[Bug runtime/13210] New: vdso runtime tracking is broken

[Bug runtime/13210] New: vdso runtime tracking is broken

[mjw at redhat dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13210

             Bug #: 13210
           Summary: vdso runtime tracking is broken
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: runtime
        AssignedTo: systemtap@sourceware.org
        ReportedBy: mjw@redhat.com
    Classification: Unclassified


vdso tracking is broken, which means user backtraces of kernel probes that get
accessed through the vdso are broken (since they cannot unwind through the
vdso).

This seems to be caused by buildid checking, which always fails.
-DDEBUG_TASK_FINDER_VMA will show:

_stp_vma_init:211: registering vmcb (_stap_target: 32045)
_stp_vma_exec_cb:88: tsk 32045:32045 , register_p: 1, process_p: 1
_stp_vma_match_vdso:32: tsk: 32045 vdso: 0x2c5000
_stp_vma_match_vdso:44: notes_addr vdso32-int80.so: 0x2c51d8
_stp_vma_match_vdso:57: darn, not equal (rc=0) at 0 (40 != 132)
[...]

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

[Bug runtime/13210] vdso runtime tracking is broken

[mjw at redhat dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13210

--- Comment #1 from Mark Wielaard <mjw at redhat dot com> 2011-09-21 10:47:31 UTC ---
Added a testcase for this that uses umodname(uaddr()) to check whether the vma
tracker at least got the name (of either the vdso or another vma that initiated
the syscall).

commit d0d7bd5fd8ebe136c7afca43e157f04c1abed243
Author: Mark Wielaard <mjw@redhat.com>
Date:   Wed Sep 21 12:38:24 2011 +0200

    PR13210 vma/vdso tracking is broken testcase

    Add a generic testcase to see if we can track the vma from which a syscall
    was made. This might or might not come in through the vdso on the
particular
    (compat) arch, but should always at least be able to get the vma name.
    Currently it fails some of the tests in different ways for different arches
    and/or kernel versions.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

[Bug runtime/13210] vdso runtime tracking is broken

[mjw at redhat dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13210

Mark Wielaard <mjw at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #2 from Mark Wielaard <mjw at redhat dot com> 2011-09-27 15:09:54 UTC ---
commit 31d50137fa93d9ff81d76584a870e664c0e3560a
Author: Mark Wielaard <mjw@redhat.com>
Date:   Tue Sep 27 16:21:15 2011 +0200

    _stp_vma_match_vdso: Don't subtract build_id_len from notes_addr.

    Older (pre-0.138) elfutils would erroneously return the address of the
    build id note address with the length added. Newer ones never do this.
    And we only support elfutils 0.142+ now anyway. Fixes vma_vdso.exp.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.