[cfe-dev] Clang-Cl - support for the FAcs flag?

[Nico Weber via cfe-dev]

Yes: If you also pass /Z7 to clang-cl, then the assembly will contain debug
information, containing line markers. You can also pass the assembly
through a demangler (MSVC's undname, or e.g.https://github.com/nico/demumble)
to unmangle function names.

$ cat test.cc
int f(int a, int b) {
  int c = a + b;
  int d = a * b;
  return c / d;
}

$ bin/clang-cl /Z7 /c /FA test.cc && cat test.asm | demumble
...
"int __cdecl f(int,int)":                          # @"\01int __cdecl
f(int,int)"
...
.cv_loc 0 1 1 0 is_stmt 0       # test.cc:1:0
.cv_fpo_proc "int __cdecl f(int,int)" 8
# %bb.0:                                # %entry
push ebp
.cv_fpo_pushreg ebp
mov ebp, esp
.cv_fpo_setframe ebp
sub esp, 16
.cv_fpo_stackalloc 16
.cv_fpo_endprologue
mov eax, dword ptr [ebp + 12]
mov ecx, dword ptr [ebp + 8]
Ltmp0:
.cv_loc 0 1 2 0                 # test.cc:2:0
mov edx, dword ptr [ebp + 8]
add edx, dword ptr [ebp + 12]
mov dword ptr [ebp - 4], edx
.cv_loc 0 1 3 0                 # test.cc:3:0
mov edx, dword ptr [ebp + 8]
imul edx, dword ptr [ebp + 12]
mov dword ptr [ebp - 8], edx
.cv_loc 0 1 4 0                 # test.cc:4:0
mov edx, dword ptr [ebp - 4]
mov dword ptr [ebp - 12], eax # 4-byte Spill
mov eax, edx
cdq
idiv dword ptr [ebp - 8]
mov dword ptr [ebp - 16], ecx # 4-byte Spill
add esp, 16
pop ebp
ret

See the `# test.cc:N` comments at the end of the .cv_loc lines.


On Wed, Mar 7, 2018 at 8:05 AM, <bjoern.gaier at horiba.com> wrote:

> I don't think I'm good enough to make a patch for clang.
> Is there at least a way to have the line numbers of the source code in the
> assembly output?
>
>
>
> From:        Nico Weber <thakis at chromium.org>
> To:        Reid Kleckner <rnk at google.com>
> Cc:        bjoern.gaier at horiba.com, cfe-dev <cfe-dev at lists.llvm.org>
> Date:        06.03.2018 19:33
> Subject:        Re: [cfe-dev] Clang-Cl - support for the FAcs flag?
> Sent by:        thakis at google.com
> ------------------------------
>
>
>
> But patches are of course gladly accepted, if you want to give it a try
> yourself :-)
>
> Other ways to get what you want:
> * If you use vim, whodis (*https://github.com/sgraham/whodis*
> <https://github.com/sgraham/whodis>) can show the assembly of the current
> function and it color-codes which parts of the function become which
> assembly. It works reasonably well. I haven't tried it on Windows yet, but
> in a chrome/win cross build on linux it worked well.
> * Ideally we'd add clang-cl support to *godbolt.org* <http://godbolt.org/>.
> I looked at this for a tiny bit a few months ago (
> *https://github.com/mattgodbolt/compiler-explorer/issues/204*
> <https://github.com/mattgodbolt/compiler-explorer/issues/204>) but got
> distracted (and I'm not super likely to get re-tracted…uh…back-on-track
> soon – so this would be a cool project for someone too)
>
> On Tue, Mar 6, 2018 at 12:54 PM, Reid Kleckner via cfe-dev <
> *cfe-dev at lists.llvm.org* <cfe-dev at lists.llvm.org>> wrote:
> I don't think there are any plans to add it. We don't have support for the
> equivalent GCC flags yet either (-Wa,-adhln, I think?), so this isn't just
> a matter of hooking things up. The feature would need to be implemented
> from scratch.
>
>
> On Tue, Mar 6, 2018 at 12:59 AM via cfe-dev <*cfe-dev at lists.llvm.org*
> <cfe-dev at lists.llvm.org>> wrote:
> Hello Clang-Dev-List-People,
>
> I use clang-cl integrated to Visual Studio to compile my projects. This is
> working fine. I wanted to ask, if there are plans for supporting the /FAcs
> flag of the visual studio compiler? Or at least is there a work around?
>
> For those how doesn't know the flag:
> It generates assembly output, but also mixes the C/C++ source code as
> comments into it. This really helps to find points of interest.
>
> Kind regards
> Björn
>
> Als GmbH eingetragen im Handelsregister Bad Homburg v.d.H. HRB 9816,
> USt.ID-Nr. DE 114 165 789
> Geschäftsführer: Dr. Hiroshi Nakamura, Dr. Robert Plank, Markus Bode,
> Heiko Lampert, Hiroshi Kawamura, Takashi Nagano, Takeshi Fukushima.
>
> _______________________________________________
> cfe-dev mailing list
> *cfe-dev at lists.llvm.org* <cfe-dev at lists.llvm.org>
> *http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev*
> <http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev>
>
> _______________________________________________
> cfe-dev mailing list
> *cfe-dev at lists.llvm.org* <cfe-dev at lists.llvm.org>
> *http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev*
> <http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev>
>
>
>
>
> Als GmbH eingetragen im Handelsregister Bad Homburg v.d.H. HRB 9816,
> USt.ID-Nr. DE 114 165 789
> Geschäftsführer: Dr. Hiroshi Nakamura, Dr. Robert Plank, Markus Bode,
> Heiko Lampert, Hiroshi Kawamura, Takashi Nagano, Takeshi Fukushima.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20180307/16dd9b05/attachment.html>