From: Michal Hocko <mhocko@suse.com>
To: Pavel Tatashin <pasha.tatashin@soleen.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
linux-mm <linux-mm@kvack.org>,
Andrew Morton <akpm@linux-foundation.org>,
Vlastimil Babka <vbabka@suse.cz>,
David Hildenbrand <david@redhat.com>,
Oscar Salvador <osalvador@suse.de>,
Dan Williams <dan.j.williams@intel.com>,
Sasha Levin <sashal@kernel.org>,
Tyler Hicks <tyhicks@linux.microsoft.com>,
Joonsoo Kim <iamjoonsoo.kim@lge.com>,
mike.kravetz@oracle.com, Steven Rostedt <rostedt@goodmis.org>,
Ingo Molnar <mingo@redhat.com>, Jason Gunthorpe <jgg@ziepe.ca>,
Peter Zijlstra <peterz@infradead.org>,
Mel Gorman <mgorman@suse.de>,
Matthew Wilcox <willy@infradead.org>,
David Rientjes <rientjes@google.com>,
John Hubbard <jhubbard@nvidia.com>
Subject: Re: [PATCH 5/6] mm: honor PF_MEMALLOC_NOMOVABLE for all allocations
Date: Fri, 4 Dec 2020 09:43:12 +0100 [thread overview]
Message-ID: <20201204084312.GA25569@dhcp22.suse.cz> (raw)
In-Reply-To: <CA+CK2bB-BC-5Szs1Piv3O=OGxQbJSGWzgMmDUtDewrCqEoNaXw@mail.gmail.com>
On Thu 03-12-20 10:15:41, Pavel Tatashin wrote:
> On Thu, Dec 3, 2020 at 4:17 AM Michal Hocko <mhocko@suse.com> wrote:
> >
> > On Wed 02-12-20 00:23:29, Pavel Tatashin wrote:
> > [...]
> > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > > index 611799c72da5..7a6d86d0bc5f 100644
> > > --- a/mm/page_alloc.c
> > > +++ b/mm/page_alloc.c
> > > @@ -3766,20 +3766,25 @@ alloc_flags_nofragment(struct zone *zone, gfp_t gfp_mask)
> > > return alloc_flags;
> > > }
> > >
> > > -static inline unsigned int current_alloc_flags(gfp_t gfp_mask,
> > > - unsigned int alloc_flags)
> > > +static inline unsigned int cma_alloc_flags(gfp_t gfp_mask,
> > > + unsigned int alloc_flags)
> > > {
> > > #ifdef CONFIG_CMA
> > > - unsigned int pflags = current->flags;
> > > -
> > > - if (!(pflags & PF_MEMALLOC_NOMOVABLE) &&
> > > - gfp_migratetype(gfp_mask) == MIGRATE_MOVABLE)
> > > + if (gfp_migratetype(gfp_mask) == MIGRATE_MOVABLE)
> > > alloc_flags |= ALLOC_CMA;
> > > -
> > > #endif
> > > return alloc_flags;
> > > }
> > >
> > > +static inline gfp_t current_gfp_checkmovable(gfp_t gfp_mask)
> > > +{
> > > + unsigned int pflags = current->flags;
> > > +
> > > + if ((pflags & PF_MEMALLOC_NOMOVABLE))
> > > + return gfp_mask & ~__GFP_MOVABLE;
> > > + return gfp_mask;
> > > +}
> > > +
> >
> > It sucks that we have to control both ALLOC and gfp flags. But wouldn't
> > it be simpler and more straightforward to keep current_alloc_flags as is
> > (module PF rename) and hook the gfp mask evaluation into current_gfp_context
> > and move it up before the first allocation attempt?
>
> We could do that, but perhaps as a separate patch? I am worried about
> hidden implication of adding extra scope (GFP_NOIO|GFP_NOFS) to the
> fast path.
Why?
> Also, current_gfp_context() is used elsewhere, and in some
> places removing __GFP_MOVABLE from gfp_mask means that we will need to
> also change other things. For example [1], in try_to_free_pages() we
> call current_gfp_context(gfp_mask) which can reduce the maximum zone
> idx, yet we simply set it to: reclaim_idx = gfp_zone(gfp_mask), not to
> the newly determined gfp_mask.
Yes and the direct reclaim should honor the movable zone restriction.
Why should we reclaim ZONE_MOVABLE when the allocation cannot really
allocate from it? Or have I misunderstood your concern?
>
> [1] https://soleen.com/source/xref/linux/mm/vmscan.c?r=2da9f630#3239
>
>
> All scope flags
> > should be applicable to the hot path as well. It would add few cycles to
> > there but the question is whether that would be noticeable over just
> > handling PF_MEMALLOC_NOMOVABLE on its own. The cache line would be
> > pulled in anyway.
>
> Let's try it in a separate patch? I will add it in the next version of
> this series.
Separate patch or not is up to you. But I do not see a strong reason why
this cannot be addressed in a single one.
--
Michal Hocko
SUSE Labs
next prev parent reply other threads:[~2020-12-04 8:44 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-02 5:23 [PATCH 0/6] prohibit pinning pages in ZONE_MOVABLE Pavel Tatashin
2020-12-02 5:23 ` [PATCH 1/6] mm/gup: perform check_dax_vmas only when FS_DAX is enabled Pavel Tatashin
2020-12-02 16:22 ` Ira Weiny
2020-12-02 18:15 ` Pavel Tatashin
2020-12-02 16:29 ` Jason Gunthorpe
2020-12-02 18:16 ` Pavel Tatashin
2020-12-03 7:59 ` John Hubbard
2020-12-03 14:52 ` Pavel Tatashin
2020-12-02 5:23 ` [PATCH 2/6] mm/gup: don't pin migrated cma pages in movable zone Pavel Tatashin
2020-12-02 16:31 ` David Hildenbrand
2020-12-02 18:17 ` Pavel Tatashin
2020-12-03 8:01 ` John Hubbard
2020-12-03 8:46 ` Michal Hocko
2020-12-03 14:58 ` Pavel Tatashin
2020-12-02 5:23 ` [PATCH 3/6] mm/gup: make __gup_longterm_locked common Pavel Tatashin
2020-12-02 16:31 ` Ira Weiny
2020-12-02 16:33 ` Ira Weiny
2020-12-02 18:19 ` Pavel Tatashin
2020-12-03 0:03 ` Pavel Tatashin
2020-12-03 8:03 ` John Hubbard
2020-12-03 15:02 ` Pavel Tatashin
2020-12-02 5:23 ` [PATCH 4/6] mm cma: rename PF_MEMALLOC_NOCMA to PF_MEMALLOC_NOMOVABLE Pavel Tatashin
2020-12-03 8:04 ` John Hubbard
2020-12-03 15:02 ` Pavel Tatashin
2020-12-03 8:57 ` Michal Hocko
2020-12-03 15:02 ` Pavel Tatashin
2020-12-02 5:23 ` [PATCH 5/6] mm: honor PF_MEMALLOC_NOMOVABLE for all allocations Pavel Tatashin
2020-12-03 8:17 ` John Hubbard
2020-12-03 15:06 ` Pavel Tatashin
2020-12-03 16:51 ` John Hubbard
2020-12-03 9:17 ` Michal Hocko
2020-12-03 15:15 ` Pavel Tatashin
2020-12-04 8:43 ` Michal Hocko [this message]
2020-12-04 8:54 ` Michal Hocko
2020-12-04 16:07 ` Pavel Tatashin
2020-12-02 5:23 ` [PATCH 6/6] mm/gup: migrate pinned pages out of movable zone Pavel Tatashin
2020-12-02 16:35 ` Jason Gunthorpe
2020-12-03 0:19 ` Pavel Tatashin
2020-12-03 1:08 ` Jason Gunthorpe
2020-12-03 1:34 ` Pavel Tatashin
2020-12-03 14:17 ` Jason Gunthorpe
2020-12-03 16:40 ` Pavel Tatashin
2020-12-03 16:59 ` Jason Gunthorpe
2020-12-03 17:14 ` Pavel Tatashin
2020-12-03 19:15 ` Pavel Tatashin
2020-12-03 19:36 ` Jason Gunthorpe
2020-12-04 16:24 ` Pavel Tatashin
2020-12-04 17:06 ` Jason Gunthorpe
2020-12-04 20:05 ` Daniel Jordan
2020-12-04 20:16 ` Pavel Tatashin
2020-12-08 2:27 ` Daniel Jordan
2020-12-04 20:52 ` Jason Gunthorpe
2020-12-08 2:48 ` Daniel Jordan
2020-12-08 13:24 ` Jason Gunthorpe
2020-12-03 8:22 ` John Hubbard
2020-12-03 15:55 ` Pavel Tatashin
2020-12-04 4:13 ` Joonsoo Kim
2020-12-04 17:43 ` Pavel Tatashin
2020-12-07 7:13 ` Joonsoo Kim
2020-12-04 4:02 ` [PATCH 0/6] prohibit pinning pages in ZONE_MOVABLE Joonsoo Kim
2020-12-04 15:55 ` Pavel Tatashin
2020-12-04 16:10 ` Jason Gunthorpe
2020-12-04 17:50 ` Pavel Tatashin
2020-12-04 18:01 ` David Hildenbrand
2020-12-04 18:10 ` Pavel Tatashin
2020-12-07 7:12 ` Joonsoo Kim
2020-12-07 12:13 ` 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=20201204084312.GA25569@dhcp22.suse.cz \
--to=mhocko@suse.com \
--cc=akpm@linux-foundation.org \
--cc=dan.j.williams@intel.com \
--cc=david@redhat.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=jgg@ziepe.ca \
--cc=jhubbard@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mike.kravetz@oracle.com \
--cc=mingo@redhat.com \
--cc=osalvador@suse.de \
--cc=pasha.tatashin@soleen.com \
--cc=peterz@infradead.org \
--cc=rientjes@google.com \
--cc=rostedt@goodmis.org \
--cc=sashal@kernel.org \
--cc=tyhicks@linux.microsoft.com \
--cc=vbabka@suse.cz \
--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).