On Wed, May 23, 2018 at 08:49:47AM -0700, Martin KaFai Lau wrote:
> On Wed, May 23, 2018 at 12:41:14PM +0200, Daniel Borkmann wrote:
> > [ +Martin ]
> >
> > On 05/21/2018 08:59 AM, Sirio Balmelli wrote:
> > > On arm32, 'cd tools/testing/selftests/bpf && make' fails with:
> > >
> > > libbpf.c:80:10: error: format ‘%ld’ expects argument of type ‘long int’,
> > > but argument 4 has type ‘int64_t {aka long long int}’ [-Werror=format=]
> > > (func)("libbpf: " fmt, ##__VA_ARGS__); \
> > > ^
> > > libbpf.c:83:30: note: in expansion of macro ‘__pr’
> > > #define pr_warning(fmt, ...) __pr(__pr_warning, fmt, ##__VA_ARGS__)
> > > ^~~~
> > > libbpf.c:1072:3: note: in expansion of macro ‘pr_warning’
> > > pr_warning("map:%s value_type:%s has BTF type_size:%ld !=
> > > value_size:%u\n",
> > >
> > > To fix, include 'inttypes.h' and change format directive to 'PRIi64'.
> > >
> > > Signed-off-by: Sirio Balmelli <si...@b-ad.ch>
> >
> > Thanks for the fix! One minor comment below.
> >
> > > ---
> > > tools/lib/bpf/libbpf.c | 6 ++++--
> > > 1 file changed, 4 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> > > index 3dbe217bf23e..e2cc8f10c188 100644
> > > --- a/tools/lib/bpf/libbpf.c
> > > +++ b/tools/lib/bpf/libbpf.c
> > > @@ -48,6 +48,8 @@
> > > #include "bpf.h"
> > > #include "btf.h"
> > >
> > > +#include <inttypes.h> /* PRIi64 */
> > > +
> > > #ifndef EM_BPF
> > > #define EM_BPF 247
> > > #endif
> > > @@ -1042,7 +1044,7 @@ static int bpf_map_find_btf_info(struct bpf_map
> > > *map, const struct btf *btf)
> > > }
> > >
> > > if (def->key_size != key_size) {
> > > - pr_warning("map:%s key_type:%s has BTF type_size:%ld !=
> > > key_size:%u\n",
> > > + pr_warning("map:%s key_type:%s has BTF type_size:%"PRIi64" !=
> > > key_size:%u\n",
> > > map->name, name, key_size, def->key_size);
> > > return -EINVAL;
> > > }
> > > @@ -1069,7 +1071,7 @@ static int bpf_map_find_btf_info(struct bpf_map
> > > *map, const struct btf *btf)
> > > }
> > >
> > > if (def->value_size != value_size) {
> > > - pr_warning("map:%s value_type:%s has BTF type_size:%ld !=
> > > value_size:%u\n",
> > > + pr_warning("map:%s value_type:%s has BTF type_size:%"PRIi64" !=
> > > value_size:%u\n",
> > > map->name, name, value_size, def->value_size);
> >
> > I don't think we need the PRIi64 in here. Could you just change it to
> > 'type_size:%u'
> > and then cast the 'key_size' resp. 'value_size' to unsigned int? We know at
> > this
> > point that the type size value is positive anyway, so we only want to show
> > the size
> > mismatch so simple cast should suffice. Thanks!
> Good point. 'unsigned int' is enough because the t->size (returned
> by btf_type_size()) is a u32.
>
> >
> > > return -EINVAL;
> > > }
> > >
> >
Thank you very much; respinning patch.
