Subject: Re: port-i386/6986
To: None <port-i386-maintainer@netbsd.org, gnats-admin@netbsd.org,>
From: Aaron J. Grier <agrier@poofygoof.com>
List: netbsd-bugs
Date: 03/02/2006 18:50:04
The following reply was made to PR port-i386/6986; it has been noted by GNATS.

From: "Aaron J. Grier" <agrier@poofygoof.com>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: port-i386/6986
Date: Thu, 2 Mar 2006 10:49:08 -0800

 we have a regression in the netbsd-3 branch.
 
 typed by hand since scp/ssh won't run.  !
 
 NetBSD 3.0_STABLE (GENERIC_TINY) #0: Thu Feb 16 22:01:02 PST 2006
 	aaron@mordor.unix.fryenet:/var/obj/amd/orthanc/usr/projects/NetBSD/3/src/sys/arch/i386/compile/GENERIC_TINY
 total memory = 7808 KB
 avail memory = 6032 KB
 mainbus0 (root)
 cpu0 at mainbus0: (uniprocessor)
 cpu: Intel 386DX (386-class)
 isa0 at mainbus0
 [... no fpu on this machine ...]
 
 # ping -c 1 gateway
 PING gateway.fryenet (10.0.0.1): 56 data bytes
 64 bytes from 10.0.0.1: icmp_seq=0 ttl=255 time=6.916 ms
 
 ----gateway.fryenet PING statistics----
 1 packets transmitted, 1 packets received, 0.0% packet loss
 math_emulate: 0xd9fa not implemented
 Illegal instruction (core dumped)
 
 GENERIC_TINY has
 options         MATH_EMULATE
 
 bla.c shown in the PR runs fine.  running gdb on ping shows the
 following:
 
 Program received signal SIGILL, Illegal instruction.
 0x0804b228 in summary ()
 (gdb) x/i $pc
 0x804b228 <summary+224>:	fsqrt
 (gdb) disassemble summary
 [...]
 0x804b224 <summary+220>:	fxch   %st(1)
 0x804b226 <summary+222>:	fld    %st(0)
 0x804b228 <summary+224>:	fsqrt
 
 I can't seem to get a test case that generates a fsqrt call, and don't
 have a verbose build log to see how gcc was invoked on ping.c.
 
 -- 
   Aaron J. Grier | "Not your ordinary poofy goof." | agrier@poofygoof.com
               "silly brewer, saaz are for pils!"  --  virt