On Tue, 20 Jun 2023, Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote: > So, what does prevent you from using GENMASK_ULL()? > > Another point, you may teach GENMASK() to issue a warning if hi and/or lo > bigger than BITS_PER_LONG. What good does that do if you want the warning for a fixed size different from unsigned long or long long? Worse, sizeof(long) depends on arch, while the GENMASK you want depends on the use case. > I still don't see the usefulness of that churn. This thread is turning into a prime example of why drivers and subsystems reinvent their own wheels instead of trying to get generally useful stuff merged in kernel headers. :p BR, Jani. -- Jani Nikula, Intel Open Source Graphics Center
