I have prepared and submitted or1ksim pkg file to the NetBSD project. http://www.pkgsrc.org/ The package contains patches form the original source from date 2005-03-04T00:00:00+0 of ork1sim cvs. I have attched the pkg Some minor things had to be changed in order to compile or1ksim. I have done this changes: traget_cpu=or32 was replaced by SIM_ARCH=or32, or32 is now default value. changed default config file to or1ksim.cfg I have added system wide config file ( or1ksim.cfg ) which is searched last. It is placed at ${PREFIX}/etc. I have upgrade autoconf / automake / aclocal. DCCFGR and ICCFGR are now functioning as expected. Fixed GPIO initialization bug. --enable-ethphy does not function in the NetBSD port there is a need to rewrite it using bpf , Linux now have similar mechanism called lpf. pkgsrc needs ftp or http access for source files tarballs, currently the pkg sources is fetched using cvs. Does opencores have ftp server for source tarball ? Best Regards. Nir Mor -------------- next part -------------- A non-text attachment was scrubbed... Name: or1ksim.tar.gz Type: application/gzip Size: 81541 bytes Desc: not available Url : http://www.opencores.org/forums/openrisc/attachments/20050308/3e5cfa84/or1ksim.tar-0001.bin

Could you maybe elaborate what submitting or1ksim pkg file to the NetBSD project
means? (I'm just curious) Is it a binary package (like rpm) or is it a like the
LSM?

It would probably be better if you would use the stable branch (stable_0_1_x) of
or1ksim. The so called development branch (HEAD) should have no problems now
but that might change when I get to submitting my dynamic recompiler work
(hopefully during the weekend). The fact that the development branch is cvs
HEAD should be reflected on the `Architechture simulator' web page and the
download link fixed such that when you click to download the `stable' simulator
it downloads the `stable_0_1_x' branch.

The patches that you have included in the pkg file is stuff that I don't believe
should be in a port. You should submit the patches to the mailing list for
review and inclusion into cvs. I have looked at them, but they are not unified
diffs which makes them very hard to reveiw. Could you resubmit them (to the
list this time) in unified diff form?

nog (for whom it's high time go and learn about the caches).