Linux DTrace development list
 help / color / mirror / Atom feed
From: Kris Van Hees <kris.van.hees@oracle.com>
To: eugene.loh@oracle.com
Cc: dtrace@lists.linux.dev, dtrace-devel@oss.oracle.com
Subject: Re: [PATCH 11/38] USDT module names may contain dots; but forbid "." and ".." names
Date: Thu, 18 Jul 2024 15:23:40 -0400	[thread overview]
Message-ID: <ZplrvGStIyiZeWYH@oracle.com> (raw)
In-Reply-To: <20240627053455.21567-12-eugene.loh@oracle.com>

On Thu, Jun 27, 2024 at 01:34:28AM -0400, eugene.loh@oracle.com wrote:
> From: Eugene Loh <eugene.loh@oracle.com>
> 
> Signed-off-by: Eugene Loh <eugene.loh@oracle.com>

Reviewed-by: Kris Van Hees <kris.van.hees@oracle.com>

> ---
>  dtprobed/dof_stash.c | 30 ++++++++++++++----------------
>  1 file changed, 14 insertions(+), 16 deletions(-)
> 
> diff --git a/dtprobed/dof_stash.c b/dtprobed/dof_stash.c
> index 62418b66..625572d5 100644
> --- a/dtprobed/dof_stash.c
> +++ b/dtprobed/dof_stash.c
> @@ -231,6 +231,20 @@ make_probespec_name(const char *prov, const char *mod, const char *fn,
>  {
>  	char *ret;
>  
> +	/*
> +	 * Ban "." and ".." as name components.  Obviously names
> +	 * containing dots are commonplace (shared libraries,
> +	 * for instance), but allowing straight . and .. would
> +	 * have obviously horrible consequences.  They can't be
> +	 * filenames anyway, and you can't create them with
> +	 * dtrace -h because they aren't valid C identifier names.
> +	 */
> +	if (strcmp(prov, ".") == 0 || strcmp(prov, "..") == 0 ||
> +	    strcmp(mod, ".") == 0 || strcmp(mod, "..") == 0 ||
> +	    strcmp(fn, ".") == 0 || strcmp(fn, "..") == 0 ||
> +	    strcmp(prb, ".") == 0 || strcmp(prb, "..") == 0)
> +		return NULL;
> +
>  	if (asprintf(&ret, "%s:%s:%s:%s", prov, mod, fn, prb) < 0) {
>  		fuse_log(FUSE_LOG_ERR, "dtprobed: out of memory making probespec\n");
>  		return NULL;
> @@ -589,22 +603,6 @@ dof_stash_write_parsed(pid_t pid, dev_t dev, ino_t ino, dt_list_t *accum)
>  							    mod, fun, prb)) == NULL)
>  				goto err_provider;
>  
> -			/*
> -			 * Ban "." and ".." as name components.  Obviously names
> -			 * containing dots are commonplace (shared libraries,
> -			 * for instance), but allowing straight . and .. would
> -			 * have obviously horrible consequences.  They can't be
> -			 * filenames anyway, and you can't create them with
> -			 * dtrace -h because they aren't valid C identifier
> -			 * names.
> -			 */
> -			op = "probe name validation";
> -			probe_err = parsedfn;
> -
> -			if (strcmp(parsedfn, ".") == 0 ||
> -			    strcmp(parsedfn, "..") == 0)
> -				goto err_provider;
> -
>  			op = "probe module";
>  
>  			if ((mod_dir = make_state_dirat(prov_dir, mod, op, 0)) < 0)
> -- 
> 2.18.4
> 

  reply	other threads:[~2024-07-18 19:23 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-27  5:34 eugene.loh
2024-06-27  5:34 ` [PATCH 01/38] Move comment closer to the code it describes eugene.loh
2024-06-27  5:34 ` [PATCH 02/38] Move dt_spec_buf_data_t and dt_spec_buf_t into dt_consume.c eugene.loh
2024-07-18  6:54   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 03/38] Get rid of apparently orphaned status[2] eugene.loh
2024-07-18  6:59   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 04/38] Get rid of apparently orphaned bufdesc stuff eugene.loh
2024-07-18 18:28   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 05/38] Get rid of unneeded enabling_defines.h eugene.loh
2024-07-18 18:35   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 06/38] Get rid of unused dtrace_repldesc_t eugene.loh
2024-07-18 18:34   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 07/38] Clean up prp/pprp/uprp variable names eugene.loh
2024-07-18 18:48   ` Kris Van Hees
2024-07-18 20:19     ` Eugene Loh
2024-06-27  5:34 ` [PATCH 08/38] Fix comment in dt_probe.c eugene.loh
2024-07-18 18:49   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 09/38] Fix comments that hardwire DBUF_ offsets eugene.loh
2024-07-18 19:04   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 10/38] Fix comments in dt_cg.c eugene.loh
2024-07-18 19:28   ` Kris Van Hees
2024-07-18 20:29     ` Eugene Loh
2024-06-27  5:34 ` [PATCH 11/38] USDT module names may contain dots; but forbid "." and ".." names eugene.loh
2024-07-18 19:23   ` Kris Van Hees [this message]
2024-06-27  5:34 ` [PATCH 12/38] USDT module names may contain dots; remove incorrect check eugene.loh
2024-07-18 19:24   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 13/38] Hide dtrace_actdesc_t until it is needed eugene.loh
2024-07-18 20:02   ` Kris Van Hees
2024-07-18 21:06     ` Eugene Loh
2024-07-18 21:28       ` Kris Van Hees
2024-07-18 22:36         ` Eugene Loh
2024-06-27  5:34 ` [PATCH 14/38] Remove orphaned dtrace_hdl_t component dt_maxformat eugene.loh
2024-07-18 20:03   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 15/38] Remove orphaned dtrace_hdl_t component dt_prov_usdt eugene.loh
2024-07-18 20:03   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 16/38] Move dt_probe_clause_t to be available outside of dt_probe.c eugene.loh
2024-07-18 20:19   ` Kris Van Hees
2024-06-27  5:34 ` [PATCH 17/38] Add a provider-specific probe_add_clause handle eugene.loh
2024-07-18 20:49   ` Kris Van Hees
2024-07-19  4:00     ` Eugene Loh
2024-06-27  5:34 ` [PATCH 18/38] Add a provider-specific probe_add_clause for underlying probes eugene.loh
2024-07-18 20:50   ` Kris Van Hees
2024-07-19  4:00     ` Eugene Loh

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=ZplrvGStIyiZeWYH@oracle.com \
    --to=kris.van.hees@oracle.com \
    --cc=dtrace-devel@oss.oracle.com \
    --cc=dtrace@lists.linux.dev \
    --cc=eugene.loh@oracle.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