linux-kernel - RE: [PATCH 1/7] debugobjects: make fixup functions return bool instead of int

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]

Date:	Fri, 22 Apr 2016 08:54:24 +0000
From:	"Du, Changbin" <changbin.du@...el.com>
To:	Thomas Gleixner <tglx@...utronix.de>
CC:	Andrew Morton <akpm@...ux-foundation.org>,
	Jonathan Corbet <corbet@....net>,
	"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
	"josh@...htriplett.org" <josh@...htriplett.org>,
	Steven Rostedt <rostedt@...dmis.org>,
	Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
	"jiangshanlai@...il.com" <jiangshanlai@...il.com>,
	John Stultz <john.stultz@...aro.org>,
	Tejun Heo <tj@...nel.org>,
	"borntraeger@...ibm.com" <borntraeger@...ibm.com>,
	"dchinner@...hat.com" <dchinner@...hat.com>,
	"linux-doc@...r.kernel.org" <linux-doc@...r.kernel.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: RE: [PATCH 1/7] debugobjects: make fixup functions return bool
 instead of int

Hi,
> On Fri, 22 Apr 2016, changbin.du@...el.com wrote:
> > From: "Du, Changbin" <changbin.du@...el.com>
> >
> > The object debugging infrastructure core provides some fixup callbacks
> > for the subsystem who use it. These callbacks are called from the debug
> > code whenever a problem in debug_object_init is detected. And
> > debugobjects core suppose them returns 1 when the fixup was successful,
> > otherwise 0. So the return type is boolean.
> >
> > A bad thing is that debug_object_fixup use the return value for
> > arithmetic operation. It confused me that what is the reall return
> 
> What's bad about that? The fact that it's used for arithmethic operation or
> that it confused you?
> 
It confused me because this is not a common usage. I was confused that what
does he fixup function return? A countable value? But doc says return fixed
or not!
	if (fixup)
		fixed = fixup(addr, state);
	debug_objects_fixups += fixed;
In common,for int return 0 indicates success, negative for fail, positive for something
countable. So I think it is better follow this rule. Here is not of countable, it is Boolean.
So why not this?
	if (fixup && fixup(addr, state))
		debug_objects_fixups++;

> > Reading over the whole code, I found some place do use the return value
> > incorrectly(see next patch). So why use bool type instead?
> 
> Patches which fix a problem need to come first not in the middle of a revamp
> series.
> 
Thanks, I am first know of this.

> > +	bool (*fixup_init)(void *addr, enum debug_obj_state state);
> > +	bool (*fixup_activate)(void *addr, enum debug_obj_state state);
> > +	bool (*fixup_destroy)(void *addr, enum debug_obj_state state);
> > +	bool (*fixup_free)(void *addr, enum debug_obj_state state);
> > +	bool (*fixup_assert_init)(void *addr, enum debug_obj_state state);
> 
> So this change will introduce a gazillion of compile warnings because the
> callbacks in the various usage sites are still having 'int' return type.
> 
No, I modified all the code who use debugojects API.

> Thanks,
> 
> 	tglx

Thanks,
Du, Changbin