This is the mail archive of the
java-discuss@sourceware.cygnus.com
mailing list for the Java project.
RE: Projects
- To: "Boehm, Hans" <hans_boehm at hp dot com>, Jeff Sturm <jsturm at sigma6 dot com>
- Subject: RE: Projects
- From: Jon Olson <olson at mmsi dot com>
- Date: Fri, 5 May 2000 17:06:58 -0700
- Cc: Tom Tromey <tromey at cygnus dot com>, green at cygnus dot com, java-discuss at sourceware dot cygnus dot com
- Organization: Modular Mining Systems
- References: <140D21516EC2D3119EE700902787664456E7E4@hplex1.hpl.hp.com>
On Fri, 05 May 2000, Boehm, Hans wrote:
>This assumes that calls to _Jv_MonitorEnter() and _Jv_MonitorExit() nest
>correctly. That's certainly true for Java source files. It used to not be
>guaranteed for .class files. Is it now?
>
>Hans
>
Hmmm... I've compiled many thousands of lines of Java files and
haven't yet found a .class file that didn't nest properly. In fact, I
put code into the gcj byte code verifier to abort if they ever didn't nest
properly. Originally it tripped a few times when I was trying to
get the logic right to munch all the try/catch blocks that release monitors,
but I haven't had it happen for about a year's worth of Java compilation.
I'd sure be interested to see a valid .class file in which monitors don't
properly nest with exceptions.
--
Jon Olson, Modular Mining Systems
3289 E. Hemisphere Loop
Tucson, AZ 85706
INTERNET: olson@mmsi.com
PHONE: (520)746-9127
FAX: (520)889-5790