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;
Date:   Fri, 7 Jul 2023 09:10:28 -0400
From:   Peter Xu <peterx@redhat.com>
To:     Axel Rasmussen <axelrasmussen@google.com>
Cc:     Alexander Viro <viro@zeniv.linux.org.uk>,
        Andrew Morton <akpm@linux-foundation.org>,
        Brian Geffon <bgeffon@google.com>,
        Christian Brauner <brauner@kernel.org>,
        David Hildenbrand <david@redhat.com>,
        Gaosheng Cui <cuigaosheng1@huawei.com>,
        Huang Ying <ying.huang@intel.com>,
        Hugh Dickins <hughd@google.com>,
        James Houghton <jthoughton@google.com>,
        "Jan Alexander Steffens (heftig)" <heftig@archlinux.org>,
        Jiaqi Yan <jiaqiyan@google.com>,
        Jonathan Corbet <corbet@lwn.net>,
        Kefeng Wang <wangkefeng.wang@huawei.com>,
        "Liam R. Howlett" <Liam.Howlett@oracle.com>,
        Miaohe Lin <linmiaohe@huawei.com>,
        Mike Kravetz <mike.kravetz@oracle.com>,
        "Mike Rapoport (IBM)" <rppt@kernel.org>,
        Muchun Song <muchun.song@linux.dev>,
        Nadav Amit <namit@vmware.com>,
        Naoya Horiguchi <naoya.horiguchi@nec.com>,
        Ryan Roberts <ryan.roberts@arm.com>,
        Shuah Khan <shuah@kernel.org>,
        Suleiman Souhlal <suleiman@google.com>,
        Suren Baghdasaryan <surenb@google.com>,
        "T.J. Alumbaugh" <talumbau@google.com>,
        Yu Zhao <yuzhao@google.com>,
        ZhangPeng <zhangpeng362@huawei.com>, linux-doc@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
        linux-mm@kvack.org, linux-kselftest@vger.kernel.org
Subject: Re: [PATCH v3 1/8] mm: make PTE_MARKER_SWAPIN_ERROR more general
Message-ID: <ZKgOxIeKN0wEmkih@x1n>
References: <20230706225037.1164380-1-axelrasmussen@google.com>
 <20230706225037.1164380-2-axelrasmussen@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20230706225037.1164380-2-axelrasmussen@google.com>
Precedence: bulk

On Thu, Jul 06, 2023 at 03:50:29PM -0700, Axel Rasmussen wrote:
> diff --git a/mm/madvise.c b/mm/madvise.c
> index 886f06066622..59e954586e2a 100644
> --- a/mm/madvise.c
> +++ b/mm/madvise.c
> @@ -660,7 +660,7 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
>  				free_swap_and_cache(entry);
>  				pte_clear_not_present_full(mm, addr, pte, tlb->fullmm);
>  			} else if (is_hwpoison_entry(entry) ||
> -				   is_swapin_error_entry(entry)) {
> +				   is_error_swp_entry(entry)) {

is_error_swp_entry() made me think whether we should call this marker as
ERROR at all - it's too generic, is_error_swp_entry() can be easily read as
"this swap entry is not legal".  Can we come up with a less generic one?

PTE_MARKER_PAGE_POISONED / PTE_MARKER_POISONED / PTE_MARKER_PAGE_ERR / ...?

We do use poisoned only in real bad physical pages before, but I think we
can still keep using it when applying it to a pte.  I think it's fine to
call a pte poisoned if it's not legal to access, just like a poisoned page.

>  				pte_clear_not_present_full(mm, addr, pte, tlb->fullmm);
>  			}
>  			continue;

The code change across the whole patch look good to me otherwise, thanks.

-- 
Peter Xu