Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
Sender: Minchan Kim <minchan.kim@gmail.com>
Date:   Wed, 11 May 2022 16:28:02 -0700
From:   Minchan Kim <minchan@kernel.org>
To:     John Hubbard <jhubbard@nvidia.com>
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        linux-mm <linux-mm@kvack.org>,
        LKML <linux-kernel@vger.kernel.org>,
        "Paul E . McKenney" <paulmck@kernel.org>,
        John Dias <joaodias@google.com>,
        David Hildenbrand <david@redhat.com>
Subject: Re: [PATCH v4] mm: fix is_pinnable_page against on cma page
Message-ID: <YnxGgo9iCsnOOBHE@google.com>
References: <bb6e76f1-2cbc-893d-c8ab-3ecd3fcae2a5@nvidia.com>
 <Ynr+wTCQpyh8+vOD@google.com>
 <2ffa7670-04ea-bb28-28f8-93a9b9eea7e8@nvidia.com>
 <YnwupNzDNv7IbjRQ@google.com>
 <54b5d177-f2f4-cef2-3a68-cd3b0b276f86@nvidia.com>
 <Ynw6mauQuNhrOAHy@google.com>
 <f7bcda22-8ffe-67be-36bc-fcde58d8884a@nvidia.com>
 <YnxCCPZUfgQNXSg6@google.com>
 <8f083802-7ab0-15ec-b37d-bc9471eea0b1@nvidia.com>
 <YnxDhdb4YCo19Qx0@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <YnxDhdb4YCo19Qx0@google.com>
Precedence: bulk

On Wed, May 11, 2022 at 04:15:17PM -0700, Minchan Kim wrote:
> On Wed, May 11, 2022 at 04:13:10PM -0700, John Hubbard wrote:
> > On 5/11/22 16:08, Minchan Kim wrote:
> > > > OK, so the code checks the wrong item each time. But the code really
> > > > only needs to know "is either _CMA or _ISOLATE set?". And so you
> > > 
> > > Yes.
> > > 
> > > > can just sidestep the entire question by writing it like this:
> > > > 
> > > > int mt = get_pageblock_migratetype(page);
> > > > 
> > > > if (mt & (MIGRATE_ISOLATE | MIGRATE_CMA))
> > > > 	return false;
> > > 
> > > I am confused. Isn't it same question?
> > > 
> > >                                                      set_pageblock_migratetype(MIGRATE_ISOLATE)
> > > if (get_pageblock_migrate(page) & MIGRATE_CMA)
> > > 
> > >                                                      set_pageblock_migratetype(MIGRATE_CMA)
> > > 
> > > if (get_pageblock_migrate(page) & MIGRATE_ISOLATE)
> > 
> > Well no, because the "&" operation is a single operation on the CPU, and
> > isn't going to get split up like that.
> 
> Oh, if that's true, yeah, I could live with it.
> 
> Thanks, let me post next revision with commenting about that.

This is delta to confirm before posting next revision.

Are you okay with this one?

diff --git a/include/linux/mm.h b/include/linux/mm.h
index cbf79eb790e0..7b2df6780552 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1626,14 +1626,14 @@ static inline bool page_needs_cow_for_dma(struct vm_area_struct *vma,
 static inline bool is_pinnable_page(struct page *page)
 {
 #ifdef CONFIG_CMA
+       int mt = get_pageblock_migratetype(page);
+
        /*
-        * use volatile to use local variable mt instead of
-        * refetching mt value.
+        * "&" operation would prevent compiler split up
+        * get_pageblock_migratetype two times for each
+        * condition check: refetching mt value two times.
         */
-       int __mt = get_pageblock_migratetype(page);
-       int mt = __READ_ONCE(__mt);
-
-       if (mt == MIGRATE_CMA || mt == MIGRATE_ISOLATE)
+       if (mt & (MIGRATE_ISOLATE | MIGRATE_CMA))
                return false;
 #endif