public inbox for systemtap@sourceware.org
 help / color / mirror / Atom feed
* [Bug runtime/10599] New: Always print module/map names for kernel/executable addresses
@ 2009-09-03 19:07 mjw at redhat dot com
  2009-09-15 16:50 ` [Bug runtime/10599] " prasad at linux dot vnet dot ibm dot com
                   ` (3 more replies)
  0 siblings, 4 replies; 5+ messages in thread
From: mjw at redhat dot com @ 2009-09-03 19:07 UTC (permalink / raw)
  To: systemtap

When the context tapset modname, [u]symname, [u]symdata or the backtrace
addresses mapped to symbol names (all this translates to _stp_symbol_print,
_stp_usymbol_print or _stp_symbol_snprint in runtime/syms.c) we only print the
module/library name associated with the address if we are currently tracking it.
Which means, we only print the module/library map name if we already have it in
our _stp_modules list.

Better vma tracking (bug #10228) means we get events for each loaded kernel
module and each mmap event (plus associated backing [shared library] file). We
should build a table of this so we can easily add support for at least printing
<module/map-name>+offset from the sym.c symbol_print functions.

Also slightly related, to print <vdso>+offset, bug #10080.

-- 
           Summary: Always print module/map names for kernel/executable
                    addresses
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: runtime
        AssignedTo: systemtap at sources dot redhat dot com
        ReportedBy: mjw at redhat dot com
 BugsThisDependsOn: 10080,10228


http://sourceware.org/bugzilla/show_bug.cgi?id=10599

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug runtime/10599] Always print module/map names for kernel/executable addresses
  2009-09-03 19:07 [Bug runtime/10599] New: Always print module/map names for kernel/executable addresses mjw at redhat dot com
@ 2009-09-15 16:50 ` prasad at linux dot vnet dot ibm dot com
  2010-05-06 18:03 ` mjw at redhat dot com
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 5+ messages in thread
From: prasad at linux dot vnet dot ibm dot com @ 2009-09-15 16:50 UTC (permalink / raw)
  To: systemtap



-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |prasad at linux dot vnet dot
                   |                            |ibm dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=10599

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug runtime/10599] Always print module/map names for kernel/executable addresses
  2009-09-03 19:07 [Bug runtime/10599] New: Always print module/map names for kernel/executable addresses mjw at redhat dot com
  2009-09-15 16:50 ` [Bug runtime/10599] " prasad at linux dot vnet dot ibm dot com
@ 2010-05-06 18:03 ` mjw at redhat dot com
  2010-06-29 20:32 ` mjw at redhat dot com
  2010-07-09 17:39 ` mjw at redhat dot com
  3 siblings, 0 replies; 5+ messages in thread
From: mjw at redhat dot com @ 2010-05-06 18:03 UTC (permalink / raw)
  To: systemtap



-- 
Bug 10599 depends on bug 10228, which changed state.

Bug 10228 Summary: Add more vma-tracking for user space symbol/backtraces
http://sourceware.org/bugzilla/show_bug.cgi?id=10228

           What    |Old Value                   |New Value
----------------------------------------------------------------------------
             Status|ASSIGNED                    |NEW
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

http://sourceware.org/bugzilla/show_bug.cgi?id=10599

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug runtime/10599] Always print module/map names for kernel/executable addresses
  2009-09-03 19:07 [Bug runtime/10599] New: Always print module/map names for kernel/executable addresses mjw at redhat dot com
  2009-09-15 16:50 ` [Bug runtime/10599] " prasad at linux dot vnet dot ibm dot com
  2010-05-06 18:03 ` mjw at redhat dot com
@ 2010-06-29 20:32 ` mjw at redhat dot com
  2010-07-09 17:39 ` mjw at redhat dot com
  3 siblings, 0 replies; 5+ messages in thread
From: mjw at redhat dot com @ 2010-06-29 20:32 UTC (permalink / raw)
  To: systemtap


------- Additional Comments From mjw at redhat dot com  2010-06-29 14:39 -------
commit 82ea0c96760eba1a1f521c007b9353e7e568906b
Author: Mark Wielaard <mjw@redhat.com>
Date:   Tue Jun 29 16:16:02 2010 +0200

    PR10599 Always print modname when available (and requested).
    
    * runtime/sym.c (_stp_umod_lookup): Also return vm_end if available.
      (_stp_kallsyms_lookup): Don't try to handle 32bit-on-64bit signedness,
      _stp_umod_lookup already does. If available return modname, size and
      offset.
      (_stp_print_addr): Always print modname when available (and requested).
      (_stp_symbol_snprint): Likewise.
    * runtime/unwind.c (unwind): Adjust _stp_umod_lookup call.
    * tapset/ucontext.stp (umodname): Likewise.


$ stap -e 'probe timer.profile { if (pid() != 0)
 { log(execname() . ": " . usymdata(uaddr())); } }'
sh: 0x00000033dc7215e8 libc-2.12.so+0x1215e8/0x37f000
sh: 0x00000033dc6a6e47 libc-2.12.so+0xa6e47/0x175000
gcc: 0x0000000000414dbd gcc+0x14dbd/0x3f000
gcc: 0x00000033dc6d3ad5 libc-2.12.so+0xd3ad5/0x37f000
cc1: 0x00000033dc6d3fb0 libc-2.12.so+0xd3fb0/0x37f000
[...]

Backtraces that cannot continue because module wasn't loaded with -d will now
end with address+modname whenever possible.
process("/bin/ls").function("ferror_unlocked@/usr/include/bits/stdio.h:133")
 0x0000000000411fcb : close_stream+0x1b/0x90 [ls]
 0x000000000040ba4d : close_stdout+0x3d/0xc0 [ls]
 0x00000033dc635ff2 [libc-2.12.so+0x35ff2/0x37f000]

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


http://sourceware.org/bugzilla/show_bug.cgi?id=10599

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug runtime/10599] Always print module/map names for kernel/executable addresses
  2009-09-03 19:07 [Bug runtime/10599] New: Always print module/map names for kernel/executable addresses mjw at redhat dot com
                   ` (2 preceding siblings ...)
  2010-06-29 20:32 ` mjw at redhat dot com
@ 2010-07-09 17:39 ` mjw at redhat dot com
  3 siblings, 0 replies; 5+ messages in thread
From: mjw at redhat dot com @ 2010-07-09 17:39 UTC (permalink / raw)
  To: systemtap



-- 
Bug 10599 depends on bug 10080, which changed state.

Bug 10080 Summary: track vdso for process symbols/backtrace
http://sourceware.org/bugzilla/show_bug.cgi?id=10080

           What    |Old Value                   |New Value
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

http://sourceware.org/bugzilla/show_bug.cgi?id=10599

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2010-07-09 17:39 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-09-03 19:07 [Bug runtime/10599] New: Always print module/map names for kernel/executable addresses mjw at redhat dot com
2009-09-15 16:50 ` [Bug runtime/10599] " prasad at linux dot vnet dot ibm dot com
2010-05-06 18:03 ` mjw at redhat dot com
2010-06-29 20:32 ` mjw at redhat dot com
2010-07-09 17:39 ` mjw at redhat dot com

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).