This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Minor update on gctest struggles
- From: "Ranjit Mathew" <rmathew4lists at hotmail dot com>
- To: "GCJ" <java at gcc dot gnu dot org>
- Date: Wed, 5 Mar 2003 20:45:47 +0530
- Subject: Minor update on gctest struggles
- Reply-to: "Ranjit Mathew" <rmathew at hotmail dot com>
Hi,
A quick update on the GC test program (gctest) problems before
I leave for the day: in short, it seems to work as far as the
gctest program can tell me for *single-threaded tests* only. :-(
The access violation errors that I always seem to be getting was
due to the fact that the GC_allocate_ml Win32 critical-section was
not being initialised in the *second* main of "tests/test.c". A
simple call to GC_init( ) fixed this. (The first main was not
being selected for MinGW explicitly with a "#if defined(MSWIN32) &&
!defined(__MINGW32__)" for some reason.)
That out of the way, the program kept failing with messages like:
"Found nonzero at 0xe4c800 - allocator is broken" and
"List reversal produced incorrect list - collector is broken"
I set NTEST to 0 (single threaded) and the program seemed
to work, to my great relief!
I added "-mthreads" to the GC makefile build options, reset NTEST
to 2, rebuilt the whole thing and it fails again. :-(
So I must now somehow figure out these failures...
BTW, I notice that for BCC, "-DSMALL_CONFIG" is defined and (more
importantly) only single-threaded tests are done.
Anyway, I'll try again tomorrow.
Ranjit.
--
Ranjit Mathew Email: rmathew AT hotmail DOT com
Bangalore, INDIA. Web: http://ranjitmathew.tripod.com/