4971401 was submitted to track the issues observed when running vmark with
the new BCI/jvmti based hprof. Two primary issues were observed - a
performance regression when compared the old JVMPI based hprof and a
core dump. This bug tracks the core observed in that bug.
---- called from signal handler with signal 11 (SIGSEGV) ------
[10] JvmtiEnv::GetLineNumberTable(0x33e78, 0xf49291f0, 0xd977f488, 0xd977f48c,
0xfe0, 0x7f06db), at 0xfefd1314
[11] jvmti_GetLineNumberTable(0x33e78, 0x5c9c48, 0xd977f488, 0xd977f48c,
0xf49291f0, 0x5c9c48), at 0xfef8157c
[12] getLineNumberTable(0xac4, 0xd977f48c, 0xd977f488, 0x800, 0x0,
0xfea37fa4), at 0xfea3e2a4
[13] getLineNumber(0x1cc66e, 0x0, 0x1, 0x2df4c, 0xfea3e410, 0xfea558ac), at
0xfea3e3f0
[14] jvmti_find_or_create(0xfea56bf8, 0x30e89, 0x4, 0x957fc8, 0x1cc66e, 0x0),
at 0xfea39ce8
[15] trace_get_current(0x5c9d00, 0x20, 0x30e89, 0x6, 0x800, 0x957f88), at
0xfea3a4e8
[16] get_current(0x5c9d00, 0xd977f738, 0xfea390dc, 0x0, 0x800, 0x2e8e8), at
0xfea27000
[17] event_newarray(0x5c9d00, 0xd977f738, 0xd977f734, 0x30000135, 0xfea558ac,
0x800), at 0xfea27228
[18] Java_sun_tools_hprof_Tracker_nativeNewArray(0x5c9d00, 0x800, 0xd977f738,
0xd977f734, 0xfea56bf8, 0x0), at 0xfea3ac20
- relates to
-
JDK-4971401 NEWHPROF: vmark/vtest hang when running with tiger b32 -server -Xrunhprof
-
- Closed
-