This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug target/64795] [5 regression] too many memory references for `lea'
- From: "ubizjak at gmail dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Mon, 26 Jan 2015 16:09:24 +0000
- Subject: [Bug target/64795] [5 regression] too many memory references for `lea'
- Auto-submitted: auto-generated
- References: <bug-64795-4 at http dot gcc dot gnu dot org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64795
UroÅ Bizjak <ubizjak at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |ASSIGNED
Last reconfirmed| |2015-01-26
Assignee|unassigned at gcc dot gnu.org |ubizjak at gmail dot com
Ever confirmed|0 |1
--- Comment #2 from UroÅ Bizjak <ubizjak at gmail dot com> ---
The failed insn is:
#(insn 8 6 13 2 (set (mem/f/j/c:DI (symbol_ref:DI ("_ZZ4mainE1a") [flags 0x2]
<var_decl 0x7fe4d00f2900 a>) [0 a.pf+0 S8 A64])
# (symbol_ref:DI ("_ZL3fn21s") [flags 0x3] <function_decl 0x7fe4d02dd0d8
fn2>)) t.ii:10 89 {*movdi_internal}
# (nil))
leaq _ZL3fn21s(%rip), _ZZ4mainE1a(%rip) # 8 *movdi_internal/6
[length = 8]
Patch that tightens lea constraints in testing:
Index: i386.md
===================================================================
--- i386.md (revision 220092)
+++ i386.md (working copy)
@@ -2208,7 +2208,8 @@
(const_string "ssecvt")
(eq_attr "alternative" "21,22,23,24")
(const_string "mskmov")
- (match_operand 1 "pic_32bit_operand")
+ (and (match_operand 0 "register_operand")
+ (match_operand 1 "pic_32bit_operand"))
(const_string "lea")
]
(const_string "imov")))
@@ -2361,7 +2362,8 @@
(const_string "ssemov")
(eq_attr "alternative" "13,14")
(const_string "mskmov")
- (match_operand 1 "pic_32bit_operand")
+ (and (match_operand 0 "register_operand")
+ (match_operand 1 "pic_32bit_operand"))
(const_string "lea")
]
(const_string "imov")))