cvs commit: src/sys/amd64/amd64 machdep.c trap.c vm_machdep.c src/sys/amd64/conf DEFAULTS src/sys/amd64/linux32 linux32_sysvec.c src/sys/arm/arm trap.c src/sys/arm/conf AVILA src/sys/arm/xscale/i8134x crb_machdep.c src/sys/compat/freebsd32 ...

[Peter Wemm]

On Wed, Mar 12, 2008 at 3:26 AM, Robert Watson <rwatson at freebsd.org> wrote:
>
>  On Wed, 12 Mar 2008, Jeff Roberson wrote:
>
>  >  Remove kernel support for M:N threading.
>  >
>  >  While the KSE project was quite successful in bringing threading to
>  >  FreeBSD, the M:N approach taken by the kse library was never developed
>  >  to its full potential.  Backwards compatibility will be provided via
>  >  libmap.conf for dynamically linked binaries and static binaries will
>  >  be broken.
>
>  This will likely require us to grow a significantly more mature approach to
>  compat libraries, especially for 5.x where I seem to recall libthr was of
>  mixed productionness.  Once these changes have settled, the right approach is
>  probably to bring up a 5.x chroot on an 8.x kernel and fault in problems.

Neither libkse nor libthr were fully functional for the most part of
5.x's lifetime.  The canonical bugfix (for amd64 and to a lesser
extent, i386) was to use libmap to redirect everything to libc_r.  It
was required if you wanted to use things like mozilla / firefox / etc
without losing your sanity.

Maybe that changed in later 5.x, but IMHO the functionality /
reliability bar wasn't that high.

>  While 5.x isn't widely hacked on currently, it is used pretty extensively in
>  our user base (a sample of Java binary downloads a few months ago, for
>  example, showed 5.x/i386 to be the most widely used platform for Java), and I
>  know I've had a lot of ... feedback ... about the fact that our upgrade path
>  from 5.x to 6.x has gotten less functional over time.

Java is the one big exception. We have to care about that one.

>  Anyhow, once things settle a bit, let's see where we are.

Yes, there are lots of options.  We've needed to have a better plan
for older library interfaces for a while.  I recall a few instances
where we had security issues that were preserved via the compat
packages for too long, etc.

Anyway.. Lets find and fix the problems as they turn up.

-- 
Peter Wemm - peter at wemm.org; peter at FreeBSD.org; peter at yahoo-inc.com
"All of this is for nothing if we don't go to the stars" - JMS/B5
"If Java had true garbage collection, most programs would delete
themselves upon execution." -- Robert Sewell