Scope of tracing for gsoc project

[Sanskar Khandelwal]

On Wed, Feb 10, 2021 at 12:11 PM Nenad Obradović <
nenad.obradovic1985 at gmail.com> wrote:

> Couple of weeks ago I was trying to port RTEMS on HIFive Unleashed (dev.
> board). I stopped doing it because of lack of time.
> I wanted to merge HiFIVEs FSBL and RTEMS riscv Start.s so that the image
> is able to run from a monolithic elf file from QSPI memory and it is
> debuggable from FreedomStudio easily. I managed to get to the point where
> tasks are created and run, but for some reason memory allocation breaks and
> it crashes (on fileio example) and I left it for time when I'm available
> again (probably spring/summer).
> If you are interested in that code, I can just send it to you.
>

I am interested in this and please send, I hope it will help me understand
more about this task also if you can share some resource to study it will
be big help.


>
> On Tue, 9 Feb 2021 at 13:13, Joel Sherrill <joel at rtems.org> wrote:
>
>>
>>
>> On Tue, Feb 9, 2021, 4:01 AM Sanskar Khandelwal <kdsanskar07 at gmail.com>
>> wrote:
>>
>>> Hello there,
>>>
>>> I am interested in contributing to a few topics but I don't know what is
>>> the current status and future enhancements you are looking for so if you
>>> can guide me it would be a big help.
>>>
>>>  1. Tracing
>>>            1. #3028 :  Run-Time Tracing
>>>
>>
>> This works for getting a snapshot of tracing information from the target
>> system, converting it to the Eclipse input format, and displaying it. We
>> have a utility in rtems tools which listens to the target from the network
>> and writes a file
>>
>> What's missing now is live tracing. Look that up for Linux Trace Toolkit.
>> They have a utility which listens to the target and writes the live stream
>> for Eclipse. For RTEMS, our utility or another one derived from it would
>> have to listen to the target and write the stream. The target code to send
>> the data may need tinkering to do it continuously.
>>
>> Live streaming should be a matter of assembling pieces and filling in
>> gaps. You would want to start by using Eclipse to see streams from RTEMS as
>> it does them today and live stream from Linux.
>>
>>
>>>            2 .#3326 : Eclipse Target Communication Framework Support
>>>
>>
>> I worked some on TCF this past year and what is going to need work isn't
>> worth the payoff. We would need to rework libdebugger to integrate with the
>> gdb command processor it provides.
>>
>>
>>>  2.  #4162 : SiFive RISC-V HiFive Unleashed BSP (Qemu)
>>>
>>
>> I don't know how much can be reused from other BSPs. Would need to verify
>> how much of the board is simulated by qemu and what peripherals are porting
>> versus new
>>
>
I researched(i.e googling) about this project while doing so I learned a
lot of things too, It was very interesting though but the thing I struggled
is that I did not know what to look for actually I wanted to find the
answer for the above questions you put, if you can tell me little more
about this questions i can look them for you and then we can decide whether
I should move ahead in this direction.


>
>> We do not have a RISC-V BSP with networking so this is nice. We need to
>> assess whether this is doable or how much in GSoC timeframe
>>
>
Since I am new in this I don't know about the doable part but I can assure
you that I will work as hard as I can, and also if you think this can not
be done in a gsoc timeframe I can do some of it in gsoc then some after it.

So I want to know what are actually the tasks for this project and also
while I am learning more about this, can I do some other things.
I saw you float some tickets related to wafblocker can I do one of them.


>
>> Good project though
>>
>>
>>  3.  #3302 : Build System conversion of BSP Config (.cfg) files to
>>> pkg-config (.pc) files
>>>
>>
>> I don't know the status of this. I thought we had these working. At least
>> I don't know what's broken.
>>
>>>
>>> _______________________________________________
>> devel mailing list
>> devel at rtems.org
>> http://lists.rtems.org/mailman/listinfo/devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20210211/358fb40b/attachment.html>