This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c/76783] [7 Regression] wrong code with conditional vector assignment @ -Og
- From: "rguenth at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Mon, 15 Aug 2016 13:37:55 +0000
- Subject: [Bug c/76783] [7 Regression] wrong code with conditional vector assignment @ -Og
- Authentication-results: sourceware.org; auth=none
- Auto-submitted: auto-generated
- References: <bug-76783-4@http.gcc.gnu.org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=76783
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Last reconfirmed| |2016-08-15
Component|rtl-optimization |c
Target Milestone|--- |7.0
Ever confirmed|0 |1
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed. Note that I think the vector construction doesn't work already.
From
the .gimple dump:
foo (unsigned int c, v64u128 v)
{
u128 D.1767;
v64u128 u;
try
{
if (c != 0) goto <D.1764>; else goto <D.1765>;
<D.1764>:
u = { 0, 0, 0, 0 };
goto <D.1766>;
<D.1765>:
u = { 0, 1, 0, 0 };
<D.1766>:
u.0_1 = u;
_2 = v + u.0_1;
u = _2;
D.1767 = BIT_FIELD_REF <u, 128, 128>;
return D.1767;
here the upper __int128 in the vector is obviously zero. Somehow a mix-up
somewhere.