All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jan Kara <jack@suse.cz>
To: Dave Hansen <dave.hansen@intel.com>
Cc: Jan Kara <jack@suse.cz>, Andi Kleen <ak@linux.intel.com>,
	linux-mm@kvack.org, Mel Gorman <mgorman@suse.de>,
	"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
	linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH 0/7 v1] Speed up page cache truncation
Date: Wed, 11 Oct 2017 23:06:13 +0200	[thread overview]
Message-ID: <20171011210613.GQ3667@quack2.suse.cz> (raw)
In-Reply-To: <e596a6d7-4858-8fe6-c315-8a285748a31a@intel.com>

On Wed 11-10-17 10:34:47, Dave Hansen wrote:
> On 10/11/2017 01:06 AM, Jan Kara wrote:
> >>> when rebasing our enterprise distro to a newer kernel (from 4.4 to 4.12) we
> >>> have noticed a regression in bonnie++ benchmark when deleting files.
> >>> Eventually we have tracked this down to a fact that page cache truncation got
> >>> slower by about 10%. There were both gains and losses in the above interval of
> >>> kernels but we have been able to identify that commit 83929372f629 "filemap:
> >>> prepare find and delete operations for huge pages" caused about 10% regression
> >>> on its own.
> >> It's odd that just checking if some pages are huge should be that
> >> expensive, but ok ..
> > Yeah, I was surprised as well but profiles were pretty clear on this - part
> > of the slowdown was caused by loads of page->_compound_head (PageTail()
> > and page_compound() use that) which we previously didn't have to load at
> > all, part was in hpage_nr_pages() function and its use.
> 
> Well, page->_compound_head is part of the same cacheline as the rest of
> the page, and the page is surely getting touched during truncation at
> _some_ point.  The hpage_nr_pages() might cause the cacheline to get
> loaded earlier than before, but I can't imagine that it's that expensive.

Then my intuition matches yours ;) but profiles disagree. That being said
I'm not really expert in CPU microoptimizations and profiling so feel free
to gather perf profiles yourself before and after commit 83929372f629 and
get better explanation of where the cost is - I would be really curious
what you come up with because the explanation I have disagrees with my
intuition as well...
								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2017-10-11 21:06 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-10 15:19 [PATCH 0/7 v1] Speed up page cache truncation Jan Kara
2017-10-10 15:19 ` Jan Kara
2017-10-10 15:19 ` [PATCH 1/7] mm: Speedup cancel_dirty_page() for clean pages Jan Kara
2017-10-10 15:19   ` Jan Kara
2017-10-10 15:19 ` [PATCH 2/7] mm: Refactor truncate_complete_page() Jan Kara
2017-10-10 15:19   ` Jan Kara
2017-10-10 15:19 ` [PATCH 3/7] mm: Factor out page cache page freeing into a separate function Jan Kara
2017-10-10 15:19   ` Jan Kara
2017-10-10 15:19 ` [PATCH 4/7] mm: Move accounting updates before page_cache_tree_delete() Jan Kara
2017-10-10 15:19   ` Jan Kara
2017-10-10 15:19 ` [PATCH 5/7] mm: Move clearing of page->mapping to page_cache_tree_delete() Jan Kara
2017-10-10 15:19   ` Jan Kara
2017-10-10 15:19 ` [PATCH 6/7] mm: Factor out checks and accounting from __delete_from_page_cache() Jan Kara
2017-10-10 15:19   ` Jan Kara
2017-10-10 15:19 ` [PATCH 7/7] mm: Batch radix tree operations when truncating pages Jan Kara
2017-10-10 15:19   ` Jan Kara
2017-10-11  7:39   ` Mel Gorman
2017-10-17 23:05   ` Andrew Morton
2017-10-17 23:05     ` Andrew Morton
2017-10-18 10:44     ` Jan Kara
2017-10-10 17:25 ` [PATCH 0/7 v1] Speed up page cache truncation Andi Kleen
2017-10-10 17:25   ` Andi Kleen
2017-10-11  8:06   ` Jan Kara
2017-10-11 16:51     ` Andi Kleen
2017-10-11 17:34     ` Dave Hansen
2017-10-11 17:59       ` Mel Gorman
2017-10-11 18:37         ` Andi Kleen
2017-10-11 21:06       ` Jan Kara [this message]
2017-10-11 21:24         ` Dave Chinner
2017-10-12  9:09           ` Mel Gorman
2017-10-12 14:07           ` 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=20171011210613.GQ3667@quack2.suse.cz \
    --to=jack@suse.cz \
    --cc=ak@linux.intel.com \
    --cc=dave.hansen@intel.com \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@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.