RE: GDB 6.2.91 available

["Dave Korn" <dk at artimi dot com>]

> -----Original Message-----
> From: gdb-owner On Behalf Of Andrew Cagney
> Sent: 08 November 2004 21:44
> To: gdb-testers; gdb
> Subject: GDB 6.2.91 available
> 
> See: ftp://sources.redhat.com/pub/gdb/branch/gdb-6.2.91.tar.bz2
> I expect to roll it into the real thing in about 24 hours.
> 
> Andrew


Results from today: note that there are at least a few potential bogons amongst
them, because after 45 minutes of letting sigstep hog 100% of my cpu while every
single test on it failed slowly one-by-one after timing out I killed the
executable.  However, since this caused the remaining sigstep tests to fail, and
since they were very probably going to fail in the end anyway, it may well not
have caused any actual difference in the end results.  

Also some of the java tests went wrong because my firewall popped up just when I
was in the middly of typing 'end results' and swallowed the 'n' keypress, thus
denying one of the testcases access to the network.  I hate it when that
happens.  (I hate it even more when a requester pops up and swallows a keypress
that ends up _allowing_ something I hadn't wanted to permit, but there ya go).


                === gdb Summary ===

# of expected passes            9163
# of unexpected failures        488
# of unexpected successes       2
# of expected failures          48
# of known failures             22
# of unresolved testcases       52
# of untested testcases         2
# of unsupported tests          15
/davek/patch-gnu/gdb/obj/6.2.91/gdb/testsuite/../../gdb/gdb version  6.3 -nx


Most recent results I could find in the -testers archive:

http://sources.redhat.com/ml/gdb-testers/2004-q2/msg00159.html

=== gdb Summary ===

# of expected passes		9035
# of unexpected failures	369
# of unexpected successes	3
# of expected failures	61
# of unknown successes	10
# of known failures		22
# of unresolved testcases	12
# of unsupported tests	8
/aaronwl/cs/debug/gdb/build/gdb-6.1.1/gdb/testsuite/../../gdb/gdb version 6.1.1
-nx

Judging by the list of .stackdump files, I reckon a lot of the failures were
caused by the test executables failing.

gdb/testsuite/annota3.exe.stackdump     gdb/testsuite/gdb.exe.stackdump
gdb/testsuite/basicclass.exe.stackdump  gdb/testsuite/ld.exe.stackdump
gdb/testsuite/bitfields.exe.stackdump   gdb/testsuite/sigbpt.exe.stackdump
gdb/testsuite/break.exe.stackdump       gdb/testsuite/signull.exe.stackdump
gdb/testsuite/funcargs.exe.stackdump

There was a segv on line 414 of macrotab.c which caused gdb to crash and leave
behind that stack dump.  More details available on request.



    cheers, 
      DaveK
-- 
Can't think of a witty .sigline today....