Re: [RFC] Remove last occurences of target_{insert/remove}_watchpoint

[Pedro Alves <pedro at codesourcery dot com>]

On Wednesday 06 May 2009 17:54:25, Joel Brobecker wrote:
> > I've just ran this version through sparc solaris 8, and
> > it survived without regressions (although it looks like solaris test
> > results got much worse somewhere along these last weeks -- around
> > 1300 fails), and watchpoints still work.  Joel, would you like to
> > take a look at this and/or ran it on mips-irix?
> 
> I won't have much time in the next few weeks, unfortunately :-(.
> But I did run it through the testsuite. The results are really abysmal
> but no regression.

Okay, I'll check it in then.


>         # of expected passes            8361
>         # of unexpected failures        2114
>         # of unexpected successes       4
>         # of expected failures          33
>         # of known failures             29
>         # of unresolved testcases       33
>         # of untested testcases         72
>         # of unsupported tests          30

It seems most failures on solaris are related to a single problem, e.g., 

p/c fun()
$1 = 49 '<error reading variable>
(gdb) FAIL: gdb.base/call-sc.exp: p/c fun(); call call-sc-tc

>grep "error reading variable" gdb.log  | wc -l
1312

Looks like things are *really* broken.  That's around
as many failures as instances of that error:

# of expected passes            10933
# of unexpected failures        1337
# of expected failures          41
# of known failures             54
# of unresolved testcases       9
# of untested testcases         13
# of unsupported tests          67

Does this ring a bell?  I'll look at this at some point,
but probably only a couple of weeks or more from now.

-- 
Pedro Alves