From: Jan Kara <jack@suse.cz>
To: Matthew Wilcox <willy@infradead.org>
Cc: <linux-fsdevel@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>, Jan Kara <jack@suse.cz>
Subject: [PATCH 14/23] workingset: Use xas_store_noinit() to clear shadow entry
Date: Wed, 22 Apr 2020 17:02:47 +0200 [thread overview]
Message-ID: <20200422150256.23473-15-jack@suse.cz> (raw)
In-Reply-To: <20200422150256.23473-1-jack@suse.cz>
Shadow entries have no marks. Use xas_store_noinit() to clear the entry
so avoid unneeded initialization of the xarray marks. This provides a
nice boost to truncate numbers. Sample benchmark showing time to
truncate 128 files 1GB each on machine with 64GB of RAM (so about half
of entries are shadow entries):
AVG STDDEV
Vanilla 4.825s 0.036
Patched 4.516s 0.014
So we can see about 6% reduction in overall truncate time.
Signed-off-by: Jan Kara <jack@suse.cz>
---
mm/truncate.c | 2 +-
mm/workingset.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/mm/truncate.c b/mm/truncate.c
index dd9ebc1da356..baef636564cc 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -39,7 +39,7 @@ static inline void __clear_shadow_entry(struct address_space *mapping,
xas_set_update(&xas, workingset_update_node);
if (xas_load(&xas) != entry)
return;
- xas_store(&xas, NULL);
+ xas_store_noinit(&xas, NULL);
mapping->nrexceptional--;
}
diff --git a/mm/workingset.c b/mm/workingset.c
index 474186b76ced..6a492140057f 100644
--- a/mm/workingset.c
+++ b/mm/workingset.c
@@ -537,7 +537,7 @@ static enum lru_status shadow_lru_isolate(struct list_head *item,
* We could store a shadow entry here which was the minimum of the
* shadow entries we were tracking ...
*/
- xas_store(&xas, NULL);
+ xas_store_noinit(&xas, NULL);
__inc_lruvec_slab_state(node, WORKINGSET_NODERECLAIM);
out_invalid:
--
2.16.4
next prev parent reply other threads:[~2020-04-22 15:04 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-22 15:02 [PATCH 0/23 v2] mm: Speedup page cache truncation Jan Kara
2020-04-22 15:02 ` [PATCH 01/23] xarray: Remove stale comment Jan Kara
2020-04-22 15:02 ` [PATCH 02/23] xarray: Provide xas_erase() and xas_store_noinit() helpers Jan Kara
2020-04-22 15:02 ` [PATCH 03/23] xarray: Use xas_store_noinit() in __xa_store, __xa_insert, __xa_alloc Jan Kara
2020-04-22 15:02 ` [PATCH 04/23] xarray: Switch xa_store_range() to use xas_store_noinit() Jan Kara
2020-04-22 15:02 ` [PATCH 05/23] xarray: Use xas_erase() in __xa_erase() Jan Kara
2020-04-22 15:02 ` [PATCH 06/23] xarray: Explicitely set XA_FREE_MARK in __xa_cmpxchg() Jan Kara
2020-04-22 15:02 ` [PATCH 07/23] xarray: Switch __xa_cmpxchg() to use xas_store_noinit() Jan Kara
2020-04-22 15:02 ` [PATCH 08/23] dax: Use xas_erase() in __dax_invalidate_entry() Jan Kara
2020-04-22 15:02 ` [PATCH 09/23] dax: Use dax_store_noinit() in grab_mapping_entry() Jan Kara
2020-04-22 15:02 ` [PATCH 10/23] dax: Convert xas_store() to xas_store_noinit() Jan Kara
2020-04-22 15:02 ` [PATCH 11/23] mm: Use xas_erase() in page_cache_delete_batch() Jan Kara
2020-04-22 15:02 ` [PATCH 12/23] mm: Use xas_erase() in collapse_file() Jan Kara
2020-04-22 15:02 ` [PATCH 13/23] mm: Use xas_store_noinit() when storing non-NULL Jan Kara
2020-04-22 15:02 ` Jan Kara [this message]
2020-04-22 15:02 ` [PATCH 15/23] swap: Use xas_erase() when removing page from swap cache Jan Kara
2020-04-22 15:02 ` [PATCH 16/23] idr: Use xas_erase() in ida_destroy() Jan Kara
2020-04-22 15:02 ` [PATCH 17/23] idr: Use xas_erase() in ida_free() Jan Kara
2020-04-22 15:02 ` [PATCH 18/23] idr: Convert xas_store() to xas_store_noinit() Jan Kara
2020-04-22 15:02 ` [PATCH 19/23] testing: Use xas_erase() to remove entries from xarray Jan Kara
2020-04-22 15:02 ` [PATCH 20/23] testing: Use xas_store_noinit() for non-NULL entries Jan Kara
2020-04-22 15:02 ` [PATCH 21/23] testing: Introduce xa_erase_order() and use it Jan Kara
2020-04-22 15:02 ` [PATCH 22/23] testing: Switch xa_store_order() to xas_store_noinit() Jan Kara
2020-04-22 15:02 ` [PATCH 23/23] xarray: Remove xas_store() Jan Kara
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=20200422150256.23473-15-jack@suse.cz \
--to=jack@suse.cz \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=willy@infradead.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).