LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: Ralf Baechle <ralf@linux-mips.org>
Cc: Atsushi Nemoto <anemo@mba.ocn.ne.jp>,
	linux-mips@linux-mips.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Optimize generic get_unaligned / put_unaligned implementations.
Date: Wed, 14 Feb 2007 20:39:03 -0800	[thread overview]
Message-ID: <20070214203903.8d013170.akpm@linux-foundation.org> (raw)
In-Reply-To: <20070214214226.GA17899@linux-mips.org>

On Wed, 14 Feb 2007 21:42:26 +0000 Ralf Baechle <ralf@linux-mips.org> wrote:

> Time for a little bit of dead horse flogging.
> 
> On Mon, Mar 06, 2006 at 05:05:52PM -0800, Andrew Morton wrote:
> 
> > > --- a/include/asm-generic/unaligned.h
> > > +++ b/include/asm-generic/unaligned.h
> > > @@ -78,7 +78,7 @@ static inline void __ustw(__u16 val, __u
> > >  
> > >  #define __get_unaligned(ptr, size) ({		\
> > >  	const void *__gu_p = ptr;		\
> > > -	__typeof__(*(ptr)) val;			\
> > > +	__u64 val;				\
> > >  	switch (size) {				\
> > >  	case 1:					\
> > >  		val = *(const __u8 *)__gu_p;	\
> > > @@ -95,7 +95,7 @@ static inline void __ustw(__u16 val, __u
> > >  	default:				\
> > >  		bad_unaligned_access_length();	\
> > >  	};					\
> > > -	val;					\
> > > +	(__typeof__(*(ptr)))val;		\
> > >  })
> > >  
> > >  #define __put_unaligned(val, ptr, size)		\
> > 
> > I worry about what impact that change might have on code generation. 
> > Hopefully none, if gcc is good enough.
> > 
> > But I cannot think of a better fix.
> 
> It does inflate the code but back then we agreed to go for Atsushi's patch
> because it was fairly obviously correct.  This patch obviously is less
> obvious but generates fairly decent, works for arbitrary data types and
> cuts down the size of unaligned.h from 122 lines to 44 so it must be good.
> 
> ...
>
> +#define get_unaligned(ptr)						\
> +({									\
> +	const struct {							\
> +		union {							\
> +			const int __un_foo[0];				\
> +			const __typeof__(*(ptr)) __un;			\
> +		} __un __attribute__ ((packed));			\
> +	} * const __gu_p = (void *) (ptr);				\
> +									\
> +	__gu_p->__un.__un;						\
>  })

Can someone please tell us how this magic works?  (And it does appear to
work).

It seems to assuming that the compiler will assume that members of packed
structures can have arbitrary alignment, even if that alignment is obvious.

Which makes sense, but I'd like to see chapter-and-verse from the spec or
from the gcc docs so we can rely upon it working on all architectures and
compilers from now until ever more.

IOW: your changlogging sucks ;)

  reply	other threads:[~2007-02-15  4:42 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20050830104056.GA4710@linux-mips.org>
     [not found] ` <20060306.203218.69025300.nemoto@toshiba-tops.co.jp>
2006-03-07  1:05   ` [PATCH] 64bit unaligned access on 32bit kernel Andrew Morton
2006-03-07  2:03     ` Atsushi Nemoto
2006-03-07 18:09     ` Ralf Baechle
2006-03-08  4:58       ` Atsushi Nemoto
2006-03-08  5:12         ` Andrew Morton
2007-02-14 21:42     ` [PATCH] Optimize generic get_unaligned / put_unaligned implementations Ralf Baechle
2007-02-15  4:39       ` Andrew Morton [this message]
2007-02-15  8:35         ` Marcel Holtmann
2007-02-15 14:34         ` Ralf Baechle
2007-02-15 21:53           ` Andrew Morton
2007-02-15 22:18             ` Ralf Baechle
2007-02-15 23:05               ` Jeremy Fitzhardinge
2007-02-15 23:38               ` Andrew Morton
2007-02-16  0:13                 ` Jeremy Fitzhardinge
2007-02-16  0:43                 ` Ralf Baechle
2007-02-16  1:27                   ` Andrew Morton
2007-02-16  1:59                     ` Ralf Baechle
2007-02-20 13:50                     ` Pavel Machek

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20070214203903.8d013170.akpm@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=anemo@mba.ocn.ne.jp \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mips@linux-mips.org \
    --cc=ralf@linux-mips.org \
    --subject='Re: [PATCH] Optimize generic get_unaligned / put_unaligned implementations.' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).