public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: David Miller <davem@davemloft.net>
Cc: rostedt@goodmis.org, richm@oldelvet.org.uk,
	609371@bugs.debian.org, ben@decadent.org.uk,
	sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org,
	fweisbec@gmail.com, mingo@redhat.com
Subject: Re: Bug#609371: linux-image-2.6.37-trunk-sparc64: module scsi_mod: Unknown relocation: 36
Date: Wed, 19 Jan 2011 17:33:39 -0500	[thread overview]
Message-ID: <20110119223339.GD23544@Krystal> (raw)
In-Reply-To: <20110119.142137.184823805.davem@davemloft.net>

* David Miller (davem@davemloft.net) wrote:
> From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
> Date: Wed, 19 Jan 2011 17:13:27 -0500
> 
> > Hrm, I'd like to see what kind of ill-conceived 32-bit architecture would
> > generate a unaligned access for a 32-bit aligned u64. Do you have examples in
> > mind ? By definition, the memory accesses should be at most 32-bit, no ? AFAIK,
> > gcc treats u64 as two distinct reads on all 32-bit architectures.
> 
> Sparc 32-bit has 64-bit loads and stores, GCC uses them because the ABI
> specifies that every structure is at least 8 byte aligned.

Ah, that's the answer I was looking for, thanks!

> 
> > gcc on my sparc64 box (32-bit userland) disagrees with you here ;) Using
> > gcc (Debian 4.3.3-14) 4.3.3, here is a demonstration that, indeed, "packed"
> > generates aweful code, but that "packed, aligned(4 or 8)" generates pretty
> > decent code:
> 
> Amazing, if this works then do it.
> 
> But please document this fully with comments and such :-)

I will, I will! ;)

So I guess we go for the following. Is it verbose enough ?

/*
 * __u64_packed_aligned:
 *
 * Forces gcc to use the u64 type alignment, up-aligning or down-aligning the
 * target type if necessary. The memory accesses to the target structure are
 * efficient (does not require bytewise memory accesses) and the atomic pointer
 * update guarantees required by RCU are kept. u64 is considered as the largest
 * type that can generate a trap for unaligned accesses (u64 on sparc32 needs to
 * be aligned on 64-bit).
 *
 * Specifying both "packed" and "aligned" generates decent code (without the
 * bytewise memory accesses generated by simply using "packed"), and forces
 * gcc to down-align the structure alignment to the alignment of a u64 type.
 *
 * This alignment should be used for both structure definitions and declarations
 * (as *both* the type and variable attribute) when using the "section"
 * attribute to generate arrays of structures.
 */
#define __u64_packed_aligned \
        __attribute__((__packed__, __aligned__(__alignof__(long long))))

Thanks,

Mathieu

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com

  reply	other threads:[~2011-01-19 22:33 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20110113.155700.102679408.davem@davemloft.net>
     [not found] ` <4D302B2F.7030108@oldelvet.org.uk>
     [not found]   ` <4D3074FE.3030707@oldelvet.org.uk>
2011-01-16  5:17     ` Bug#609371: linux-image-2.6.37-trunk-sparc64: module scsi_mod: Unknown relocation: 36 David Miller
2011-01-16 14:17       ` Richard Mortimer
2011-01-16 19:39         ` David Miller
2011-01-17 14:11           ` Steven Rostedt
2011-01-17 14:37             ` Bastian Blank
2011-01-17 19:35             ` Mathieu Desnoyers
2011-01-18  6:36               ` David Miller
2011-01-18  5:34             ` David Miller
2011-01-18  6:00               ` David Miller
2011-01-18  6:08                 ` David Miller
2011-01-18 16:46                   ` Mathieu Desnoyers
2011-01-18 17:33                     ` Steven Rostedt
2011-01-18 18:16                       ` Steven Rostedt
2011-01-18 18:26                         ` Steven Rostedt
2011-01-18 20:13                           ` Mathieu Desnoyers
2011-01-18 20:22                             ` Steven Rostedt
2011-01-19  5:08                               ` Mathieu Desnoyers
2011-01-19  5:16                                 ` David Miller
2011-01-19 15:10                                   ` Mathieu Desnoyers
2011-01-19 16:14                                     ` Sam Ravnborg
2011-01-19 16:18                                       ` Mathieu Desnoyers
2011-01-19  6:32                                 ` David Miller
2011-01-19  7:20                                   ` David Miller
2011-01-19 15:33                                     ` Mathieu Desnoyers
2011-01-19 21:40                                       ` David Miller
2011-01-19 22:00                                         ` Steven Rostedt
2011-01-19 22:09                                           ` David Miller
2011-01-19 22:21                                           ` Mathieu Desnoyers
2011-01-19 22:23                                             ` David Miller
2011-01-19 22:32                                             ` Sam Ravnborg
2011-01-19 22:34                                               ` Mathieu Desnoyers
2011-01-19 22:13                                         ` Mathieu Desnoyers
2011-01-19 22:21                                           ` David Miller
2011-01-19 22:33                                             ` Mathieu Desnoyers [this message]
2011-01-20  0:41                                               ` David Miller
2011-01-21  0:04                                                 ` Mathieu Desnoyers
2011-01-21 18:06                                                   ` Richard Mortimer
2011-01-21 18:52                                                     ` Mathieu Desnoyers
2011-01-21 19:15                                                       ` Mathieu Desnoyers
2011-01-21 20:14                                                       ` Richard Mortimer
2011-01-21 20:40                                                         ` Mathieu Desnoyers
2011-01-21 22:50                                                           ` Richard Mortimer
2011-01-22 18:42                                                             ` Richard Mortimer
2011-01-22 18:53                                                               ` Mathieu Desnoyers
2011-01-19 15:46                                     ` Steven Rostedt
2011-01-19 16:15                                       ` Mathieu Desnoyers
2011-01-19 18:13                                         ` Steven Rostedt
2011-01-19 18:20                                           ` Mathieu Desnoyers
2011-01-19 21:44                                             ` David Miller
2011-01-19 22:15                                               ` Mathieu Desnoyers
2011-01-19 22:22                                                 ` David Miller
2011-01-19 15:11                                   ` Mathieu Desnoyers
2011-01-19 15:27                   ` Richard Mortimer
2011-01-17  6:07       ` David Miller
2011-01-17  9:05         ` Jesper Nilsson
2011-02-01  5:11           ` David Miller
2011-02-01 10:03             ` Jesper Nilsson
2011-01-17 10:22         ` Richard Mortimer
2011-01-17 14:15           ` Steven Rostedt
2011-01-18  6:35             ` David Miller
2011-01-18 17:30               ` Steven Rostedt
2011-01-17 19:46           ` R_SPARC_13 (Re: Bug#609371: linux-image-2.6.37-trunk-sparc64: module scsi_mod: Unknown relocation: 36) Richard Mortimer
2011-01-17 21:02             ` R_SPARC_13 David Miller
2011-01-17 23:34               ` R_SPARC_13 Richard Mortimer
2011-01-18  0:18                 ` R_SPARC_13 David Miller
2011-01-18  0:37                 ` R_SPARC_13 David Miller
2011-01-18  1:28                   ` R_SPARC_13 Richard Mortimer
2011-01-18  6:50                   ` R_SPARC_13 David Miller
2011-01-18 10:52                     ` R_SPARC_13 Richard Mortimer
2011-01-18 13:23                       ` R_SPARC_13 Richard Mortimer
2011-01-18 21:00                         ` R_SPARC_13 David Miller
2011-01-19  4:12                           ` R_SPARC_13 David Miller
2011-01-17 14:39         ` Bug#609371: linux-image-2.6.37-trunk-sparc64: module scsi_mod: Unknown relocation: 36 Bernhard R. Link
2011-01-18  5:24           ` David Miller
2011-01-18  9:26             ` Jesper Nilsson
2011-01-18  6:27           ` David Miller
2011-01-18 17:05             ` Steven Rostedt

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=20110119223339.GD23544@Krystal \
    --to=mathieu.desnoyers@efficios.com \
    --cc=609371@bugs.debian.org \
    --cc=ben@decadent.org.uk \
    --cc=davem@davemloft.net \
    --cc=fweisbec@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=richm@oldelvet.org.uk \
    --cc=rostedt@goodmis.org \
    --cc=sparclinux@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox