From owner-cvs-all Sun Feb 25 0:46:39 2001 Delivered-To: cvs-all@freebsd.org Received: from earth.backplane.com (earth-nat-cw.backplane.com [208.161.114.67]) by hub.freebsd.org (Postfix) with ESMTP id A4CD837B491; Sun, 25 Feb 2001 00:46:33 -0800 (PST) (envelope-from dillon@earth.backplane.com) Received: (from dillon@localhost) by earth.backplane.com (8.11.2/8.9.3) id f1P8jp412754; Sun, 25 Feb 2001 00:45:51 -0800 (PST) (envelope-from dillon) Date: Sun, 25 Feb 2001 00:45:51 -0800 (PST) From: Matt Dillon Message-Id: <200102250845.f1P8jp412754@earth.backplane.com> To: Peter Wemm Cc: Bruce Evans , Jake Burkholder , cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/i386/i386 db_trace.c exception.s machdep.c trap.c src/sys/i386/include asnames.h References: <200102250736.f1P7aVi01775@mobile.wemm.org> Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG :> :> This breaks profiling and pessimizes Xsyscall a little. Please back this :> out. : :Are you sure? The X*syscall stubs dont seem to be profiled.. Yes, that's what the FAKE_MCOUNT() macro does. :Obviously I could easily be misunderstanding things, but the FAKE_MCOUNT :pulls the caller off the stack, right? How does this all tie in? :Would this change imply that the cpu time of the lcall version would be :applied to the int0x80 version? FAKE_MCOUNT tells __mcount where the caller trapframe really started. It doesn't pull anything off the stack. If you have two entry points both intersecting at the same FAKE_MCOUNT without using the special macros to handle the jump then the profiling code will get really confused in regards to the entry point. -Matt : :Cheers, :-Peter :-- :Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message