BZ #5: Early SIGQUIT causes fatal assert in vm/jit/codegen.inc:728
Status fields:
creation_ts: | 2005-07-20 12:50 |
component: | unspecified |
version: | unspecified |
rep_platform: | i386 |
op_sys: | Linux |
bug_status: | RESOLVED |
resolution: | FIXED |
reporter: | greenrd@greenrd.org |
The attached test case rapidly fires SIGQUITs at the VM. The timing needed to
trigger this bug may need to be calibrated for different machines - on my
machine, a value of startupDelay of 0.1 reproduces this bug in about 25% of test
runs. (In the other 75% of test runs, the SIGQUIT comes too early, before a
signal handler has been installed.) The bug looks like this:
Full thread dump CACAO 0.92+cvs:
Cannot find Java function at 0x8071ade
cacao: ../../../../src/vm/jit/codegen.inc:728: codegen_findmethod: Assertion `0'
failed.
Exited due to signal 6
Attachment id=3
date: | 2005-07-20 12:52 |
desc: | Rapid SIGQUIT generator |
type: | text/x-python |
download: | automatic.py |
Attachment id=4
date: | 2005-07-20 12:53 |
desc: | HelloWorld in Java |
type: | text/x-java |
download: | HelloWorld.java |