[llvm-dev] Where's the optimiser gone? (part 5.a): missed tail calls, and more...

[Stefan Kanthak via llvm-dev]

Compile the following functions with "-O3 -target amd64"
(see <https://godbolt.org/z/5xqYhH>):

__int128 div(__int128 foo, __int128 bar)
{
    return foo / bar;
}

On the left the generated code; on the right the expected,
properly optimised code:

div: # @div
    push  rbp         |
    mov   rbp, rsp    |
    call  __divti3    |    jmp   __divti3
    pop   rbp         |
    ret               |


__int128 mod(__int128 foo, __int128 bar)
{
    return foo % bar;
}

mod: # @mod
    push  rbp         |
    mov   rbp, rsp    |
    call  __modti3    |    jmp   __modti3
    pop   rbp         |
    ret               |


__int128 mul(__int128 foo, __int128 bar)
{
    return foo * bar;
}

mul: # @mul
    push  rbp
    mov   rbp, rsp
    mov   r8, rdx
    mov   rax, rdx
    mul   rdi
    imul  rsi, r8
    add   rdx, rsi
    imul  rcx, rdi
    add   rdx, rcx
    pop   rbp
    ret