Received-SPF: pass (google.com: domain of linux-kernel+bounces-42675-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Message-ID: <df8f1bf4-819a-4b2a-927d-e97fe196cdf6@arm.com>
Date: Mon, 29 Jan 2024 12:15:07 +0000
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH net-next 1/7] dma: compile-out DMA sync op calls when not
 used
Content-Language: en-GB
To: Alexander Lobakin <aleksander.lobakin@intel.com>,
 Christoph Hellwig <hch@lst.de>
Cc: "David S. Miller" <davem@davemloft.net>,
 Eric Dumazet <edumazet@google.com>, Jakub Kicinski <kuba@kernel.org>,
 Paolo Abeni <pabeni@redhat.com>, Marek Szyprowski
 <m.szyprowski@samsung.com>, Joerg Roedel <joro@8bytes.org>,
 Will Deacon <will@kernel.org>,
 Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
 "Rafael J. Wysocki" <rafael@kernel.org>,
 Magnus Karlsson <magnus.karlsson@intel.com>,
 Maciej Fijalkowski <maciej.fijalkowski@intel.com>,
 Alexander Duyck <alexanderduyck@fb.com>, bpf@vger.kernel.org,
 netdev@vger.kernel.org, iommu@lists.linux.dev, linux-kernel@vger.kernel.org
References: <20240126135456.704351-1-aleksander.lobakin@intel.com>
 <20240126135456.704351-2-aleksander.lobakin@intel.com>
 <20240129061136.GD19258@lst.de>
 <4e23d103-ea1c-4fd3-852e-f7e2ec9170ad@intel.com>
From: Robin Murphy <robin.murphy@arm.com>
In-Reply-To: <4e23d103-ea1c-4fd3-852e-f7e2ec9170ad@intel.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit

On 2024-01-29 11:07 am, Alexander Lobakin wrote:
> From: Christoph Hellwig <hch@lst.de>
> Date: Mon, 29 Jan 2024 07:11:36 +0100
> 
>> On Fri, Jan 26, 2024 at 02:54:50PM +0100, Alexander Lobakin wrote:
>>> Some platforms do have DMA, but DMA there is always direct and coherent.
>>> Currently, even on such platforms DMA sync operations are compiled and
>>> called.
>>> Add a new hidden Kconfig symbol, DMA_NEED_SYNC, and set it only when
>>> either sync operations are needed or there is DMA ops or swiotlb
>>> enabled. Set dma_need_sync() and dma_skip_sync() (stub for now)
>>> depending on this symbol state and don't call sync ops when
>>> dma_skip_sync() is true.
>>> The change allows for future optimizations of DMA sync calls depending
>>> on compile-time or runtime conditions.
>>
>> So the idea of compiling out the calls sounds fine to me.  But what
>> is the point of the extra indirection through the __-prefixed calls?
> 
> Because dma_sync_* ops are external functions, not inlines, and in the
> next patch I'm adding a check there.
> 
>>
>> And if we need that (please document it in the commit log), please
>> make the wrappers proper inline functions and not macros.

In fact those wrappers could perhaps subsume the existing stub 
definitions, by starting with a refactor along these lines:

static inline dma_sync_x(...)
{
	if (IS_ENABLED(CONFIG_NEED_DMA_SYNC))
		__dma_sync_x(...);
}

Cheers,
Robin.