[Bug target/19162] [4.0 regression] ICE while building libobjc's sendmsg.c

2004-12-26 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-12-27 00:18 --- Could you test the patch in and report back the results so I send in the patch since I don't have access to arm-linux and arm-elf will doe

[Bug target/19162] [4.0 regression] ICE while building libobjc's sendmsg.c

2004-12-26 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-12-27 00:17 --- Created an attachment (id=7827) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7827&action=view) patch which should fix this This is easiest patch which will fix the bug, basically we need a movxf and

[Bug target/19162] [4.0 regression] ICE while building libobjc's sendmsg.c

2004-12-26 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-12-26 22:47 --- The problem is that there is no movxf. So this is definitely a target problem. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19162 --- You are receiving this mail because: --- You reported the

[Bug target/19162] [4.0 regression] ICE while building libobjc's sendmsg.c

2004-12-26 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-12-26 22:26 --- We are trying to emit the following RTL: (Set (reg:XF 115) (reg:XF 16 f0)) So maybe this bug is in the middle-end and related to the subreg work. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19162 -

[Bug target/19162] [4.0 regression] ICE while building libobjc's sendmsg.c

2004-12-26 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-12-26 22:21 --- The bug is in untyped_call. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19162 --- You are receiving this mail because: --- You reported the bug, or are watching the reporter.

[Bug target/19162] [4.0 regression] ICE while building libobjc's sendmsg.c

2004-12-26 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-12-26 22:19 --- Here is the reduced testcase: int *f(void) { return __builtin_apply (0,0,0); } We are just going into an infinite loop: #254 0x002d6c04 in operand_subword_force (op=0x4167a020, offset=0, mode=XFmode) at

[Bug target/19162] [4.0 regression] ICE while building libobjc's sendmsg.c

2004-12-26 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-12-26 21:36 --- Confirmed, this bug is related to __builtin_apply, I was waiting for this to be filed. I wonder if there is something which can be done without disabling libobjc for arm-linux or just using libffi. --

Processed: lower severity of gcc-3.3 report

2004-12-26 Thread Debian Bug Tracking System
Processing commands for [EMAIL PROTECTED]: > severity 286840 important Bug#286840: gcc-3.3: ia64: build failures with -fno-omit-frame-pointer Severity set to `important'. > tags 286840 + fixed-upstream Bug#286840: gcc-3.3: ia64: build failures with -fno-omit-frame-pointer There were no tags set.

Bug#173513: [Bug tree-optimization/9079] [tree-ssa] Inline constant function pointers

2004-12-26 Thread sxanth at ceid dot upatras dot gr
-- What|Removed |Added CC||sxanth at ceid dot upatras ||dot gr http://gcc.gnu.org/bugzilla

Want a new COMPUTER

2004-12-26 Thread Yolanda Holman
Special 0ffer on M1cros0ft sOftware Leading SOftware such as: Bundle 1: Wind0ws X.P PrO + 0ffice X'P Pr0 for as low as 8o$ full report Bundle 2: Macromedia Dreamwaver MX 20o4 + Flash MX 20o4 - 1OO D0||ars Bundle 3: Adobe - PhOtosh0p 7, Premiere 7, Illustrator 10 - 12o D0\\ars The offer