random: use rejection sampling for uniform bounded random integers

Until the very recent commits, many bounded random integers were
calculated using `get_random_u32() % max_plus_one`, which not only
incurs the price of a division -- indicating performance mostly was not
a real issue -- but also does not result in a uniformly distributed
output if max_plus_one is not a power of two. Recent commits moved to
using `prandom_u32_max(max_plus_one)`, which replaces the division with
a faster multiplication, but still does not solve the issue with
non-uniform output.

For some users, maybe this isn't a problem, and for others, maybe it is,
but for the majority of users, probably the question has never been
posed and analyzed, and nobody thought much about it, probably assuming
random is random is random. In other words, the unthinking expectation
of most users is likely that the resultant numbers are uniform.

So we implement here an efficient way of generating uniform bounded
random integers. Through use of compile-time evaluation, and avoiding
divisions as much as possible, this commit introduces no measurable
overhead. At least for hot-path uses tested, any potential difference
was lost in the noise. On both clang and gcc, code generation is pretty
small.

The new function, get_random_u32_below(), lives in random.h, rather than
prandom.h, and has a "get_random_xxx" function name, because it is
suitable for all uses, including cryptography.

In order to be efficient, we implement a kernel-specific variant of
Daniel Lemire's algorithm from "Fast Random Integer Generation in an
Interval", linked below. The kernel's variant takes advantage of
constant folding to avoid divisions entirely in the vast majority of
cases, works on both 32-bit and 64-bit architectures, and requests a
minimal amount of bytes from the RNG.

Link: https://arxiv.org/pdf/1805.10941.pdf
Cc: stable@vger.kernel.org # to ease future backports that use this api
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

1 files changed, 40 insertions, 0 deletions

diff --git a/include/linux/random.h b/include/linux/random.h
index 147a5e0d0b8e..3a82c0a8bc46 100644
--- a/include/linux/random.h
+++ b/include/linux/random.h
@@ -51,6 +51,46 @@ static inline unsigned long get_random_long(void)
 #endif
 }
 
+u32 __get_random_u32_below(u32 ceil);
+
+/*
+ * Returns a random integer in the interval [0, ceil), with uniform
+ * distribution, suitable for all uses. Fastest when ceil is a constant, but
+ * still fast for variable ceil as well.
+ */
+static inline u32 get_random_u32_below(u32 ceil)
+{
+	if (!__builtin_constant_p(ceil))
+		return __get_random_u32_below(ceil);
+
+	/*
+	 * For the fast path, below, all operations on ceil are precomputed by
+	 * the compiler, so this incurs no overhead for checking pow2, doing
+	 * divisions, or branching based on integer size. The resultant
+	 * algorithm does traditional reciprocal multiplication (typically
+	 * optimized by the compiler into shifts and adds), rejecting samples
+	 * whose lower half would indicate a range indivisible by ceil.
+	 */
+	BUILD_BUG_ON_MSG(!ceil, "get_random_u32_below() must take ceil > 0");
+	if (ceil <= 1)
+		return 0;
+	for (;;) {
+		if (ceil <= 1U << 8) {
+			u32 mult = ceil * get_random_u8();
+			if (likely(is_power_of_2(ceil) || (u8)mult >= (1U << 8) % ceil))
+				return mult >> 8;
+		} else if (ceil <= 1U << 16) {
+			u32 mult = ceil * get_random_u16();
+			if (likely(is_power_of_2(ceil) || (u16)mult >= (1U << 16) % ceil))
+				return mult >> 16;
+		} else {
+			u64 mult = (u64)ceil * get_random_u32();
+			if (likely(is_power_of_2(ceil) || (u32)mult >= -ceil % ceil))
+				return mult >> 32;
+		}
+	}
+}
+
 /*
  * On 64-bit architectures, protect against non-terminated C string overflows
  * by zeroing out the first byte of the canary; this leaves 56 bits of entropy.