linux-kernel - Re: [PATCH 03/16] dt-bindings: spi: Add spi peripheral specific property

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: <a92a9107-66d6-adb2-e11b-2e9bccda0c81@broadcom.com>
Date:   Wed, 11 Jan 2023 11:48:16 -0800
From:   William Zhang <william.zhang@...adcom.com>
To:     Mark Brown <broonie@...nel.org>
Cc:     Rob Herring <robh+dt@...nel.org>,
        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,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 03/16] dt-bindings: spi: Add spi peripheral specific
 property



On 01/10/2023 02:01 PM, Mark Brown wrote:
> On Mon, Jan 09, 2023 at 12:18:09PM -0800, William Zhang wrote:
> 
>> This setting is set per spi message for particular chip select of the device
>> when starting the message through bcm63xx_hsspi_set_clk function and restore
>> to default(clock gating) when message is done through
>> bcm63xx_hsspi_restore_clk_gate.
> 
> In that case I am extremely confused about what the feature is supposed
> to do.  The description says:
> 
> +  brcm,no-clk-gate:
> +    $ref: /schemas/types.yaml#/definitions/flag
> +    description:
> +      Some SPI device such as Broadcom ISI based voice daughtercard requires
> +SPI
> +      clock running even when chip select is deasserted. By default the
> +      controller turns off or gate the clock when cs is not active to save
> +      power. This flag tells the controller driver to keep the clock running
> +      when chip select is not active.
> 
> 
> which to me sounds like the clock should never be turned off and instead
> left running at all times.  Switching back to clock gating after sending
> the message doesn't seem to correspond to the above at all, the message
> being done would normally also be the point at which chip select is
> deasserted.
> 
This feature is used by our voice team and as far I can tell, it is used 
to keep clock running between the transfers within the same message. 
But now that we have prepend mode to combine to one transfer or dummy 
workaround to keep cs always active between transfers, this indeed does 
not seems right.   I will have to talk to the voice team why this is 
still needed and get back here.

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