From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gdb-patches-return-60158-listarch-gdb-patches=sources.redhat.com@sourceware.org>
Received: (qmail 11772 invoked by alias); 20 Nov 2008 08:33:10 -0000
Received: (qmail 11726 invoked by uid 22791); 20 Nov 2008 08:33:08 -0000
X-Spam-Check-By: sourceware.org
Received: from ti-out-0910.google.com (HELO ti-out-0910.google.com) (209.85.142.189)     by sourceware.org (qpsmtpd/0.31) with ESMTP; Thu, 20 Nov 2008 08:32:30 +0000
Received: by ti-out-0910.google.com with SMTP id d10so178712tib.12         for <gdb-patches@sourceware.org>; Thu, 20 Nov 2008 00:32:27 -0800 (PST)
Received: by 10.110.69.5 with SMTP id r5mr2547597tia.36.1227169947467;         Thu, 20 Nov 2008 00:32:27 -0800 (PST)
Received: by 10.110.103.3 with HTTP; Thu, 20 Nov 2008 00:32:27 -0800 (PST)
Message-ID: <daef60380811200032j122a90d6y9bf673ef19c7f174@mail.gmail.com>
Date: Thu, 20 Nov 2008 16:37:00 -0000
From: teawater <teawater@gmail.com>
To: "Pedro Alves" <alves.ped@gmail.com>
Subject: Re: [RFA] Resubmit process record and replay, 7/10
Cc: gdb-patches@sourceware.org
In-Reply-To: <200811200250.22173.alves.ped@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
References: <daef60380811160023ne49a1ebt60058c5348dc4bc4@mail.gmail.com> 	 <200811200240.47107.pedro@codesourcery.com> 	 <200811200250.22173.alves.ped@gmail.com>
X-IsSubscribed: yes
Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm
Precedence: bulk
List-Id: <gdb-patches.sourceware.org>
List-Subscribe: <mailto:gdb-patches-subscribe@sourceware.org>
List-Archive: <http://sourceware.org/ml/gdb-patches/>
List-Post: <mailto:gdb-patches@sourceware.org>
List-Help: <mailto:gdb-patches-help@sourceware.org>, <http://sourceware.org/ml/#faqs>
Sender: gdb-patches-owner@sourceware.org
X-SW-Source: 2008-11/txt/msg00549.txt.bz2

Thanks Pedro,


Agree with your idea in record_wait part.  Actually, I plan to do it
after P record check-in in the before.
Now, I will do it  at once. :)

On the other hand, could you let me keep the code in linux-nat.c.
Because I think it can make linux-nat record speed up.

What about write a special function to check the breakpoint? Then both
linux-nat and record_wait can call it.


Hui

On Thu, Nov 20, 2008 at 10:50, Pedro Alves <alves.ped@gmail.com> wrote:
> On Thursday 20 November 2008 02:40:46, Pedro Alves wrote:
>
>> Can you please try moving all this blob...
>
> [...]
>
>> ... to the record target?  It seems to be interested in getting
>> *all* events, instead of letting linux_nat_wait filter some.
>
> Sorry, reading back, I don't think I explained myself that well.  I meant
> something like:
>
> ptid_t
> record_wait (ptid_t ptid, target_waitstatus *status)
> {
>  while (1)
>  {
>     eptid = beneath->to_wait (ptid, &status);
>
>     if (status.kind == TARGET_WAITKIND_TRAP)
>      {
>         if breakpoint_here (pc - pc_adjustement)
>           {
>             got breakpoint;
>             break;
>           }
>         record_message ();
>         beneath->resume (step);
>      }
>      else
>        break probably.
>  }
>
>  if (got breakpoint)
>    adjust_pc;
>  return eptid;
> }
>
> --
> Pedro Alves
>