Received-SPF: pass (google.com: domain of linux-kernel+bounces-64887-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Precedence: bulk
Mime-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset=UTF-8
Date: Wed, 14 Feb 2024 09:51:40 +0100
Message-Id: <CZ4O9QACM45B.2HA0L2O4QL5PL@bootlin.com>
Cc: <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>
To: "Tudor Ambarus" <tudor.ambarus@linaro.org>, "Mark Brown"
 <broonie@kernel.org>
From: =?utf-8?q?Th=C3=A9o_Lebrun?= <theo.lebrun@bootlin.com>
Subject: Re: [PATCH] spi: spi-mem: add statistics support to ->exec_op()
 calls
References: <20240209-spi-mem-stats-v1-1-dd1a422fc015@bootlin.com>
 <b0844e5a-ee4b-4608-99a1-877660e01d57@linaro.org>
 <CZ41HDHS7WX6.6MJL1O2PBVW1@bootlin.com>
 <7b3c08b0-80ed-4409-96d4-d55b938df6f4@linaro.org>
In-Reply-To: <7b3c08b0-80ed-4409-96d4-d55b938df6f4@linaro.org>

Hello,

On Wed Feb 14, 2024 at 9:00 AM CET, Tudor Ambarus wrote:
> On 2/13/24 15:00, Th=C3=A9o Lebrun wrote:
> > On Tue Feb 13, 2024 at 1:39 PM CET, Tudor Ambarus wrote:
> >>>  /**
> >>>   * spi_mem_exec_op() - Execute a memory operation
> >>>   * @mem: the SPI memory
> >>> @@ -339,8 +383,12 @@ int spi_mem_exec_op(struct spi_mem *mem, const s=
truct spi_mem_op *op)
> >>>  		 * read path) and expect the core to use the regular SPI
> >>>  		 * interface in other cases.
> >>>  		 */
> >>> -		if (!ret || ret !=3D -ENOTSUPP || ret !=3D -EOPNOTSUPP)
> >>> +		if (!ret || ret !=3D -ENOTSUPP || ret !=3D -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, whi=
ch
> >> is typically 256 bytes. And since there are SPI NORs that can run at 4=
00
> >> MHz, I guess some performance penalty shouldn't be excluded.
> >=20
> > I did my testing on a 40 MHz octal SPI NOR with most reads being much
> > bigger than 256 bytes, so I probably didn't have the fastest setup
> > indeed.
>
> yeah, reads are bigger, the entire flash can be read with a single read o=
p.
>
> >=20
> > What shape would that take? A spi-mem DT prop? New field in the SPI
> > statistics sysfs directory?
> >=20
>
> I think I'd go with a sysfs entry, it provides flexibility. But I guess
> we can worry about this if we have some numbers, and I don't have, so
> you're fine even without the opt-out option.

Some ftrace numbers:
- 48002 calls to spi_mem_add_op_stats();
- min 1.053000=C2=B5s;
- avg 1.175652=C2=B5s;
- max 16.272000=C2=B5s.

Platform is Mobileye EyeQ5. Cores are Imagine Technologies I6500-F. I
don't know the precision of our timer but we might be getting close to
what is measurable.

Thanks,

--
Th=C3=A9o Lebrun, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com