Linux-Trace-Devel Archive on lore.kernel.org
 help / color / Atom feed
From: Steven Rostedt <rostedt@goodmis.org>
To: vincent.donnefort@arm.com
Cc: linux-trace-devel@vger.kernel.org
Subject: Re: [PATCH 1/2] trace-cmd: Enable kptr_restrict
Date: Mon, 11 Nov 2019 17:47:54 -0500
Message-ID: <20191111174754.47651068@gandalf.local.home> (raw)
In-Reply-To: <1573123866-348262-1-git-send-email-vincent.donnefort@arm.com>


Hi Vincent!


On Thu,  7 Nov 2019 10:51:05 +0000
vincent.donnefort@arm.com wrote:

> From: Vincent Donnefort <vincent.donnefort@arm.com>
> 
> kptr_restrict might prevent trace-cmd from accessing /proc/kallsyms,
> leading to a trace without the kernel function names resolved.
> 
> Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
> 
> diff --git a/lib/trace-cmd/trace-output.c b/lib/trace-cmd/trace-output.c
> index 41932ee..3c4f306 100644
> --- a/lib/trace-cmd/trace-output.c
> +++ b/lib/trace-cmd/trace-output.c
> @@ -674,6 +674,39 @@ static int read_event_files(struct tracecmd_output *handle,
>  	return ret;
>  }
>  
> +static void set_proc_kptr_restrict(int reset)
> +{
> +	char *path = "/proc/sys/kernel/kptr_restrict";

I believe this is a relatively new file. We should do a stat to see if
it exists, and if it does not, simply fail silently. I don't think we
want to give a warning if the file doesn't exist because the kernel
doesn't have it.

> +	static char saved = 'X';
> +	int fd, ret = -1;
> +	char buf;
> +
> +	fd = open(path, O_RDONLY);
> +	if (fd < 0)
> +		goto err;
> +
> +	if (reset) {
> +		buf = saved;
> +	} else {
> +		if (read(fd, &buf, 1) < 0)
> +			goto err;
> +		saved = buf;
> +		buf = '0';
> +	}
> +	close(fd);
> +

Perhaps if reset is true and buf == 'X', we should simply exit, as it
would appear that we never put anything into buf. And probably should
make the 'X' a macro:

#define KPTR_UNINITIALIZED	'X'

	static char saved = KPTR_UNINITIALIZED;

[..]

	if (reset && buf == KPTR_UNINITIALIZED)
		return;


-- Steve

> +	fd = open(path, O_WRONLY);
> +	if (fd < 0)
> +		goto err;
> +	if (write(fd, &buf, 1) > 0)
> +		ret = 0;
> +err:
> +	if (fd > 0)
> +		close(fd);
> +	if (ret)
> +		warning("can't set kptr_restrict");
> +}
> +
>  static int read_proc_kallsyms(struct tracecmd_output *handle,
>  			      const char *kallsyms)
>  {
> @@ -698,12 +731,16 @@ static int read_proc_kallsyms(struct tracecmd_output *handle,
>  	endian4 = convert_endian_4(handle, size);
>  	if (do_write_check(handle, &endian4, 4))
>  		return -1;
> +
> +	set_proc_kptr_restrict(0);
>  	check_size = copy_file(handle, path);
>  	if (size != check_size) {
>  		errno = EINVAL;
>  		warning("error in size of file '%s'", path);
> +		set_proc_kptr_restrict(1);
>  		return -1;
>  	}
> +	set_proc_kptr_restrict(1);
>  
>  	return 0;
>  }


      parent reply index

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-07 10:51 vincent.donnefort
2019-11-07 10:51 ` [PATCH 2/2] trace-cmd: Add an option to set saved_cmdlines_size vincent.donnefort
2019-11-11 22:52   ` Steven Rostedt
2019-11-12 12:03     ` Vincent Donnefort
2019-11-12 14:40       ` Steven Rostedt
2019-11-12 19:01         ` Vincent Donnefort
2019-11-12 19:09           ` Steven Rostedt
2019-11-13 12:14             ` [PATCH v2 1/2] trace-cmd: Enable kptr_restrict vincent.donnefort
2019-11-13 12:14               ` [PATCH v2 2/2] trace-cmd: Add an option to set saved_cmdlines_size vincent.donnefort
2019-11-13 12:27             ` [PATCH " Vincent Donnefort
2019-11-13 15:08               ` Steven Rostedt
2019-11-12 12:03     ` [PATCH v2 1/2] trace-cmd: Enable kptr_restrict vincent.donnefort
2019-11-12 12:03       ` [PATCH v2 2/2] trace-cmd: Add an option to set saved_cmdlines_size vincent.donnefort
2019-11-11 22:47 ` Steven Rostedt [this message]

Reply instructions:

You may reply publically 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=20191111174754.47651068@gandalf.local.home \
    --to=rostedt@goodmis.org \
    --cc=linux-trace-devel@vger.kernel.org \
    --cc=vincent.donnefort@arm.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

Linux-Trace-Devel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-trace-devel/0 linux-trace-devel/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-trace-devel linux-trace-devel/ https://lore.kernel.org/linux-trace-devel \
		linux-trace-devel@vger.kernel.org
	public-inbox-index linux-trace-devel

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-trace-devel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git