Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 26 Jul 2018 11:17:32 +0200
From:   Christoph Hellwig <hch@lst.de>
To:     Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Cc:     linux-snps-arc@lists.infradead.org, linux-kernel@vger.kernel.org,
        linux-arch@vger.kernel.org,
        Vineet Gupta <Vineet.Gupta1@synopsys.com>,
        Alexey Brodkin <Alexey.Brodkin@synopsys.com>, hch@lst.de
Subject: Re: [PATCH 3/4] ARC: refactor arch/arc/mm/dma.c
Message-ID: <20180726091731.GC24209@lst.de>
References: <20180724101001.31965-1-Eugeniy.Paltsev@synopsys.com> <20180724101001.31965-4-Eugeniy.Paltsev@synopsys.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180724101001.31965-4-Eugeniy.Paltsev@synopsys.com>
User-Agent: Mutt/1.5.17 (2007-11-01)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Jul 24, 2018 at 01:10:00PM +0300, Eugeniy Paltsev wrote:
> Refactoring, no functional change intended.

Might be worth explaining a bit why you are refactoring it (i.e. what is the
benefit).

> 
> Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
> ---
>  arch/arc/mm/dma.c | 28 ++++++++++------------------
>  1 file changed, 10 insertions(+), 18 deletions(-)
> 
> diff --git a/arch/arc/mm/dma.c b/arch/arc/mm/dma.c
> index b693818cd8e5..46584c7c2858 100644
> --- a/arch/arc/mm/dma.c
> +++ b/arch/arc/mm/dma.c
> @@ -27,30 +27,24 @@ void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
>  	struct page *page;
>  	phys_addr_t paddr;
>  	void *kvaddr;
> -	int need_coh = 1, need_kvaddr = 0;
> +	bool need_coh = !(attrs & DMA_ATTR_NON_CONSISTENT);
>  
>  	page = alloc_pages(gfp, order);
>  	if (!page)
>  		return NULL;
>  
>  	/* This is linear addr (0x8000_0000 based) */
>  	paddr = page_to_phys(page);
>  
>  	*dma_handle = paddr;
>  
> +	/*
> +	 * - A coherent buffer needs MMU mapping to enforce non-cachability
> +	 * - A highmem page needs a virtual handle (hence MMU mapping)
> +	 *   independent of cachability.
> +	 * kvaddr is kernel Virtual address (0x7000_0000 based)
> +	 */
> +	if (PageHighMem(page) || need_coh) {

dma_alloc_attrs clears __GFP_HIGHMEM from the passed in gfp mask, so
you'll never get a highmem page here.

That also means you can merge this conditional with the one for the cache
writeback and invalidation and kill the need_coh flag entirely.

>  		kvaddr = ioremap_nocache(paddr, size);
>  		if (kvaddr == NULL) {
>  			__free_pages(page, order);
> @@ -81,11 +75,9 @@ void arch_dma_free(struct device *dev, size_t size, void *vaddr,
>  {
>  	phys_addr_t paddr = dma_handle;
>  	struct page *page = virt_to_page(paddr);
> -	int is_non_coh = 1;
> -
> -	is_non_coh = (attrs & DMA_ATTR_NON_CONSISTENT);
> +	bool is_coh = !(attrs & DMA_ATTR_NON_CONSISTENT);
>  
> -	if (PageHighMem(page) || !is_non_coh)
> +	if (PageHighMem(page) || is_coh)
>  		iounmap((void __force __iomem *)vaddr);
>  

Same here.

Also if you clean this up it would be great to take the per-device pfn offset
into account, even if that isn't used anywhere on arc yet, that is call
phys_to_dma and dma_to_phys to convert to an from the dma address.