I've stamped on a couple of kernel bugs today. First one:
adb <an a.out>
main?i
:r
Will panic with a "still refs" complaint in deref_pset() or thereabouts.
Fix is in the treatment of copy-on-write pset's during a fork().
Second, if you adb a file, run it, debug it, and so forth, you will lose a
couple bits of memory after everything exit()'s. The proc p_prefs hash list
was not being cleaned up on server exit().
If either is giving some of you grief send me a note and I'll assemble a
patch. 1.3.3 is imminent, in any case.
Andy
Received on Sat Dec 31 15:38:08 1994
This archive was generated by hypermail 2.1.8 : Thu Sep 22 2005 - 15:12:11 PDT