https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84723
Martin Sebor <msebor at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Keywords| |ice-on-valid-code
Status|UNCONFIRMED |NEW
Last reconfirmed| |2018-03-06
CC| |msebor at gcc dot gnu.org
Ever confirmed|0 |1
Known to fail| |8.0
--- Comment #1 from Martin Sebor <msebor at gcc dot gnu.org> ---
Confirmed. The first revision to ICE is r250736 (gcc 8.0.0), which is also
where attribute naked was added.
r250736 | uros | 2017-07-31 06:22:41 -0400 (Mon, 31 Jul 2017) | 23 lines
PR target/25967
* config/i386/i386.c (ix86_function_naked): New function.
(ix86_can_use_return_insn_p): Return false for naked functions.
(ix86_expand_prologue): Skip prologue for naked functions.
(ix86_expand_epilogue): Skip epilogue for naked functions
and emit trap instruction.
(ix86_warn_func_return): New function.
(ix86_attribute_table): Add "naked" attribute specification.
(TARGET_WARN_FUNC_RETURN): Define.
* doc/extend.texi (x86 Function Attributes) <naked>: Document it.
Another test case:
__attribute__((target_clones ("avx", "default")))
__attribute__((noclone))
void foo (void) { }
The create_target_clone() function assumes the call to
cgraph_node::create_version_clone_with_body() succeeds, which isn't necessarily
the case.
But mutually exclusive attributes should be detected earlier on, during
parsing, so if target_clones is incompatible with naked, GCC should issue a
warning and drop the latter attribute.
