Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Pedro Alves <palves@redhat.com>
To: Yao Qi <qiyaoltc@gmail.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH 04/12] Delete reinsert breakpoints from forked child
Date: Tue, 14 Jun 2016 11:40:00 -0000	[thread overview]
Message-ID: <d9e32b96-41b2-e8e1-dfeb-dd26525caa77@redhat.com> (raw)
In-Reply-To: <86porkt3yy.fsf@gmail.com>

On 06/14/2016 12:16 PM, Yao Qi wrote:

>> >
>>> >> I think
>>> >> they've already covered by gdb.base/step-over-syscall.exp.
>> >
>> > In that case, shouldn't we be extending that test instead?
> OK, I extend step-over-syscall.exp by setting different
> combinations of follow-fork and detach-on-fork modes, and it still can
> trigger GDBserver internal errors.  How about the patch below?  

Fine with me.  One suggestion below.

> +    foreach_with_prefix detach-on-fork {"on" "off"} {
> +	foreach_with_prefix follow-fork {"parent" "child"} {
> +	    foreach syscall { "fork" "vfork" "clone" } {
> +
> +		if { $syscall != "vfork"
> +		     || ${follow-fork} != "parent"
> +		     || ${detach-on-fork} != "off" } {
> +		    # Both vforked child process and parent process are
> +		    # under GDB's control, but GDB follows the parent
> +		    # process only, which can't be run until vforked child
> +		    # finishes.  Skip the test in this scenario.
> +		    break_cond_on_syscall $syscall ${follow-fork} ${detach-on-fork}
> +		}

I'd suggest reversing the condition logic, making it match the
comment, like this:

		if { $syscall == "vfork"
		     && ${follow-fork} == "parent"
		     && ${detach-on-fork} == "off" } {
		    # Both vforked child process and parent process are
		    # under GDB's control, but GDB follows the parent
		    # process only, which can't be run until vforked child
		    # finishes.  Skip the test in this scenario.
                    continue
		}

	        break_cond_on_syscall $syscall ${follow-fork} ${detach-on-fork}


This way makes it easier to add more "skip" conditions in the future too.  E.g.,

		if { $syscall == "vfork"
		     && ${follow-fork} == "parent"
		     && ${detach-on-fork} == "off" } {
		    # Both vforked child process and parent process are
		    # under GDB's control, but GDB follows the parent
		    # process only, which can't be run until vforked child
		    # finishes.  Skip the test in this scenario.
                    continue
		}

+		if { $syscall == "whatnot"
+		     && ${follow-fork} == "parent"} {
+		    # For whatever reason, this shouldn't be tested either.
+                   continue
+		}

	        break_cond_on_syscall $syscall ${follow-fork} ${detach-on-fork}

> +	    }
> +	}
> +    }
>  }

Thanks,
Pedro Alves


  reply	other threads:[~2016-06-14 11:40 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-02  9:31 [PATCH 00/12 V2] Use reinsert breakpoint for vCont;s Yao Qi
2016-06-02  9:31 ` [PATCH 07/12] Create sub classes of 'struct breakpoint' Yao Qi
2016-06-13 15:09   ` Pedro Alves
2016-06-02  9:31 ` [PATCH 09/12] Make reinsert_breakpoint thread specific Yao Qi
     [not found]   ` <71a5322e-41e3-9e23-df73-e14b14c1d656@redhat.com>
2016-06-14 12:52     ` Yao Qi
2016-06-14 12:57       ` Pedro Alves
2016-06-02  9:31 ` [PATCH 08/12] Refactor clone_all_breakpoints Yao Qi
2016-06-13 15:14   ` Pedro Alves
2016-06-02  9:31 ` [PATCH 01/12] Switch to current thread in finish_step_over Yao Qi
2016-06-13 14:25   ` Pedro Alves
2016-06-02  9:31 ` [PATCH 03/12] Step over exit with reinsert breakpoints Yao Qi
2016-06-13 14:37   ` Pedro Alves
2016-06-13 14:52     ` Yao Qi
2016-06-13 15:01       ` Pedro Alves
2016-06-17  9:50         ` Yao Qi
2016-06-02  9:31 ` [PATCH 06/12] Pass breakpoint type in set_breakpoint_at Yao Qi
2016-06-13 15:07   ` Pedro Alves
2016-06-02  9:31 ` [PATCH 05/12] Handle reinsert breakpoints for vforked child Yao Qi
2016-06-13 15:07   ` Pedro Alves
2016-06-02  9:31 ` [PATCH 10/12] Switch current_thread to lwp's thread in install_software_single_step_breakpoints Yao Qi
2016-06-13 15:26   ` Pedro Alves
2016-06-02  9:31 ` [PATCH 11/12] Use reinsert_breakpoint for vCont;s Yao Qi
2016-06-13 15:55   ` Pedro Alves
2016-06-14 13:14     ` Yao Qi
2016-06-14 15:48       ` Pedro Alves
2016-06-15 16:41         ` Yao Qi
2016-06-17 15:10           ` Pedro Alves
2016-06-20 18:09             ` Antoine Tremblay
2016-06-02  9:31 ` [PATCH 04/12] Delete reinsert breakpoints from forked child Yao Qi
2016-06-13 15:02   ` Pedro Alves
2016-06-13 16:53     ` Yao Qi
2016-06-13 17:29       ` Pedro Alves
2016-06-14 11:17         ` Yao Qi
2016-06-14 11:40           ` Pedro Alves [this message]
2016-06-17  9:53             ` Yao Qi
2016-06-02  9:31 ` [PATCH 02/12] More assert checks on reinsert breakpoint Yao Qi
2016-06-13 14:25   ` Pedro Alves
2016-06-02  9:31 ` [PATCH 12/12] Support vCont s and S actions with software single step Yao Qi
2016-06-13 15:56   ` Pedro Alves

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=d9e32b96-41b2-e8e1-dfeb-dd26525caa77@redhat.com \
    --to=palves@redhat.com \
    --cc=gdb-patches@sourceware.org \
    --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