[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Problems compiling gcc on OpenBSD/sparc
- To: tech_(_at_)_openbsd_(_dot_)_org
- Subject: Re: Problems compiling gcc on OpenBSD/sparc
- From: David <david_(_at_)_southern_(_dot_)_com>
- Date: Sun, 28 Jan 1996 16:04:55 +0000 (GMT)
- Reply-to: david_(_at_)_mono_(_dot_)_org
Update on this...
The same source with the same -O option compiled fine on my
ss2 'SUNW,Sun 4/75 (RT601 or L64811 v2 @ 40 MHz, TMS390C602A FPU)'.
The 'enquire' from the ss2 panics the sun4/300.
(Final moments of 4/300 below)
This leave me with a 92k binary that can panic a 4/300 on demand...
cool...
David
PROPERTIES OF DOUBLE
Base = 2
Significant base digits = 53 (= at least 15 decimal digits)
Arithmetic rounds towards nearest
Tie breaking rounds to even
Smallest x such that 1.0-base**x != 1.0 = -53
Smallest panic: fpu_cleanup 1
Stopped at 0xf80ab980: jmpl [%o7 + 0x8], %g0
db> x such tchaot 1n.0t-x != 1.0 55115123125783
db>9 e-17
Smallest
db> x s
db> uch
db> thact o1.n0+btas
syncing disks... e**x != 1.0 = -52
Sm7 7 allest x such that 1.0+x != 1.0 = 1.1102230246251568e-16
(Above number + 1.0) - 1.0 = 2.25 204460492503131e-16
Number of bits used for exponent = 11
Minimum normalised exponent = -1022
Minimum normalised positive number = 2.2250738585072014e-308
On Sun, 28 Jan 1996, David wrote:
> When I try to compile the gcc in the OpenBSD/sparc source tree
> it panics when running ./enquire.
> This is on a 'SUN-4/300 series (L64812 or ACT8847 FPU)'.
>
> Anyone have any ideas on this? - I'm going to try compiling it
> with the default '-g' instead to see if its spcifically the -O
> thats causing the problem - I'll try compiling it on an ss2 as
> well...
>
> David
>
> ./enquire -f > tmp-float.h
> panic: fpu_cleanup 1
> Stopped at 0xf80ab980: jmpl [%o7 + 0x8], %g0
> db>
> db> trace
> ?(f86a7100, f8615800, 1000, 3, 9a5, 2000) at 0xf809b9fc
> ?(f86a7100, 11401084, 40eb35c, f9f51fb0, ffffffff, a) at 0xf80a95d0
> ?(4109730, 20, 4, 4109708, 81cf41, 0) at 0xf8006680
> db> cont
> syncing disks... 8 8 6 3 done
> Frame pointer is at 0xf9f51cd0
> Call traceback:
> pc = f80a1fd0 args = (f80a1c00, 11001fe0, f80bd000, f9f51d50, 4ce0, c)
> fp = f9f51d38
> pc = f801bf50 args = (100, f80b8c00, 1, 4e08, 10, 100000) fp = f9f51da0
> pc = f809b9fc args = (100, 0, 9a5, 0, 1f, 22170) fp = f9f51e08
> pc = f80a95d0 args = (f86a7100, f8615800, 1000, 3, 9a5, 2000) fp =
> f9f51ee8
> pc = f8006680 args = (f86a7100, 11401084, 40eb35c, f9f51fb0, ffffffff,
> a) fp = f9f51f50
> pc = 0 args = (4109730, 20, 4, 4109708, 81cf41, 0) fp = f7fff3f0
>
> dumping to dev 701, offset 16
>
>
> David/abs (david_(_at_)_mono_(_dot_)_org)
>
> +44 181 888 8949 System Manager, Southern Studios Ltd, Box 59, London N22 1AR
> >=- Microsoft asks you where you want to go. Unix gets you there -=<
> <<< Monochrome - Largest UK Internet BBS - telnet mono.org >>>
>
>
Visit your host, monkey.org