Results of 20070419 testing

Mike Mason mmlnx@us.ibm.com
Fri Apr 20 17:35:00 GMT 2007


I'm working on a fix for the socket test failure.  Should be checked in today.  

The problem is that do_sock_write() is called in one place instead of two as of 2.6.19 and, thus, gets optimized away. Same for do_sock_read().  Is there any way to mark a symbol so that the compiler doesn't optimize it away?

Mike

William Cohen wrote:
> Looking over the systemtap results.
> 
> fc7 i686
> Kernel: Linux 2.6.20-1.3071.fc7PAE #1 SMP Sun Apr 15 22:52:54 EDT 2007 
> i686 i686 i386 GNU/Linux
> 
> Testsuite summary of failed tests
> FAIL: buildok/lket.stp
>     can't find local $task for tapset/rpc.stp sunrpc.sched.release_task
>     no $return for tapset/rpc.stp  sunrpc.sched.new_task
> FAIL: buildok/process_test.stp
>     couldn't find  kernel.function("handle_signal")
> FAIL: buildok/socket.stp
>     couldn't find   kernel.function("do_sock_write").return
> FAIL: systemtap.samples/lket(stap timeout)
>     took too long to build (timed out before pass 2
> FAIL: systemtap.stress/current.stp startup (timeout)
>     took too long to build (timed out before pass 2
>         === systemtap Summary ===
> 
> # of expected passes        318
> # of unexpected failures    5
> # of expected failures        129
> # of known failures        4
> # of untested testcases        7
> runtest completed at Thu Apr 19 07:04:46 2007
> 
> 
> fc7 x86_64
> Kernel: Linux 2.6.20-1.3071.fc7 #1 SMP Sun Apr 15 22:52:43 EDT 2007 
> x86_64 x86_64 x86_64 GNU/Linux
> 
> Testsuite summary of failed tests
> FAIL: buildok/lket.stp
>     can't find local $task for tapset/rpc.stp sunrpc.sched.release_task
>     no $return for tapset/rpc.stp  sunrpc.sched.new_task
> FAIL: buildok/process_test.stp
>     couldn't find  kernel.function("handle_signal")
> FAIL: buildok/socket.stp
>     couldn't find   kernel.function("do_sock_write").return
> FAIL: systemtap.samples/lket(semantic error)
>     can't find local $task for tapset/rpc.stp sunrpc.sched.release_task
> FAIL: transport fill staging buffer - relayfs (0)
> FAIL: systemtap.stress/current.stp startup (timeout)
>     All the following 32-bit syscall test failed because
>      missing missing glibc-devel*.i386 on machine:
> /usr/include/gnu/stubs.h:7:27: error: gnu/stubs-32.h: No such file or 
> directory
> 
> FAIL: 32-bit access
> FAIL: 32-bit acct
> FAIL: 32-bit alarm
> FAIL: 32-bit chmod
> FAIL: 32-bit clock
> FAIL: 32-bit dir
> FAIL: 32-bit forkwait
> FAIL: 32-bit itimer
> FAIL: 32-bit link
> FAIL: 32-bit mmap
> FAIL: 32-bit mount
> FAIL: 32-bit net1
> FAIL: 32-bit openclose
> FAIL: 32-bit readwrite
> FAIL: 32-bit rt_signal
> FAIL: 32-bit sendfile
> FAIL: 32-bit stat
> FAIL: 32-bit statfs
> FAIL: 32-bit swap
> FAIL: 32-bit sync
> FAIL: 32-bit timer
> FAIL: 32-bit trunc
> FAIL: 32-bit uid
> FAIL: 32-bit umask
> FAIL: 32-bit unlink
>         === systemtap Summary ===
> 
> # of expected passes        315
> # of unexpected failures    31
> # of expected failures        129
> # of known failures        4
> # of untested testcases        7
> # of unsupported tests        5
> runtest completed at Thu Apr 19 05:30:16 2007
> 
> 
> rhel4u4 i686
> crashed, last teset listed in log was:
> Running 
> /home/wcohen/stap_testing_200704190830/src/testsuite/systemtap.samples/lket.exp 
> ...
> 
> 
> 
> rhel4u4 x86_64
> Kernel: Linux 2.6.9-42.0.10.ELsmp #1 SMP Fri Feb 16 17:13:42 EST 2007 
> x86_64 x86_64 x86_64 GNU/Linux
> 
> Testsuite summary of failed tests
> FAIL: systemtap.base/timers.stp shutdown (eof)
>     in log looks like test ran ok. problem with the timers.exp check?
> FAIL: buildok/process_test.stp
>     couldn't find $sig local var for kernel.function("handle_signal")
>         in tapset/signal.stp
> FAIL: buildok/socket.stp
>     can't find probe point kernel.function("do_sock_write").return
>     can't find probe point socket.do_write.return
>     can't find probe point kernel.function("do_sock_read").return
>     can't find probe point socket.do_read.return
>     can't find probe point kernel.function("do_sock_write")
>     can't find probe point socket.do_write
>     can't find probe point kernel.function("do_sock_read")
>     can't find probe point socket.do_read
> FAIL: systemtap.samples/tcptest.stp shutdown (eof)
>     in log looks like test ran ok. problem with the tcptest.exp check?
>         === systemtap Summary ===
> 
> # of expected passes        344
> # of unexpected failures    4
> # of expected failures        129
> # of known failures        4
> # of untested testcases        7
> # of unsupported tests        5
> runtest completed at Thu Apr 19 05:23:12 2007
> 
> rhel4u4 ia64
> Date: 200704190830
> User: wcohen
> Kernel: Linux 2.6.9-42.0.10.EL #1 SMP Fri Feb 16 17:01:51 EST 2007 ia64 
> ia64 ia64 GNU/Linux
> 
> Testsuite summary of failed tests
> FAIL: systemtap.base/kmodule.stp startup (eof)
>     stap: 
> /home/wcohen/rh-rpms/BUILD/elfutils-0.125/libdwfl/offline.c:77: 
> dwfl_offline_section_address: Assertion `main_shdr->sh_addr != 0' failed.
> FAIL: probefunc:kernel.function("context_switch").inline compilation
>     can't find probe point kernel.function("context_switch").inline
> FAIL: buildok/process_test.stp
>     couldn't find $sig local var for kernel.function("handle_signal")
>         in tapset/signal.stp
> FAIL: buildok/scsi.stp
>     stap: 
> /home/wcohen/rh-rpms/BUILD/elfutils-0.125/libdwfl/offline.c:77: 
> dwfl_offline_section_address: Assertion `main_shdr->sh_addr != 0' failed.
>     stap: 
> /home/wcohen/rh-rpms/BUILD/elfutils-0.125/libdwfl/offline.c:77: 
> dwfl_offline_section_address: Assertion `main_shdr->sh_addr != 0' failed.
> FAIL: buildok/socket.stp
>     can't find probe point kernel.function("do_sock_write").return
>     can't find probe point socket.do_write.return
>     can't find probe point kernel.function("do_sock_read").return
>     can't find probe point socket.do_read.return
>     can't find probe point kernel.function("do_sock_write")
>     can't find probe point socket.do_write
>     can't find probe point kernel.function("do_sock_read")
>     can't find probe point socket.do_read
> FAIL: buildok/syscall.stp
>     a number of sys_* functions not found:
>     sys_alarm, sys_fadvise64, sys_gethostname, sys_getpgrp, sys_llseek,
>     sys_nice, sys_old_getrlimit, sys_pause, sys_rt_sigreturn,
>     sys_rt_sigsuspend, sys_signal, sys_sigprocmask, sys_utime,...
> FAIL: systemtap.samples/lket(pass1)
>     WARNING: cannot find module nfs debuginfo: No such file or directory
>     stap: 
> /home/wcohen/rh-rpms/BUILD/elfutils-0.125/libdwfl/offline.c:77: 
> dwfl_offline_section_address: Assertion `main_shdr->sh_addr != 0' failed.
> FAIL: systemtap.samples/tcptest.stp shutdown (eof)
>     in log looks like test ran ok. problem with the tcptest.exp check?
> FAIL: conversions.stp 0xffffffffffffffff (6)
> FAIL: systemtap.stress/current.stp startup (eof)
>     stap: 
> /home/wcohen/rh-rpms/BUILD/elfutils-0.125/libdwfl/offline.c:77: 
> dwfl_offline_section_address: Assertion `main_shdr->sh_addr != 0' failed.
> The syscall tests failed because
>     unable to find $fildes for tapset/syscalls2.stp:252:21
> FAIL: access
> FAIL: acct
> FAIL: alarm
> FAIL: chmod
> FAIL: clock
> FAIL: dir   
> FAIL: forkwait
> FAIL: itimer
> FAIL: link
> FAIL: mmap
> FAIL: mount
> FAIL: net1
> FAIL: openclose
> FAIL: readwrite
> FAIL: rt_signal
> FAIL: sendfile
> FAIL: stat
> FAIL: statfs
> FAIL: swap
> FAIL: sync
> FAIL: timer
> FAIL: trunc
> FAIL: uid
> FAIL: umask
> FAIL: unlink
>         === systemtap Summary ===
> 
> # of expected passes        282
> # of unexpected failures    35
> # of unexpected successes    2
> # of expected failures        127
> # of known failures        4
> # of untested testcases        7
> # of unsupported tests        3
> runtest completed at Thu Apr 19 05:53:07 2007



More information about the Systemtap mailing list