From: Jason A. Donenfeld
> Sent: 07 October 2022 18:56
...
> > Given these kinds of less mechanical changes, it may make sense to split
> > these from the "trivial" conversions in a treewide patch. The chance of
> > needing a revert from the simple 1:1 conversions is much lower than the
> > need to revert by-hand changes.
> >
> > The Cocci script I suggested in my v1 review gets 80% of the first
> > patch, for example.
>
> I'll split things up into a mechanical step and a non-mechanical step.
> Good idea.
I'd also do something about the 'get_random_int() & 3' cases.
(ie remainder by 2^n-1)
These can be converted to 'get_random_u8() & 3' (etc).
So they only need one random byte (not 4) and no multiply.
Possibly something based on (the quickly typed, and not C):
#define get_random_below(val) [
if (builtin_constant(val))
BUILD_BUG_ON(!val || val > 0x100000000ull)
if (!(val & (val - 1)) {
if (val <= 0x100)
return get_random_u8() & (val - 1);
if (val <= 0x10000)
return get_random_u16() & (val - 1);
return get_random_u32() & (val - 1);
}
}
BUILD_BUG_ON(sizeof (val) > 4);
return ((u64)get_random_u32() * val) >> 32;
}
get_random_below() is a much better name than prandom_u32_max().
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)
