All of lore.kernel.org
 help / color / mirror / Atom feed
From: "J. Bruce Fields" <bfields@fieldses.org>
To: NeilBrown <neilb@suse.de>
Cc: linux-nfs@vger.kernel.org
Subject: Re: [PATCH 0/9] Some improvements to request deferral and related code
Date: Fri, 11 Sep 2009 17:07:29 -0400	[thread overview]
Message-ID: <20090911210729.GG5701@fieldses.org> (raw)
In-Reply-To: <20090909062539.20462.67466.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>

On Wed, Sep 09, 2009 at 04:32:54PM +1000, NeilBrown wrote:
> Hi,
>  here again are those patches from request deferral improvement set
>  that have not yet been applied.
> 
>  I had previously missed the cache lookups in svcauth_gss, and have now
>  made them work correctly.
> 
>  I also discovered that for NFSv4.1 and later, the current deferral
>  scheme is disabled once the request gets into the NFS layer (i.e. out
>  of the RPC layer).  I don't know that 4.0 doesn't disable it as well
>  - maybe the problems it can cause are more severe with 4.1?

Yes, I think we'd need special handling of the sequence operation, at
least.  I can't recall exactly what the issue was, but we just decided
to give up on deferrals entirely until they were handled in some better
way....

> With
>  this patch set included it might be acceptable to disable the current
>  scheme for 4.0.

Great, thanks.

--b.

> 
>  I believe this is ready for wider testing and probably mainline
>  inclusion.
> 
> Thanks,
> NeilBrown
> 
> 
> ---
> 
> NeilBrown (9):
>       sunrpc: close connection when a request is irretrievably lost.
>       sunrpc/cache: change deferred-request hash table to use hlist.
>       nfsd/idmap: drop special request deferal in favour of improved default.
>       sunrpc/cache: retry cache lookups that return -ETIMEDOUT
>       sunrpc/cache: allow threads to block while waiting for cache update.
>       sunrpc/cache: avoid variable over-loading in cache_defer_req
>       sunrpc/cache: use list_del_init for the list_head entries in cache_deferred_req
>       sunrpc/cache: simplify cache_fresh_locked and cache_fresh_unlocked.
>       sunrpc/cache: change cache_defer_req to return -ve error, not boolean.
> 
> 
>  fs/nfsd/export.c                  |   18 ++++++
>  fs/nfsd/nfs4idmap.c               |  105 ++++------------------------------
>  include/linux/sunrpc/cache.h      |    5 +-
>  include/linux/sunrpc/svcauth.h    |   10 ++-
>  net/sunrpc/auth_gss/svcauth_gss.c |   29 ++++++---
>  net/sunrpc/cache.c                |  115 ++++++++++++++++++++++++-------------
>  net/sunrpc/svc.c                  |    3 +
>  net/sunrpc/svc_xprt.c             |   11 ++++
>  net/sunrpc/svcauth_unix.c         |   34 +++++++++--
>  9 files changed, 176 insertions(+), 154 deletions(-)
> 
> -- 
> Signature
> 

      parent reply	other threads:[~2009-09-11 21:07 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-09-09  6:32 [PATCH 0/9] Some improvements to request deferral and related code NeilBrown
     [not found] ` <20090909062539.20462.67466.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-09-09  6:32   ` [PATCH 3/9] sunrpc/cache: use list_del_init for the list_head entries in cache_deferred_req NeilBrown
     [not found]     ` <20090909063254.20462.7969.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-09-18 15:48       ` J. Bruce Fields
2009-09-09  6:32   ` [PATCH 6/9] sunrpc/cache: retry cache lookups that return -ETIMEDOUT NeilBrown
     [not found]     ` <20090909063254.20462.41616.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-12-02 22:11       ` J. Bruce Fields
2009-12-03 16:57         ` J. Bruce Fields
2009-12-04  4:38           ` Neil Brown
     [not found]             ` <20091204153845.1ec83de5-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-12-05  1:17               ` J. Bruce Fields
2009-12-15  6:27                 ` Neil Brown
     [not found]                   ` <20091215172729.5e1d0190-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2010-02-01 17:11                     ` J. Bruce Fields
2010-02-02 21:33                       ` Neil Brown
2009-09-09  6:32   ` [PATCH 4/9] sunrpc/cache: avoid variable over-loading in cache_defer_req NeilBrown
     [not found]     ` <20090909063254.20462.68582.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-09-18 21:24       ` J. Bruce Fields
2009-09-09  6:32   ` [PATCH 5/9] sunrpc/cache: allow threads to block while waiting for cache update NeilBrown
     [not found]     ` <20090909063254.20462.99277.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-12-02 20:59       ` J. Bruce Fields
2009-12-02 21:23         ` Trond Myklebust
2009-12-02 21:50           ` Trond Myklebust
2009-09-09  6:32   ` [PATCH 2/9] sunrpc/cache: simplify cache_fresh_locked and cache_fresh_unlocked NeilBrown
2009-09-09  6:32   ` [PATCH 7/9] nfsd/idmap: drop special request deferal in favour of improved default NeilBrown
     [not found]     ` <20090909063254.20462.80299.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-12-02 22:18       ` J. Bruce Fields
2009-09-09  6:32   ` [PATCH 1/9] sunrpc/cache: change cache_defer_req to return -ve error, not boolean NeilBrown
     [not found]     ` <20090909063254.20462.57204.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-09-11 21:03       ` J. Bruce Fields
2009-09-09  6:32   ` [PATCH 8/9] sunrpc/cache: change deferred-request hash table to use hlist NeilBrown
2009-09-09  6:32   ` [PATCH 9/9] sunrpc: close connection when a request is irretrievably lost NeilBrown
2009-09-11 21:07   ` J. Bruce Fields [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=20090911210729.GG5701@fieldses.org \
    --to=bfields@fieldses.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=neilb@suse.de \
    /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.