All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
To: linux-kernel@vger.kernel.org
Cc: andrea@suse.de, akpm@osdl.org
Subject: [PATCH] x86 bitops.h commentary on instruction reordering
Date: Thu, 5 Aug 2004 17:06:22 -0300	[thread overview]
Message-ID: <20040805200622.GA17324@logos.cnet> (raw)

Hi, 

Back when we were discussing the need for a memory barrier in sync_page(), 
it came to me (thanks Andrea!) that the bit operations can be perfectly 
reordered on architectures other than x86. 

I think the commentary on i386 bitops.h is misleading, its worth 
to note that that these operations are not guaranteed not to be 
reordered on different architectures. 

clear_bit() already does that:

 * clear_bit() is atomic and may not be reordered.  However, it does
 * not contain a memory barrier, so if it is used for locking purposes,
 * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
 * in order to ensure changes are visible on other processors.

What you think of the following


--- a/include/asm-i386/bitops.h.orig	2004-08-05 17:35:40.639336160 -0300
+++ b/include/asm-i386/bitops.h	2004-08-05 17:35:12.486616024 -0300
@@ -30,7 +30,12 @@
  * @addr: the address to start counting from
  *
  * This function is atomic and may not be reordered.  See __set_bit()
- * if you do not require the atomic guarantees.
+ * if you do not require the atomic guarantees. 
+ * 
+ * Note: there are no guarantees that this function will not be reordered 
+ * on non x86 architectures, so if you are writting portable code, 
+ * make sure not to rely on its reordering guarantees.
+ * 
  * Note that @nr may be almost arbitrarily large; this function is not
  * restricted to acting on a single-word quantity.
  */
@@ -109,7 +114,8 @@ static inline void __change_bit(int nr, 
  * @nr: Bit to change
  * @addr: Address to start counting from
  *
- * change_bit() is atomic and may not be reordered.
+ * change_bit() is atomic and may not be reordered. It may be
+ * reordered on other architectures than x86.
  * Note that @nr may be almost arbitrarily large; this function is not
  * restricted to acting on a single-word quantity.
  */
@@ -127,6 +133,7 @@ static inline void change_bit(int nr, vo
  * @addr: Address to count from
  *
  * This operation is atomic and cannot be reordered.  
+ * It may be reordered on other architectures than x86.
  * It also implies a memory barrier.
  */
 static inline int test_and_set_bit(int nr, volatile unsigned long * addr)
@@ -165,7 +172,8 @@ static inline int __test_and_set_bit(int
  * @nr: Bit to clear
  * @addr: Address to count from
  *
- * This operation is atomic and cannot be reordered.  
+ * This operation is atomic and cannot be reordered. 
+ * It can be reorderdered on other architectures other than x86.
  * It also implies a memory barrier.
  */
 static inline int test_and_clear_bit(int nr, volatile unsigned long * addr)

             reply	other threads:[~2004-08-05 21:26 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-08-05 20:06 Marcelo Tosatti [this message]
2004-08-06 14:17 ` [PATCH] x86 bitops.h commentary on instruction reordering Vladislav Bolkhovitin
2004-08-06 14:33   ` Marcelo Tosatti
2004-08-06 15:36     ` Vladislav Bolkhovitin
2004-08-06 15:53       ` Marcelo Tosatti
2004-08-06 16:52         ` Vladislav Bolkhovitin
2004-08-06 17:09           ` Marcelo Tosatti
2004-08-06 22:33             ` H. Peter Anvin
2004-08-07  1:01               ` Zwane Mwaikambo
2004-08-07 22:16               ` Alan Cox
2004-08-09 15:14             ` Vladislav Bolkhovitin
2004-08-09 15:50               ` Marcelo Tosatti
2004-08-09 17:35                 ` Vladislav Bolkhovitin
     [not found]                   ` <20040809183437.GD6361@logos.cnet>
2004-08-09 20:12                     ` Vladislav Bolkhovitin
2004-08-09 21:21                       ` Alan Cox
2004-08-10 12:28                         ` Vladislav Bolkhovitin
2004-08-09 11:58         ` Maciej W. Rozycki
     [not found] <2pY7Y-W1-7@gated-at.bofh.it>
     [not found] ` <2qdJL-3sh-27@gated-at.bofh.it>
     [not found]   ` <2qePx-4eM-21@gated-at.bofh.it>
     [not found]     ` <2qf8S-4pI-31@gated-at.bofh.it>
     [not found]       ` <2qg4V-54c-17@gated-at.bofh.it>
     [not found]         ` <2qgoh-5og-29@gated-at.bofh.it>
     [not found]           ` <2qhkn-649-41@gated-at.bofh.it>
     [not found]             ` <2rkgh-zP-45@gated-at.bofh.it>
     [not found]               ` <2rlFk-1wC-27@gated-at.bofh.it>
     [not found]                 ` <2rmhX-20I-17@gated-at.bofh.it>
     [not found]                   ` <2roWw-40d-19@gated-at.bofh.it>
     [not found]                     ` <2roWw-40d-17@gated-at.bofh.it>
2004-08-09 21:23                       ` Andi Kleen
2004-08-10 12:26                         ` Vladislav Bolkhovitin

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=20040805200622.GA17324@logos.cnet \
    --to=marcelo.tosatti@cyclades.com \
    --cc=akpm@osdl.org \
    --cc=andrea@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.