From: sumanth <sumanth.gundapneni@redpinesignals.com>
To: Michael Snyder <msnyder@vmware.com>
Cc: gdb <gdb@sourceware.org>
Subject: Re: TRACE implementation
Date: Fri, 23 Oct 2009 15:54:00 -0000 [thread overview]
Message-ID: <4AE132F6.9010800@redpinesignals.com> (raw)
In-Reply-To: <4ADF17B1.6030100@vmware.com>
Hi Michael,
Very pleased with the information you provided.
Still I have certain doubts in mind.
Tracepoints can b implemented through either remote
stubs or gdbserver or gdb agents (byte-coding).
Is this
true?..................................................
which can be best suited to me as I am accessing my
fpga/processor through a parallel port jtag. I dont want to
burden my processor by running a server on it.
The problem here , I googled in net and I didn't
even find a single prototype to have a fell of how to trace data(trace
points) on my target .
Thanks,
Sumanth G
Michael Snyder wrote:
> sumanth wrote:
>> Hi ,
>>
>> I am talking about "gdb tracepoints" only. I am asking for
>> "How can I implement them in GDB ?"
>
> Great. Thanks for the clarification.
> Partial answer: you don't have to implement them in gdb.
> They're already implemented in gdb.
>
>> means, where I can start with. Heard something about remote stubs
>
> Yes, that's correct. GDB already knows how to tell a stub
> to collect trace data. You need to implement a stub that
> understands the trace commands that gdb will send to it.
>
>> Michael , Can you be a bit more brief on "published API" which
>> you mentioned in your last response.
>
> Sure. First, if you haven't seen this already, you probably
> want to look at it: http://sourceware.org/gdb/talks/esc-west-1999
> There is a white paper, a slide presentation, and other documents.
> It's all old, but not obsolete, since very little additional
> development has been done since then.
>
> Second, the remote protocol messages that your stub will
> need to implement are specified here:
>
> http://sourceware.org/gdb/current/onlinedocs/gdb_37.html#SEC683
>
> I should let you know that some folks at Code Sourcery have
> recently been doing some work with tracepoints. If you were
> interested in hiring someone to do the work for you on contract,
> either I or one of the other maintainers could probably put you
> in contact with someone privately.
>
> Good luck,
> Michael
>
>
>
>
next prev parent reply other threads:[~2009-10-23 4:46 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-20 9:06 sumanth
2009-10-20 19:33 ` Michael Snyder
2009-10-21 5:41 ` sumanth
2009-10-21 14:26 ` Michael Snyder
2009-10-21 14:52 ` Nicholas Mc Guire
2009-10-23 15:54 ` sumanth [this message]
2009-10-23 16:48 ` Michael Snyder
2009-10-24 14:13 ` sumanth
[not found] <F7CE05678329534C957159168FA70DEC5157AEBFC9@EUSAACMS0703.eamcs.ericsson.se>
2009-10-21 14:22 ` Dominique Toupin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4AE132F6.9010800@redpinesignals.com \
--to=sumanth.gundapneni@redpinesignals.com \
--cc=gdb@sourceware.org \
--cc=msnyder@vmware.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox