http://pike.ida.liu.se/pub/pike/beta/7.6.108/Pike-v7.6.108.tar.gz http://pike.ida.liu.se/pub/pike/beta/7.6.108/Pike-v7.6.108.1-win32.exe
Changes since last beta: o Fixed mktime to work better with out of bounds time fields. [bug 4326] o Fixed support for big endian 64-bit. o Fixed -t (tracing) on win32. o Fixed GL on win32. (aka AIDO starts and can play media streams.)
On darwin-8.9.1-i386, make verify fails on one test (of 150430, 59 skipped):
/Users/xenofarm/beta/Pike-v7.6.108/src/testsuite.in:10265: Test 10430 (shift 1) failed to return proper results.
o->a(): ({ /* 2 elements */ -1, "invalid memory access of location 00000ac0 eip=92b14a67\n# can not save log file, dump to screen..\n\nlocal time = mon apr 16 06:52:00 2007\n\n# this javanativecrash log describes the java state at a native crash in a java application.\n# the corresponding native state can be found in the crash log generated by crashreporter.\n\n# if this error is reproducible, please report it with the following information: \n# 1. provide the steps to reproduce, a test case, and any relevant information\n# 2. this javanativecrash_pid<num>.crash.log (java state)\n# 3. the corresponding <name>.crash.log (native state; generated by crashreporter)\n# file report at: http://bugreport.apple.com/%5Cn%5Cn# an unexpected java error has been detected by hotspot virtual machine:\n#\n# java vm: java hotspot(tm) client vm (1.5.0_07-87 mixed mode, sharing)\n#\n# bus error (0xa) at pc=0x92b14a67, pid=3997, tid=29303808\n\n---------------" [...]
hi,
is that a windows problem or accross all systems? anything that makes pike crash is a serious problem for applications like sTeam where users are allowed to run pike code.
is there any way to catch crashes like these? can a pike process prevent the loading of specific cmods so that such crashes can be avoided?
greetings, martin.
Nope. That's the main testsuite, and if that is the only failure that means that Java verified ok. The error is that you get a SIGSEGV, the JVM just catches that in order to give you a Java backtrace (which is irrelevant in this case).
More of a data point than objection, but that is splitting hairs. :-)
I'd say go, though it wouldn't hurt with more input on how things work on different OSes. http://pike.ida.liu.se/development/pikefarm/7.6.xml (build 662) says there exist green builds for linux ia64, sunos sparc, lots of yellow ones, and some reds. Should someone knowledgeable have time to write up a summaric overview of those (ranging from "known bad client setup" to "encountered error worth postponing release"), that's of course helpful.
pike-devel@lists.lysator.liu.se
-
Johan Sundstr�m (Achtung Liebe!) ＠ Pike (-) developers forum -
Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) ＠ Pike (-) developers forum
-
Martin Bähr -
Peter Bortas ＠ Pike developers forum