All of lore.kernel.org
 help / color / mirror / Atom feed
From: NeilBrown <neilb@suse.com>
To: lustre-devel@lists.lustre.org
Subject: [lustre-devel] [PATCH 18/29] lustre: osc_cache: avoid unnecessary tests.
Date: Wed, 09 Jan 2019 17:24:02 +1100	[thread overview]
Message-ID: <154701504226.26726.13771066469284363266.stgit@noble> (raw)
In-Reply-To: <154701488711.26726.17363928508883972338.stgit@noble>

These tests (have we reached page limit) are not necessary, as the
next call to try_to_add_extent_for_io() will perform the
same test, and fail if necessary.

Having the tests don't add any clarity to the code, and are unlikely
to help performance as they are likely to fail more often than they
succeed.

So discard them.

Signed-off-by: NeilBrown <neilb@suse.com>
---
 drivers/staging/lustre/lustre/osc/osc_cache.c |    6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/staging/lustre/lustre/osc/osc_cache.c b/drivers/staging/lustre/lustre/osc/osc_cache.c
index 14e6efb4b80b..5666f384ac6c 100644
--- a/drivers/staging/lustre/lustre/osc/osc_cache.c
+++ b/drivers/staging/lustre/lustre/osc/osc_cache.c
@@ -1990,8 +1990,6 @@ static unsigned int get_write_extents(struct osc_object *obj,
 			return data.erd_page_count;
 		EASSERT(ext->oe_nr_pages <= data.erd_max_pages, ext);
 	}
-	if (data.erd_page_count == data.erd_max_pages)
-		return data.erd_page_count;
 
 	while (!list_empty(&obj->oo_urgent_exts)) {
 		ext = list_entry(obj->oo_urgent_exts.next,
@@ -1999,8 +1997,6 @@ static unsigned int get_write_extents(struct osc_object *obj,
 		if (!try_to_add_extent_for_io(cli, ext, &data))
 			return data.erd_page_count;
 	}
-	if (data.erd_page_count == data.erd_max_pages)
-		return data.erd_page_count;
 
 	/*
 	 * One key difference between full extents and other extents: full
@@ -2015,8 +2011,6 @@ static unsigned int get_write_extents(struct osc_object *obj,
 		if (!try_to_add_extent_for_io(cli, ext, &data))
 			break;
 	}
-	if (data.erd_page_count == data.erd_max_pages)
-		return data.erd_page_count;
 
 	ext = first_extent(obj);
 	while (ext) {

  parent reply	other threads:[~2019-01-09  6:24 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-09  6:24 [lustre-devel] [PATCH 00/29] assorted osc cleanups NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 01/29] lustre: osc_cache: discard oe_intree NeilBrown
2019-01-10  1:57   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 06/29] lustre: osc: use overlapped() consistently NeilBrown
2019-01-10  2:01   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 09/29] lustre: osc: remove test on 'found' being an error NeilBrown
2019-01-10  2:07   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 05/29] lustre: osc: convert oe_refc and oe_users to kref and refcount_ NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 10/29] lustre: osc_cache: avoid list_for_each_entry_safe when clearing list NeilBrown
2019-01-10  2:10   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 02/29] lustre: osc_cache: use assert_spin_locked() NeilBrown
2019-01-10  1:56   ` Andreas Dilger
2019-01-10  5:04     ` NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 07/29] lustre: osc: convert a while loop to for NeilBrown
2019-01-10  2:04   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 11/29] lustre: osc_cache: simplify osc_wake_cache_waiters() NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 04/29] lustre: osc: simplify list manipulation NeilBrown
2019-01-10  1:58   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 12/29] lustre: osc_cache: avoid confusing variable reuse NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 08/29] lustre: osc: simplify osc_extent_find() NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 03/29] lustre: osc: simplify osc_extent_wait() NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 13/29] lustre: osc_cache: change osc_enter_cache_try to return bool NeilBrown
2019-01-09  6:24 ` NeilBrown [this message]
2019-01-09  6:24 ` [lustre-devel] [PATCH 21/29] lustre: osc_cache: don't drop a lock we didn't take - two NeilBrown
2019-01-10  2:03   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 20/29] lustre: osc_cache: don't drop a lock we didn't take NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 17/29] lustre: osc_cache: simplify list walk in get_write_extents() NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 24/29] lustre: osc_cache: change need_release to bool NeilBrown
2019-01-10  2:43   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 19/29] lustre: osc_cache: convert while to for in get_write_extents() NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 27/29] lustre: osc_cache: white-space and other checkpatch fixes NeilBrown
2019-01-10  2:12   ` Andreas Dilger
2019-01-11  0:48     ` NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 15/29] lustre: osc_cache: change osc_make_rpc() to return bool NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 25/29] lustre: remove cl_page_cancel() NeilBrown
2019-01-10  3:15   ` Andreas Dilger
2019-01-09  6:24 ` [lustre-devel] [PATCH 22/29] lustre: osc_cache: osc_prep_async_page() has meaningless return NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 14/29] lustre: osc_cache: convert cl_cache_waiters to a wait_queue NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 16/29] lustre: osc_cache: use osc_makes_hprpc() more consistently NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 23/29] lustre: osc_cache: remove 'transient' arg from osc_enter_cache_try NeilBrown
2019-01-10  3:02   ` Andreas Dilger
2019-01-10  4:04     ` NeilBrown
2019-01-11  0:27     ` NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 28/29] lustre: osc_request: assorted white-space and check-patch fixes NeilBrown
2019-01-10  2:19   ` Andreas Dilger
2019-01-10  5:25     ` NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 26/29] lustre: osc_cache: simplify osc_page_gang_lookup() NeilBrown
2019-01-10  2:40   ` Andreas Dilger
2019-01-11  1:11     ` NeilBrown
2019-01-11  3:54       ` Andreas Dilger
2019-01-30  3:02         ` NeilBrown
2019-01-09  6:24 ` [lustre-devel] [PATCH 29/29] lustre: centralize handling of PTLRPCD_SET NeilBrown
2019-01-10  2:23   ` Andreas Dilger

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=154701504226.26726.13771066469284363266.stgit@noble \
    --to=neilb@suse.com \
    --cc=lustre-devel@lists.lustre.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 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.