This is the mail archive of the
java-patches@gcc.gnu.org
mailing list for the Java project.
RE: patch WIN-00: boehm-gc/configure.in support for win32
- From: "Boehm, Hans" <hans_boehm at hp dot com>
- To: "'Adam Megacz'" <patches at lists dot megacz dot com>, java-patches at gcc dot gnu dot org
- Date: Thu, 31 Jan 2002 10:28:34 -0800
- Subject: RE: patch WIN-00: boehm-gc/configure.in support for win32
Thanks.
The first part of this seems completely uncontroversial, and I added it to
my version of the collector.
The second part is clearly also right in this situation. But I'm a little
concerned that someone targeting a really memory constrained environment
will want at least the -DSMALL_CONFIG again.
Perhaps NO_CLOCK should be added only for specific platforms (i.e. we take
it out for now, as in Adam's patch)? Perhaps there should be a libgcj-wide
optimize-for-small-memory-devices flag which turns on -DSMALL_CONFIG?
Hans
> -----Original Message-----
> From: Adam Megacz [mailto:patches@lists.megacz.com]
> Sent: Thursday, January 31, 2002 12:40 AM
> To: java-patches@gcc.gnu.org
> Subject: patch WIN-00: boehm-gc/configure.in support for win32
>
>
>
> 2002-01-31 Adam Megacz <adam@xwt.org>
>
> * gcc/boehm-gc/configure.in: support for win32, saner
> cross-compile options
>
>
> *** configure.in Sat Dec 15 23:12:04 2001
> --- configure.in Thu Jan 31 00:30:46 2002
> *************** case "$THREADS" in
> *** 108,114 ****
> ;;
> esac
> ;;
> ! decosf1 | irix | mach | os2 | solaris | win32 | dce | vxworks)
> AC_MSG_ERROR(thread package $THREADS not yet supported)
> ;;
> *)
> --- 108,117 ----
> ;;
> esac
> ;;
> ! win32)
> ! AC_DEFINE(GC_WIN32_THREADS)
> ! ;;
> ! decosf1 | irix | mach | os2 | solaris | dce | vxworks)
> AC_MSG_ERROR(thread package $THREADS not yet supported)
> ;;
> *)
> *************** AC_DEFINE(GC_GCJ_SUPPORT)
> *** 226,237 ****
> AC_DEFINE(ATOMIC_UNCOLLECTABLE)
>
> dnl This is something of a hack. When cross-compiling we turn off
> ! dnl some functionality. We also enable the "small" configuration.
> ! dnl These is only correct when targetting an embedded
> system. FIXME.
> if test -n "${with_cross_host}"; then
> AC_DEFINE(NO_SIGSET)
> - AC_DEFINE(NO_CLOCK)
> - AC_DEFINE(SMALL_CONFIG)
> AC_DEFINE(NO_DEBUGGING)
> fi
>
> --- 229,238 ----
> AC_DEFINE(ATOMIC_UNCOLLECTABLE)
>
> dnl This is something of a hack. When cross-compiling we turn off
> ! dnl some functionality. These is only correct when targetting an
> ! dnl embedded system. FIXME.
> if test -n "${with_cross_host}"; then
> AC_DEFINE(NO_SIGSET)
> AC_DEFINE(NO_DEBUGGING)
> fi
> EOF
>