Hi Holger,
On Thu, 4 Jul 2013 10:29:46 +0200, Holger Brunck
<holger.bru...@keymile.com> wrote:
> If we have the following entry in cramfs:
> => cramfsls
> -rw-r--r-- 1922689 uImage
>
> cramfsload would also succeed if we try to do:
> => cramfsload uImage_1
> CRAMFS load complete: 1922689 bytes loaded to 0x100000
>
> The old code succeeds if the begin of the filename we search matches
> with a filename stored in cramfs. But the searched file may have
> additional characters and is therfore not the file we are looking for.
> So compare also the length of the filename we search and the
> filename we found in cramfs. This leads to:
> => cramfsload uImage_1
> can't find corresponding entry
> CRAMFS LOAD ERROR<0> for uImage_1!
>
> which is the behaviour we want.
> Signed-off-by: Holger Brunck <holger.bru...@keymile.com>
> cc: Wolfgang Denk <w...@denx.de>
> ---
Can't the commit message above be summarized as follows?
---8<---
"cramfsload uImage_1" succeeds even though the actual file is named
"uImage".
Fix file name comparison when one name is the prefix of the other.
---8<---
The demonstrative part of the commit message can go here, below the
commit message delimiter '---'.
> fs/cramfs/cramfs.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/fs/cramfs/cramfs.c b/fs/cramfs/cramfs.c
> index 910955d..e578a1e 100644
> --- a/fs/cramfs/cramfs.c
> +++ b/fs/cramfs/cramfs.c
> @@ -126,7 +126,8 @@ static unsigned long cramfs_resolve (unsigned long begin,
> unsigned long offset,
> namelen--;
> }
>
> - if (!strncmp (filename, name, namelen)) {
> + if (!strncmp(filename, name, namelen) &&
> + (namelen == strlen(filename))) {
> char *p = strtok (NULL, "/");
>
> if (raw && (p == NULL || *p == '\0'))
Amicalement,
--
Albert.
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot
