[Bug tapsets/10204] New: Place userspace markers in systemtap itself

[Bug tapsets/10204] New: Place userspace markers in systemtap itself

[jistone at redhat dot com]

Perhaps we should eat our own dogfood, and start placing userspace markers in
our own programs.  Some simple candidates off the top of my head:

= stap
  - start/end of each phase
  - launching other processes [stap_system()]
  - various cache activity

= staprun
  - loading/unloading the module

= stapio
  - send/recv control messages
  - recv data messages

Not all of these would necessarily need markers, but they could all go into a
stap.* tapset in whatever form makes sense.

-- 
           Summary: Place userspace markers in systemtap itself
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: tapsets
        AssignedTo: systemtap at sources dot redhat dot com
        ReportedBy: jistone at redhat dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=10204

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

[Bug tapsets/10204] Place userspace markers in systemtap itself

[ksebasti at redhat dot com]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|systemtap at sources dot    |ksebasti at redhat dot com
                   |redhat dot com              |
             Status|NEW                         |ASSIGNED


http://sourceware.org/bugzilla/show_bug.cgi?id=10204

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

[Bug tapsets/10204] Place userspace markers in systemtap itself

[ksebasti at redhat dot com]

------- Additional Comments From ksebasti at redhat dot com  2009-07-31 20:56 -------
Committed as of 67a31a9f57db3f8108e31522b3cf5a6d7d604f22. After talking it over
a bit with fche, due to how the tapsets are processed (at make install time vs
configure time) I commented out the stapio probe in the tapset for now. The
corresponding static marker is still committed however.

I'll leave the bug open but reassign it.

(In reply to comment #8)
> > probe staprun.send_control_message =
> process("/usr/local/bin/staprun").mark("send__ctlmsg") {
> 
> Isn't this probe really in stapio?

The marker is in the send_request function, which is in
runtime/staprun/common.c, so I guess not :)

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|ksebasti at redhat dot com  |systemtap at sources dot
                   |                            |redhat dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=10204

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

[Bug tapsets/10204] Place userspace markers in systemtap itself

[jistone at redhat dot com]

------- Additional Comments From jistone at redhat dot com  2009-07-31 22:18 -------
> (In reply to comment #8)
> > > probe staprun.send_control_message =
> > process("/usr/local/bin/staprun").mark("send__ctlmsg") {
> > 
> > Isn't this probe really in stapio?
> 
> The marker is in the send_request function, which is in
> runtime/staprun/common.c, so I guess not :)

Here "common" means shared by both stapio and staprun, so the mark should
actually manifest in both.  staprun's use of send_request is only for loading
kernel module relocations; stapio's messages are more interesting IMO.

-- 


http://sourceware.org/bugzilla/show_bug.cgi?id=10204

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

[Bug tapsets/10204] Place userspace markers in systemtap itself

[fche at redhat dot com]

------- Additional Comments From fche at redhat dot com  2009-08-05 19:23 -------
committed

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


http://sourceware.org/bugzilla/show_bug.cgi?id=10204

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