From: "J. Bruce Fields" <bfields@fieldses.org>
To: NeilBrown <neilb@suse.de>
Cc: linux-nfs@vger.kernel.org
Subject: Re: [PATCH 03/12] sunrpc/cache: simplify cache_fresh_locked and cache_fresh_unlocked.
Date: Tue, 4 Aug 2009 11:45:39 -0400 [thread overview]
Message-ID: <20090804154539.GL14249@fieldses.org> (raw)
In-Reply-To: <20090804052238.15929.17142.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
On Tue, Aug 04, 2009 at 03:22:38PM +1000, NeilBrown wrote:
> The extra call to cache_revisit_request in cache_fresh_unlocked is not
> needed, as should have been fairly clear at the time of
> commit 4013edea9a0b6cdcb1fdf5d4011e47e068fd6efb
>
> If there are requests to be revisited, then we can be sure that
> CACHE_PENDING is set, so the second call is sufficient.
Yes: the only cache_make_upcall() caller (cache_check()) sets
CACHE_PENDING first. And a grep for CACHE_PENDING verifies that
everyone that clears it immediately calls either cache_revisit_request()
or cache_dequeue(). OK!
Thanks, applied.--b.
>
> So remove the first call.
> Then remove the 'new' parameter,
> then remove the return value for cache_fresh_locked which is only used
> to provide the value for 'new'.
>
> Signed-off-by: NeilBrown <neilb@suse.de>
> ---
>
> net/sunrpc/cache.c | 23 ++++++++++-------------
> 1 files changed, 10 insertions(+), 13 deletions(-)
>
> diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
> index 44f4516..c1f897c 100644
> --- a/net/sunrpc/cache.c
> +++ b/net/sunrpc/cache.c
> @@ -103,18 +103,16 @@ EXPORT_SYMBOL_GPL(sunrpc_cache_lookup);
>
> static void cache_dequeue(struct cache_detail *detail, struct cache_head *ch);
>
> -static int cache_fresh_locked(struct cache_head *head, time_t expiry)
> +static void cache_fresh_locked(struct cache_head *head, time_t expiry)
> {
> head->expiry_time = expiry;
> head->last_refresh = get_seconds();
> - return !test_and_set_bit(CACHE_VALID, &head->flags);
> + set_bit(CACHE_VALID, &head->flags);
> }
>
> static void cache_fresh_unlocked(struct cache_head *head,
> - struct cache_detail *detail, int new)
> + struct cache_detail *detail)
> {
> - if (new)
> - cache_revisit_request(head);
> if (test_and_clear_bit(CACHE_PENDING, &head->flags)) {
> cache_revisit_request(head);
> cache_dequeue(detail, head);
> @@ -130,7 +128,6 @@ struct cache_head *sunrpc_cache_update(struct cache_detail *detail,
> */
> struct cache_head **head;
> struct cache_head *tmp;
> - int is_new;
>
> if (!test_bit(CACHE_VALID, &old->flags)) {
> write_lock(&detail->hash_lock);
> @@ -139,9 +136,9 @@ struct cache_head *sunrpc_cache_update(struct cache_detail *detail,
> set_bit(CACHE_NEGATIVE, &old->flags);
> else
> detail->update(old, new);
> - is_new = cache_fresh_locked(old, new->expiry_time);
> + cache_fresh_locked(old, new->expiry_time);
> write_unlock(&detail->hash_lock);
> - cache_fresh_unlocked(old, detail, is_new);
> + cache_fresh_unlocked(old, detail);
> return old;
> }
> write_unlock(&detail->hash_lock);
> @@ -165,11 +162,11 @@ struct cache_head *sunrpc_cache_update(struct cache_detail *detail,
> *head = tmp;
> detail->entries++;
> cache_get(tmp);
> - is_new = cache_fresh_locked(tmp, new->expiry_time);
> + cache_fresh_locked(tmp, new->expiry_time);
> cache_fresh_locked(old, 0);
> write_unlock(&detail->hash_lock);
> - cache_fresh_unlocked(tmp, detail, is_new);
> - cache_fresh_unlocked(old, detail, 0);
> + cache_fresh_unlocked(tmp, detail);
> + cache_fresh_unlocked(old, detail);
> cache_put(old, detail);
> return tmp;
> }
> @@ -224,8 +221,8 @@ int cache_check(struct cache_detail *detail,
> cache_revisit_request(h);
> if (rv == -EAGAIN) {
> set_bit(CACHE_NEGATIVE, &h->flags);
> - cache_fresh_unlocked(h, detail,
> - cache_fresh_locked(h, get_seconds()+CACHE_NEW_EXPIRY));
> + cache_fresh_locked(h, get_seconds()+CACHE_NEW_EXPIRY);
> + cache_fresh_unlocked(h, detail);
> rv = -ENOENT;
> }
> break;
>
>
next prev parent reply other threads:[~2009-08-04 15:45 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-08-04 5:22 [PATCH 00/12] Some improvements to request deferral and related code NeilBrown
[not found] ` <20090804051145.15929.11356.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-04 5:22 ` [PATCH 03/12] sunrpc/cache: simplify cache_fresh_locked and cache_fresh_unlocked NeilBrown
[not found] ` <20090804052238.15929.17142.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-04 15:45 ` J. Bruce Fields [this message]
2009-08-04 5:22 ` [PATCH 01/12] sunrpc/cache: rename queue_loose to cache_dequeue NeilBrown
[not found] ` <20090804052238.15929.91015.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-04 14:05 ` J. Bruce Fields
2009-08-04 5:22 ` [PATCH 02/12] sunrpc/cache: make sure deferred requests eventually get revisited NeilBrown
[not found] ` <20090804052238.15929.74402.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-04 15:02 ` J. Bruce Fields
2009-08-04 5:22 ` [PATCH 04/12] sunrpc/cache: recheck cache validity after cache_defer_req NeilBrown
[not found] ` <20090804052238.15929.56800.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-04 20:42 ` J. Bruce Fields
2009-08-06 4:57 ` Neil Brown
[not found] ` <19066.25248.283061.383233-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-25 21:50 ` J. Bruce Fields
2009-08-26 0:42 ` Neil Brown
2009-08-04 5:22 ` [PATCH 07/12] sunrpc/cache: allow thread to block while waiting for cache update NeilBrown
2009-08-04 5:22 ` [PATCH 08/12] sunrpc/cache: retry cache lookups that return -ETIMEDOUT NeilBrown
2009-08-04 5:22 ` [PATCH 06/12] sunrpc/cache: avoid variable over-loading in cache_defer_req NeilBrown
[not found] ` <20090804052239.15929.87201.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-04 20:47 ` J. Bruce Fields
2009-08-06 4:35 ` Neil Brown
2009-08-04 5:22 ` [PATCH 09/12] nfsd/idmap: drop special request deferal in favour of improved default NeilBrown
2009-08-04 5:22 ` [PATCH 10/12] sunrpc: fix memory leak in unix_gid cache NeilBrown
[not found] ` <20090804052239.15929.71459.stgit-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-04 20:55 ` J. Bruce Fields
2009-08-04 5:22 ` [PATCH 11/12] sunrpc/cache: change deferred-request hash table to use hlist NeilBrown
2009-08-04 5:22 ` [PATCH 05/12] sunrpc/cache: use list_del_init for the list_head entries in cache_deferred_req NeilBrown
2009-08-04 5:22 ` [PATCH 12/12] sunrpc: close connection when a request is irretrievably lost NeilBrown
2009-08-04 14:04 ` [PATCH 00/12] Some improvements to request deferral and related code J. Bruce Fields
2009-08-07 4:13 ` Neil Brown
[not found] ` <19067.43518.105153.247173-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2009-08-10 15:05 ` J. Bruce Fields
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=20090804154539.GL14249@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox