OpenBSD problems with Alpha patches (was Re: [Sbcl-help] Re: CMUCL Installation)

[William H. N. <wil...@ai...>]

On Wed, May 16, 2001 at 04:02:34PM +0200, Hannah Schroeter wrote:
> Yes, but as "compensation" it seems that it doesn't work on BSD,
> especially OpenBSD, anymore.

(I didn't realize there was another OpenBSD user out there! Welcome!)
 
> I have 0.6.11.41 running, and newer sources from the CVS fail to compile
> with that host, especially in the C runtime system.
> 
> There are problems that the source uses the names "st_mtime" ...
> in an own structure to map the OS specific struct stat to a common
> structure, however, on OpenBSD, struct stat contains:
> 
> 	...
>         struct  timespec st_atimespec;  /* time of last access */
>         struct  timespec st_mtimespec;  /* time of last data modification */
>         struct  timespec st_ctimespec;  /* time of last file status change */
> 	...
[etc.]

I ran into this OpenBSD problem last week. I made a fix for it, and
even checked it into CVS in the "flaky1" branch. It's a slightly
unsatisfactory fix, since I left FILE-LENGTH stubbed out for now, but
all the other stat stuff, including all the stat stuff that SBCL
itself needs, still works. Unfortunately I had some GC problems too,
and in trying to fix them seem to have screwed everything up, not only
for OpenBSD but for Linux too. Even more unfortunately, in trying to
fix those problems I've been bewildered for days. So the "flaky1"
changes haven't been merged into the main branch, because then no one,
Linux or *BSD, would be able to compile from CVS.:-(

Unscrewing this at least for Linux is currently my first SBCL
development priority. (Unscrewing it for OpenBSD would be ordinarily
be just as high a priority, since I want to develop on my OpenBSD
laptop, except that some of the confusion over the last week has been
caused by OpenBSD 2.8 kernel crashes, while Linux has remained
reliable. At this point I certainly don't need any more confusion, so
I'm focusing on Linux.)

After that, OpenBSD remains a reasonably high priority. Unfortunately,
the problems begun last week have been a rather humbling experience --
it's remarkable how confused I've gotten.. -- and I don't know whether
I'll get tripped up by more OpenBSD kernel bugs, so I don't have much
of a guess whether the OpenBSD fixes will be done today or this
weekend or what.

-- 
William Harold Newman <wil...@ai...>
"Tweak alpha so it sends SIGBUS for unaligned access, and does NOT
do a fixup. This encourages people to fix their code." -- a commit
note from <http://www.OpenBSD.org/plus29.html>
PGP key fingerprint 85 CE 1C BA 79 8D 51 8C  B9 25 FB EE E0 C3 E5 7C