[llvm-bugs] [Bug 42291] New: [X86] TZCNT should be a separate CPU feature?

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=42291

            Bug ID: 42291
           Summary: [X86] TZCNT should be a separate CPU feature?
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Backend: X86
          Assignee: unassignedbugs at nondot.org
          Reporter: lebedev.ri at gmail.com
                CC: craig.topper at gmail.com, llvm-bugs at lists.llvm.org,
                    llvm-dev at redking.me.uk, spatel+llvm at rotateright.com

Currently there are FeaturePOPCNT, FeatureLZCNT.
Nominally, POPCNT and LZCNT are part of ABM instruction set,
and TZCNT is part of BMI1.

But i'm seeing something weird here:

 ./bin/llvm-exegesis --opcode-name=TZCNT32rr --benchmarks-file=- --mode=latency
Check generated assembly with: /usr/bin/objdump -d /tmp/snippet-fe658c.o
---
mode:            latency
key:             
  instructions:    
    - 'TZCNT32rr R11D R11D'
  config:          ''
  register_initial_values: 
    - 'R11D=0x0'
cpu_name:        amdfam10
llvm_triple:     x86_64-unknown-linux-gnu
num_repetitions: 10000
measurements:    
  - { key: latency, value: 4.0207, per_snippet_value: 4.0207 }
error:           ''
info:            Repeating a single explicitly serial instruction
assembled_snippet:
41BB00000000F3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBF3450FBCDBC3
...

That CPU does not support BMI1, yet TZCNT32rr works?

$ ./bin/clang -mbmi /tmp/test.cpp && ./a.out ; echo $?
6
$ cat /tmp/test.cpp 
#include <x86intrin.h>

int main(int argc, char* argv[]) {
    return _tzcnt_u32(0b1000000); // yes, i tried with different patterns
}
$ ./bin/clang -mbmi /tmp/test.cpp -o - -S
        .text
        .file   "test.cpp"
        .globl  main                    # -- Begin function main
        .p2align        4, 0x90
        .type   main, at function
main:                                   # @main
        .cfi_startproc
# %bb.0:                                # %entry
        pushq   %rbp
        .cfi_def_cfa_offset 16
        .cfi_offset %rbp, -16
        movq    %rsp, %rbp
        .cfi_def_cfa_register %rbp
        movl    $0, -8(%rbp)
        movl    %edi, -12(%rbp)
        movq    %rsi, -24(%rbp)
        movl    $64, -4(%rbp)
        movl    -4(%rbp), %eax
        tzcntl  %eax, %eax
        popq    %rbp
        .cfi_def_cfa %rsp, 8
        retq
.Lfunc_end0:
        .size   main, .Lfunc_end0-main
        .cfi_endproc
                                        # -- End function

        .ident  "clang version 9.0.0 (https://github.com/llvm-mirror/clang.git
fecd5a1c92f86f9dd15e8aea516c0e7d21f46dac)
(https://github.com/llvm-mirror/llvm.git
50613bd4fdcec438dd605506550709322262cf5b)"
        .section        ".note.GNU-stack","", at progbits
        .addrsig

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20190616/50ccc069/attachment-0001.html>