Received-SPF: pass (google.com: domain of linux-kernel+bounces-64841-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Message-ID: <d5d71068-eff7-4a1d-ad2f-948bf0313b9c@linaro.org>
Date: Wed, 14 Feb 2024 07:42:57 +0000
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH] spi: spi-mem: add statistics support to ->exec_op() calls
Content-Language: en-US
To: Mark Brown <broonie@kernel.org>
Cc: =?UTF-8?Q?Th=C3=A9o_Lebrun?= <theo.lebrun@bootlin.com>,
 linux-spi@vger.kernel.org, linux-kernel@vger.kernel.org,
 Dhruva Gole <d-gole@ti.com>, Gregory CLEMENT <gregory.clement@bootlin.com>,
 Vladimir Kondratiev <vladimir.kondratiev@mobileye.com>,
 Thomas Petazzoni <thomas.petazzoni@bootlin.com>,
 Tawfik Bayouk <tawfik.bayouk@mobileye.com>
References: <20240209-spi-mem-stats-v1-1-dd1a422fc015@bootlin.com>
 <b0844e5a-ee4b-4608-99a1-877660e01d57@linaro.org>
 <69f023bb-2844-496f-9f3f-4e2e80129d33@sirena.org.uk>
From: Tudor Ambarus <tudor.ambarus@linaro.org>
In-Reply-To: <69f023bb-2844-496f-9f3f-4e2e80129d33@sirena.org.uk>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit


On 2/13/24 15:28, Mark Brown wrote:
> On Tue, Feb 13, 2024 at 12:39:02PM +0000, Tudor Ambarus wrote:
>> On 2/9/24 13:51, Théo Lebrun wrote:
> 
>>> +		if (!ret || ret != -ENOTSUPP || ret != -EOPNOTSUPP) {
>>> +			spi_mem_add_op_stats(ctlr->pcpu_statistics, op, ret);
>>> +			spi_mem_add_op_stats(mem->spi->pcpu_statistics, op, ret);
>>> +
> 
>> Would be good to be able to opt out the statistics if one wants it.
> 
>> SPI NORs can write with a single write op maximum page_size bytes, which
>> is typically 256 bytes. And since there are SPI NORs that can run at 400
>> MHz, I guess some performance penalty shouldn't be excluded.
> 
> If we can cope with this sort of statistics collection in the networking
> fast path we can probably cope with it for SPI too, the immediate
> recording is all per CPU so I'd like to see some numbers showing that
> it's a problem before worrying about it too much.  Even the people doing
> things like saturating CAN buses haven't been raising it as a concern
> for the regular SPI data path.  We could add a Kconfig if it's an issue.

Ok, we can deal with it afterwards if it'll become an issue.