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;
Message-ID: <269e4c6e-d6ee-bace-9fab-a9dcb4268d5a@redhat.com>
Date:   Wed, 29 Jun 2022 11:59:26 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.9.0
Content-Language: en-US
To:     Alex Sierra <alex.sierra@amd.com>, jgg@nvidia.com
Cc:     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
References: <20220629035426.20013-1-alex.sierra@amd.com>
 <20220629035426.20013-4-alex.sierra@amd.com>
From:   David Hildenbrand <david@redhat.com>
Organization: Red Hat
Subject: Re: [PATCH v7 03/14] mm: handling Non-LRU pages returned by
 vm_normal_pages
In-Reply-To: <20220629035426.20013-4-alex.sierra@amd.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 29.06.22 05:54, Alex Sierra wrote:
> With DEVICE_COHERENT, we'll soon have vm_normal_pages() return
> device-managed anonymous pages that are not LRU pages. Although they
> behave like normal pages for purposes of mapping in CPU page, and for
> COW. They do not support LRU lists, NUMA migration or THP.
> 
> Callers to follow_page that expect LRU pages, are also checked for
> device zone pages due to DEVICE_COHERENT type.

Can we rephrase that to (because zeropage)

"Callers to follow_page() currently don't expect ZONE_DEVICE pages,
however, with DEVICE_COHERENT we might now return ZONE_DEVICE. Check for
ZONE_DEVICE pages in applicable users of follow_page() as well."


[...]

>  		/*
> diff --git a/mm/memory.c b/mm/memory.c
> index 7a089145cad4..e18555af9024 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -624,6 +624,13 @@ struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr,
>  		if (is_zero_pfn(pfn))
>  			return NULL;
>  		if (pte_devmap(pte))
> +/*
> + * NOTE: New uers of ZONE_DEVICE will not set pte_devmap() and will have

s/uers/users/

> + * refcounts incremented on their struct pages when they are inserted into
> + * PTEs, thus they are safe to return here. Legacy ZONE_DEVICE pages that set
> + * pte_devmap() do not have refcounts. Example of legacy ZONE_DEVICE is
> + * MEMORY_DEVICE_FS_DAX type in pmem or virtio_fs drivers.
> + */

[...]

> diff --git a/mm/mprotect.c b/mm/mprotect.c
> index ba5592655ee3..e034aae2a98b 100644
> --- a/mm/mprotect.c
> +++ b/mm/mprotect.c
> @@ -95,7 +95,7 @@ static unsigned long change_pte_range(struct mmu_gather *tlb,
>  					continue;
>  
>  				page = vm_normal_page(vma, addr, oldpte);
> -				if (!page || PageKsm(page))
> +				if (!page || is_zone_device_page(page) || PageKsm(page))
>  					continue;
>  
>  				/* Also skip shared copy-on-write pages */

In -next/-mm there is now an additional can_change_pte_writable() that
calls vm_normal_page() --  added by me. I assume that that is indeed
fine because we can simply map device coherent pages writable.

Besides the nits, LGTM

Acked-by: David Hildenbrand <david@redhat.com>

-- 
Thanks,

David / dhildenb