public inbox for systemtap@sourceware.org
 help / color / mirror / Atom feed
From: Masami Hiramatsu <mhiramat@redhat.com>
To: Roland McGrath <roland@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>, lkml <linux-kernel@vger.kernel.org>,
	        systemtap <systemtap@sources.redhat.com>,
	        DLE <dle-develop@lists.sourceforge.net>
Subject: Re: [PATCH -tip 2/3] Add coredump tracepoint
Date: Sat, 14 Nov 2009 00:00:00 -0000	[thread overview]
Message-ID: <4AFDF305.20903@redhat.com> (raw)
In-Reply-To: <20091113233912.192A3100E@magilla.sf.frob.com>

Roland McGrath wrote:
> I can't really see what this has to do with "sched" to warrant that name.
> But, whatever.
>
> Note that you put the tracepoint where it won't get called in the various
> cases where no dump is really being made because of RLIMIT_CORE or file
> failures.  I suspect you would like to get those reported.  (Perhaps
> especially so, since there won't be any file around to notice later.)

Exactly, yes.

> Also, it seems nice to give the tracepoint the chance to look at the actual
> open file in case a fancy one wants to do that.

Ah, that's very nice to me! thanks!

>
> e.g.
>
> diff --git a/fs/exec.c b/fs/exec.c
> index ba112bd..0000000 100644
> --- a/fs/exec.c
> +++ b/fs/exec.c
> @@ -1822,9 +1822,7 @@ void do_coredump(long signr, int exit_co
>   	ispipe = format_corename(corename, signr);
>   	unlock_kernel();
>
> -	if ((!ispipe)&&  (core_limit<  binfmt->min_coredump))
> -		goto fail_unlock;
> -
> +	file = NULL;
>    	if (ispipe) {
>   		if (core_limit == 0) {
>   			/*
> @@ -1845,7 +1843,7 @@ void do_coredump(long signr, int exit_co
>   				"Process %d(%s) has RLIMIT_CORE set to 0\n",
>   				task_tgid_vnr(current), current->comm);
>   			printk(KERN_WARNING "Aborting core\n");
> -			goto fail_unlock;
> +			goto nopipe;
>   		}
>
>   		dump_count = atomic_inc_return(&core_dump_count);
> @@ -1853,14 +1851,14 @@ void do_coredump(long signr, int exit_co
>   			printk(KERN_WARNING "Pid %d(%s) over core_pipe_limit\n",
>   			       task_tgid_vnr(current), current->comm);
>   			printk(KERN_WARNING "Skipping core dump\n");
> -			goto fail_dropcount;
> +			goto nopipe;
>   		}
>
>   		helper_argv = argv_split(GFP_KERNEL, corename+1,&helper_argc);
>   		if (!helper_argv) {
>   			printk(KERN_WARNING "%s failed to allocate memory\n",
>   			       __func__);
> -			goto fail_dropcount;
> +			goto nopipe;
>   		}
>
>   		core_limit = RLIM_INFINITY;
> @@ -1870,13 +1868,19 @@ void do_coredump(long signr, int exit_co
>   				&file)) {
>    			printk(KERN_INFO "Core dump to %s pipe failed\n",
>   			       corename);
> -			goto fail_dropcount;
> +			goto nopipe;
>    		}
> - 	} else
> + 	} else if (core_limit>= binfmt->min_coredump) {
>    		file = filp_open(corename,
>   				 O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag,
>   				 0600);
> -	if (IS_ERR(file))
> +	}
> +
> +nopipe:
> +	trace_process_coredump((int) signr, core_limit, mm_flags,
> +			       corename, file);
> +
> +	if (!file || IS_ERR(file))
>   		goto fail_dropcount;
>   	inode = file->f_path.dentry->d_inode;
>   	if (inode->i_nlink>  1)

-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com

  reply	other threads:[~2009-11-14  0:00 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-11-13 22:52 [PATCH -tip 1/3] Pass mm->flags to binfmt core_dump for bitflag consistency Masami Hiramatsu
2009-11-13 22:52 ` [PATCH -tip 3/3] Add get_signal tracepoint Masami Hiramatsu
2009-11-13 23:53   ` Roland McGrath
2009-11-14  0:12     ` Ingo Molnar
2009-11-16 21:52       ` Masami Hiramatsu
2009-11-16 22:11         ` Roland McGrath
2009-11-16 22:40           ` Masami Hiramatsu
2009-11-16 23:01             ` Roland McGrath
2009-11-16 23:46               ` Masami Hiramatsu
2009-11-17  6:03         ` Ingo Molnar
2009-11-17 15:25           ` Masami Hiramatsu
2009-11-14  0:29     ` Masami Hiramatsu
2009-11-13 22:52 ` [PATCH -tip 2/3] Add coredump tracepoint Masami Hiramatsu
2009-11-13 23:39   ` Roland McGrath
2009-11-14  0:00     ` Masami Hiramatsu [this message]
2009-11-14  0:02     ` Ingo Molnar
2009-11-14  0:06       ` Roland McGrath
2009-11-14  0:14         ` Ingo Molnar
2009-11-14  1:49           ` Roland McGrath
2009-11-14  0:26         ` Masami Hiramatsu
2009-11-13 23:09 ` [PATCH -tip 1/3] Pass mm->flags to binfmt core_dump for bitflag consistency Andrew Morton
2009-11-13 23:25   ` Ingo Molnar
2009-11-13 23:45     ` Masami Hiramatsu
2009-11-13 23:16 ` Roland McGrath
2009-11-13 23:23   ` Ingo Molnar
2009-11-13 23:30     ` Roland McGrath

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=4AFDF305.20903@redhat.com \
    --to=mhiramat@redhat.com \
    --cc=dle-develop@lists.sourceforge.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=roland@redhat.com \
    --cc=systemtap@sources.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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).