Improved printing for floats

[scurest]

All floats I've tried now print correctly.

See #375 (comment) for the first commit. The further changes are

• -0.0 now prints correctly, thanks @tiehuis
• you actually need 9 digits for f32s (I think 8 would be enough if you used the next digit to round, but if there's a lot of 9s at the end you'd need to carry and it'd be a pain); for f64s, I just use the whole buffer Errol outputs
• fixed some more small typos in our Errol code
• Errol has a bug, possibly? See errol3: Wrong result for 9.007199254740992e15 marcandrysco/Errol#8. I had to make the change I mentioned in that issue to get the tests to succeed.

For the last bullet, maybe we want to wait for an upstream confirmation and patch.

Tests

The tests I made are here (see the usage file at the bottom). All the tests I've run have passed. In particular the f32 tests are intended to show by exhaustion that we're correct for all f32s.

• f32 I iterate over all positive u32s, bitCasting them to f32s, skipping the NaNs and Infs, and printing the rest to stdout in zig. This is piped into a C program where I scanf with %f and check if the nth one I get bitcasts back to n. This should exhaustively test all positive f32s.

  If you want to check all f32s, just change the 0x80000000s to 0xffffffffs in both files. I didn't just because it takes about 2 hours to run already.

• rand This generates some random numbers with xoroshiro128+, bitcasts to f64, prints them, pipes them into C, scanfs them with %lf and checks if they're the same. The more times you run it, the greater your confidence should be that we're correct :)

[andrewrk]

Excellent work, @scurest.

I'm going to wait for the tests to pass, and then merge.

[scurest]

Ah, I forgot to ask: I couldn't get the zig files for the tests to compile in release mode. How do you do it?

[andrewrk]

Are you asking how to run a test in release mode?

zig test foo.zig --release-fast
or
zig test foo.zig --release-safe

[scurest]

Well, I meant compile say f32-out.zig in release mode.

If I use zig build-exe --release-safe f32-out.zig I get linker errors from lld for __fixunsdfti, __umodti3, __udivti3, etc. The u128 stuff.
If I add --library c I get a binary but that binary only prints three floats before printing "index out of bounds" and segfaulting.

[andrewrk]

If I use zig build-exe --release-safe f32-out.zig I get linker errors from lld for __fixunsdfti, __umodti3, __udivti3, etc. The u128 stuff.

Hmm. This is really surprising, I'm looking into this right now. At first glance it seems like LLVM optimization is deleting functions that are marked with linkonce linkage, which according to http://llvm.org/docs/LangRef.html#linkage-types seems like shouldn't happen.

[andrewrk]

I fixed the compiler-rt problem in master branch. The other problems you ran into I see the same problem, and it's a bug in zig. I'm looking into it. I'll merge this PR though, because this is better than status quo.