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]
Date: Tue, 9 Jan 2024 10:43:23 -0800
From: Bjorn Andersson <quic_bjorande@...cinc.com>
To: Zhenhua Huang <quic_zhenhuah@...cinc.com>
CC: <miquel.raynal@...tlin.com>, <richard@....at>, <vigneshr@...com>,
        <robh+dt@...nel.org>, <krzysztof.kozlowski+dt@...aro.org>,
        <conor+dt@...nel.org>, <linux-mtd@...ts.infradead.org>,
        <devicetree@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <quic_tingweiz@...cinc.com>
Subject: Re: [PATCH] dt-bindings: mtd: Change the schema for nodename which
 includes "sram"

On Thu, Jan 04, 2024 at 07:23:02PM +0800, Zhenhua Huang wrote:
> Node name which includes "sram" not only represents MTD devices, but also
> lots of sram devices(eg, qcom,imem.yaml, rules in folder sram/*).
> 
> To avoid the conflicts, change the schema as:
>  - if node name includes "sram", must select "compatible" to match
> (I have listed all "comptible" string in mtd.yaml by searching
> drivers/mtd/* to find applicable drivers)
>  - if node name is nand/flash, use "nodename" to select.
> 

You're right, it doesn't seem appropriate for the mtd binding to be
selected for nodes intended to match e.g. the qcom,imem binding.

> Fixes: 7bdc671822e9 ("dt-bindings: mtd: physmap: Reuse the generic definitions")
> Signed-off-by: Zhenhua Huang <quic_zhenhuah@...cinc.com>
> ---
> Hello,
> 
> Tested a few devicetree nodes, which confirms:
> "qcom,imem.yaml" which in sram/ not matches with mtd.yaml anymore.
> All nodes include string "sram" must have "compatible" which listed in
> mtd.yaml to be matched.
> 
> Current I just modify the rule for "sram" as it is definitely conflicting with
> rules in sram/*. I have not much backgrounds on nand/flash whether they may have
> similar conflicts.
> 
>  Documentation/devicetree/bindings/mtd/mtd.yaml | 24 +++++++++++++++++++++---
>  1 file changed, 21 insertions(+), 3 deletions(-)
> 
> diff --git a/Documentation/devicetree/bindings/mtd/mtd.yaml b/Documentation/devicetree/bindings/mtd/mtd.yaml
> index f322290..1704437 100644
> --- a/Documentation/devicetree/bindings/mtd/mtd.yaml
> +++ b/Documentation/devicetree/bindings/mtd/mtd.yaml
> @@ -10,10 +10,28 @@ maintainers:
>    - Miquel Raynal <miquel.raynal@...tlin.com>
>    - Richard Weinberger <richard@....at>
>  
> -properties:
> -  $nodename:

The mtd binding is $ref'ed by other bindings, similar to how we do with
other bindings of common properties.

So, I think the problem with mtd is that this ($nodename) turns into a
"select" automatically, which causes this binding to match nodes by
name, rather than just those $ref'ing it.


We should be able to avoid the automatically created select from the
$nodename, and rely on $ref, by just adding:

select: false

Regards,
Bjorn

> -    pattern: "^(flash|.*sram|nand)(@.*)?$"
> +select:
> +  if:
> +    properties:
> +      $nodename:
> +        pattern: "^.*sram(@.*)?$"
> +  then:
> +    properties:
> +      compatible:
> +        enum:
> +          - arm,vexpress-psram
> +          - cypress,cy7c1019dv33-10zsxi
> +          - microchip,48l640
> +          - mtd-ram
> +  else:
> +    properties:
> +      $nodename:
> +        pattern: "^(flash|nand)(@.*)?$"
>  
> +  required:
> +    - $nodename
> +
> +properties:
>    label:
>      description:
>        User-defined MTD device name. Can be used to assign user friendly
> -- 
> 2.7.4
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ