llvm / llvm/llvm-project

[x86-64] Conditional increment on non-carry non-overflow => speculative increment+cmov

Open
#207,987 12 comments 0 reactions 0 assignees View on GitHub
backend:X86 missed-optimization
Dominant language
LLVM
Stars
40.5k
Forks
18.7k
PR merge metrics
PR metrics pending

Description

[C Godbolt](https://c.godbo.lt/#g:!((g:!((g:!((h:codeEditor,i:(filename:'1',fontScale:14,fontUsePx:'0',j:1,lang:___c,selection:(endColumn:1,endLineNumber:2,positionColumn:1,positionLineNumber:2,selectionStartColumn:1,selectionStartLineNumber:2,startColumn:1,startLineNumber:2),source:'%23include+%3Cstdint.h%3E%0Auint64_t+foo(uint64_t+a,+uint64_t+b)+%7B%0A++++return+a+%2B+((b+%26+63)+!!%3D+0)%3B%0A%7D%0A'),l:'5',n:'0',o:'C+source+%231',t:'0')),k:48.46943717911461,l:'4',n:'0',o:'',s:0,t:'0'),(g:!((h:compiler,i:(compiler:cclang_trunk,filters:(b:'0',binary:'1',binaryObject:'1',commentOnly:'0',debugCalls:'1',demangle:'0',directives:'0',execute:'1',intel:'0',libraryCode:'0',trim:'1',verboseDemangling:'0'),flagsViewOpen:'1',fontScale:14,fontUsePx:'0',j:1,lang:___c,libs:!(),options:'-O3+-march%3Dznver5',overrides:!(),selection:(endColumn:1,endLineNumber:1,positionColumn:1,positionLineNumber:1,selectionStartColumn:1,selectionStartLineNumber:1,startColumn:1,startLineNumber:1),source:1),l:'5',n:'0',o:'+x86-64+clang+(trunk)+(Editor+%231)',t:'0')),k:51.53056282088541,l:'4',m:100,n:'0',o:'',s:0,t:'0')),l:'2',n:'0',o:'',t:'0')),version:4)

```c
#include
uint64_t foo(uint64_t a, uint64_t b) {
return a + ((b & 63) != 0);
}
```

Compiles to:

```asm
foo:
and esi, 63
mov rax, rdi
cmp rsi, 1
sbb rax, -1
ret
```

Could be:

```asm
foo:
lea rax, [rdi + 1] ; speculative a + 1
test esi, 63 ; ZF = 1 iff (idx & 63) == 0 (non-destructive)
cmovz rax, rdi ; if zero, take the un-incremented a
ret
```

Contributor guide

Open the contributing guide

Research direction

Reproduce the foo example from the linked Godbolt configuration with clang trunk, -O3, and -march=znver5, then compare the emitted x86-64 assembly with the proposed lea/test/cmovz sequence. No repository file or test is named; done means the compiler produces the requested equivalent optimization for this case.

Written by the indexing model from the issue text.

Assessment

Tech stack
c
Domain
compilers, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.