From: William Zhang <william.zhang@broadcom.com>
To: Mark Brown <broonie@kernel.org>
Cc: Linux SPI List <linux-spi@vger.kernel.org>,
Broadcom Kernel List <bcm-kernel-feedback-list@broadcom.com>,
anand.gore@broadcom.com, tomer.yacoby@broadcom.com,
dan.beygelman@broadcom.com, joel.peshkin@broadcom.com,
f.fainelli@gmail.com, jonas.gorski@gmail.com,
kursad.oney@broadcom.com, dregan@mail.com,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 07/16] spi: bcm63xx-hsspi: Add polling mode support
Date: Wed, 11 Jan 2023 14:57:31 -0800 [thread overview]
Message-ID: <45f71d68-1cbb-a775-46f3-e07ccce4c8d8@broadcom.com> (raw)
In-Reply-To: <Y787JLl9Hdp0sWsC@sirena.org.uk>
[-- Attachment #1: Type: text/plain, Size: 1799 bytes --]
On 01/11/2023 02:41 PM, Mark Brown wrote:
> On Wed, Jan 11, 2023 at 12:13:43PM -0800, William Zhang wrote:
>
>> Thanks for the explanation. I saw the spi-uniphier.c and spi-bcm2835.c doing
>> the trick you mentioned(thanks Kursad for pointing out). In our case, even
>> the maximum fifo size usage(512bytes), the polling still have better
>> performance than interrupt. The MTD test result included in this patch is
>> based on maximum fifo usage. So there is no benefit to switch to interrupt
>> based on transfer size.
>
>> Does the spi framework has any requirement on how much time that the
>> driver's transfer_one function can spend? I can see the polling function
>> might take quite some time in busy loop if the clock is low, for example, at
>> 100Hz(slowest clock this controller can go), it takes 512x8/100Hz ~= 41ms to
>> complete. If this is something in concern, I can do the interrupt switch
>> based on a time limit value if interrupt is available.
>
> There's no fixed limit, some hardware just doesn't have DMA. Some
> doesn't even have interrupts which is even better. If there's always a
> clear benefit for using PIO then just do that, perhaps creating a sysfs
> hook to allow people to switch to DMA if they need it (rather than
> requiring people to update their DT, this is really a runtime
> performance tradeoff rather than a description of the hardware). If
> there's a point at which the performance is very similar then it's
> probably worth switching to DMA there to lower the CPU usage, but if
> it's always faster to use PIO then just defaulting to PIO seems like the
> best option.
>
Sure. I will add the sysfs option. Interrupt node is now required in dts
but by default it will still runs at polling mode until sysfs option is
set to use interrupt.
[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4212 bytes --]
next prev parent reply other threads:[~2023-01-11 22:58 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-06 20:07 [PATCH 00/16] spi: bcm63xx-hsspi: driver and doc updates William Zhang
2023-01-06 20:07 ` [PATCH 01/16] dt-bindings: spi: Convert bcm63xx-hsspi bindings to json-schema William Zhang
2023-01-07 15:18 ` Rob Herring
2023-01-07 15:32 ` Krzysztof Kozlowski
2023-01-09 7:52 ` William Zhang
2023-01-09 8:48 ` Krzysztof Kozlowski
2023-01-06 20:07 ` [PATCH 02/16] dt-bindings: spi: Add bcmbca-hsspi controller support William Zhang
2023-01-08 14:51 ` Krzysztof Kozlowski
2023-01-09 8:27 ` William Zhang
2023-01-09 8:56 ` Krzysztof Kozlowski
2023-01-09 19:13 ` William Zhang
2023-01-10 8:40 ` Krzysztof Kozlowski
2023-01-10 22:18 ` Florian Fainelli
2023-01-11 1:08 ` William Zhang
2023-01-11 9:02 ` Krzysztof Kozlowski
2023-01-11 18:04 ` William Zhang
2023-01-11 18:12 ` Krzysztof Kozlowski
2023-01-11 18:44 ` William Zhang
2023-01-12 8:21 ` Krzysztof Kozlowski
2023-01-12 19:50 ` William Zhang
2023-01-13 7:41 ` Krzysztof Kozlowski
2023-01-14 3:17 ` William Zhang
2023-01-15 14:31 ` Krzysztof Kozlowski
2023-01-11 0:59 ` William Zhang
2023-01-11 9:01 ` Krzysztof Kozlowski
2023-01-06 20:07 ` [PATCH 03/16] dt-bindings: spi: Add spi peripheral specific property William Zhang
2023-01-06 21:14 ` Mark Brown
2023-01-07 3:27 ` William Zhang
2023-01-07 15:38 ` Rob Herring
2023-01-09 8:06 ` William Zhang
2023-01-09 19:19 ` Mark Brown
2023-01-09 20:18 ` William Zhang
2023-01-10 22:01 ` Mark Brown
2023-01-11 19:48 ` William Zhang
2023-01-08 14:52 ` Krzysztof Kozlowski
2023-01-09 8:27 ` William Zhang
2023-01-06 20:07 ` [PATCH 04/16] ARM: dts: broadcom: bcmbca: Add spi controller node William Zhang
2023-01-06 20:07 ` [PATCH 05/16] arm64: " William Zhang
2023-01-06 20:07 ` [PATCH 06/16] spi: bcm63xx-hsspi: Endianness fix for ARM based SoC William Zhang
2023-01-07 7:44 ` kernel test robot
2023-01-07 21:21 ` kernel test robot
2023-01-07 21:52 ` kernel test robot
2023-01-07 23:23 ` kernel test robot
2023-01-11 6:33 ` kernel test robot
2023-01-06 20:07 ` [PATCH 07/16] spi: bcm63xx-hsspi: Add polling mode support William Zhang
2023-01-06 21:47 ` Mark Brown
2023-01-07 3:35 ` William Zhang
2023-01-09 19:06 ` Mark Brown
2023-01-09 20:10 ` William Zhang
2023-01-10 22:49 ` Mark Brown
2023-01-11 20:13 ` William Zhang
2023-01-11 22:41 ` Mark Brown
2023-01-11 22:57 ` William Zhang [this message]
2023-01-06 20:08 ` [PATCH 08/16] spi: bcm63xx-hsspi: Handle cs_change correctly William Zhang
2023-01-06 20:08 ` [PATCH 09/16] spi: bcm63xx-hsspi: Fix multi-bit mode setting William Zhang
2023-01-06 20:08 ` [PATCH 10/16] spi: bcm63xx-hsspi: Make dummy cs workaround as an option William Zhang
2023-01-12 18:08 ` Mark Brown
2023-01-18 23:09 ` William Zhang
2023-01-19 13:09 ` Mark Brown
2023-01-06 20:08 ` [PATCH 11/16] spi: bcm63xx-hsspi: Add prepend feature support William Zhang
2023-01-06 22:00 ` Mark Brown
2023-01-07 3:52 ` William Zhang
2023-01-09 19:31 ` Mark Brown
2023-01-09 20:43 ` William Zhang
2023-01-10 21:18 ` Mark Brown
2023-01-11 19:42 ` William Zhang
2023-01-12 16:57 ` Mark Brown
2023-01-06 20:08 ` [PATCH 12/16] spi: bcm63xx-hsspi: Add clock gate disable option support William Zhang
2023-01-06 20:08 ` [PATCH 13/16] spi: spi-mem: Allow controller supporting mem_ops without exec_op William Zhang
2023-01-06 20:08 ` [PATCH 14/16] spi: bcm63xx-hsspi: prepend: Disable spi mem dual io read op support William Zhang
2023-01-06 22:07 ` Mark Brown
2023-01-07 3:57 ` William Zhang
2023-01-06 20:08 ` [PATCH 15/16] spi: bcmbca-hsspi: Add driver for newer HSSPI controller William Zhang
2023-01-07 22:02 ` kernel test robot
2023-01-08 2:25 ` kernel test robot
2023-01-06 20:08 ` [PATCH 16/16] MAINTAINERS: Add entry for Broadcom Broadband SoC HS SPI drivers William Zhang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=45f71d68-1cbb-a775-46f3-e07ccce4c8d8@broadcom.com \
--to=william.zhang@broadcom.com \
--cc=anand.gore@broadcom.com \
--cc=bcm-kernel-feedback-list@broadcom.com \
--cc=broonie@kernel.org \
--cc=dan.beygelman@broadcom.com \
--cc=dregan@mail.com \
--cc=f.fainelli@gmail.com \
--cc=joel.peshkin@broadcom.com \
--cc=jonas.gorski@gmail.com \
--cc=kursad.oney@broadcom.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-spi@vger.kernel.org \
--cc=tomer.yacoby@broadcom.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).