On Tue, 2017-08-01 at 03:40 +0900, OGAWA Hirofumi wrote: > Andy Shevchenko <[email protected]> writes: > > > We may use hex2bin() instead of custom approach. > > > > Signed-off-by: Andy Shevchenko <[email protected]> > > >
> > + fill = hex2bin(hc, ip + 1, 2); > > + if (fill) > > + return fill; > > This should not use random errno (in this case, it is -1 (EPERM)). You perhaps missed the side note I put after --- line. It reflects this change. > > > + *op++ = hc[1]; > > + *op++ = hc[0]; > > Maybe, originally endian bug? No problem reported. And as you noticed quite ago it is __be16 originally as it goes hi-lo. -- Andy Shevchenko <[email protected]> Intel Finland Oy
