The libMicro on the following Solaris versions (uname -a): SunOS shemesh 5.11 snv_15 i86pc i386 i86pc SunOS maadim 5.10 Generic i86pc i386 i86pc When running "./bench > out.txt", the tests longjmp and siglongjmp eat CPU but does not finish after 15 minutes of CPU eaten.
Two consecutive pstack's: 12934: bin/longjmp -E -C 200 -L -S -W -N longjmp -I 10 d0e94b2c longjmp (d0e60000, 80473e0, d0fd2000, d0fd2000, d0fd0080, d0f39000) + 1c 08052576 worker_thread (d0e60000, 0, d0d9a000, 80479c4, 0, 0) + c0 08052688 worker_process (8a96c2df, 100594, 0, 137f0c7f, 706d742f, 62696c2f) + 7a 080522bc actual_main (b, 80479c4, d0fde0a4, d0ef7315, d0e92581, 8053de2) + 618 08053c47 main (b, 80479c4, 80479f4) + 17 08051ae0 _start (b, 8047ae8, 8047af4, 8047af7, 8047afa, 8047afe) + 80 12934: bin/longjmp -E -C 200 -L -S -W -N longjmp -I 10 08051c7e benchmark (d0e60000, 80473e0, d0fd2000, d0fd2000, d0fd0080, d0f39000) + 1d 08052576 worker_thread (d0e60000, 0, d0d9a000, 80479c4, 0, 0) + c0 08052688 worker_process (8a96c2df, 100594, 0, 137f0c7f, 706d742f, 62696c2f) + 7a 080522bc actual_main (b, 80479c4, d0fde0a4, d0ef7315, d0e92581, 8053de2) + 618 08053c47 main (b, 80479c4, 80479f4) + 17 08051ae0 _start (b, 8047ae8, 8047af4, 8047af7, 8047afa, 8047afe) + 80 The header of one of the output files: !Libmicro_#: 0.3.0 !Options: -E -C 200 -L -S -W !Machine_name: shemesh !OS_name: SunOS !OS_release: 5.11 !OS_build: snv_15 !Processor: i86pc !#CPUs: 1 !CPU_MHz: 2600MHz !CPU_NAME: Intel(r) Pentium(r) 4 CPU 2.60GHz !IP_address: 10.1.1.135 !Run_by: alexey !Date: 08/18/05 14:15 !Compiler: gcc !Compiler Ver.: 3.4.3 !sizeof(long): 4 !extra_CFLAGS: -m32 !TimerRes: 1 nsecs The header of second of the output files: !Libmicro_#: 0.3.0 !Options: -E -C 200 -L -S -W !Machine_name: maadim !OS_name: SunOS !OS_release: 5.10 !OS_build: Generic !Processor: i86pc !#CPUs: 2 !CPU_MHz: 1800MHz !CPU_NAME: AMD Opteron(tm) Processor 244 !IP_address: 10.1.1.136 !Run_by: alexey !Date: 08/18/05 14:17 !Compiler: gcc !Compiler Ver.: 3.4.3 !sizeof(long): 4 !extra_CFLAGS: -m32 !TimerRes: 1 nsecs Seem to be strange. This message posted from opensolaris.org _______________________________________________ perf-discuss mailing list perf-discuss@opensolaris.org
