Closed
Bug 120725
Opened 23 years ago
Closed 21 years ago
On Mac9.1, JS crashing on tests/ecma/TypeConversion/9.4-2.js
Categories
(Core :: JavaScript Engine, defect)
Tracking
()
RESOLVED
WONTFIX
People
(Reporter: pschwartau, Assigned: khanson)
Details
(4 keywords)
As recently as a week or so ago, this test passed just fine on my Mac. Suddenly it is crashing! Only on my Mac; passes fine on WinNT, Linux. Will research further to isolate the line in the testcase that provokes the crash. Patrick took a look at this and found that memory was being allocated in blocks that overlapped...
Reporter | ||
Comment 1•23 years ago
|
||
Here is the path to the test: mozilla/js/tests/ecma/TypeConversion/9.4-2.js
Summary: On Mac9.1, JS crashing on tests/ecma/TypeConversion/9.4-2.js → On Mac9.1, JS crashing on tests/ecma/TypeConversion/9.4-2.js
Reporter | ||
Comment 2•23 years ago
|
||
I have reduced the testcase down to this one line, which crashes my Mac: Date(8.64e15); With this proviso: this does not crash for me in the interactive shell. I have to make a test file with this one line in it, save it into the mozilla:js:tests:ecma:TypeConversion directory, and run the test driver on it three times in a row. Only on the third time does my Mac crash. Note: every time one runs the test driver on this test, it will load the utility file mozilla:js:tests:ecma:shell.js before loading the testcase. What my findings indicate is that some sort of random amount of memory allocation is taking place by running the driver three times, which eventually exposes the memory allocation-overlap problem Patrick saw. Note the recent change to jsnum.c to fix bug 120083, "JavaScript toInt32 conversion doesn't match ECMAScript definition" http://bonsai.mozilla.org/cvslog.cgi?file=mozilla/js/src/jsnum.c And the recent change to jsdate.c to fix bug 118266, "JS Date type mixed with document.cookie considered harmful" http://bonsai.mozilla.org/cvslog.cgi?file=mozilla/js/src/jsdate.c
Reporter | ||
Comment 3•23 years ago
|
||
Never mind the Date() function above; this one line is all I need to crash: 8.64e15; Again, I must save this in a test file, say "TEST.js", save it to the mozilla:js:tests:ecma:TypeConversion directory, and run the test driver on it three times in a row. This is the command line I'm using in the MPW shell (note the space and non-standard 'f' after 'MacPerl'; the -I option to Perl simply tells it where your Perl libraries are): perl -I"Yang:Home:Mozilla tools:MacPerl ƒ:lib" jsDriver.pl -e smdebug -f TEST.html -k -l ecma:TypeConversion:TEST.js
Comment 4•22 years ago
|
||
By the definitions on <http://bugzilla.mozilla.org/bug_status.html#severity> and <http://bugzilla.mozilla.org/enter_bug.cgi?format=guided>, crashing and dataloss bugs are of critical or possibly higher severity. Only changing open bugs to minimize unnecessary spam. Keywords to trigger this would be crash, topcrash, topcrash+, zt4newcrash, dataloss.
Severity: normal → critical
Comment 5•21 years ago
|
||
This bug is targeted at a Mac classic platform/OS, which is no longer supported by mozilla.org. Please re-target it to another platform/OS if this bug applies there as well or resolve this bug. I will resolve this bug as WONTFIX in four weeks if no action has been taken. To filter this and similar messages out, please filter for "mac_cla_reorg".
Comment 6•21 years ago
|
||
noMoMacOS9 support. /be
Status: NEW → RESOLVED
Closed: 21 years ago
Resolution: --- → WONTFIX
You need to log in
before you can comment on or make changes to this bug.
Description
•