public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Pedro Alves <palves@redhat.com>
To: Patrick Palka <patrick@parcs.ath.cx>, Yao Qi <qiyaoltc@gmail.com>
Cc: "gdb-patches@sourceware.org" <gdb-patches@sourceware.org>
Subject: Re: [PATCH] Unset attach_flag when running a new process
Date: Thu, 13 Aug 2015 16:30:00 -0000	[thread overview]
Message-ID: <55CCC619.3040207@redhat.com> (raw)
In-Reply-To: <CA+C-WL9DD9GgNqQkB5N1Qv7OAGRTtn2z7N0KNt9dqhMmaNgT_A@mail.gmail.com>

On 08/02/2015 09:20 PM, Patrick Palka wrote:
> On Thu, Jul 30, 2015 at 12:24 PM, Yao Qi <qiyaoltc@gmail.com> wrote:
>> On 30/07/15 16:28, Patrick Palka wrote:
>>>
>>> +  /* Unset attach_flag, it may be set if the previous process associated
>>> with
>>> +     the inferior was attached to.  */
>>> +  current_inferior ()->attach_flag = 0;
>>
>>
>> It is better to do such reset in target.c:target_pre_inferior, which
>> is created for such purpose, as far as I know.
> 
> That makes sense, although I am a bit hesitant about such a change,
> since target_pre_inferior has a lot of callers.  

You mean, the whole lot of 3 callers.  :-)

infcmd.c:542:  target_pre_inferior (from_tty);
infcmd.c:2603:  target_pre_inferior (from_tty);
target.c:2191:  target_pre_inferior (from_tty);

> Is it OK, for
> instance, to assume that current_inferior points to the right inferior
> when target_pre_inferior gets called?  Currently current_inferior is
> not called directly or indirectly from target_pre_inferior.

Should be.  It's called from the run command, and from the
attach command.  Those assume that the current inferior is
the inferior that we're about to run/attach.

The call in target_preopen is more dubious.  That should probably
be iterating over all inferiors and calling target_pre_inferior
on each.  For the case where you have multiple inferiors, and then
connect to a remote target with "target remote".



> +set test "kill process"
> +send_gdb "kill\n"
> +gdb_expect {
> +    -re ".y or n. $" {
> +	send_gdb "y\n"
> +	exp_continue
> +    }
> +    -re "$gdb_prompt $" {
> +	pass $test
> +    }
> +    default {
> +	fail $test
> +    }
> +}

Avoid raw gdb_expect.  You can write:

        gdb_test "kill" \
            "" \
            "kill process" \
            "Kill the program being debugged.*y or n. $" \
            "y"

> +
> +set test "restart process"
> +gdb_test "start" "Starting program.*Temporary breakpoint .*" $test
> +
> +set test "attempt kill via quit"
> +send_gdb "quit\n"
> +set ok 0
> +# The quit prompt should warn about killing the process, not about detaching the
> +# process, since this process was not attached to.
> +gdb_expect {

Please use gdb_test_multiple.  That catches internal errors and such.

> +    -re "will be killed.*.y or n. $" {
> +	set ok 1
> +	send_gdb "n\n"
> +	exp_continue
> +    }
> +    -re "will be detached.*.y or n. $" {
> +	send_gdb "n\n"
> +	exp_continue
> +    }
> +    -re "$gdb_prompt $" {
> +	if $ok {
> +	    pass $test
> +	} else {
> +	    fail $test
> +	}

gdb_assert $ok $test


> +    }
> +    default {
> +	fail $test
> +    }
> +}
> +
> +remote_exec host "kill $testpid"
> 

Thanks,
Pedro Alves

  reply	other threads:[~2015-08-13 16:30 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-30  1:24 Patrick Palka
2015-07-30 15:28 ` Patrick Palka
2015-07-30 16:24   ` Yao Qi
2015-08-02 20:20     ` Patrick Palka
2015-08-13 16:30       ` Pedro Alves [this message]
2015-08-13 17:19         ` Patrick Palka
2015-08-17 13:01           ` Patrick Palka
2015-08-26 16:05             ` Patrick Palka
2015-08-26 17:06             ` Pedro Alves
2015-08-27  0:56               ` Patrick Palka
2015-09-16 16:34             ` Doug Evans
2015-08-13 15:12   ` Patrick Palka

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=55CCC619.3040207@redhat.com \
    --to=palves@redhat.com \
    --cc=gdb-patches@sourceware.org \
    --cc=patrick@parcs.ath.cx \
    --cc=qiyaoltc@gmail.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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).