Subject: Bug#215314: FTBFS on some platforms (SableVM*test* versions) - msg#00032

Previous Message by Date: click to view message preview

Re: Improved SableVM site - proof of a concept

Thanks, Done! GBP On Fri, 2003-10-10 at 05:38, Chris Pickett wrote: > Perfect! > > You might want to fix *firendly* and *etinne* on the first line of the > "Documentation" page. > > Cheers, > Chris > > Grzegorz B. Prokopski wrote: > > >Hi! > > > >Thanks for ideas, all should be fixed now. Please check out > >updated site and report if you find any problems or have new > >ideas. > > > > GBP > > > >>From SVN log: > >* Links to other pages now open in the same window (_top) > >* Added news about inline engine and gcc-3.3 > >* Linked to SF's BTS and our "project site" at SF. > > > > > >On Wed, 2003-10-08 at 23:57, David Bélanger wrote: > > > > > >>#3 please > >> > >>If I want a new Window, I will middle click... (I always middle click > >>all the links anyway). > >> > >>Greg, > >> > >>I looked at the source and I think you would need to specify the target. > >>I guess by default links will open in the same frame. > >> > >>With special target _top, it will do option 3. > >><a href="....." target="_top">...</a> > >> > >>David > >> > >>On Wed, Oct 08, 2003 at 10:33:10PM -0400, Chris Pickett wrote: > >> > >> > >>>Hi, > >>> > >>>In my browser (Mozilla), when I click on the 1.0.9 release link at the > >>>top, it opens in a very small frame that you need to scroll to get to > >>>the project release files. It's not even obvious what's happened, it > >>>just looks like the top of the page has been replaced with the top of > >>>the sourceforge page. > >>> > >>>How about one of the following: > >>> > >>>1) making it open in a new window (or popup) > >>>2) making it jump to the right position in the download page > >>>3) switching the whole browser window to the sourceforge page (i.e. > >>>don't link exclusively in the frame). > >>> > >>>Of course, it works fine if I middle-click and open the link in a tab, > >>>but not everybody has or uses tabbed browsing. > >>> > >>>Personally I would choose option (3) since this is the default hyperlink > >>>behaviour. > >>> > >>>Cheers, > >>>Chris > >>> > >>> > >>> > >>> > >>> > >>>------------------------------------------------------- > >>>This SF.net email is sponsored by: SF.net Giveback Program. > >>>SourceForge.net hosts over 70,000 Open Source Projects. > >>>See the people who have HELPED US provide better services: > >>>Click here: http://sourceforge.net/supporters.php > >>>_______________________________________________ > >>>Sablevm-developer mailing list > >>>Sablevm-developer@xxxxxxxxxxxxxxxxxxxxx > >>>https://lists.sourceforge.net/lists/listinfo/sablevm-developer > >>> > >>> > > > > ------------------------------------------------------- > This SF.net email is sponsored by: SF.net Giveback Program. > SourceForge.net hosts over 70,000 Open Source Projects. > See the people who have HELPED US provide better services: > Click here: http://sourceforge.net/supporters.php > _______________________________________________ > Sablevm-developer mailing list > Sablevm-developer@xxxxxxxxxxxxxxxxxxxxx > https://lists.sourceforge.net/lists/listinfo/sablevm-developer -- Grzegorz B. Prokopski <gadek@xxxxxxxxxx> Debian GNU/Linux http://www.debian.org SableVM - LGPLed JVM http://www.sablevm.org ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. SourceForge.net hosts over 70,000 Open Source Projects. See the people who have HELPED US provide better services: Click here: http://sourceforge.net/supporters.php

Next Message by Date: click to view message preview

Regressions in the trunk (1.0.9 probably too)?

Hi! You may find it interesting. During different things I've done during cleanup of my sandbox version (looong story) I was unable to track down why did I still get these EXCEPTIONS running bytecodes test suite. So I compiled current trunk w/ gcc-3.2: SableVM version 1.0.9 - signal based exception detection - copying garbage collection - bidirectional object layout - inline-threaded interpreter Running BTF gave this: + exec /home/greg/sablevm-work/usr/bin/sablevm -Y -c /usr/share/java/junit.jar:.: BytecodeVerificationTest ..................EXCEPTION: iconst_m1 failed ....................... ............EXCEPTION: dcmpg failed when comparing with NaN as value1 EXCEPTION: dcmpg failed when comparing with NaN as value2 ...............EXCEPTION: idiv failed when dividing Integer.MAX_VALUE by -1 ......... Time: 0.519 OK (77 tests) I get the same result for direct and switched w/ signals enabled. I've just tested switched WITHOUT signals as the safest one and SableVM version 1.0.9 - copying garbage collection - bidirectional object layout - switch threaded interpreter + exec /home/greg/sablevm-work/usr/bin/sablevm -Y -c /usr/share/java/junit.jar:.: BytecodeVerificationTest ..................EXCEPTION: iconst_m1 failed ....................... ............EXCEPTION: dcmpg failed when comparing with NaN as value1 EXCEPTION: dcmpg failed when comparing with NaN as value2 Floating point exception [program stops executing at this point] Just to be absolutely sure - I recompiled switched w/o signals with gcc-2.95. But the result remains the same. Interesting - isn't it? ;-) I haven't looked into this any deeper as I want to stick to my "cleanup" list as close as possible. But it was important for my cleaning to know whether this regression [*] happened in *my* version or not. GBP [*] I am assuming that it *used to* work in previous versions but I might be wrong. We really need to check sth. like sablevm 1.0.0 performance and corectness. At some, not too distant, point. -- Grzegorz B. Prokopski <gadek@xxxxxxxxxx> Debian GNU/Linux http://www.debian.org SableVM - LGPLed JVM http://www.sablevm.org ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. SourceForge.net hosts over 70,000 Open Source Projects. See the people who have HELPED US provide better services: Click here: http://sourceforge.net/supporters.php

Previous Message by Thread: click to view message preview

thread-specific data

Hi (not necessarily Etienne), This seems fairly obvious from the source, but I just wanted to check. There is one thread per _svmt_JNIEnv, right? When the thread dies, the _svmt_JNIEnv that contains to it dies also, right? No switching of threads within an _svmt_JNIEnv? I am asking because I want to have some thread-specific data, and I need to know if it's okay to put it in _svmt_JNIEnv, or if I must put it in _svmt_thread . . . Thanks, Chris ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. SourceForge.net hosts over 70,000 Open Source Projects. See the people who have HELPED US provide better services: Click here: http://sourceforge.net/supporters.php

Next Message by Thread: click to view message preview

Regressions in the trunk (1.0.9 probably too)?

Hi! You may find it interesting. During different things I've done during cleanup of my sandbox version (looong story) I was unable to track down why did I still get these EXCEPTIONS running bytecodes test suite. So I compiled current trunk w/ gcc-3.2: SableVM version 1.0.9 - signal based exception detection - copying garbage collection - bidirectional object layout - inline-threaded interpreter Running BTF gave this: + exec /home/greg/sablevm-work/usr/bin/sablevm -Y -c /usr/share/java/junit.jar:.: BytecodeVerificationTest ..................EXCEPTION: iconst_m1 failed ....................... ............EXCEPTION: dcmpg failed when comparing with NaN as value1 EXCEPTION: dcmpg failed when comparing with NaN as value2 ...............EXCEPTION: idiv failed when dividing Integer.MAX_VALUE by -1 ......... Time: 0.519 OK (77 tests) I get the same result for direct and switched w/ signals enabled. I've just tested switched WITHOUT signals as the safest one and SableVM version 1.0.9 - copying garbage collection - bidirectional object layout - switch threaded interpreter + exec /home/greg/sablevm-work/usr/bin/sablevm -Y -c /usr/share/java/junit.jar:.: BytecodeVerificationTest ..................EXCEPTION: iconst_m1 failed ....................... ............EXCEPTION: dcmpg failed when comparing with NaN as value1 EXCEPTION: dcmpg failed when comparing with NaN as value2 Floating point exception [program stops executing at this point] Just to be absolutely sure - I recompiled switched w/o signals with gcc-2.95. But the result remains the same. Interesting - isn't it? ;-) I haven't looked into this any deeper as I want to stick to my "cleanup" list as close as possible. But it was important for my cleaning to know whether this regression [*] happened in *my* version or not. GBP [*] I am assuming that it *used to* work in previous versions but I might be wrong. We really need to check sth. like sablevm 1.0.0 performance and corectness. At some, not too distant, point. -- Grzegorz B. Prokopski <gadek@xxxxxxxxxx> Debian GNU/Linux http://www.debian.org SableVM - LGPLed JVM http://www.sablevm.org ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. SourceForge.net hosts over 70,000 Open Source Projects. See the people who have HELPED US provide better services: Click here: http://sourceforge.net/supporters.php