Bug 1032854

Summary: aarch64: ICE in emit_move_insn, в expr.c:3486 with double __attribute__ ((vector_size (8)))
Product: [Fedora] Fedora Reporter: Petr Machata <pmachata>
Component: gccAssignee: Jakub Jelinek <jakub>
Status: CLOSED UPSTREAM QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: rawhideCC: blc, jakub, law, mnewsome
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2014-01-16 14:56:58 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 922257    

Description Petr Machata 2013-11-21 02:59:10 UTC
Description of problem:

# LANG=C gcc x.c
x.c: In function 'fun_vec_double_8':
x.c:4:3: internal compiler error: in emit_move_insn, at expr.c:3486
   return ret;
   ^
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://bugzilla.redhat.com/bugzilla> for instructions.
Preprocessed source stored into /tmp/ccc768lu.out file, please attach this to your bugreport.

# cat /tmp/ccc768lu.out
// /usr/libexec/gcc/aarch64-redhat-linux/4.8.1/cc1 -quiet x.c -quiet -dumpbase x.c -auxbase x -o - -frandom-seed=0
# 1 "x.c"
# 1 "<command-line>"
# 1 "/usr/include/stdc-predef.h" 1 3 4
# 1 "<command-line>" 2
# 1 "x.c"
typedef double __attribute__ ((vector_size (8))) vec_double_8_t;
vec_double_8_t fun_vec_double_8 (void) {
  vec_double_8_t ret = { 3.5 };
  return ret;
}

Version-Release number of selected component (if applicable):
gcc-4.8.1-10.x1.fc19.aarch64

How reproducible:
Consistently.

Steps to Reproduce:
1. As indicated.
2.
3.

Actual results:
ICE.

Expected results:
Water?

Additional info:
There is http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56548 with similar symptoms, but that should have been fixed by 4.8, and the reproducer shown there doesn't do it for me.  Must be some other code with the same symptoms.

Comment 1 Jakub Jelinek 2014-01-16 14:56:58 UTC
Filed PR59843.  Why are you trying that BTW, single element vectors don't make any sense...