netdev - Re: [PATCH] sctp: Clean up type-punning in sctp_cmd

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: <20121026132858.GB25087@hmsreliant.think-freely.org>
Date:	Fri, 26 Oct 2012 09:28:58 -0400
From:	Neil Horman <nhorman@...driver.com>
To:	David Laight <David.Laight@...LAB.COM>
Cc:	netdev@...r.kernel.org, Vlad Yasevich <vyasevich@...il.com>,
	"David S. Miller" <davem@...emloft.net>, linux-sctp@...r.kernel.org
Subject: Re: [PATCH] sctp: Clean up type-punning in sctp_cmd_t union

On Fri, Oct 26, 2012 at 10:00:13AM +0100, David Laight wrote:
> > Subject: [PATCH] sctp: Clean up type-punning in sctp_cmd_t union
> ...
> > +#define SCTP_NULL_BYTE 0xAA
> >  static inline sctp_arg_t SCTP_NULL(void)
> >  {
> > -	sctp_arg_t retval; retval.ptr = NULL; return retval;
> > +	sctp_arg_t retval;
> > +	memset(&retval, SCTP_NULL_BYTE, sizeof(sctp_arg_t));
> > +	return retval;
> >  }
> 
> You really don't want to be taking the address of a local
> variable that would normally be held in a register.
> It stops the compiler doing a lot of optimisations.
> In this case the structure is also being returned by value,
> not nice except that all(?) modern ABI do pass small structures
> in registers.
> 
Thank you dave, I hadn't considered the implications of missed optimizations
when storing to a register here.

> An assignment of some member of the union to NULL would
> seem most appropriate.
> OTOH the code that uses this must be someones 'bright idea (tm)'
> that probably wasn't such a good idea after all.
> 
nobody uses it currently, Its just that some state machine side effects don't
require an argument, but you still have to pass one in, so I was trying to
enhance this to make it obvious in a backtrace should someone inadvertently use
it in the future.  I'm going to revert it though, a big fat zero is probably
just as clear.
Neil

> 	David
> 
> 
> 
> 
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html