git clone https://github.com/RobertCNelson/ti-linux-kernel-dev.git
cd ti-linux-kernel-dev
checkout ti-linux-4.1.y
./build-kernel.sh

When you see menuconfig, enable samples/rpmsg under "Kernel Hacking”. Press 
exit and allow the build to continue. The sample code will be in 
KERNEL/samples/rpmsg.

Regards,
John




> On Oct 27, 2015, at 2:58 PM, Rick Mann <[email protected]> wrote:
> 
> Sorry for the dumb question, but where's the kernel repo and branch that this 
> patch applies to (or where it has already been applied)? And where are these 
> examples, in the -ti branches?
> 
>> On Oct 27, 2015, at 14:40 , John Syne <[email protected]> wrote:
>> 
>> Hi Rick,
>> 
>> In samples/rpmsg, you will find a way to test pruss-pru/virtio_bus. You need 
>> to configure your kernel to build the sample/rpmsg. Use "make menuconfig” 
>> and under "Kernel hacking”/“Sample kernel code”, enable “Build rpmsg client 
>> sample”. 
>> 
>> Next, you need to build the PRU firmware from"
>> 
>> https://git.ti.com/pru-software-support-package
>> 
>> If you are using BBB, build examples/am335x/PRU_RPMsg_Echo_Interrupt0” and 
>> the relabel the output to am335x-pru0-fw and place it in the /lib/firmware 
>> folder. 
>> 
>> 
>> After you install the kernel module, run the sample with:
>> 
>> modprobe rpmsg_client_sample
>> 
>> This will transmit 100 “hello world” message to the PRU which receives an 
>> interrupt, copies the message and then sends the same message back to Linux. 
>> You will see this in dmesg. 
>> 
>> 
>> Regards,
>> John
>> 
>> 
>> 
>> 
>>> On Oct 27, 2015, at 2:16 PM, Rick Mann <[email protected]> wrote:
>>> 
>>> 
>>>> On Oct 27, 2015, at 11:23 , John Syne <[email protected]> wrote:
>>>> 
>>>> Hi Rich,
>>>> 
>>>> Here is the update Robert added to his repo yesterday.
>>>> 
>>>> https://github.com/RobertCNelson/ti-linux-kernel-dev/blob/393f5ea242a86544063ec2af63ef270094bfa1da/patches/pru/0003-Add-rpmsg_pru-support.patch
>>> 
>>> Thanks for this, John. I can't wait to give it a shot. Is there an example 
>>> somewhere of how to use it? One of the challenges I face is modifying the 
>>> LEDScape code to use this instead of pruss. However, it might force me to 
>>> just write my own PRU code to do what I do with the PRU. In both cases, 
>>> I'll probably need an example or other documentation on how to actually 
>>> make it work.
>>> 
>>> Thanks,
>>> Rick
>>> 
>>>> 
>>>> Regards,
>>>> John
>>>> 
>>>> 
>>>> 
>>>> 
>>>>> On Oct 26, 2015, at 8:50 PM, William Hermans <[email protected]> wrote:
>>>>> 
>>>>> I have it working Rick on . . .
>>>>> $ uname -a
>>>>> Linux beaglebone 4.1.9-bone-rt-r16 #1 Thu Oct 1 06:19:41 UTC 2015 armv7l 
>>>>> GNU/Linux
>>>>> 
>>>>> This kernel was just an apt-get install. But, it's not a TI specific 
>>>>> kernel if that's what you're wanting. As far as I can tell though, it's 
>>>>> identical except that the pru stuff was re-enabled by Robert at our 
>>>>> request some days ago. Couple weeks maybe ?
>>>>> 
>>>>> 
>>>>> On Mon, Oct 26, 2015 at 7:29 PM, Rick Mann <[email protected]> wrote:
>>>>> Hi John,
>>>>> 
>>>>> Did you ever resolve this?
>>>>> 
>>>>>> On Oct 19, 2015, at 11:08 , John Syne <[email protected]> wrote:
>>>>>> 
>>>>>> Hi Rich,
>>>>>> 
>>>>>> I have this working on BBB using V4.1.6-bone15, but I should receive an 
>>>>>> x15 sometime today and it will take me a few days to do testing. There 
>>>>>> is however one hick-up that needs input from Pantelis to resolve. When 
>>>>>> BBB first boots, remoteproc/virtio_rpmsg_bus load automatically and the 
>>>>>> PRU firmware loads as before. The sample code under samples/rpmsg work 
>>>>>> work perfectly. However, if I uninstall remoteproc.ko and then reinstall 
>>>>>> it, I get a kernel exception.
>>>>>> 
>>>>>> Investigating further, I found that _omap_device_notifier_call sets 
>>>>>> pdev->archdata.od = NULL when the pruss_remoteproc is uninstalled, but 
>>>>>> when I install the pruss_remoteproc, pruss_probe tries to run 
>>>>>> omap_device_deassert_hardreset which references archdata.od which is is 
>>>>>> now NULL.
>>>>>> 
>>>>>> Looking at TI’s V3.14 kernel, it does not set archdata.od = NULL, so I 
>>>>>> can uninstall and install the KO without issue.
>>>>>> 
>>>>>> In the V4.1 code, if I comment out lines 57 & 60 [1], pruss_remoteproc 
>>>>>> uninstalls and installs without error.
>>>>>> 
>>>>>> My question is how do I recreate archdata.od before pruss_probe executes 
>>>>>> omap_device_deassert_hardreset? Given that this patch was authored by 
>>>>>> Pantelis, I figured you would be the best person to answer my question.
>>>>>> 
>>>>>> [1] 
>>>>>> https://github.com/RobertCNelson/linux-dev/blob/master/patches/bbb_overlays/0006-arm-omap-Proper-cleanups-for-omap_device.patch
>>>>>> 
>>>>>> 
>>>>>> Regards,
>>>>>> John
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>>> On Oct 18, 2015, at 7:11 PM, Robert Nelson <[email protected]> 
>>>>>>> wrote:
>>>>>>> 
>>>>>>> On Sun, Oct 18, 2015 at 8:52 PM, Rick Mann <[email protected]> 
>>>>>>> wrote:
>>>>>>>> 
>>>>>>>>> On Oct 18, 2015, at 17:11 , Robert Nelson <[email protected]> 
>>>>>>>>> wrote:
>>>>>>>>> 
>>>>>>>>> On Sun, Oct 18, 2015 at 7:05 PM, Rick Mann <[email protected]> 
>>>>>>>>> wrote:
>>>>>>>>>> Pruss is still broken in the 4.1.10-ti-r21 kernel, right?
>>>>>>>>>> 
>>>>>>>>>> Does anyone know if the new remoteproc thing can be used to download 
>>>>>>>>>> existing firmware that was built with pruss in mind, and how I could 
>>>>>>>>>> modify my code to use that instead?
>>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> another user got it working with remoteproc, he's just waiting for a
>>>>>>>>> beta-x15 to arrive so that they will work the same on both boards..
>>>>>>>> 
>>>>>>>> Oh, do you remember who that was?
>>>>>>> 
>>>>>>> Yeap, he will post it when he's ready.
>>>>>>> 
>>>>>>> Regards,
>>>>>>> 
>>>>>>> --
>>>>>>> Robert Nelson
>>>>>>> https://rcn-ee.com/
>>>>>>> 
>>>>>>> --
>>>>>>> For more options, visit http://beagleboard.org/discuss
>>>>>>> ---
>>>>>>> You received this message because you are subscribed to the Google 
>>>>>>> Groups "BeagleBoard" group.
>>>>>>> To unsubscribe from this group and stop receiving emails from it, send 
>>>>>>> an email to [email protected].
>>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>> 
>>>>>> --
>>>>>> For more options, visit http://beagleboard.org/discuss
>>>>>> ---
>>>>>> You received this message because you are subscribed to the Google 
>>>>>> Groups "BeagleBoard" group.
>>>>>> To unsubscribe from this group and stop receiving emails from it, send 
>>>>>> an email to [email protected].
>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>> 
>>>>> 
>>>>> --
>>>>> Rick Mann
>>>>> [email protected]
>>>>> 
>>>>> 
>>>>> --
>>>>> For more options, visit http://beagleboard.org/discuss
>>>>> ---
>>>>> You received this message because you are subscribed to the Google Groups 
>>>>> "BeagleBoard" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send an 
>>>>> email to [email protected].
>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>> 
>>>>> 
>>>>> -- 
>>>>> For more options, visit http://beagleboard.org/discuss
>>>>> --- 
>>>>> You received this message because you are subscribed to the Google Groups 
>>>>> "BeagleBoard" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send an 
>>>>> email to [email protected].
>>>>> For more options, visit https://groups.google.com/d/optout.
>>>> 
>>>> 
>>>> -- 
>>>> For more options, visit http://beagleboard.org/discuss
>>>> --- 
>>>> You received this message because you are subscribed to the Google Groups 
>>>> "BeagleBoard" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send an 
>>>> email to [email protected].
>>>> For more options, visit https://groups.google.com/d/optout.
>>> 
>>> 
>>> -- 
>>> Rick Mann
>>> [email protected]
>>> 
>>> 
>>> -- 
>>> For more options, visit http://beagleboard.org/discuss
>>> --- 
>>> You received this message because you are subscribed to the Google Groups 
>>> "BeagleBoard" group.
>>> To unsubscribe from this group and stop receiving emails from it, send an 
>>> email to [email protected].
>>> For more options, visit https://groups.google.com/d/optout.
>> 
>> -- 
>> For more options, visit http://beagleboard.org/discuss
>> --- 
>> You received this message because you are subscribed to the Google Groups 
>> "BeagleBoard" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected].
>> For more options, visit https://groups.google.com/d/optout.
> 
> 
> -- 
> Rick Mann
> [email protected]
> 
> 
> -- 
> For more options, visit http://beagleboard.org/discuss
> --- 
> You received this message because you are subscribed to the Google Groups 
> "BeagleBoard" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.

-- 
For more options, visit http://beagleboard.org/discuss
--- 
You received this message because you are subscribed to the Google Groups 
"BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to