[Bug jit/123004] New: [14/15/16 Regression] jit.dg/test-cold-attribute.c failing since r16-1575

[siddhesh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123004

            Bug ID: 123004
           Summary: [14/15/16 Regression] jit.dg/test-cold-attribute.c
                    failing since r16-1575
           Product: gcc
           Version: 16.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: jit
          Assignee: dmalcolm at gcc dot gnu.org
          Reporter: siddhesh at gcc dot gnu.org
                CC: antoyo at gcc dot gnu.org
  Target Milestone: ---

Since r16-1575:

commit b7e9ac6549ec5c73ea1a7ae7514ad1a049e801f1
Author: H.J. Lu <hjl.tools@gmail.com>
Date:   Sun May 25 07:40:29 2025 +0800

    x86: Enable *mov<mode>_(and|or) only for -Oz

jit.dg/test-cold-attribute.c has been failing:

```
PASS: output-of-test-cold-attribute.c.s exists
FAIL: output-of-test-cold-attribute.c.s output pattern test, is         .file  
"fake.c"
        .text
        .section        .text.unlikely,"ax",@progbits
        .globl  t
        .type   t, @function
t:
.LFB0:
        .cfi_startproc
        movl    $-1, %eax
        ret
        .cfi_endproc
.LFE0:
        .size   t, .-t
        .ident  "GCC: (GNU) 14.3.1 20251022 (Red Hat 14.3.1-4)"
        .section        .note.GNU-stack,"",@progbits
, should match orl
```

It looks like the test needs to be fixed up similar to
gcc.target/i386/cold-attribute-4.c, but it doesn't look like gccjit has a way
to provide a string optimization level option, so I suppose that needs to be
improved to allow the test to be fixed.  Setting component to jit for that.

The fix was backported at least as far as gcc14.