Technology and Trends
 Linux Activists Mailing List Archives
From: (Orest Zborowski COMP)
Subject: x11v1.1 available
Date: Thu, 2 Jul 92 23:24:38 GMT

x11v1.1 is available on this is the latest x386 release
of x11r5, running under linux0.96b (patch2) and gcc2.2.2, both of which
you'll need to get (both are on banjo as well). you'll need patch2 to get
linus' fixes of the serial port handling - it fixes logitech mice and
others which need cs7 support. gcc2.2.2 is required for the shared libs.
you should have also applied the socket patch i posted which fix some
errors in the unix domain sockets. this patch allows xfig to run (its neat).

i haven't heard but one bad experience with x11v1.1, so i'm opening it up
for general consumption. if there are errors in it, i'll make a new version.

the packaging is the same as for x11v1.0:
        xbase   server, xinit, twm, misc fonts, everything you'll need to
                run x11 in a minimal configuration.
        xextra  extra clients and demos that i brought down from the x11r5
        xprog   include files, static and shared libs for x11. these are
                essential if you're using gcc2.2.2.
        x75dpi  contains the 75dpi fonts, needed if you're running openlook,
                or want more fonts. i don't think that its changed from
                x11v1.0, so you don't have to get them again.

here are some of the additions in x11v1.1:
Notes from the 1.0 to 1.1 release
Tue Jun 30 22:50:23 EDT 1992

o  bugfix applied for dash handling (x386-1.2e).

o  bugfix applied for missing ban switching during blt (x386-1.2e).

o  et4000 support for 16 clocks, to incorporate higher clock values
   no more need for vndor "hiclock" (x386-1.2e).

o  bugfix applied for trashing video state during server reset.

o  led switching fixed.

o  Xcofig modified to include all available options for easier handling.

o  server/clients/demos compiled with gcc2.2.2 and -O6 opimization.

o  imake cpp invocation problem rectified (using cpp -imake).

o  "xfig bug" fixed with corrections to kernel socet code.

the 'vendor "hiclock"' hack is no longer needed. the et4000 driver can now
detect up to 16 clocks, so you should be ok, but may need to fix your
Xconfig to put dummy clock values in the first 8 slots.

please let me know if there are any problems with this!

zorst (orest zborowski)