Subject: port-sparc64/13501: memcpy/memset userland lossage on sparc64
To: None <gnats-bugs@gnats.netbsd.org>
From: None <martin@duskware.de>
List: netbsd-bugs
Date: 07/18/2001 16:52:46
>Number:         13501
>Category:       port-sparc64
>Synopsis:       memcpy/memset userland lossage on sparc64
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    port-sparc64-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Jul 18 07:50:01 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Martin Husemann
>Release:        1.5W, July 18 2001
>Organization:
	
>Environment:
	
System: NetBSD setting-sun.duskware.de 1.5W NetBSD 1.5W (SETTINGSUN) #0: Wed Jul 18 12:25:57 MEST 2001 martin@setting-sun.duskware.de:/usr/src/sys/arch/sparc64/compile/SETTINGSUN sparc64
Architecture: sparc64
Machine: sparc64
>Description:

I'm running a current kernel (sources cvs updated this morning). After building
and installing a new libc, I get random errors from the toolchain, like

cc -O -pipe  -Wall -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Wno-uninitialized   -c /usr/src/lib/libbz2/../../dist/bzip2/blocksort.c
cpp: output pipe has been closed
{standard input}: Assembler messages:
{standard input}:0: Warning: end of file not at end of a line; newline inserted
{standard input}:731: Error: Illegal operands
cc: Internal compiler error: program cc1 got fatal signal 10
*** Error code 1

This started with the first enhancend (VIS) memcpy/memset versions in libc.
If I restore an old libc (from the day before we got an assembly version of
memcpy/bcopy) all is fine again.

>How-To-Repeat:

Boot a -current kernel, make includes, build & install lib/csu, build and
install lib/libc - try to build the rest of the system.

>Fix:

n/a
>Release-Note:
>Audit-Trail:
>Unformatted: