linux-kernel - Re: [PATCH 2/4] x86/pvops: target CREATE_TRACE

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <alpine.DEB.2.00.0904171330190.14919@gandalf.stny.rr.com>
Date:	Fri, 17 Apr 2009 13:33:54 -0400 (EDT)
From:	Steven Rostedt <rostedt@...dmis.org>
To:	Jeremy Fitzhardinge <jeremy@...p.org>
cc:	mathieu.desnoyers@...ymtl.ca, Ingo Molnar <mingo@...e.hu>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Jeremy Fitzhardinge <jeremy.fitzhardinge@...rix.com>
Subject: Re: [PATCH 2/4] x86/pvops: target CREATE_TRACE_POINTS to particular
 subsystems



On Fri, 17 Apr 2009, Jeremy Fitzhardinge wrote:

> Steven Rostedt wrote:
> > On Fri, 17 Apr 2009, Jeremy Fitzhardinge wrote:
> > 
> >   
> > > Steven Rostedt wrote:
> > >     
> > > > Ah yes! It needs to be:
> > > > 
> > > > #ifdef CONFIG_IRQ_TRACE_POINTS
> > > > #undef CONFIG_IRQ_TRACE_POINTS
> > > > #include <trace/define_trace.h>
> > > > #endif
> > > > 
> > > > Otherwise we get into the recursion again.
> > > >         
> > > We should probably also move the #define TRACE_SYS in there as well
> > > (without
> > > the #undef), as it should only have one definition at a time...
> > >     
> > 
> > Actually, I'm kind of against that. Just because as it stands, the
> > TRACE_SYSTEM macro is up at the top, and it is easy to see.
> >   
> 
> Yes, but it means that if you're in the middle of CREATE_FOO_TRACE_POINTS and
> foo.h happens to include bar.h, suddenly TRACE_SUBSYSTEM becomes bar...

How so?

> 
> > Actually, we could do (from the top of the file)
> > 
> > #ifdef CONFIG_IRQ_TRACE_POINTS
> > #undef CONFIG_IRQ_TRACE_POINTS

The first thing the file does is undef CONFIG_FOO_TRACE_POINTS, anything 
else that gets incuded, will not take this path.

> > 
> > #define TRACE_SYSTEM irq
> > 
> > #include <trace/define_trace.h>
> > 
> > #elif !defined(_TRACE_IRQ_H) || defined(TRACE_HEADER_MULTI_READ)
> > #define _TRACE_IRQ_H
> > 
> > #include <linux/tracepoint.h>
> > #include <linux/interrupt.h>
> > 
> > 
> > [...]
> > 
> > #endif
> >   
> 
> That's slightly different from what we have now.  At the moment its
> 
>    #if !defined(_TRACE_IRQ_H)...
>    ...
>    #endif
> 
>    #ifdef CREATE_IRQ_TRACE_POINTS
>    ...
>    #endif
> 
> So we get both the main part of the file and the CREATE_X_TRACE_POINTS parts.
> Your suggestion makes them exclusive:
> 
>    #ifdef CREATE_IRQ_TRACE_POINTS
>    ...
>    #elif !defined(_TRACE_IRQ_H)...
>    ...
>    #endif
>      
> Does that make a difference?

Hmm, I was about to disagree, but I think we need it separate. Because, we 
still need it to do the tracepoint conversion, and we also need to call 
tracepoint.h and TRACE_EVENT before calling define_trace.h.

OK, scrap that idea ;-)

-- Steve

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/