Jakob Stoklund Olesen

Yes, I expect we will need mitigations for all variants of Spectre. All of these attacks depend on a high-resolution timer being available to malicious code, so a first-level mitigation...

*Re: Heap bounds checking* I haven't measured the performance of using `lfence` to guard bounds checks myself, but from what I hear from people who did, results are very sad....

Thanks, Jonathan. I'll leave it open to track remaining work.

Intel has `i32x4.clz` and `i64x2.clz`, but only in AVX-512. I don't see vectorized versions of the other instructions.

The f64x2 version could be included too. ARMv7 would need to do two scalar additions anyway, and ARMv8/SSE3 have the instruction. MIPS and POWER do not have these instructions.

The function of `std::mt19937` is completely specified by the C++ standard, so it should be portable. On the other hand, the operation of `uniform_int_distribution` is apparently *not* fully specified, so...

Loop optimizations in LLVM will generally ignore irreducible control flow and not attempt to optimize it. The loop analysis they're based on will only recognize natural loops, so you just...

Is this the same as #237?

I agree. The spec is fine, we should just fix the polyfill and tests.

Oops, you are right. Fixed.