stage1/xgcc -Bstage1/ -B/opt/gnu/gcc/gcc-3.4.4/hppa1.1-hp-hpux11.11/bin/ -c -g
-O2 -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototype
s -pedantic -Wno-long-long -Wold-style-definition -DHAVE_CONFIG_H -I. -I.
-I../../gcc/gcc -I../../gcc/gcc/. -I../../gcc/gcc/../include ../../gcc/gcc/cpp
exp.c -o cppexp.o
../../gcc/gcc/cppexp.c: In function `num_negate':
../../gcc/gcc/cppexp.c:1076: error: unrecognizable insn:
(insn 188 115 0 ../../gcc/gcc/cppexp.c:1073 (sequence [
(jump_insn 165 115 82 (parallel [
(set (pc)
(label_ref 187))
(set (reg:QI 24 %r24 [116])
(const_int 1 [0x1]))
]) 230 {*pa.md:8757} (nil)
(expr_list:REG_BR_PRED (const_int 4 [0x4])
(nil)))
(insn/s:TI 82 165 0 (set (mem/s:QI (plus:SI (reg/v/f:SI 3 %r3 [orig:
95 num ] [95])
(const_int 9 [0x9])) [22 num.overflow+0 S1 A8])
(reg:QI 24 %r24 [116])) 57 {*pa.md:3003} (nil)
(nil))
]) -1 (nil)
(nil))
../../gcc/gcc/cppexp.c:1076: internal compiler error: in insn_default_length, at
insn-attrtab.c:825
Please submit a full bug report,
with preprocessed source if appropriate.
Bah, I know what's wrong. The regression was caused by the following
fix:
2004-11-22 John David Anglin <[EMAIL PROTECTED]>
PR rtl-optimization/14838
* emit-rtl.c (get_first_nonnote_insn): Don't assume first insn is a
note.
(get_last_nonnote_insn): Don't assume last insn is a note.
The above fix doesn't handle the case when the first/last insn is a
sequence.
Patch in testing.
--
Summary: [3.3/3.4/4.0 Regression] cppexp.c:1076: error:
unrecognizable insn
Product: gcc
Version: 3.4.4
Status: UNCONFIRMED
Severity: critical
Priority: P2
Component: middle-end
AssignedTo: danglin at gcc dot gnu dot org
ReportedBy: danglin at gcc dot gnu dot org
CC: gcc-bugs at gcc dot gnu dot org
GCC build triplet: hppa1.1-hp-hpux*
GCC host triplet: hppa1.1-hp-hpux*
GCC target triplet: hppa1.1-hp-hpux*
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18730
