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/110414] New: [14 Regression] Dead Code Elimination Regression since r14-1127-g9e2017ae6ac Date: Mon, 26 Jun 2023 15:06:40 +0000 [thread overview] Message-ID: <bug-110414-4@http.gcc.gnu.org/bugzilla/> (raw) https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110414 Bug ID: 110414 Summary: [14 Regression] Dead Code Elimination Regression since r14-1127-g9e2017ae6ac 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: --- https://godbolt.org/z/5E795dPvY Given the following code: void foo(void); static int a, b, e, l, m, n; static int *c = &a, *j; static int **d = &c; static int ***f; void __assert_fail(char *, char *, int, char *) __attribute__((__noreturn__)); static void g(char) { if (c == &b || c == &a) ; else __assert_fail("", "", 3, __PRETTY_FUNCTION__); } static void h(); static int *i(); static short k() { j = i(); h(); **f = &e; if (c == &b || c == &a || c == 0) __assert_fail("", "", 6, __PRETTY_FUNCTION__); return 0; } static void h() { if (l) { for (; l; l = l - 8) *d = &m; if (c == &a) ; else __assert_fail("", "", 9, __PRETTY_FUNCTION__); } if (c == &b || c == &a || c == 0) ; else { foo(); __assert_fail("", "", 8, __PRETTY_FUNCTION__); } } static int *i() { if (c == &b || c == &a) ; else { __builtin_unreachable(); __assert_fail("", "", 6, __PRETTY_FUNCTION__); } return &n; } int main() { g(a && k()); } gcc-trunk -Os does not eliminate the call to foo: main: pushq %rcx cmpl $0, a(%rip) je .L2 movl l(%rip), %edx testl %edx, %edx je .L3 movl %edx, %eax movl $__PRETTY_FUNCTION__.1, %ecx movq $m, c(%rip) shrl $3, %eax imull $-8, %eax, %eax addl %edx, %eax movl $9, %edx movl %eax, l(%rip) jmp .L7 .L3: movq c(%rip), %rax cmpq $a, %rax sete %dl cmpq $b, %rax sete %cl orb %cl, %dl jne .L4 testq %rax, %rax je .L4 call foo movl $__PRETTY_FUNCTION__.1, %ecx movl $8, %edx .L7: movl $.LC0, %esi movq %rsi, %rdi call __assert_fail .L4: movl $__PRETTY_FUNCTION__.2, %ecx movl $6, %edx jmp .L7 .L2: movq c(%rip), %rax cmpq $a, %rax je .L5 cmpq $b, %rax je .L5 movl $__PRETTY_FUNCTION__.0, %ecx movl $3, %edx jmp .L7 .L5: xorl %eax, %eax popq %rdx ret gcc-13.1.0 -Os eliminates the call to foo: main: pushq %rcx movq c(%rip), %rdx cmpq $a, %rdx sete %al cmpq $b, %rdx sete %dl orl %edx, %eax cmpl $0, a(%rip) je .L2 movl l(%rip), %edx testl %edx, %edx je .L3 movl %edx, %eax movl $__PRETTY_FUNCTION__.1, %ecx movq $m, c(%rip) shrl $3, %eax imull $-8, %eax, %eax addl %edx, %eax movl $9, %edx movl %eax, l(%rip) jmp .L7 .L2: testb %al, %al jne .L5 movl $__PRETTY_FUNCTION__.0, %ecx movl $3, %edx .L7: movl $.LC0, %esi movq %rsi, %rdi call __assert_fail .L3: movl $__PRETTY_FUNCTION__.2, %ecx movl $6, %edx jmp .L7 .L5: xorl %eax, %eax popq %rdx ret Bisects to r14-1127-g9e2017ae6ac
next reply other threads:[~2023-06-26 15:06 UTC|newest] Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top 2023-06-26 15:06 theodort at inf dot ethz.ch [this message] 2023-06-27 6:33 ` [Bug tree-optimization/110414] " rguenth at gcc dot gnu.org 2023-06-27 9:37 ` rguenth at gcc dot gnu.org 2024-03-08 15:30 ` law at gcc dot gnu.org 2024-05-07 7:40 ` [Bug tree-optimization/110414] [14/15 " rguenth 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-110414-4@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: linkBe 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).