All kernel modules, libraries and packages in thunk are compiled for MIPS32_R1 32BIT. So, all kernel modules working only for MIPS32_R1 32BIT except some libraries and packages will work both MIPS32_R1 and MIPS32_R2.

If you compile kernel modules for rt-n16 yourself then the issue will solve.
Please note the Openwrt are not native support for rt-n16 router.
Thank to gkashperko who providing the patch to get the RT-N16 up and running with OpenWrt.

I've got trunk (HEAD 93005b12c529da797699c24c58243379ffed1efd) compiled including the patches. Quite a journey..

Anyone interested in those packages, hands up. Can also provide a quilt series for easy application of the latest patches (plus a kernel git patch for an undecl SSB_PCI build error).

Changed 5 years ago by anonymous

comment:352 follow-up: Changed 5 years ago by banderols@…

What is the difference between images from www.znau.edu.ua/temp/asus-rt-n16/ ?

Now there are 3 images:
openwrt-brcm4716-glibc_nptl-w_extras-squashfs.trx
openwrt-brcm4716-squashfs.trx
openwrt-brcm4716-w_extras-squashfs.trx

comment:353 in reply to: ↑ 352 Changed 5 years ago by anonymous

Replying to banderols@…:

What is the difference between images from www.znau.edu.ua/temp/asus-rt-n16/ ?

Now there are 3 images:
openwrt-brcm4716-glibc_nptl-w_extras-squashfs.trx
openwrt-brcm4716-squashfs.trx
openwrt-brcm4716-w_extras-squashfs.trx

I am wondering myself what the different are between the three builds. I have been dying to try openwrt in my box but apparently support has been limited till now. Let's keep plugging, we should get this fully supported soon. This is the king of routers.

comment:354 Changed 5 years ago by anonymous

any idea why i get "invalid architecture" error messages when using this repo with georges firmware?

http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/

comment:355 Changed 5 years ago by jon.hermansen@…

anonymous: You'll need to edit /etc/opkg.conf, adding arch directives. See this post for more info.

comment:356 follow-up: Changed 5 years ago by Krey@…

r28753 work unstable. WiFi traffic over 15MB\s cause router to reboot

comment:357 in reply to: ↑ 356 Changed 5 years ago by anonymous

Replying to Krey@…:

r28753 work unstable. WiFi traffic over 15MB\s cause router to reboot

Can someone comment on this, is there a fix why the reboots occur? Krey can you check the logs and see if anything is showing there.

comment:358 Changed 5 years ago by Krey@…

I don't have a uart cable, but i can try to disable watchdog

comment:359 Changed 5 years ago by en1gma

i have a belkin n300 share (F7D3302 v1/v1000)
it has 8MB/64MB and same cpu
dd-wrt status page shows its a "Broadcom BCM4716 chip rev 1" but i think its really BMC4718 like rt-n16

is there something i can do to help test?
im pretty new at this but not a complete noob...i made a jtag cable a while back for a different router and added an sd card to it....this router i think already has the header for it.

if i can be any help let me know.
im interested

comment:360 Changed 5 years ago by anonymous

Hi all

I have built latest trunk r29288 with George's patches and uploaded the squash-fs and almost all the packages (in the ImageBuilder tar) to megaupload. This file contains the links to all the others as the spam filter smacked me down. As always, this build comes with no warranty and I will not be responsible for lost time or equipment damage. Use at your own risk.

http://www.megaupload.com/?d=LL9T5H12

NOTE: THIS BUILD BOOTS TO 192.168.1.5 <<<<< NOTE THE 5

Fixes required for RT-N16 against OpenWrt trunk 29288

package/mac80211/Makefile: Fix dependency from failed patch hunk

/trunk/target/linux/generic/patches-3.0/102-ehci_hcd_ignore_oc.patch: Remove line below

unsigned frame_index_bug:1; /* MosChip (AKA NetMos) */

Apply fix from this thread:
http://permalink.gmane.org/gmane.linux.kernel.wireless.general/73918*

MD5 Checksums

c0a1a2a021224f18b88f4e710e55d508 *OpenWrt-ImageBuilder-brcm4716-for-Linux-x86_64.tar.bz2

dfb28c527e6e861474ebfbc7cfac5277 *OpenWrt-SDK-brcm4716-for-Linux-x86_64-gcc-4.5-linaro_uClibc-0.9.32.tar.bz2

f16cd91d360783cc1d25d453cf7f1bff *OpenWrt-Toolchain-brcm4716-for-mipsel-gcc-4.5-linaro_uClibc-0.9.32.tar.bz2

f97ba4b16e55916d7020ee6b9e8ac5e5 *openwrt-brcm4716-rootfs.tar.gz

54805f371ea93cdcf55d39e3f3277648 *openwrt-brcm4716-squashfs.trx

comment:361 Changed 5 years ago by anonymous

It will be great to see real support for this router, rather than a whole bunch of patches from different places.

comment:362 follow-up: Changed 5 years ago by anonymous

Keep dreaming, with the license issues it won't happen.

comment:363 in reply to: ↑ 362 Changed 5 years ago by anonymous

Replying to anonymous:

Keep dreaming, with the license issues it won't happen.

Does the issue deals with broadcom drivers?

comment:364 Changed 5 years ago by anonymous

Yes, the bgmac gigabit ethernet driver has licence issues hence the need for George and his patchset. I myself don't understand how all the other drivers could be released with a licence except for this one.

comment:365 follow-up: Changed 5 years ago by anonymous

Hi,

can anybody help me with this:

# cat /etc/opkg.conf
src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay
arch all 100
arch brcm4716 200
arch brcm47xx 300
# opkg install ntpdate
Installing ntpdate (4.2.6p4-3) to root...
Downloading http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/ntpdate_4.2.6p4-3_brcm47xx.ipk.
Configuring ntpdate.
# which ntpdate
/usr/sbin/ntpdate
# /usr/sbin/ntpdate
-ash: /usr/sbin/ntpdate: not found

?

comment:366 in reply to: ↑ 365 Changed 5 years ago by anonymous

Replying to anonymous:

Hi,

can anybody help me with this:

# cat /etc/opkg.conf
src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay
arch all 100
arch brcm4716 200
arch brcm47xx 300
# opkg install ntpdate
Installing ntpdate (4.2.6p4-3) to root...
Downloading http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/ntpdate_4.2.6p4-3_brcm47xx.ipk.
Configuring ntpdate.
# which ntpdate
/usr/sbin/ntpdate
# /usr/sbin/ntpdate
-ash: /usr/sbin/ntpdate: not found

?

src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages[[BR]]
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay
arch all 1
arch mipsel 200
arch brcm47xx 10

comment:367 Changed 5 years ago by anonymous

I am trying to build but it builds over 3.0.9 with fail. How to build it with 3.0.4?

comment:368 follow-up: Changed 5 years ago by anonymous

is there any build with luci? I can't make it running

comment:369 in reply to: ↑ 368 Changed 5 years ago by anonymous

Replying to anonymous:

is there any build with luci? I can't make it running

A build with luci would be nice for many of us.

comment:370 Changed 5 years ago by anonymous

There is an active thread over on the OpenWrt forums, where messages can be edited and such: forum.openwrt.org/viewtopic.php?id=33560

comment:371 Changed 5 years ago by anonymous

[build@localhost trunk]$ make V=99 -j4
make[1]: Entering directory `/home/build/trunk'
make[2]: Entering directory `/home/build/trunk'
make[3]: Entering directory `/home/build/trunk/target/linux'
make[4]: Entering directory `/home/build/trunk/target/linux/brcm4716'
if [ -s "/home/build/trunk/build_dir/linux-brcm4716/linux-3.0.4/patches/series" ]; then (cd "/home/build/trunk/build_dir/linux-brcm4716/linux-3.0.4"; if quilt --quiltrc=- next >/dev/null 2>&1; then quilt --quiltrc=- push -a; else quilt --quiltrc=- top >/dev/null 2>&1; fi ); fi
Applying patch platform/0001-bcma-move-parsing-of-EEPROM-into-own-function.patch
patching file drivers/bcma/scan.c
Hunk #1 FAILED at 200.
Hunk #2 FAILED at 236.
Hunk #3 FAILED at 351.
3 out of 3 hunks FAILED -- rejects in file drivers/bcma/scan.c
Patch platform/0001-bcma-move-parsing-of-EEPROM-into-own-function.patch does not apply (enforce with -f)
make[4]: * home/build/trunk/build_dir/linux-brcm4716/linux-3.0.4/.quilt_checked Error 1
make[4]: Leaving directory `/home/build/trunk/target/linux/brcm4716'
make[3]:
* [compile] Error 2
make[3]: Leaving directory `/home/build/trunk/target/linux'
make[2]: * [target/linux/compile] Error 2
make[2]: Leaving directory `/home/build/trunk'
make[1]:
* home/build/trunk/staging_dir/target-mipsel_uClibc-0.9.32/stamp/.target_compile Error 2
make[1]: Leaving directory `/home/build/trunk'
make: * [world] Error 2

any ideas? build on latest trunk fails

comment:372 Changed 5 years ago by azdps

I haven't been able to build the latest trunk version as well. I believe it has to do with the added support for Broadcom's bcma driver in the latest trunk versions. Georges patches will needed to be modified to accommodate the changes. Not sure if that's the sole issue though.

comment:373 Changed 5 years ago by anonymous

George has updated his patchset for the newest trunk posted yesterday... Thanks George!

comment:374 Changed 5 years ago by anonymous

A build with luci would be nice for many of us.

comment:375 Changed 5 years ago by jianyu

Strongly agree with the luci build-in. I got rt-n16 bricked several times after install luci, no matter which version - 10.03, 10.03.1 or snapshot.

comment:376 Changed 5 years ago by azdps

luci should be easy to install after you install Georges firmware. after you log in via ssh to your router issue the following commands at the command prompt:

opkg update
opkg install luci
/etc/init.d/uhttpd enable
/etc/init.d/uhttpd start
reboot

after the router is rebooted, luci should be a accessible by visiting the following website address using internet explorer or firefox etc:

http://192.168.1.1

you may need to change 192.168.1.1 to whatever your router ip address is.

comment:377 follow-up: Changed 5 years ago by azdps

luci should be easy to install after you install Georges firmware. after you log in via ssh to your router issue the following commands at the command prompt:

opkg update
opkg install luci
/etc/init.d/uhttpd enable
/etc/init.d/uhttpd start
reboot

after the router is rebooted, luci should be a accessible by visiting the following website address using internet explorer or firefox etc:

http://192.168.1.1

you may need to change 192.168.1.1 to whatever your router ip address is.

comment:378 in reply to: ↑ 377 Changed 5 years ago by anonymous

Replying to azdps:

luci should be easy to install after you install Georges firmware. after you log in via ssh to your router issue the following commands at the command prompt:

opkg update
opkg install luci
/etc/init.d/uhttpd enable
/etc/init.d/uhttpd start
reboot

after the router is rebooted, luci should be a accessible by visiting the following website address using internet explorer or firefox etc:

http://192.168.1.1

you may need to change 192.168.1.1 to whatever your router ip address is.

comment:379 follow-ups: Changed 5 years ago by anonymous

can some one post step by step to build the image fot this RT N 16?
all this patches drive me crazy here.

comment:380 in reply to: ↑ 379 Changed 5 years ago by anonymous

Replying to anonymous:

can some one post step by step to build the image for this RT N 16?
all this patches drive me crazy here.

comment:381 in reply to: ↑ 379 Changed 5 years ago by anonymous

Replying to anonymous:

can some one post step by step to build the image fot this RT N 16?
all this patches drive me crazy here.

See http://www.znau.edu.ua/temp/asus-rt-n16/openwrt-bcm4716-README.txt for instructions.

svn co svn://svn.openwrt.org/openwrt/trunk/
patch -p1 < path_to_downloaded_patch (for every patch)

But latest trunk does not patched correctly with latest patches anyway.

comment:382 Changed 5 years ago by razvi

http://openwrt.razvi.ro/attitude_adjustment_asus_rt-n16_svn_r29617/

I will post a version with luci included. I just installed openwrt-brcm4716-jffs2-128k.trx and it's running very smooth.

comment:383 Changed 5 years ago by anonymous

was anyone able to build openwrt with recent backfire_10.03.1 ?
002-openwrt4716-TARGET_brcm4716-deps.patch is failing for me

comment:384 Changed 5 years ago by anonymous

this work for me:

replace manually
*

DEPENDS:=@(!(TARGET_avr32+ DEPENDS:=@(!(TARGET_avr32
TARGET_ep93xxTARGET_ps3TARGET_pxcabLINUX_3_2)BROKEN)
TARGET_ep93xxTARGET_ps3TARGET_pxcabLINUX_3_2TARGET_brcm4716)BROKEN)

*

in /package/mac80211/Makefile after patch -p1<002-openwrt4716-TARGET_brcm4716-deps.patch

my trunk version is revision 29754

comment:385 Changed 5 years ago by anonymous

Replying to razvi:

http://openwrt.razvi.ro/attitude_adjustment_asus_rt-n16_svn_r29617/

I will post a version with luci included. I just installed openwrt-brcm4716-jffs2-128k.trx and it's running very smooth.

Thank You. I wait for long time such a version. Please tell me where i download a such version. There is something that does not work?

comment:387 Changed 5 years ago by jianyu

Same question again: anyone made 802.11n working? I'm curios how dd-wrt did that.

comment:388 Changed 5 years ago by anonymous

So, is someone working to integrate this in openwrt trunk?

comment:389 Changed 5 years ago by razvi

802.11n is working.

config 'wifi-device' 'wl0'

option 'type' 'broadcom'
option 'hwmode' '11ng'

comment:390 Changed 5 years ago by jow

Did you measure the phy rate on a client?
Internally the broadcom setup script just checks whether the mode ends with "g" and sets it as gmode, there is no handling of 11n at all in the driver integration.

comment:391 Changed 5 years ago by camden.lindsay+openwrt@…

Hello-

I have one of these units that i recently replaced with something else. I would be happy to do testing, soldering, anything needed to help with the progress here. Or ship it to a developer as a donation.

Please let me know what i can do to help :)

comment:392 Changed 5 years ago by anonymous

I am running razvi version posted above, anyone knows why I'm getting a lot of TX packet errors/RX frames lost on the wl0 interface and yet the link remains/seems stable

wl0 Link encap:Ethernet HWaddr XX:XX:XX:XX:XX:XX

inet addr:aaa.bbb.ccc.ddd Bcast:aa.bb.cc.dd Mask:255.255.255.0
inet6 addr: XX:XX:XX:XX:XX:XX/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1215 errors:0 dropped:8 overruns:0 frame:842053
TX packets:1139 errors:12 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:120662 (117.8 KiB) TX bytes:434007 (423.8 KiB)
Interrupt:3 Base address:0x1000

comment:393 Changed 5 years ago by anonymous

would this latest build support kmod-macvlan? Otherwise what would be the right package type (brcm47xx?) to chose the correct repository to apply kmod-macvlan.
That's a function I need to apply multiple Mac addresses to the ame physical network interface.
Thanks

comment:394 Changed 5 years ago by anonymous

is 802.11n supported & working ?

comment:395 Changed 5 years ago by anonymous

Hi,

I have an Asus RT-N16.

I would like to know which version of razvi i can install on it.
And what is the difference between brcm4716-jffs2-128k.trx, brcm4716-jffs2-64k.trx and brcm4716-squashfs.trx.
Another question: i can install one of these versions from web gui router?
Now i run Tomato Shibby.

Thanks,

comment:396 Changed 5 years ago by bozo29

Hi,

Same question.

Asus RT-N16.
I understand that i can install brcm4716-jffs2-128k from razvi. Is it o.k?
I can install from web gui, from Tomato?
And where i can read about differences between 128K, 64k or squashfs?

Thank you in advance.

comment:397 follow-ups: Changed 5 years ago by smanet

Hello,
i want to know if at current time in the trunk these services works:
Wireless N
IPV6
USB (at least storage)
I'm trying to build my firmware but I'm scared that something will not work.

Thank you in advance and great work!

comment:398 Changed 5 years ago by anonymous

Hi all,

I have installed "openwrt-brcm4716-jffs2-128k.trx" from "r29906" and I see that working kernel is: "3.0.17".
Can anyone tell me what would be the consequences if I would run:

opkg update
opkg install kernel -- or -- opkg upgrade kernel

(because there seems to be a package "kernel 3.0.18")

Since RT-N16 is not fully supported yet, would that generate all sorts of problems? (modules missing, modules not loading, ...)

On the other hand, I actually need "kmod-dummy" and looks like: opkg install kmod-dummy is installing one for 3.0.18 that doesn't want to load in the current 3.0.17 running kernel.

Many thanks

comment:400 in reply to: ↑ 399 ; follow-up: Changed 5 years ago by anonymous

Replying to razvi:

http://openwrt.razvi.ro/attitude_adjustment_asus_rt-n16_svn_r30420/

Can you please, tell me if i could install "openwrt-brcm4716-jffs2-128k.trx", directly from Tomato web ui? On Asus RT-N16.

I ask it here, because RT-N16 does not fully support by OpenWRT and tutorials from the official website could not help me.

Thanks.

comment:401 Changed 5 years ago by xray

Yes, it's flashed directly from tomato-ext web-interface.

I trying to build from source with your patches, but it fails to apply generic patches. Is there a way to fix something?
Error:

Applying /home/xray/Devel/OpenWRT/trunk_backup/target/linux/generic/patches-3.0/102-ehci_hcd_ignore_oc.patch using plaintext: 
patching file drivers/usb/host/ehci-hcd.c
patching file drivers/usb/host/ehci-hub.c
Hunk #1 succeeded at 656 (offset 1 line).
Hunk #2 succeeded at 881 (offset 1 line).
patching file drivers/usb/host/ehci.h
Hunk #1 FAILED at 138.
1 out of 1 hunk FAILED -- saving rejects to file drivers/usb/host/ehci.h.rej
Patch failed!  Please fix /home/xray/Devel/OpenWRT/trunk_backup/target/linux/generic/patches-3.0/102-ehci_hcd_ignore_oc.patch!

comment:402 Changed 5 years ago by anonymous

Okay, sorry, seems that I found solution about failed patch...

comment:403 follow-up: Changed 5 years ago by anonymous

There is conflict of two patches now:
package/broadcom-wl/009-fix_compile_3_2.patch
from trunk, and
package/broadcom-wl/009-linux-3.2-ndo_set_rx_mode.patch
from razvi

comment:404 in reply to: ↑ 400 Changed 5 years ago by razvi

Replying to anonymous:

Can you please, tell me if i could install "openwrt-brcm4716-jffs2-128k.trx", directly from Tomato web ui? On Asus RT-N16.

I ask it here, because RT-N16 does not fully support by OpenWRT and tutorials from the official website could not help me.

Thanks.

use ASUS's recovery mode firmware restauration tool

comment:405 in reply to: ↑ 403 Changed 5 years ago by anonymous

Replying to anonymous:

There is conflict of two patches now:
package/broadcom-wl/009-fix_compile_3_2.patch
from trunk, and
package/broadcom-wl/009-linux-3.2-ndo_set_rx_mode.patch
from razvi

I inspected the patches. It seems (to me) that the two patches do the same thing, just via a different method. I simply deleted "package/broadcom-wl/009-linux-3.2-ndo_set_rx_mode.patch" and it fixed the build issues for me.

comment:406 Changed 5 years ago by anonymous

I deleted the other one, from tunk. All built ok.

comment:407 in reply to: ↑ 397 Changed 5 years ago by anonymous

Replying to smanet:

Hello,
i want to know if at current time in the trunk these services works:
Wireless N

No - only turbo mode

IPV6

Can't fully test, the ISP null-routes the 6to4 relay (192.88.99.1). But the option is there, the router gets the address and correctly configures radvd.

USB (at least storage)

Yes

I'm trying to build my firmware but I'm scared that something will not work.

By default, it won't work. You need to go to advanced settings and choose gcc-4.5 with Linaro patches as the compiler. Otherwise, lua will segfault, and thus the web interface will be unusable. There are surely other miscompilations caused by gcc-4.6.

Thank you in advance and great work!

comment:408 in reply to: ↑ 397 Changed 5 years ago by anonymous

Replying to smanet:

IPV6

Tested by trying to ping6 another 6to4 host. Does not work by default due to a bug that is not specific to ASUS RT-N16. Namely, the default firewall blocks protocol-41 packets. "Solved" by this command:

iptables -I INPUT 1 -i pppoe-wan --proto 41 -j ACCEPT

comment:410 Changed 5 years ago by anonymous

You guys are making great progress on the device....hopefully it gets integrated in the liast of working devices by openwrt....that would be awesome.

comment:411 Changed 5 years ago by anonymous

Why OpenWRT developers just try to contact ASUS, explain the problem and try to get it offically solved with ASUS help?

RT-N16 (and other ASUS models) are already officially supported by Tomato and DD-WRT. More than that "ASUS highly values the DIY community" and officially endorse DD-WRT:
http://promos.asus.com/US/ASUS_DD-WRT/index.htm

comment:412 Changed 5 years ago by jow

The OpenWrt developers already contacted Broadcom, they already explained the problem, Broadcom already told they're looking into it and we're still waiting for any progress - so its not just a matter of contacting ASUS.

comment:414 Changed 5 years ago by gregd72002

Is there any issue with DHCP client/server on the recent builds? Whenever I compile the recent trunk with George patches the router does not obtain ip from the modem neither my computer can obtain address from the router?

comment:415 follow-up: Changed 5 years ago by anonymous

is wlan-n working or not? anyone got wlan connection rates above 54mbit?

comment:416 in reply to: ↑ 415 Changed 5 years ago by anonymous

Replying to anonymous:

is wlan-n working or not? anyone got wlan connection rates above 54mbit?

Yes, you can get more than 54mbit for sure (i.e. g turbo or possibly n) but you should compile it yourself.

So, is anyone else having problem with DHCP on custom builds?
Razvi, are u using recent George patches or do you have some additional customisations for u do not observe this issue on 31118? Would you mind posing your .config for compiling image for asus rt n16?

comment:417 Changed 5 years ago by anonymous

Hi.

I've been trying to set this up for my RT-N16 and tried jffs2-128k of various builds (30711, 31118 and 31134).
I send the file through tftp and it all looks good, but after that I cannot even ping 192.168.1.1 (Destination Host Unreachable), having static IP of 192.168.1.2/24 on my PC.

I can flash back to stock and DD-WRT and they work, so...

Any help is appreciated; I definitely want to give OpenWRT a shot.

comment:418 Changed 5 years ago by anonymous

wlan-n is not working. There was a patch posted, but it doesn't work for me either: https://lists.openwrt.org/pipermail/openwrt-devel/2012-March/014356.html + https://lists.openwrt.org/pipermail/openwrt-devel/2012-March/014357.html

As for the "unable to connect" issue (that actually looks like "packets bigger than ~100 bytes are dropped"), I confirm it with the 2012-03-09T19-52 release and confirm that it doesn't exist with 2012-02-10T15-42.

comment:419 Changed 4 years ago by anonymous

anyone else tested this 802.11n patch?

comment:420 Changed 4 years ago by anonymous

any news when will this router get finally support from openwrt? This patching situation is driving many of us nuts.

comment:421 Changed 4 years ago by jow

No changes on the underlying legal issues.

comment:422 Changed 4 years ago by anonymous

use tomatousb instead xD

comment:423 Changed 4 years ago by assid@…

Has there been any more updates? This device is technically superior to any thing else in the same price point. Am waiting like the other hundreds of people :/
Hope theres a breakthrough soon.

comment:424 follow-ups: Changed 4 years ago by hauke

Broadcom changed the license of the Ethernet driver to the ISC License (BSD license which is compatible with the GPL) in the last major version of their SDK. The GPL tarball of the Asus rt-ac66u contains this code and we are working on a driver base on this code.

comment:425 in reply to: ↑ 424 Changed 4 years ago by anonymous

Replying to hauke:

Broadcom changed the license of the Ethernet driver to the ISC License (BSD license which is compatible with the GPL) in the last major version of their SDK. The GPL tarball of the Asus rt-ac66u contains this code and we are working on a driver base on this code.

2 years of waiting... but now, there is a light in the tunnel!! Keep it up guys, I'm still hoping for a time I can download the sources and execute simply 'make' to get the image! ;)

comment:426 follow-up: Changed 4 years ago by anonymous

Has anyone had any luck building OpenWRT for this router recently? Using the patches from http://www.znau.edu.ua/temp/asus-rt-n16/2012-05-28T17-30/ results in the following build error. Any ideas how to fix this?

Applying patch platform/020-bcma-move-parallel-flash-into-a-union.patch
patching file arch/mips/bcm47xx/nvram.c
patching file drivers/bcma/driver_mips.c
Hunk #1 FAILED at 189.
1 out of 1 hunk FAILED -- rejects in file drivers/bcma/driver_mips.c
patching file include/linux/bcma/bcma_driver_chipcommon.h
Hunk #1 succeeded at 122 (offset 5 lines).
Hunk #2 succeeded at 418 with fuzz 2 (offset 36 lines).
Hunk #3 succeeded at 433 (offset 36 lines).
Hunk #4 succeeded at 462 (offset 36 lines).
Patch platform/020-bcma-move-parallel-flash-into-a-union.patch does not apply (enforce with -f)
make[4]: * share/openwrt2/trunk/build_dir/linux-brcm4716/linux-3.3.7/.quilt_checked Error 1
make[4]: Leaving directory `/share/openwrt2/trunk/target/linux/brcm4716'
make[3]:
* [compile] Error 2
make[3]: Leaving directory `/share/openwrt2/trunk/target/linux'
make[2]: * [target/linux/compile] Error 2
make[2]: Leaving directory `/share/openwrt2/trunk'
make[1]:
* share/openwrt2/trunk/staging_dir/target-mipsel_uClibc-0.9.33.2/stamp/.target_compile Error 2
make[1]: Leaving directory `/share/openwrt2/trunk'
make: * [world] Error 2

comment:427 in reply to: ↑ 424 Changed 4 years ago by dvb15

Replying to hauke:

Broadcom changed the license of the Ethernet driver to the ISC License (BSD license which is compatible with the GPL) in the last major version of their SDK. The GPL tarball of the Asus rt-ac66u contains this code and we are working on a driver base on this code.

So, any luck in making a working driver?

comment:428 Changed 4 years ago by anonymous

Are there patches against the latest revision? Since there is a new kernel opkg is not able to install new kernel modules any more :-(

comment:428 Changed 4 years ago by anonymous

Are there patches against the latest revision? Since there is a new kernel opkg is not able to install new kernel modules any more :-(

comment:429 Changed 4 years ago by elf.openwrt@…

hauke (and others)

http://support.asus.com/download.aspx?SLanguage=en&p=11&s=2&m=RT-N16&os=&hashedid=WAa6AQFncrceRBEo
looks like fully GPLed source for the RTN16. :) From my (extremely cursory) examination so far, I've not found the ever-present object (.o) files, nor anything that clearly resembled precompiled firmware, etc. (On a further note, they seem to have done this with all of the routers I've checked so far, if there are other outstanding models.) Also, the CFE code has been partially released, but the Changelog/release notes might the most useful part for debugging, fi you spend the time to go through them... all kinds of aggravating issues are dealt with in some detail. Hope this helps. If this turns out to be actual GPLed RTN16 code, what will the timeline look like for fully supported?

Thanks!

-elf

comment:430 Changed 4 years ago by stevemurphymsu@…

While it does not look like George has released another patch, I was able to compile against r32870, which has been running very well for about a month using the following procedure:

  1. Instead of applying patch 000, I ran cp -r target/linux/brcm47xx target/linux/brcm4716
  2. Applied patch 001 (worked, except created the issue in step 4 below)
  3. Applied patch 002 (worked except for usb.mk patch) and manually applied the change below)

--- openwrt.r31937.orig/package/kernel/modules/usb.mk 2012-05-28 17:24:16.885317026 +0300
+++ openwrt.r31937/package/kernel/modules/usb.mk 2012-05-28 17:25:31.521620154 +0300
@@ -994,3 +994,17 @@ endef

$(eval $(call KernelPackage,usb-brcm47xx))

+define KernelPackage/usb-brcm4716
+ SUBMENU:=$(USB_MENU)
+ TITLE:=Support for USB on bcm4716
+ DEPENDS:=@USB_SUPPORT @TARGET_brcm4716
+ KCONFIG:= \
+ CONFIG_USB_HCD_SSB
+ FILES:= \
+ $(LINUX_DIR)/drivers/usb/host/ssb-hcd.ko
+ AUTOLOAD:=$(call AutoLoad,19,ssb-hcd,1)
+ $(call AddDepends/usb)
+endef
+
+$(eval $(call KernelPackage,usb-brcm4716))
+

  1. In order to get openwrt to build I had to remove the following from target/linux/brcm4716/patches-3.3/982-openwrt4716-ssb-force-core-ctl-and-state-ops-usage.patch:

@@ -1565,7 +1559,7 @@ static int b43legacy_request_firmware(st

int err;

/* do dummy read */

  • ssb_read32(dev->dev, SSB_TMSHIGH);

+ ssb_core_state_flags(dev->dev);

if (!fw->ucode) {

if (rev == 2)

filename = "ucode2";

comment:431 in reply to: ↑ 426 Changed 4 years ago by anonymous

Replying to anonymous:

Has anyone had any luck building OpenWRT for this router recently? Using the patches from http://www.znau.edu.ua/temp/asus-rt-n16/2012-05-28T17-30/ results in the following build error. Any ideas how to fix this?

Use trunk r31958

comment:432 Changed 4 years ago by shinton@…

Does b43/b43legacy drivers work with the RT-N16? Anyone with experience trying this?

comment:433 Changed 4 years ago by anonymous

Hello. Unfortunately both images - build by razvi and build by George, suffer terrible CPU HOG (kernel ksoftirqd process) when WAN transfers are greater than 2-3 megabyses/sec.

comment:434 Changed 4 years ago by chhatc <cchhat01@…>

hi hauke and others:

there is a post on the openwrt forum discussing the progress that is being made to support the broadcom based ethernet driver based on the new relaxed drivers by broadcom.

https://forum.openwrt.org/viewtopic.php?id=39581

Last i heard, Zajec on that thread was working on a driver. However we haven't heard anything for a while now? is there any plan to move this work along.
I am very keen on seeing some of the newer/powerful broadcom based hardware come to life under the openwrt firmware.
I myself have a Linksys E4200 V1 which can be used as a testing medium as well...

Please let me know.

Thanks,
chhatc

comment:435 follow-up: Changed 4 years ago by anonymous

I have latest build openwrt installed from http://www.znau.edu.ua/temp/asus-rt-n16/2012-11-12T11-43/ But can't install packages.

cat /etc/opkg.conf
src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay
arch all 1
arch mipsel 200
arch brcm47xx 10

opkg install tcpdump
...
...
Package wireless-tools version 29-5 has no valid architecture, ignoring.
Package kmod-scsi-core version 3.3.8-1 has no valid architecture, ignoring.
Package ppp-mod-pppoe version 2.4.5-6 has no valid architecture, ignoring.
Package kmod-ipt-nat version 3.3.8-1 has no valid architecture, ignoring.
Installing tcpdump (4.2.1-2) to root...
Downloading http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/tcpdump_4.2.1-2_brcm47xx.ipk.
Collected errors:

  • satisfy_dependencies_for: Cannot satisfy the following dependencies for tcpdump:
  • libc * libc *
  • opkg_install_cmd: Cannot install package tcpdump.

What I am doing wrong?

comment:436 in reply to: ↑ 435 Changed 4 years ago by anonymous

Replying to anonymous:

I have latest build openwrt installed from http://www.znau.edu.ua/temp/asus-rt-n16/2012-11-12T11-43/ But can't install packages.

cat /etc/opkg.conf
src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay
arch all 1
arch mipsel 200
arch brcm47xx 10

opkg install tcpdump
...
...
Package wireless-tools version 29-5 has no valid architecture, ignoring.
Package kmod-scsi-core version 3.3.8-1 has no valid architecture, ignoring.
Package ppp-mod-pppoe version 2.4.5-6 has no valid architecture, ignoring.
Package kmod-ipt-nat version 3.3.8-1 has no valid architecture, ignoring.
Installing tcpdump (4.2.1-2) to root...
Downloading http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages/tcpdump_4.2.1-2_brcm47xx.ipk.
Collected errors:

  • satisfy_dependencies_for: Cannot satisfy the following dependencies for tcpdump:
  • libc * libc *
  • opkg_install_cmd: Cannot install package tcpdump.

What I am doing wrong?

Ok. here is worked config:

src/gz snapshots http://downloads.openwrt.org/snapshots/trunk/brcm47xx/packages
dest root /
dest ram /tmp
lists_dir ext /var/opkg-lists
option overlay_root /overlay
arch all 1
arch mipsel 200
arch brcm4716 20
arch brcm47xx 10

comment:437 in reply to: ↑ description ; follow-up: Changed 4 years ago by intelminer@…

Replying to Coool:
Tried building from source. Latest patch-set applied fine

Started building, eventually erroring out with

make[1]: Entering directory `/home/intelminer/openwrt'
make[2]: Entering directory `/home/intelminer/openwrt'
make[2]: Entering directory `/home/intelminer/openwrt'
rm -rf /home/intelminer/openwrt/staging_dir/target-mipsel_uClibc-0.9.33.2/root-brcm4716
make[3]: Entering directory `/home/intelminer/openwrt/target/linux'
make[4]: Entering directory `/home/intelminer/openwrt/target/linux/brcm4716'
if [ -s "/home/intelminer/openwrt/build_dir/target-mipsel_uClibc-0.9.33.2/linux-brcm4716/linux-3.3.8/patches/series" ]; then (cd "/home/intelminer/openwrt/build_dir/target-mipsel_uClibc-0.9.33.2/linux-brcm4716/linux-3.3.8"; if quilt --quiltrc=- next >/dev/null 2>&1; then quilt --quiltrc=- push -a; else quilt --quiltrc=- top >/dev/null 2>&1; fi ); fi
make[2]: Leaving directory `/home/intelminer/openwrt'
Applying patch platform/060-ssb-add-serial-flash-driver.patch
patching file drivers/ssb/Kconfig
patching file drivers/ssb/Makefile
patching file drivers/ssb/driver_chipcommon_sflash.c
patching file drivers/ssb/driver_mipscore.c
patching file drivers/ssb/main.c
Hunk #1 succeeded at 19 (offset 1 line).
Hunk #2 succeeded at 541 (offset 6 lines).
patching file drivers/ssb/ssb_private.h
Hunk #1 succeeded at 242 with fuzz 2 (offset 31 lines).
patching file include/linux/ssb/ssb_driver_chipcommon.h
Hunk #4 FAILED at 604.
1 out of 4 hunks FAILED -- rejects in file include/linux/ssb/ssb_driver_chipcommon.h
Patch platform/060-ssb-add-serial-flash-driver.patch does not apply (enforce with -f)
make[4]: * home/intelminer/openwrt/build_dir/target-mipsel_uClibc-0.9.33.2/linux-brcm4716/linux-3.3.8/.quilt_checked Error 1
make[4]: Leaving directory `/home/intelminer/openwrt/target/linux/brcm4716'
make[3]:
* [compile] Error 2
make[3]: Leaving directory `/home/intelminer/openwrt/target/linux'
make[2]: * [target/linux/compile] Error 2
make[2]: Leaving directory `/home/intelminer/openwrt'
make[1]:
* home/intelminer/openwrt/staging_dir/target-mipsel_uClibc-0.9.33.2/stamp/.target_compile Error 2
make[1]: Leaving directory `/home/intelminer/openwrt'

Anyone got any ideas?

comment:438 in reply to: ↑ 437 Changed 4 years ago by intelminer@…

Replying to intelminer@…:

Replying to Coool:

rm -rf'd trunk, re-cloned it, seemed to build correctly, JSON mirror however is offline, build exits due to being unable to download it

comment:439 Changed 4 years ago by evgenyz@…

What blocks adding www.znau.edu.ua/temp/asus-rt-n16 patch set into trunk? I'm very interested and maybe can provide some help.

comment:440 Changed 4 years ago by hauke

  • Resolution set to fixed
  • Status changed from assigned to closed

The Asus rt-n16 and most of the other recent bcm47xx devices should work with trunk now. We have an Ethernet driver for them. Wifi with b43 does not support ieee8011n rates, but n-phy devices and no 5GHz. broadcom-wl should work.

Make sure you build an image with CONFIG_PACKAGE_kmod-bgmac selected or use the image builder to select this package.

If there are still problems please open a new ticket with the problem you are facing, do not reopen this ticket.

comment:441 Changed 4 years ago by gregd72002@…

Are you saying it is over now? What am I supposed to do now?
I've been here almost every day for last 3 years...

comment:442 Changed 4 years ago by anonymous

Welcome to open source purism gregd72002 ;p.

comment:443 Changed 4 years ago by anonymous

gregd72002. I suggest you wait for a binary build. If you want to bite the bullet yourself you need to pull down 'trunk' from the SVN repository to a Linux host, then follow the build instructions. You need to select the correct top level targets using 'make menuconfig' as the build instructions specify, and add the kmod-bgmac as noted in the ticket resolutions details. I will be doing this process later and I assume George will too, so if either of us gets a working build you could use that.

comment:444 Changed 4 years ago by sherca

I am glad to know it is done, thanks for all the work !

Wich output should be used ? openwrt-brcm47xx-squashfs.trx ?

I used Broadcom BCM947xx/953xx as Target System and BCM4705/BCM4785, BCM43xx WiFi (wl, proprietary) as Target Profile.

I understood that no patch is needed now, or am I mistaking ?

Thanks again

comment:445 follow-up: Changed 4 years ago by anonymous

Just built it from trunk #35310. No patches needed. It boots, ether comes up, wifi comes up. I'm doing a re-build with more options, after I have a decent set of binaries I'll post them somewhere and put a link here, and post my config file. Sherca, you also need the kmod-bgmac and the USB options for a sane build.

comment:446 Changed 4 years ago by anonymous

Also, I'm building with B43 not wl

comment:447 Changed 4 years ago by anonymous

Why would you use b43? It's stupid to do that, you won't get n support which for a router is a pretty big deal. There's a reason all rt-n16 builds used wl... Don't reinvent the wheel :)

comment:448 in reply to: ↑ description Changed 4 years ago by intelminer@…

Replying to Coool:

Ooh. Finally!

Going to take a crack at building this over the weekend, and put up binaries + packages etc

comment:449 in reply to: ↑ 445 Changed 4 years ago by anonymous

Replying to anonymous:

Thanks for that. But if I don't need USB, it is not mandatory to add it, right ? It will be ok without it ?

And wich output binary have you used ? openwrt-brcm47xx-squashfs.trx ?

comment:450 Changed 4 years ago by evgenyz@…

Is support only available in trunk or in AA branch as well (will it be if no)?

comment:451 Changed 4 years ago by cchhat01@…

So before I bite as well, I want to ask... will the TRX work for a E4200 v1 as well considering its a Broadcom 4716/4718?

Thanks,
C

comment:452 Changed 4 years ago by anonymous

The RT-N16 has a mips32 R2 processor as opposed to R1 like most brcm47xx. As such, shouldn't a new target be created for the device in openwrt that will select R2?

comment:453 Changed 4 years ago by anonymous

The RT-N16 has a mips32 R2 processor as opposed to R1 like most brcm47xx. As such, shouldn't a new target be created for the device in openwrt that will select R2?

comment:454 Changed 4 years ago by intelminer@…

Built and flashed OpenWRT (compiled from source)

Device came up, seemed to "work" (IE: could connect to via the Web UI)

Wi-Fi would connect, though DHCP didn't seem to pass-through so the devices

Upon rebooting the device however, it is inaccessible.

If anyone whom has theirs hooked up to a JTAG would like to let me flash builds to it/test it, I'd be extremely grateful

comment:455 Changed 4 years ago by hauke

The snapshot releases are now containing the bgmac Ethernet driver and are working at least with my Netgear WNDR3400 and they should also work with the Asus RT-N16. Be aware that the snapshot releases are automatically generated and not tested before being uploaded to the server.

http://downloads.openwrt.org/snapshots/trunk/brcm47xx/

Attitude Adjustment does not have support for these devices like the Asus RT-N16 and will not get support for these devices, Barrier Breaker will be the first stable release with support for these devices.

Last edited 4 years ago by hauke (previous) (diff)

comment:456 Changed 4 years ago by anonymous

Doesn't work, stuck at "starting program at 0x80001000". Incorrect CFE mappings I guess? How to fix this? Could a profile for the rt-n16 be added to trunk?

comment:457 Changed 4 years ago by asusfan@…

Seems this was premature to label the rt-n16 as working in openwrt. Next steps should be to create a dedicated profile for the device and post more information in my opinion.

comment:458 Changed 4 years ago by asusfan@…

  • Resolution fixed deleted
  • Status changed from closed to reopened

reopen to add more info to: http://wiki.openwrt.org/toh/asus/rt-n16 and add profile for rt-n16.

comment:459 Changed 4 years ago by hauke

  • Resolution set to fixed
  • Status changed from reopened to closed

comment:460 Changed 4 years ago by rcienio@…

  • Resolution fixed deleted
  • Status changed from closed to reopened

I installed OpenWrt Barrier Breaker r35409 / LuCI Trunk (trunk+svn9662) on the rt-n16.
I have a problem with

OpenWrt kern.info kernel: [ 89.020000] Found board: "Unknown Board"

Detection has to be done differently. How to fix this?

comment:461 Changed 4 years ago by rcienio@…

I erase nvram and the problem still exists. Detection should be done differently.

comment:462 Changed 4 years ago by hauke

  • Resolution set to fixed
  • Status changed from reopened to closed

@rcienio:
Please open a new ticket for this problem and attach the output of "nvram show" and do not reopen this ticket.

comment:463 Changed 4 years ago by anonymous

Many thanks for getting the RT-N16 support in!

comment:464 Changed 4 years ago by anonymous

Hello, sorry to trouble you all.
Anyone knows how to build openwrt from trunk sources?

I‘ve tried the following steps to try to compile openwrt for rt-n16

get the sources from openwrt svn server, and run

./scripts/feeds update -a
./scripts/feeds install -a
make menuconfig

But problem comes out when doing the config, I found that there is no profile named bcm47xx or rt-n16.

The only broadcom profiles that I found is BCM2708/BCM2835, BCM63XX, and BCM947XX/BCM953XX

So How can I do the menuconfig to finish the config for rt-n16?

Thanks/

comment:465 Changed 3 years ago by anonymous

Use a target system of BCM947xx/953xx and a target profile of Broadcom SoC, all ethernet, BCM43xx wifi (wl, proprietary) and it will build just fine. You will want to remove the b43 and b43 legacy kernel modules to get the wl driver to load correctly. I have mine running two VLANS with a hidden SSID on WPA2 and a visible SSID on WEP (don't ask! ). All seems solid on r36141.

Thanks to all who made this work at last!

comment:466 follow-up: Changed 3 years ago by anonymous

Stellar, building now to give it a try!

One problem, I can't find out how to load the image to the router. I have serial cable connected, but don't know how to use the commands on CFE> to write the image to the RT-N16 and there's no tftpd running on it ... I couldn't find any documents about CFE .

Any help is appreciated.

comment:467 in reply to: ↑ 466 Changed 3 years ago by anonymous

Replying to anonymous:

Stellar, building now to give it a try!

One problem, I can't find out how to load the image to the router. I have serial cable connected, but don't know how to use the commands on CFE> to write the image to the RT-N16 and there's no tftpd running on it ... I couldn't find any documents about CFE .

Any help is appreciated.

Forgot about the recovery mode. Hit the recovery button to get the TFTPD going on the router, then flash.

Add Comment

Modify Ticket

Action
as closed .
The resolution will be deleted. Next status will be 'reopened'.
Author


E-mail address and user name can be saved in the Preferences.

 
Note: See TracTickets for help on using tickets.