From owner-freebsd-questions Fri Nov 10 19:56:01 1995 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id TAA27551 for questions-outgoing; Fri, 10 Nov 1995 19:56:01 -0800 Received: from kalypso.iqm.unicamp.br (kalypso.iqm.unicamp.br [143.106.13.10]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id TAA27536 for ; Fri, 10 Nov 1995 19:55:35 -0800 Received: (from vazquez@localhost) by kalypso.iqm.unicamp.br (8.6.11/FreBSD2.0.5-8.6.9) id BAA24067 for freebsd-questions@freebsd.org; Sat, 11 Nov 1995 01:53:10 GMT From: Pedro A M Vazquez Message-Id: <199511110153.BAA24067@kalypso.iqm.unicamp.br> Subject: Floating Exception problem with P100 To: freebsd-questions@freebsd.org Date: Sat, 11 Nov 1995 01:53:07 +0000 () Organization: Instituto de Quimica Unicamp X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 2721 Sender: owner-questions@freebsd.org Precedence: bulk Hello I got a P100 motherboard to speed up my quantum chemistry calculations but I'm having a very strange problem. The same executable (GAMESS/95) running on my old 486DX33 works fine but aborts with floating overflow with the pentium board. I've tried several input test files and all give the same result. If I mask the overflows the results are wrong on the P100 but still right on the 486. Both machines are running 2.0.5 and the program was compiled under this version too. From the boot I got the following for the P100 (it has an Award BIOS): Nov 9 04:56:03 nca /kernel: CPU: 100-MHz Pentium 815\100 (Pentium-class CPU) Nov 9 04:56:03 nca /kernel: Origin = "GenuineIntel" Id = 0x525 Stepping=5 Nov 9 04:56:03 nca /kernel: Features=0x1bf Nov 9 04:56:03 nca /kernel: real memory = 16384000 (4000 pages) Nov 9 04:56:03 nca /kernel: avail memory = 14807040 (3615 pages) Nov 9 04:56:03 nca /kernel: Probing for devices on the ISA bus: Nov 9 04:56:04 nca /kernel: sc0 at 0x60-0x6f irq 1 on motherboard Nov 9 04:56:04 nca /kernel: sc0: VGA color <16 virtual consoles, flags=0x0> Nov 9 04:56:04 nca /kernel: ed1 at 0x300-0x31f irq 5 on isa Nov 9 04:56:04 nca /kernel: ed1: address "00:20:a9:01:a4:6f, type NE2000 (16 bit) Nov 9 04:56:04 nca /kernel: sio0 at 0x3f8-0x3ff irq 4 on isa Nov 9 04:56:05 nca /kernel: sio0: type 16550A Nov 9 04:56:05 nca /kernel: sio1 at 0x2f8-0x2ff irq 3 on isa Nov 9 04:56:05 nca /kernel: sio1: type 16550A Nov 9 04:56:06 nca /kernel: lpt0 at 0x378-0x37f irq 7 on isa Nov 9 04:56:06 nca /kernel: lpt0: Interrupt-driven port Nov 9 04:56:06 nca /kernel: lp0: TCP/IP capable interface Nov 9 04:56:07 nca /kernel: fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa Nov 9 04:56:07 nca /kernel: fdc0: NEC 72065B Nov 9 04:56:07 nca /kernel: fd0: 1.44MB 3.5in Nov 9 04:56:07 nca /kernel: wdc0 at 0x1f0-0x1f7 irq 14 on isa Nov 9 04:56:07 nca /kernel: wdc0: unit 0 (wd0): Nov 9 04:56:07 nca /kernel: wd0: 516MB (1057392 sectors), 1049 cyls, 16 heads, 63 S/T, 512 B/S Nov 9 04:56:10 nca /kernel: npx0 on motherboard Nov 9 04:56:10 nca /kernel: npx0: INT 16 interface Nov 9 04:56:10 nca /kernel: Probing for devices on the pci0 bus: Nov 9 04:56:10 nca /kernel: configuration mode 1 allows 32 devices. Nov 9 04:56:10 nca /kernel: pci0:0: OPTI, device=0xc557, class=bridge [not supported] Nov 9 04:56:11 nca /kernel: pci0:1: OPTI, device=0xc558, class=bridge [not supported] Nov 9 04:56:11 nca /kernel: pci0:20: OPTI, device=0xc621, class=storage [not supported] Nov 9 04:58:49 nca /kernel: map(20): io(300) Any ideas or help ? I've checked for the Pentium bug and it seems ok (no fdiv bug) Thanks Pedro