mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Nathan Scott <nathans@sgi.com>
To: David Howells <dhowells@redhat.com>
Cc: hch@infradead.org, linux-xfs@oss.sgi.com, linux-kernel@vger.kernel.org
Subject: Re: XFS accessing arch-specific structures
Date: Tue, 13 Dec 2005 08:56:32 +1100	[thread overview]
Message-ID: <20051213085632.B7465575@wobbly.melbourne.sgi.com> (raw)
In-Reply-To: <25190.1134398352@warthog.cambridge.redhat.com>; from dhowells@redhat.com on Mon, Dec 12, 2005 at 02:39:12PM +0000

Hi there David,

On Mon, Dec 12, 2005 at 02:39:12PM +0000, David Howells wrote:
> I've got a problem in which XFS is accessing arch-specific structures, and
> thus requiring those structures to conform to its ideals. Specifically, it's
> trying to read the counter using atomic_read(), whether or not this is
> possible:
> 
>     fs/xfs/linux-2.6/sema.h:
>     #define valusema(sp)			(atomic_read(&(sp)->count))
> 
>     compile log:
> 
>       CC      fs/xfs/xfs_inode.o
>       CC      fs/xfs/xfs_inode_item.o
>     fs/xfs/xfs_inode_item.c: In function `xfs_inode_item_pushbuf':
>     fs/xfs/xfs_inode_item.c:803: error: structure has no member named `count'
>     fs/xfs/xfs_inode_item.c:825: error: structure has no member named `count'
> 
> Can you fix this please? This will not compile with all archs.
> 
> I'm told that Christoph Hellwig may have an idea or a patch that might provide
> a fix. If it's necessary to get the count on the semaphore (which it might
> be), then you should add a function to each asm/semaphore.h to retrieve it and
> use that.

I just chatted with Christoph about this.  For some reason we thought
all instances of direct count use were in debug code (or were put into
debug-only code), but we missed the one in xfs_inode_item_pushbuf.  It
only needs to know if count > 0, so we could implement that using the
down_trylock primitive, but it'd be pretty ugly.

It looks like a couple of the arch's have now got a sem_getcount, and
Christoph is planning on reviving an earlier patch to provide that API
across all architectures, and then we can change XFS to use that.

cheers.

-- 
Nathan

      reply	other threads:[~2005-12-12 21:56 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-12-12 14:39 David Howells
2005-12-12 21:56 ` Nathan Scott [this message]

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=20051213085632.B7465575@wobbly.melbourne.sgi.com \
    --to=nathans@sgi.com \
    --cc=dhowells@redhat.com \
    --cc=hch@infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-xfs@oss.sgi.com \
    /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

all inboxes | Powered by JetHome®