Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230810103332.3062143-1-ryan.roberts@arm.com>
 <20230810103332.3062143-4-ryan.roberts@arm.com> <CAOUHufZatMprwzaT+=Lb6jSw60DN_Br=Og+N6krK3OQZiGyAYA@mail.gmail.com>
In-Reply-To: <CAOUHufZatMprwzaT+=Lb6jSw60DN_Br=Og+N6krK3OQZiGyAYA@mail.gmail.com>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Thu, 10 Aug 2023 11:31:54 -0700
Message-ID: <CAHk-=wiPp3DquE4A4FaWjgmS4wx7BT4MEFj5AOQie=tnK1N6yg@mail.gmail.com>
Subject: Re: [PATCH v1 3/4] mm/mmu_gather: Remove encoded_page infrastructure
To:     Yu Zhao <yuzhao@google.com>
Cc:     Ryan Roberts <ryan.roberts@arm.com>,
        Hugh Dickins <hughd@google.com>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Will Deacon <will@kernel.org>,
        "Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Nick Piggin <npiggin@gmail.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Christian Borntraeger <borntraeger@linux.ibm.com>,
        Sven Schnelle <svens@linux.ibm.com>,
        Arnd Bergmann <arnd@arndb.de>,
        "Matthew Wilcox (Oracle)" <willy@infradead.org>,
        David Hildenbrand <david@redhat.com>,
        "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
        Yin Fengwei <fengwei.yin@intel.com>,
        Yang Shi <shy828301@gmail.com>,
        "Huang, Ying" <ying.huang@intel.com>, Zi Yan <ziy@nvidia.com>,
        linux-mm@kvack.org, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, 10 Aug 2023 at 10:35, Yu Zhao <yuzhao@google.com> wrote:
>
> Adding the original author and reviewers... They might want (need) to
> take a look at this series.

It looks fine to me. The important part is that the rmap removal has
to be done after the TLB flush, but before the page table lock is
released.

That used to be a special thing for anonymous pages and thus needed
that special flag. But if it's done for *all* pages the need to flag
pages goes away.

I see no issues with this, although obviously I might have missed something.

                Linus