Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 14 Apr 2021 19:51:57 +0100
From:   Matthew Wilcox <willy@infradead.org>
To:     David Hildenbrand <david@redhat.com>
Cc:     Andrew Morton <akpm@linux-foundation.org>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, Tejun Heo <tj@kernel.org>,
        FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>,
        Douglas Gilbert <dougg@torque.net>,
        Chris Wilson <chris@chris-wilson.co.uk>,
        Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH] mm: Optimise nth_page for contiguous memmap
Message-ID: <20210414185157.GU2531743@casper.infradead.org>
References: <20210413194625.1472345-1-willy@infradead.org>
 <67922c5b-0a7b-4eab-9fee-455acf555ebf@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <67922c5b-0a7b-4eab-9fee-455acf555ebf@redhat.com>
Precedence: bulk

On Wed, Apr 14, 2021 at 05:24:42PM +0200, David Hildenbrand wrote:
> On 13.04.21 21:46, Matthew Wilcox (Oracle) wrote:
> > +#if defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP)
> >   #define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))
> > +#else
> > +#define nth_page(page,n) ((page) + (n))
> > +#endif
> 
> For sparsemem we could optimize within a single memory section. But not sure
> if it's worth the trouble.

Not only is it not worth the trouble, I suspect it's more expensive to
test-and-branch than just unconditionally call pfn_to_page() and
page_to_pfn().  That said, I haven't measured.

SPARSEMEM_VMEMMAP is default Y, and enabled by arm64, ia64, powerpc,
riscv, s390, sparc and x86.  I mean ... do we care any more?

> Reviewed-by: David Hildenbrand <david@redhat.com>
> 
> -- 
> Thanks,
> 
> David / dhildenb
>