Float compare-for-equality and select optimization opportunity

Hi all,

I’m trying to generate code containing an ordered float compare for equality, and select. The resulting code however has an unordered compare and some Boolean logic that I think could be eliminated. In C syntax the code looks like this:

float x, y;

int a, b, c

if(x == y) // Rotate the integers

{

int t;

t = a;

a = b;

b = c;

c = t;

}

This is the resulting x86 assembly code:

movss xmm0,dword ptr [ecx+4]

ucomiss xmm0,dword ptr [ecx+8]

sete al

setnp dl

test dl,al

mov edx,edi

cmovne edx,ecx

cmovne ecx,esi

cmovne esi,edi

While I’m pleasantly surprised that my branch does get turned into several select operations as intended (cmov - conditional move – in x86), I’m confused why it uses the ucomiss instruction (unordered compare and set flags). I only used IRBuilder::CreateFCmpOEQ. It also appears to invert the conditional, for no clear reason. I think it could be rewritten as follows:

movss xmm0,dword ptr [ecx+4]

comiss xmm0,dword ptr [ecx+8]

mov edx,edi

cmove edx,ecx

cmove ecx,esi

cmove esi,edi

Compared to the original C syntax code this looks pretty straightforward. Curiously, when I replace the compare-for-equality with something like a less-than, it does generate such compact code (using comiss and cmova). And the not-equal case looks like this:

movss xmm0,dword ptr [ecx+4]

ucomiss xmm0,dword ptr [ecx+8]

mov esi,ecx

cmove esi,edx

cmovne ecx,eax

cmove edx,eax

So this generates compact code but with an unordered compare.

Anyway, it looks like the compare-for-equality case in particular is missing an optimization opportunity. It’s no big deal to me but I thought someone here might be interested…

Cheers,

Nicolas Capens

Both ZF and PF will be set if unordered, so the code below is IEEE correct…you want to generate ‘fcmp ueq’ instead of ‘fcmp oqe’

Nicolas:
The LLVM mail server is being slow, so I’m direct e-mailing my comment. The generated code is IEEE correct…use CreateFCmpUEQ for unordered or equal, which should generate your hand written version…I’m assuming that your next e-mail (min/max) is probably the same thing.

Hi Marc,

I’m a bit confused. Isn’t the standard compare (i.e. the one for a language like C) an ordered one? I tried converting some C code to LLVM C++ API code with the online demo, and it uses FCMP_OEQ…

Cheers,

Nicolas

Hi Marc,

Hi Marc,

Thanks a lot for the detailed explanation!

For the record and for my own reference: If the ZF (zero flag) is 0 we know for certain the arguments are not equal (!=). If the ZF is 1 they are either equal or unordered. The PF determines whether any argument is NaN. So for equality (==) we need to check two flags if we care about the unordered case.

I was also thrown off by LLVM’s use of ucomiss instead of comiss, but it has nothing to do with the compares themselves, only whether or not an exception is thrown for QNaNs. So for the best optimization opportunity I should always use unordered compares in LLVM, right?

Thanks again!

Nicolas