Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Phil Muldoon <pmuldoon@redhat.com>
To: Tom Tromey <tromey@redhat.com>
Cc: Doug Evans <dje@google.com>, gdb-patches@sourceware.org
Subject: Re: [patch] Add an evaluation function hook to Python breakpoints.
Date: Wed, 05 Jan 2011 20:23:00 -0000	[thread overview]
Message-ID: <m3bp3vf6c4.fsf@redhat.com> (raw)
In-Reply-To: <m3sjx7urkv.fsf@fleche.redhat.com> (Tom Tromey's message of "Wed,	05 Jan 2011 11:35:12 -0700")

Tom Tromey <tromey@redhat.com> writes:

>>>>>> "Doug" == Doug Evans <dje@google.com> writes:

I just came back from Christmas vacation today, so apologies for this
thread time-gap.

> Doug> The user will see the normal breakpoint output, and then be expected
> Doug> to know that the breakpoint is special (e.g., a rule was violated and
> Doug> that s/he needs to pay attention to what the rule was).  Is there a
> Doug> plan to have something more?  [Or am I missing something?]
>
> Yeah, we want to expose some of the breakpoint_ops stuff to python.

Yeah ultimately this was going to be series of patches (the first being
this, the ability to write conditions entirely within Python), but the
whole-story (from these threads) has probably shown we need to think
about the solution in totality, not incrementally. 

> Doug> Sorry, maybe I was too subtle.
> Doug> If one is using this facility to implement a rule checker that stops
> Doug> when the rule is violated, and one sets another breakpoint at the same
> Doug> location, when the program stops how does one know whether the rule
> Doug> was violated or not?
>
> One does not.  This is a pre-existing problem in gdb.
>
>
> Phil, let's withdraw this patch until some later time when we have
> worked out the issues.  We can continue to discuss on the archer list.
>
> I think we should at least deal with exposing to Python the bits needed
> to customize breakpoint notifications.
>
> Also we should write a couple real-life uses of this to make sure it is
> working ok.  

That's ok, no worries.  On that note, I keep a separate branch of my
Python breakpoints improvements patches over on archer within the
archer-pmuldoon-python-breakpoints branch.  That is always open for
folks to pull and comment on.  

We need to determine a concrete API plan though.  What do we need these
improvements to do? The only concrete action I saw was the ability to
override breakpoint operations.  (And probably a callback into the
python api to return the appropriate message). 

What else?

Cheers

Phil


  reply	other threads:[~2011-01-05 20:23 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-13 13:50 Phil Muldoon
2010-12-13 14:19 ` Eli Zaretskii
2010-12-13 14:47   ` Phil Muldoon
2010-12-13 15:07     ` Eli Zaretskii
2010-12-13 17:21       ` Phil Muldoon
2010-12-13 17:46         ` Eli Zaretskii
2010-12-13 14:33 ` Pedro Alves
2010-12-13 14:56   ` Phil Muldoon
2010-12-13 15:07     ` Pedro Alves
2010-12-13 20:45 ` Doug Evans
2010-12-13 21:02   ` Phil Muldoon
2010-12-14  3:31     ` Doug Evans
2010-12-14 17:18       ` Phil Muldoon
2010-12-14 17:28   ` Tom Tromey
2010-12-14 19:51     ` Phil Muldoon
2010-12-14 20:00       ` Phil Muldoon
2010-12-15 15:34     ` Phil Muldoon
2010-12-15 20:51       ` Tom Tromey
2011-01-27 12:44         ` Phil Muldoon
     [not found]           ` <AANLkTimi6ugruNAqUGHni8Kvkz+B5-s2aAkEoTY2D_gT@mail.gmail.com>
2011-01-27 21:40             ` Phil Muldoon
2011-01-28 10:42           ` Tom Tromey
2010-12-15 16:21     ` Doug Evans
2010-12-15 20:57       ` Tom Tromey
2010-12-21 17:33         ` Doug Evans
2010-12-21 20:02           ` Tom Tromey
2010-12-22 16:34             ` Doug Evans
2010-12-22 17:35               ` Tom Tromey
2010-12-28  5:53                 ` Doug Evans
2011-01-05 18:35                   ` Tom Tromey
2011-01-05 20:23                     ` Phil Muldoon [this message]
2011-01-09 20:32                       ` Doug Evans
2010-12-14 17:46   ` Pedro Alves
2010-12-14 16:35 ` Tom Tromey
2010-12-14 17:02   ` Phil Muldoon
2010-12-14 17:48     ` Tom Tromey
2010-12-14 16:42 ` Tom Tromey

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=m3bp3vf6c4.fsf@redhat.com \
    --to=pmuldoon@redhat.com \
    --cc=dje@google.com \
    --cc=gdb-patches@sourceware.org \
    --cc=tromey@redhat.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