gdb native hppa2.0w-hp-hpux11.11

[Randolph Chung <randolph at tausq dot org>]

These are the new FAILs I see with gdb HEAD vs gdb 6.1. In my gdb HEAD
tree I also have this one patch that is approved but not yet checked in:

http://sources.redhat.com/ml/gdb-patches/2004-06/msg00268.html

This is the result of filtering (PASS|FAIL) from gdb.sum from gdb-6.1 vs
gdb-HEAD, doing a diff, and then looking for ^+FAIL

      1 +FAIL: gdb.ada/null_record.exp
     21 +FAIL: gdb.base-hp/callfwmall.exp
      8 +FAIL: gdb.base-hp/reg.exp
      2 +FAIL: gdb.base/auxv.exp
      1 +FAIL: gdb.base/break.exp
      1 +FAIL: gdb.base/call-ar-st.exp
      2 +FAIL: gdb.base/call-sc.exp
      4 +FAIL: gdb.base/callfuncs.exp
      3 +FAIL: gdb.base/funcargs.exp
      2 +FAIL: gdb.base/interrupt.exp
      1 +FAIL: gdb.base/mips_pro.exp
      3 +FAIL: gdb.base/selftest.exp
      2 +FAIL: gdb.base/siginfo.exp
      1 +FAIL: gdb.base/signals.exp
      3 +FAIL: gdb.base/signull.exp
      7 +FAIL: gdb.base/sigstep.exp
     16 +FAIL: gdb.base/watchpoint.exp
      2 +FAIL: gdb.gdb/complaints.exp
      6 +FAIL: gdb.mi/mi-syn-frame.exp
      1 +FAIL: gdb.mi/mi2-file.exp
      6 +FAIL: gdb.mi/mi2-syn-frame.exp
      3 +FAIL: gdb.threads/manythreads.exp
      1 +FAIL: gdb.threads/pthread_cond_wait.exp

Michael's report showed fewer regressions; I suspect that is because it
filters out some tests that were not run with 6.1 (and is thus more
accurate).

I will do more analysis on this. I suspect the FAILs in gdb.base/call*
will be fairly easy to fix. The signal ones should also be doable. I've 
never looked at the mi code so some help there will be helpful :)

I am a bit concerned at hppa64-hpux appears to be quite broken. I'm not
sure about the status with gdb6.1.

randolph
-- 
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/