linux-kernel - Re: [PATCH 3/3] spi: dw: Drop default number of CS setting

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: Fri, 16 Feb 2024 19:00:28 +0200
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Serge Semin <fancer.lancer@...il.com>
Cc: Mark Brown <broonie@...nel.org>, 
	Alexey Malahov <Alexey.Malahov@...kalelectronics.ru>, linux-spi@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/3] spi: dw: Drop default number of CS setting

On Fri, Feb 16, 2024 at 5:36 PM Serge Semin <fancer.lancer@...il.com> wrote:
> On Thu, Feb 15, 2024 at 09:32:23PM +0200, Andy Shevchenko wrote:
> > On Thu, Feb 15, 2024 at 09:00:48PM +0300, Serge Semin wrote:
> > > DW APB/AHB SSI core now supports the procedure which automatically
> > > determines the number of native CS. Thus there is no longer point in
> > > defaulting to four CS if platform doesn't specify the real number.

the platform


..

> > > -   num_cs = 4;
> >
> > Simply update the default here?
> >
> > > -   device_property_read_u32(&pdev->dev, "num-cs", &num_cs);
>
> Do you suggest to simply:
>
> --- a/drivers/spi/spi-dw-mmio.c
> +++ b/drivers/spi/spi-dw-mmio.c
> @@ -364,8 +364,9 @@ static int dw_spi_mmio_probe(struct platform_device *pdev)
>                                      &dws->reg_io_width))
>                 dws->reg_io_width = 4;
>
> -       num_cs = 4;
> +       num_cs = 0;
>
>         device_property_read_u32(&pdev->dev, "num-cs", &num_cs);
>
> ?

Either this or do

num_cs = dw_spi_get_num_cs_from_hw(...);

What would work better WRT hardware?

..

> My idea was to make the statement looking closer to what is
> implemented for "reg-io-width" property. An alternative to what you
> suggest and to my patch can be converting the dw_spi::num_cs type to
> u32 and pass it to the device_property_read_u32() method directly:

..patch...

> What do you think? Would that be better?

I like the change, but again, are you sure it won't break any setups?
If yes, go for this!

-- 
With Best Regards,
Andy Shevchenko