On Feb 5, 2016, at 9:08 AM, Wouter van Kesteren <woutershep@xxxxxxxxx> wrote:
>
> '~0ULL' is a 'unsigned long long' that when converted to a loff_t,
> which is signed, gets turned into -1. later in vfs_setpos we have
> 'if (offset > maxsize)', which makes it always return EINVAL.
Patch itself looks reasonable, and the bug was introduced in commit
b25472f9b "new helpers: no_seek_end_llseek{,_size}()".
Wouter, you need to resubmit this with your "Signed-off-by:" line,
see linux/Documentation/SubmittingPatches for details, and the patch
should be sent "To: Al Viro" and "CC: linux-fsdevel" so it isn't lost.
When you resubmit you can add my:
Reviewed-by: Andreas Dilger <adilger@xxxxxxxxx>
Cheers, Andreas
> ---
> fs/read_write.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/fs/read_write.c b/fs/read_write.c
> index 324ec27..0c8782a 100644
> --- a/fs/read_write.c
> +++ b/fs/read_write.c
> @@ -17,6 +17,7 @@
> #include <linux/splice.h>
> #include <linux/compat.h>
> #include <linux/mount.h>
> +#include <linux/fs.h>
> #include "internal.h"
>
> #include <asm/uaccess.h>
> @@ -183,7 +184,7 @@ loff_t no_seek_end_llseek(struct file *file, loff_t offset, int whence)
> switch (whence) {
> case SEEK_SET: case SEEK_CUR:
> return generic_file_llseek_size(file, offset, whence,
> - ~0ULL, 0);
> + OFFSET_MAX, 0);
> default:
> return -EINVAL;
> }
> --
> 2.7.0
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
Cheers, Andreas
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
