Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!uwm.edu!ux1.cso.uiuc.edu!ux1.cso.uiuc.edu!aglew
From: aglew@dwarfs.csg.uiuc.edu (Andy Glew)
Newsgroups: comp.arch
Subject: Re: MP Locking Fine Grain or Course -> Coarse
Message-ID: <AGLEW.90May30143608@dwarfs.csg.uiuc.edu>
Date: 30 May 90 19:36:08 GMT
References: <21911@shamash.cdc.com>
Sender: usenet@ux1.cso.uiuc.edu (News)
Organization: University of Illinois, Computer Systems Group
Lines: 13
In-Reply-To: gdb@hare.udev.cdc.com's message of 30 May 90 16:56:45 GMT

>Kernel profiling should use a microsecond clock and cannot be
>excluded even by spl6() or spl7() calls.

But if the timer sampling tick excluded by some spl level, you can
still get useful information: if you are using flat profiling (instead
of per-procedure gprof style profiling) the blocked timer ticks
accumulate at the splx() or spl0() that unblocks the high spl.  So,
while you don't get too much detail, you can at least tell which (end
of a) critical section took up the time.
    (Of course, if you unblock only on return to the user you lose.)

--
Andy Glew, aglew@uiuc.edu