Linux-Devicetree Archive on lore.kernel.org
 help / color / Atom feed
From: Minchan Kim <minchan@kernel.org>
To: Michal Hocko <mhocko@suse.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	linux-mm <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>,
	hyesoo.yu@samsung.com, david@redhat.com, surenb@google.com,
	pullip.cho@samsung.com, joaodias@google.com, hridya@google.com,
	john.stultz@linaro.org, sumit.semwal@linaro.org,
	linux-media@vger.kernel.org, devicetree@vger.kernel.org,
	hch@infradead.org, robh+dt@kernel.org,
	linaro-mm-sig@lists.linaro.org
Subject: Re: [PATCH v4 2/4] mm: failfast mode with __GFP_NORETRY in alloc_contig_range
Date: Tue, 26 Jan 2021 11:10:18 -0800
Message-ID: <YBBpGji3BJYHidHs@google.com> (raw)
In-Reply-To: <20210126074449.GA827@dhcp22.suse.cz>

On Tue, Jan 26, 2021 at 08:44:49AM +0100, Michal Hocko wrote:
> On Mon 25-01-21 11:33:36, Minchan Kim wrote:
> > On Mon, Jan 25, 2021 at 02:12:00PM +0100, Michal Hocko wrote:
> > > On Thu 21-01-21 09:55:00, Minchan Kim wrote:
> > > > Contiguous memory allocation can be stalled due to waiting
> > > > on page writeback and/or page lock which causes unpredictable
> > > > delay. It's a unavoidable cost for the requestor to get *big*
> > > > contiguous memory but it's expensive for *small* contiguous
> > > > memory(e.g., order-4) because caller could retry the request
> > > > in different range where would have easy migratable pages
> > > > without stalling.
> > > > 
> > > > This patch introduce __GFP_NORETRY as compaction gfp_mask in
> > > > alloc_contig_range so it will fail fast without blocking
> > > > when it encounters pages needed waiting.
> > > 
> > > I am not against controling how hard this allocator tries with gfp mask
> > > but this changelog is rather void on any data and any user.
> > > 
> > > It is also rather dubious to have retries when then caller says to not
> > > retry.
> > 
> > Since max_tries is 1 with ++tries, it shouldn't retry.
> 
> OK, I have missed that. This is a tricky code. ASYNC mode should be
> completely orthogonal to the retries count. Those are different things.
> Page allocator does an explicit bail out based on __GFP_NORETRY. You
> should be doing the same.

A concern with __GFP_NOWAIT is regardless of flags passed to cma_alloc,
internal implementation of alloc_contig_range inside will use blockable
operation. See __alloc_contig_migrate_range.

If we go with __GFP_NOWAIT, we should propagate the gfp_mask inside of
__alloc_contig_migrate_range to make cma_alloc consistent with alloc_pages.
(IIUC, that's what you want - make gfp_mask consistent between cma_alloc
and alloc_pages) but I am worry about the direction will make complicate
situation since cma invovles migration context as well as target page
allocation context. Sometime, the single gfp flag could be trouble
to express both contexts all at once. 

> 
> > > 
> > > Also why didn't you consider GFP_NOWAIT semantic for non blocking mode?
> > 
> > GFP_NOWAIT seems to be low(specific) flags rather than the one I want to
> > express. Even though I said only page writeback/lock in the description,
> > the goal is to avoid costly operations we might find later so such
> > "failfast", I thought GFP_NORETRY would be good fit.
> 
> I suspect you are too focused on implementation details here. Think
> about the indended semantic. Callers of this functionality will not
> think about those (I hope because if they rely on these details then the
> whole thing will become unmaintainable because any change would require
> an audit of all existing users). All you should be caring about is to
> control how expensive the call can be. GFP_NOWAIT is not really low
> level from that POV. It gives you a very lightweight non-sleeping
> attempt to allocate. GFP_NORETRY will give you potentially sleeping but
> an opportunistic-easy-to-fail attempt. And so on. See how that is
> absolutely free of any page writeback or any specific locking.

With above reason I mentioned, I wanted to express __GFP_NORETRY as 
"opportunistic-easy-to-fail attempt" to support cma_alloc as "failfast"
for migration context.

> -- 
> Michal Hocko
> SUSE Labs

  reply index

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-21 17:54 [PATCH v4 0/4] Chunk Heap Support on DMA-HEAP Minchan Kim
2021-01-21 17:54 ` [PATCH v4 1/4] mm: cma: introduce gfp flag in cma_alloc instead of no_warn Minchan Kim
2021-01-21 18:46   ` David Hildenbrand
2021-01-21 18:50   ` Minchan Kim
2021-01-25 13:07   ` Michal Hocko
2021-01-25 19:42     ` Minchan Kim
2021-01-26  7:38       ` Michal Hocko
2021-01-26 19:12         ` Minchan Kim
2021-01-27 20:21           ` Minchan Kim
2021-01-21 17:55 ` [PATCH v4 2/4] mm: failfast mode with __GFP_NORETRY in alloc_contig_range Minchan Kim
2021-01-25 13:12   ` Michal Hocko
2021-01-25 13:13     ` Michal Hocko
2021-01-25 19:33     ` Minchan Kim
2021-01-26  7:44       ` Michal Hocko
2021-01-26 19:10         ` Minchan Kim [this message]
2021-01-27  8:14           ` Michal Hocko
2021-01-27 20:42         ` Minchan Kim
2021-01-28  7:53           ` Michal Hocko
2021-01-28 16:56             ` Minchan Kim
2021-01-21 17:55 ` [PATCH v4 3/4] dt-bindings: reserved-memory: Make DMA-BUF CMA heap DT-configurable Minchan Kim
2021-01-26  7:07   ` John Stultz
2021-01-27 20:25     ` Hridya Valsaraju
2021-02-05 22:55   ` Rob Herring
2021-01-21 17:55 ` [PATCH v4 4/4] dma-buf: heaps: add chunk heap to dmabuf heaps Minchan Kim
2021-01-26  7:07   ` Christoph Hellwig
2021-01-26 19:27     ` Minchan Kim
2021-01-26  7:32   ` John Stultz
2021-01-26 19:24     ` Minchan Kim
2021-01-26  7:46   ` Michal Hocko
2021-01-26 19:25     ` Minchan Kim
2021-01-27  8:09       ` Michal Hocko

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=YBBpGji3BJYHidHs@google.com \
    --to=minchan@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=david@redhat.com \
    --cc=devicetree@vger.kernel.org \
    --cc=hch@infradead.org \
    --cc=hridya@google.com \
    --cc=hyesoo.yu@samsung.com \
    --cc=joaodias@google.com \
    --cc=john.stultz@linaro.org \
    --cc=linaro-mm-sig@lists.linaro.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.com \
    --cc=pullip.cho@samsung.com \
    --cc=robh+dt@kernel.org \
    --cc=sumit.semwal@linaro.org \
    --cc=surenb@google.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

Linux-Devicetree Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-devicetree/0 linux-devicetree/git/0.git
	git clone --mirror https://lore.kernel.org/linux-devicetree/1 linux-devicetree/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-devicetree linux-devicetree/ https://lore.kernel.org/linux-devicetree \
		devicetree@vger.kernel.org
	public-inbox-index linux-devicetree

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-devicetree


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git