Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 31 Mar 2022 10:53:41 +0200
From:   Christoph Hellwig <hch@lst.de>
To:     Alex Sierra <alex.sierra@amd.com>
Cc:     jgg@nvidia.com, david@redhat.com, Felix.Kuehling@amd.com,
        linux-mm@kvack.org, rcampbell@nvidia.com,
        linux-ext4@vger.kernel.org, linux-xfs@vger.kernel.org,
        amd-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
        hch@lst.de, jglisse@redhat.com, apopple@nvidia.com,
        willy@infradead.org, akpm@linux-foundation.org
Subject: Re: [PATCH v2 1/3] mm: add vm_normal_lru_pages for LRU handled
 pages only
Message-ID: <20220331085341.GA22102@lst.de>
References: <20220330212537.12186-1-alex.sierra@amd.com> <20220330212537.12186-2-alex.sierra@amd.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220330212537.12186-2-alex.sierra@amd.com>
User-Agent: Mutt/1.5.17 (2007-11-01)
Precedence: bulk

> -	page = vm_normal_page(vma, addr, pte);
> +	page = vm_normal_lru_page(vma, addr, pte);

Why can't this deal with ZONE_DEVICE pages?  It certainly has
nothing do with a LRU I think.  In fact being able to have
stats that count say the number of device pages here would
probably be useful at some point.

In general I find the vm_normal_lru_page vs vm_normal_page
API highly confusing.  An explicit check for zone device pages
in the dozen or so spots that care has a much better documentation
value, especially if accompanied by comments where it isn't entirely
obvious.

>  		page = follow_page(vma, addr,
> -				FOLL_GET | FOLL_MIGRATION | FOLL_REMOTE);
> +				FOLL_GET | FOLL_MIGRATION | FOLL_REMOTE | FOLL_LRU);

Overly long line here.

> +/*
> + * NOTE: Technically this should goto check_pfn label. However, page->_mapcount
> + * is never incremented for device pages that are mmap through DAX mechanism
> + * using pmem driver mounted into ext4 filesystem. When these pages are unmap,
> + * zap_pte_range is called and vm_normal_page return a valid page with
> + * page_mapcount() = 0, before page_remove_rmap is called.
> + */

Please properly indent comments.

> + * zone, as long as the pte's are present and vm_normal_lru_page() succeeds. These
>   * pages also get pinned.

Another overly long line here.