On Mon, Nov 18, 2019 at 08:40:58AM +0100, Heinrich Schuchardt wrote:
> For GCC 9.2.1 the negative of an unsigned size_t value is again an unsigned
> size_t value, e.g. -5UL = 18446744073709551611UL.
>
> Correct the logical constraint. Add necessary type conversions.
>
> Signed-off-by: Heinrich Schuchardt <xypron.glpk@xxxxxx>
> ---
> It would be advisable to add -Wextra to the libfdt Makefile after
> correcting the 50+ -Wsign-compare problems reported.
> ---
> libfdt/fdt_ro.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/libfdt/fdt_ro.c b/libfdt/fdt_ro.c
> index adab0cb..b69806f 100644
> --- a/libfdt/fdt_ro.c
> +++ b/libfdt/fdt_ro.c
> @@ -59,9 +59,9 @@ const char *fdt_get_string(const void *fdt, int stroffset, int *lenp)
> }
> } else if (fdt_magic(fdt) == FDT_SW_MAGIC) {
> if ((stroffset >= 0)
> - || (stroffset < -fdt_size_dt_strings(fdt)))
> + || ((size_t)-stroffset > fdt_size_dt_strings(fdt)))
Hmm. Why is the explicit cast necessary? -stroffset is definitely
positive (because of the test above), so shouldn't it be promoted
correctly here?
> goto fail;
> - if ((-stroffset) < len)
> + if ((size_t)-stroffset < len)
Same here.
> len = -stroffset;
> } else {
> err = -FDT_ERR_INTERNAL;
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
Attachment:
signature.asc
Description: PGP signature
