[Bug c/114097] New: Missed register optimization in _Noreturn functions

[Bug c/114097] New: Missed register optimization in _Noreturn functions

[pskocik at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

            Bug ID: 114097
           Summary: Missed register optimization in _Noreturn functions
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pskocik at gmail dot com
  Target Milestone: ---

Consider a never-returning functions such as this:

#include<setjmp.h>
#include<stdio.h>
//_Noreturn
void noret(unsigned A, unsigned B, unsigned C, unsigned D, unsigned E, jmp_buf
Jb){

    for(;A--;) puts("A");
    for(;B--;) puts("B");
    for(;C--;) puts("C");
    for(;D--;) puts("D");
    for(;E--;) puts("E");

    longjmp(Jb,1);
}

https://godbolt.org/z/35YjrhjYq

In its prologue, gcc saves the arguments in call-preserved registers to
preserve them around the puts calls, and it does so the usual way: by (1)
pushing the old values of the call-preserved registers to the stack and (2)
actually moving the arguments into the call-preserved registers.

        pushq   %r15
        movq    %r9, %r15
        pushq   %r14
        movl    %edi, %r14d
        pushq   %r13
        movl    %esi, %r13d
        pushq   %r12
        movl    %edx, %r12d
        pushq   %rbp
        movl    %ecx, %ebp
        pushq   %rbx
        movl    %r8d, %ebx
        pushq   %rax
        //...

Since this function demonstrably never returns, step 1 can be entirely elided
as the old values of the call-preserved registers won't ever need to be
restored
(desirably, gcc does not generate the would-be-dead restoration code):


        movq    %r9, %r15
        movl    %edi, %r14d
        movl    %esi, %r13d
        movl    %edx, %r12d
        movl    %ecx, %ebp
        movl    %r8d, %ebx
        pushq   %rax
        //...

(Also desirable would be the unrealized tailoptimization of the longjmp call in
this case: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=10837)

[Bug c/114097] Missed register optimization in _Noreturn functions

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2024-02-25
            Version|unknown                     |14.0
                 CC|                            |hjl.tools at gmail dot com
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1

--- Comment #1 from H.J. Lu <hjl.tools at gmail dot com> ---
__attribute__((noreturn)) works in GCC 14:

[hjl@gnu-cfl-3 tmp]$ cat y.c
#include<setjmp.h>
#include<stdio.h>
//_Noreturn
__attribute__((noreturn))
void noret(unsigned A, unsigned B, unsigned C, unsigned D, unsigned E, jmp_buf
Jb){

    for(;A--;) puts("A");
    for(;B--;) puts("B");
    for(;C--;) puts("C");
    for(;D--;) puts("D");
    for(;E--;) puts("E");

    longjmp(Jb,1);
}
[hjl@gnu-cfl-3 tmp]$ /usr/gcc-14.0.1-x32/bin/gcc -S -O2 y.c
[hjl@gnu-cfl-3 tmp]$ cat y.s
        .file   "y.c"
        .text
        .section        .rodata.str1.1,"aMS",@progbits,1
.LC0:
        .string "A"
.LC1:
        .string "B"
.LC2:
        .string "C"
.LC3:
        .string "D"
.LC4:
        .string "E"
        .text
        .p2align 4
        .globl  noret
        .type   noret, @function
noret:
.LFB11:
        .cfi_startproc
        subq    $8, %rsp
        .cfi_def_cfa_offset 16
        movl    %esi, %r15d
        movl    %edx, %r14d
        movl    %ecx, %r13d
        movl    %r8d, %ebp
        movq    %r9, %r12
        testl   %edi, %edi
        je      .L2
        leal    -1(%rdi), %ebx
        .p2align 4,,10
        .p2align 3
.L3:
        movl    $.LC0, %edi
        call    puts
        subl    $1, %ebx
        jnb     .L3
.L2:
        leal    -1(%r15), %ebx
        testl   %r15d, %r15d
        je      .L4
        .p2align 4,,10
        .p2align 3
.L5:
        movl    $.LC1, %edi
        call    puts
        subl    $1, %ebx
        jnb     .L5
.L4:
        leal    -1(%r14), %ebx
        testl   %r14d, %r14d
        je      .L6
        .p2align 4,,10
        .p2align 3
.L7:
        movl    $.LC2, %edi
        call    puts
        subl    $1, %ebx
        jnb     .L7
.L6:
        leal    -1(%r13), %ebx
        testl   %r13d, %r13d
        je      .L8
        .p2align 4,,10
        .p2align 3
.L9:
        movl    $.LC3, %edi
        call    puts
        subl    $1, %ebx
        jnb     .L9
.L8:
        leal    -1(%rbp), %ebx
        testl   %ebp, %ebp
        je      .L10
        .p2align 4,,10
        .p2align 3
.L11:
        movl    $.LC4, %edi
        call    puts
        subl    $1, %ebx
        jnb     .L11
.L10:
        movl    $1, %esi
        movq    %r12, %rdi
        call    longjmp
        .cfi_endproc
.LFE11:
        .size   noret, .-noret
        .ident  "GCC: (GNU) 14.0.1 20240223 (experimental)"
        .section        .note.GNU-stack,"",@progbits
[hjl@gnu-cfl-3 tmp]$

[Bug c/114097] Missed register optimization in _Noreturn functions

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

--- Comment #2 from H.J. Lu <hjl.tools at gmail dot com> ---
I couldn't find a way to access the _Noreturn info in backend.

[Bug c/114097] Missed register optimization in _Noreturn functions

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

--- Comment #3 from H.J. Lu <hjl.tools at gmail dot com> ---
Created attachment 57524
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=57524&action=edit
A patch

[Bug target/114097] Missed register optimization in _Noreturn functions

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|c                           |target
   Target Milestone|---                         |14.0
           Assignee|unassigned at gcc dot gnu.org      |hjl.tools at gmail dot com

[Bug target/114097] Missed register optimization in _Noreturn functions

[pskocik at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

--- Comment #4 from Petr Skocik <pskocik at gmail dot com> ---
Excellent! Thank you very much. Didn't realize the functionality was already
there, but didn't work without an explicit __attribute((noreturn)). Now I can
get rid of my most complex assembly function which I stupidly (back then I
thought cleverly) wrote. :)

[Bug target/114097] Missed register optimization in _Noreturn functions

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

--- Comment #5 from H.J. Lu <hjl.tools at gmail dot com> ---
A patch is submitted:

https://patchwork.sourceware.org/project/gcc/list/?series=31294

[Bug target/114097] Missed register optimization in _Noreturn functions

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

--- Comment #6 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by H.J. Lu <hjl@gcc.gnu.org>:

https://gcc.gnu.org/g:bb98f71bac8aace4e685e648a81dfaf365123833

commit r14-9178-gbb98f71bac8aace4e685e648a81dfaf365123833
Author: H.J. Lu <hjl.tools@gmail.com>
Date:   Sun Feb 25 13:14:39 2024 -0800

    x86: Check interrupt instead of noreturn attribute

    ix86_set_func_type checks noreturn attribute to avoid incompatible
    attribute error in LTO1 on interrupt functions.  Since TREE_THIS_VOLATILE
    is set also for _Noreturn without noreturn attribute, check interrupt
    attribute for interrupt functions instead.

    gcc/

            PR target/114097
            * config/i386/i386-options.cc (ix86_set_func_type): Check
            interrupt instead of noreturn attribute.

    gcc/testsuite/

            PR target/114097
            * gcc.target/i386/pr114097-1.c: New test.

[Bug target/114097] Missed register optimization in _Noreturn functions

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114097

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|NEW                         |RESOLVED

--- Comment #7 from H.J. Lu <hjl.tools at gmail dot com> ---
Fixed.