LKML Archive on lore.kernel.org
 help / color / Atom feed
From: Linus Torvalds <torvalds@osdl.org>
To: Pavel Machek <pavel@ucw.cz>
Cc: Paul Mackerras <paulus@samba.org>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Linux Kernel list <linux-kernel@vger.kernel.org>,
	Andrew Morton <akpm@osdl.org>,
	David Brownell <david-b@pacbell.net>
Subject: Re: Totally broken PCI PM calls
Date: Fri, 15 Oct 2004 08:56:32 -0700 (PDT)
Message-ID: <Pine.LNX.4.58.0410150839010.3897@ppc970.osdl.org> (raw)
In-Reply-To: <20041015135955.GD2015@elf.ucw.cz>



On Fri, 15 Oct 2004, Pavel Machek wrote:
> 
> I'm trying to learn how to work with bitwise on obsolete stuff, but
> checking there is good, too, right?
> 
> Is this right way to do it?
> 
> +typedef enum pm_request __bitwise {
> +       __bitwise PM_SUSPEND, /* enter D1-D3 */
> +       __bitwise PM_RESUME,  /* enter D0 */
> +} pm_request_t;

No, "__bitwise" is a type attribute, so you'd have to do it something like 
this:

	typedef int __bitwise pm_request_t;

	enum pm_request {
		PM_SUSPEND = (__force pm_request_t) 1,
		PM_RESUME = (__force pm_request_t) 2
	};

which makes PM_SUSPEND and PM_RESUME "bitwise" integers (the "__force" is 
there because sparse will complain about casting to/from a bitwise type, 
but in this case we really _do_ want to force the conversion). And because 
the enum values are all the same type, now "enum pm_request" will be that 
type too.

And with gcc, all the __bitwise/__force stuff goes away, and it all ends 
up looking just like integers to gcc.

Quite frankly, you don't need the enum there. The above all really just 
boils down to one special "int __bitwise" type.

So the simpler way is to just do

	typedef int __bitwise pm_request_t;

	#define PM_SUSPEND ((__force pm_request_t) 1)
	#define PM_RESUME ((__force pm_request_t) 2)

and you now have all the infrastructure needed for strict typechecking. 

One small note: the constant integer "0" is special. You can use a 
constant zero as a bitwise integer type without sparse ever complaining. 
This is because "bitwise" (as the name implies) was designed for making 
sure that bitwise types don't get mixed up (little-endian vs big-endian 
vs cpu-endian vs whatever), and there the constant "0" really _is_ 
special. 

Also, because of the "bitwise" nature of bitwise types, you cannot add, 
subtract or do a lot of things with bitwise types. But you _can_ use the 
bitwise operations on them, and you can test them for equality.

So at some point we might add a separate "__opaque" type that allows no
operations at all (except for assignment and equality comparison), and
where "0" isn't special. But in the meantime, __bitwise gets you most of 
the way. Just keep in mind that sparse won't warn about use of the 
constant zero.

> Having __bitwise at every line in enum looks quite ugly to my
> eyes.

And in fact you cannot do it that way. "__bitwise" will always create a
_new_ type, so every time you use it you get a _different_ type. So to use
it sanely, you have to create _one_ typedef for each type you want to use, 
and make that one __bitwise, and that will be the only __bitwise that 
you'll ever see for that particular usage. After that, you use the 
typedef, because it is now a unique type, thanks to the __bitwise.

>	 [Where to get sparse? I tried to google for it but "sparse" is
> very common word (as in sparse matrix). And theres no
> kernel/people/linus on kernel.org...]

With BK, you can just get it from

	bk://sparse.bkbits.net/sparse

and I think DaveJ does tar-balls somewhere. If you search for "sparse 
checker linux" you'll find a number of hits..

Once you have it, just do

	make
	make install

as your regular user, and it will install sparse in your ~/bin directory. 
After that, doing a kernel make with "make C=1" will run sparse on all the 
C files that get recompiled, or with "make C=2" will run sparse on the 
files whether they need to be recompiled or not (ie the latter is fast way 
to check the whole tree if you have already built it).

		Linus

  reply index

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-10-11  0:45 Benjamin Herrenschmidt
2004-10-11  2:41 ` Linus Torvalds
2004-10-11  3:42   ` Paul Mackerras
2004-10-11  4:04     ` Linus Torvalds
2004-10-11  4:24       ` Paul Mackerras
2004-10-11  9:57         ` Pavel Machek
2004-10-11 14:42         ` Linus Torvalds
2004-10-11 14:56           ` suspend-to-RAM [was Re: Totally broken PCI PM calls] Pavel Machek
2004-10-11 15:30             ` Linus Torvalds
2004-10-11 17:39               ` Olivier Galibert
2004-10-11 18:21                 ` Pavel Machek
2004-10-11 15:53             ` Brice Goglin
2004-10-11 16:17               ` Pavel Machek
2004-10-11 17:09                 ` Brice Goglin
2004-10-11 18:23                   ` Pavel Machek
2004-10-11 18:40                     ` Brice Goglin
2004-10-11 16:47         ` Totally broken PCI PM calls David Brownell
2004-10-11 22:28           ` Benjamin Herrenschmidt
2004-10-11 22:58             ` Dmitry Torokhov
2004-10-11 23:08               ` Benjamin Herrenschmidt
2004-10-12  3:00                 ` David Brownell
2004-10-12  4:09                   ` Dmitry Torokhov
2004-10-12 16:56                     ` David Brownell
2004-10-12  9:27             ` Russell King
2004-10-12 11:24               ` Benjamin Herrenschmidt
2004-10-11  4:25     ` Linus Torvalds
2004-10-11 10:18       ` Pavel Machek
2004-10-11 10:54         ` Benjamin Herrenschmidt
2004-10-11 16:01         ` Linus Torvalds
2004-10-15 13:59           ` Pavel Machek
2004-10-15 15:56             ` Linus Torvalds [this message]
2004-10-24 20:58               ` Pavel Machek
2004-10-24 21:18                 ` Linus Torvalds
2004-10-11 16:36     ` David Brownell
2004-10-11 21:17       ` Nigel Cunningham
2004-10-11 21:37         ` David Brownell
2004-10-11 22:12           ` Stefan Seyfried
2004-10-12  2:59             ` David Brownell
2004-10-12  8:54               ` Pavel Machek
2004-10-12 10:32                 ` Stefan Seyfried
2004-10-12 18:28                 ` David Brownell
2004-10-12 20:28                   ` Stefan Seyfried
2004-10-13 13:34                     ` David Brownell
2004-10-12  1:24           ` Nigel Cunningham
2004-10-12  8:53           ` Pavel Machek
2004-10-12 18:52             ` David Brownell
2004-10-12 19:50               ` Pavel Machek
2004-10-12 22:13               ` Benjamin Herrenschmidt
2004-10-12 22:35                 ` David Brownell
2004-10-11 22:26       ` Benjamin Herrenschmidt
2004-10-11  3:45   ` Benjamin Herrenschmidt
2004-10-11  4:08     ` Linus Torvalds
2004-10-11  4:23       ` Benjamin Herrenschmidt
2004-10-11  4:32         ` Linus Torvalds
2004-10-11  4:55           ` Benjamin Herrenschmidt
2004-10-11 16:15             ` David Brownell
2004-10-11 22:22               ` Benjamin Herrenschmidt
2004-10-12  2:46                 ` David Brownell
2004-10-12  4:02                   ` Benjamin Herrenschmidt
2004-10-12 10:49                     ` Nigel Cunningham
2004-10-12 11:27                       ` Benjamin Herrenschmidt
2004-10-12 11:38                         ` Nigel Cunningham
2004-10-12 11:51               ` Pavel Machek
2004-10-11 10:08     ` Pavel Machek
2004-10-11  9:51 ` 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=Pine.LNX.4.58.0410150839010.3897@ppc970.osdl.org \
    --to=torvalds@osdl.org \
    --cc=akpm@osdl.org \
    --cc=benh@kernel.crashing.org \
    --cc=david-b@pacbell.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=paulus@samba.org \
    --cc=pavel@ucw.cz \
    /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

LKML Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/lkml/0 lkml/git/0.git
	git clone --mirror https://lore.kernel.org/lkml/1 lkml/git/1.git
	git clone --mirror https://lore.kernel.org/lkml/2 lkml/git/2.git
	git clone --mirror https://lore.kernel.org/lkml/3 lkml/git/3.git
	git clone --mirror https://lore.kernel.org/lkml/4 lkml/git/4.git
	git clone --mirror https://lore.kernel.org/lkml/5 lkml/git/5.git
	git clone --mirror https://lore.kernel.org/lkml/6 lkml/git/6.git
	git clone --mirror https://lore.kernel.org/lkml/7 lkml/git/7.git
	git clone --mirror https://lore.kernel.org/lkml/8 lkml/git/8.git
	git clone --mirror https://lore.kernel.org/lkml/9 lkml/git/9.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 lkml lkml/ https://lore.kernel.org/lkml \
		linux-kernel@vger.kernel.org
	public-inbox-index lkml

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-kernel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git