public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "theodort at inf dot ethz.ch" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/109943] New: [13/14 Regression] Missed Dead Code Elimination when using __builtin_unreachable since r13-6834-g41ade3399bd
Date: Tue, 23 May 2023 13:30:32 +0000	[thread overview]
Message-ID: <bug-109943-4@http.gcc.gnu.org/bugzilla/> (raw)

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109943

            Bug ID: 109943
           Summary: [13/14 Regression] Missed Dead Code Elimination when
                    using __builtin_unreachable since
                    r13-6834-g41ade3399bd
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: theodort at inf dot ethz.ch
  Target Milestone: ---

void foo(void);
static int b = 5;
static int *c = &b;
static char(a)(char d, char g) {
    ;
    return d % g;
}
static void e(unsigned char d, unsigned g) {
    char h = a(g, d);
    if (h)
        if ((d) >= 0) {
            foo();
        };
}
static void f(int d) { e(b, d); }
int main() {
    f(*c);
    *c = 0;
}

gcc-trunk -O3 generates:

main:
        movl    $0, b(%rip)
        xorl    %eax, %eax
        ret

However, including additional info via __builtin_unreachable results to worse
code:

void foo(void);
static int b = 5;
static int *c = &b;
static char(a)(char d, char g) {
    if ((g) <= 0) {
        __builtin_unreachable();
    }
    return d % g;
}
static void e(unsigned char d, unsigned g) {
    char h = a(g, d);
    if (h)
        if ((d) >= 0) {
            foo();
        };
}
static void f(int d) { e(b, d); }
int main() {
    f(*c);
    *c = 0;
}

gcc-trunk (and 13) -O3 generates:

main:
        movl    b(%rip), %ecx
        movsbl  %cl, %eax
        movzbl  %cl, %ecx
        cltd
        idivl   %ecx
        testl   %edx, %edx
        jne     .L11
        movl    $0, b(%rip)
        xorl    %eax, %eax
        ret
.L11:
        pushq   %rax
        call    foo
        xorl    %edx, %edx
        xorl    %eax, %eax
        movl    %edx, b(%rip)
        popq    %rcx
        ret

https://godbolt.org/z/n34Tsr3vq

gcc-12 can generate optimized code for both cases

Started with r13-6834-g41ade3399bd

             reply	other threads:[~2023-05-23 13:30 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-23 13:30 theodort at inf dot ethz.ch [this message]
2023-05-23 15:56 ` [Bug tree-optimization/109943] " rguenth at gcc dot gnu.org
2023-06-20 16:34 ` theodort at inf dot ethz.ch
2023-07-27  9:26 ` rguenth at gcc dot gnu.org
2023-08-21 23:09 ` pinskia at gcc dot gnu.org
2024-01-12 13:50 ` rguenth at gcc dot gnu.org
2024-01-12 13:52 ` rguenth at gcc dot gnu.org
2024-05-21  9:15 ` [Bug tree-optimization/109943] [13/14/15 " jakub at gcc dot gnu.org

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=bug-109943-4@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).