All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jerry Van Baren <gvb.uboot@gmail.com>
To: u-boot@lists.denx.de
Subject: [U-Boot] libfdt: Fix bugs in fdt_get_path()
Date: Sat, 30 Aug 2008 12:13:11 -0400	[thread overview]
Message-ID: <48B97197.5060703@gmail.com> (raw)
In-Reply-To: <20080829041913.GC30197@yookeroo.seuss>

FYI, I propose to pick this up for the *NEXT* release of u-boot ("Yellow 
Submarine" ;-).  My reasoning is that (1) the current window closes 
today, which makes me uncomfortable WRT putting in changes today and (2) 
this appears to be a latent bug (at worst, a misleading error message); 
I have not seen evidence that it is causing problems at the moment.

Best regards,
gvb

David Gibson wrote:
> The current implementation of fdt_get_path() has a couple of bugs,
> fixed by this patch.
> 
> First, contrary to its documentation, on success it returns the length
> of the node's path, rather than 0.  The testcase is correspondingly
> wrong, and the patch fixes this as well.
> 
> Second, in some circumstances, it will return -FDT_ERR_BADOFFSET
> instead of -FDT_ERR_NOSPACE when given insufficient buffer space.
> Specifically this happens when there is insufficient space even to
> hold the path's second last component.  This behaviour is corrected,
> and the testcase updated to check it.
> 
> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
> 
> Index: dtc/tests/get_path.c
> ===================================================================
> --- dtc.orig/tests/get_path.c	2008-08-29 14:11:09.000000000 +1000
> +++ dtc/tests/get_path.c	2008-08-29 14:11:11.000000000 +1000
> @@ -43,6 +43,8 @@ static void check_path_buf(void *fdt, co
>  	memset(buf, POISON, sizeof(buf)); /* poison the buffer */
>  
>  	len = fdt_get_path(fdt, offset, buf, buflen);
> +	verbose_printf("get_path() %s -> %d -> %s\n", path, offset, buf);
> +
>  	if (buflen <= pathlen) {
>  		if (len != -FDT_ERR_NOSPACE)
>  			FAIL("fdt_get_path([%d bytes]) returns %d with "
> @@ -51,9 +53,9 @@ static void check_path_buf(void *fdt, co
>  		if (len < 0)
>  			FAIL("fdt_get_path([%d bytes]): %s", buflen,
>  			     fdt_strerror(len));
> -		if (len != pathlen)
> -			FAIL("fdt_get_path([%d bytes]) reports length %d "
> -			     "instead of %d", buflen, len, pathlen);
> +		if (len != 0)
> +			FAIL("fdt_get_path([%d bytes]) returns %d "
> +			     "instead of 0", buflen, len);
>  		if (strcmp(buf, path) != 0)
>  			FAIL("fdt_get_path([%d bytes]) returns \"%s\" "
>  			     "instead of \"%s\"", buflen, buf, path);
> @@ -70,6 +72,8 @@ static void check_path(void *fdt, const 
>  	check_path_buf(fdt, path, pathlen, 1024);
>  	check_path_buf(fdt, path, pathlen, pathlen+1);
>  	check_path_buf(fdt, path, pathlen, pathlen);
> +	check_path_buf(fdt, path, pathlen, 0);
> +	check_path_buf(fdt, path, pathlen, 2);
>  }
>  
>  int main(int argc, char *argv[])
> Index: dtc/libfdt/fdt_ro.c
> ===================================================================
> --- dtc.orig/libfdt/fdt_ro.c	2008-08-29 14:11:11.000000000 +1000
> +++ dtc/libfdt/fdt_ro.c	2008-08-29 14:11:11.000000000 +1000
> @@ -328,9 +328,6 @@ int fdt_get_path(const void *fdt, int no
>  	for (offset = 0, depth = 0;
>  	     (offset >= 0) && (offset <= nodeoffset);
>  	     offset = fdt_next_node(fdt, offset, &depth)) {
> -		if (pdepth < depth)
> -			continue; /* overflowed buffer */
> -
>  		while (pdepth > depth) {
>  			do {
>  				p--;
> @@ -338,14 +335,16 @@ int fdt_get_path(const void *fdt, int no
>  			pdepth--;
>  		}
>  
> -		name = fdt_get_name(fdt, offset, &namelen);
> -		if (!name)
> -			return namelen;
> -		if ((p + namelen + 1) <= buflen) {
> -			memcpy(buf + p, name, namelen);
> -			p += namelen;
> -			buf[p++] = '/';
> -			pdepth++;
> +		if (pdepth >= depth) {
> +			name = fdt_get_name(fdt, offset, &namelen);
> +			if (!name)
> +				return namelen;
> +			if ((p + namelen + 1) <= buflen) {
> +				memcpy(buf + p, name, namelen);
> +				p += namelen;
> +				buf[p++] = '/';
> +				pdepth++;
> +			}
>  		}
>  
>  		if (offset == nodeoffset) {
> @@ -355,7 +354,7 @@ int fdt_get_path(const void *fdt, int no
>  			if (p > 1) /* special case so that root path is "/", not "" */
>  				p--;
>  			buf[p] = '\0';
> -			return p;
> +			return 0;
>  		}
>  	}
>  
> 

  reply	other threads:[~2008-08-30 16:13 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-08-29  4:19 libfdt: Fix bugs in fdt_get_path() David Gibson
2008-08-30 16:13 ` Jerry Van Baren [this message]
2008-08-31  8:16   ` [U-Boot] " David Gibson
2008-09-25 15:01 ` Jon Loeliger
2008-09-25 15:01   ` Jon Loeliger

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=48B97197.5060703@gmail.com \
    --to=gvb.uboot@gmail.com \
    --cc=u-boot@lists.denx.de \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.