From: Andrew Morton <akpm@linux-foundation.org>
To: Minchan Kim <minchan@kernel.org>
Cc: Chris Goldsworthy <cgoldswo@codeaurora.org>,
Laura Abbott <labbott@kernel.org>,
Oliver Sang <oliver.sang@intel.com>,
David Hildenbrand <david@redhat.com>,
John Dias <joaodias@google.com>,
Matthew Wilcox <willy@infradead.org>,
Michal Hocko <mhocko@suse.com>,
Suren Baghdasaryan <surenb@google.com>,
Vlastimil Babka <vbabka@suse.cz>,
LKML <linux-kernel@vger.kernel.org>,
lkp@lists.01.org, lkp@intel.com, ying.huang@intel.com,
feng.tang@intel.com, zhengjun.xing@intel.com,
linux-mm <linux-mm@kvack.org>
Subject: Re: [PATCH v2] mm: fs: invalidate bh_lrus for only cold path
Date: Tue, 1 Jun 2021 16:15:40 -0700 [thread overview]
Message-ID: <20210601161540.9f449314965bd94c84725481@linux-foundation.org> (raw)
In-Reply-To: <20210601145425.1396981-1-minchan@kernel.org>
On Tue, 1 Jun 2021 07:54:25 -0700 Minchan Kim <minchan@kernel.org> wrote:
> kernel test robot reported the regression of fio.write_iops[1]
> with [2].
>
> Since lru_add_drain is called frequently, invalidate bh_lrus
> there could increase bh_lrus cache miss ratio, which needs
> more IO in the end.
>
> This patch moves the bh_lrus invalidation from the hot path(
> e.g., zap_page_range, pagevec_release) to cold path(i.e.,
> lru_add_drain_all, lru_cache_disable).
This code is starting to hurt my brain.
What are the locking/context rules for invalidate_bh_lrus_cpu()?
AFAICT it offers no protection against two CPUs concurrently running
__invalidate_bh_lrus() against the same bh_lru.
So when CONFIG_SMP=y, invalidate_bh_lrus_cpu() must always and only be
run on the cpu which owns the bh_lru. In which case why does it have
the `cpu' arg?
Your new lru_add_and_bh_lrus_drain() follows these rules by calling
invalidate_bh_lrus_cpu() from a per-cpu worker or when CONFIG_SMP=n.
I think. It's all as clear as mud and undocumented. Could you please
take a look at this? Comment the locking/context rules thoroughly and
check that they are being followed? Not forgetting cpu hotplug... See if
there's a way of simplifying/clarifying the code?
The fact that swap.c has those #ifdef CONFIG_SMPs in there is a hint
that we're doing something wrong (or poorly) in there. Perhaps that's
unavoidable because of all the fancy footwork in __lru_add_drain_all().
next prev parent reply other threads:[~2021-06-01 23:15 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-01 14:54 [PATCH v2] mm: fs: invalidate bh_lrus for only cold path Minchan Kim
2021-06-01 23:15 ` Andrew Morton [this message]
2021-06-02 22:45 ` Minchan Kim
2021-06-09 20:52 ` Chris Goldsworthy
2021-06-18 22:05 ` Minchan Kim
2021-07-21 5:11 ` Chris Goldsworthy
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=20210601161540.9f449314965bd94c84725481@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=cgoldswo@codeaurora.org \
--cc=david@redhat.com \
--cc=feng.tang@intel.com \
--cc=joaodias@google.com \
--cc=labbott@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lkp@intel.com \
--cc=lkp@lists.01.org \
--cc=mhocko@suse.com \
--cc=minchan@kernel.org \
--cc=oliver.sang@intel.com \
--cc=surenb@google.com \
--cc=vbabka@suse.cz \
--cc=willy@infradead.org \
--cc=ying.huang@intel.com \
--cc=zhengjun.xing@intel.com \
/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).