linux-kernel - Re: [PATCH 07/16] spi: bcm63xx-hsspi: Add polling mode support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ec84b84b-41be-32ad-2e76-afac59a621d0@broadcom.com>
Date:   Fri, 6 Jan 2023 19:35:59 -0800
From:   William Zhang <william.zhang@...adcom.com>
To:     Mark Brown <broonie@...nel.org>
Cc:     Linux SPI List <linux-spi@...r.kernel.org>,
        Broadcom Kernel List <bcm-kernel-feedback-list@...adcom.com>,
        anand.gore@...adcom.com, tomer.yacoby@...adcom.com,
        dan.beygelman@...adcom.com, joel.peshkin@...adcom.com,
        f.fainelli@...il.com, jonas.gorski@...il.com,
        kursad.oney@...adcom.com, dregan@...l.com,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 07/16] spi: bcm63xx-hsspi: Add polling mode support



On 01/06/2023 01:47 PM, Mark Brown wrote:
> On Fri, Jan 06, 2023 at 12:07:59PM -0800, William Zhang wrote:
> 
>> Polling mode provides better throughput in general by avoiding the
>> interrupt overhead as the maximum data size one interrupt can handle is
>> only 512 bytes.
> 
>> When interrupt is not defined in the HSSPI dts node, driver switches to
>> polling mode for controller SPI message processing.  Also add driver
>> banner message when the driver is loaded successfully.
> 
> This should not be something the user selects via the DT, if the polling
> mode is better then the driver should just use it regardless of there
> being an interrupt wired up.  Generally there's some point at which the
> benefits of polling become minimal (and the costs more impactful) but if
> the DMA setup is as bad as it sounds then the driver should just ignore
> the interrupt.
> 
I agree but this is more for backward compatibility as the original 
driver uses interrupt to work with MIPS based SoC and I do not want to 
change that behavior.  For newer devices I added, they work in polling 
mode by default with the dts I supplied.  IMHO it is also good to have 
this fallback option to use interrupt in case user is sensitive to cpu 
usage.

Download attachment "smime.p7s" of type "application/pkcs7-signature" (4212 bytes)