public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: David Rees <dbr@greenhydrant.com>
To: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: linux-kernel@vger.kernel.org
Subject: Re: NFS to Irix server broken again in 2.4.9
Date: Tue, 4 Sep 2001 13:20:39 -0700	[thread overview]
Message-ID: <20010904132039.A18908@greenhydrant.com> (raw)
In-Reply-To: <15253.1002.189305.674221@barley.abo.fi> <20010904120737.A17459@greenhydrant.com> <shswv3eena8.fsf@charged.uio.no>
In-Reply-To: <shswv3eena8.fsf@charged.uio.no>; from trond.myklebust@fys.uio.no on Tue, Sep 04, 2001 at 10:03:43PM +0200

On Tue, Sep 04, 2001 at 10:03:43PM +0200, Trond Myklebust wrote:
> >>>>> " " == David Rees <dbr@greenhydrant.com> writes:
> 
>      > Previous 2.4.X kernels didn't require the 32bitclients option
>      > on the IRIX server for some reason.
> 
> That was because prior to 2.4.9 the kernel would automatically
> truncate the getdents() offsets to 32 bits. We now have true 64 bit
> offsets, and they actually get passed back to userland.
> 
> glibc-2.x's 32 bit version of readdir() still assumes that
> getdents64() syscall returns some an offset (rather than a cookie) and
> that the offset fits into 32bits on ordinary directories.
> Using '32bitclients' on these older IRIX servers sort of shoehorns
> them into the glibc assumptions in the same way the 32 bit truncation
> in kernels 2.4.[0-8] did.

Hmm, the the server I'm running is IRIX 6.5.12 which isn't that old.  IRIX
6.5.13 was only released a little while ago (with some specific IRIX/Linux
NFS fixes according to the changelog) so I assume that with 6.5.13 we won't
need the 32bitclients options.

Thanks for the info.

-Dave

      reply	other threads:[~2001-09-04 20:20 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-09-04 16:40 NFS to Irix server broken again in 2.4.9 Mikko Huhtala
2001-09-04 19:07 ` David Rees
2001-09-04 20:03   ` Trond Myklebust
2001-09-04 20:20     ` David Rees [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=20010904132039.A18908@greenhydrant.com \
    --to=dbr@greenhydrant.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=trond.myklebust@fys.uio.no \
    /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