netdev - Re: [PATCH net-next v4 1/1] ptp: clockmatrix: support 32-bit address space

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, 17 Nov 2023 16:54:19 +0000
From: Simon Horman <horms@...nel.org>
To: Min Li <lnimi@...mail.com>
Cc: richardcochran@...il.com, lee@...nel.org, linux-kernel@...r.kernel.org,
	netdev@...r.kernel.org, Min Li <min.li.xe@...esas.com>
Subject: Re: [PATCH net-next v4 1/1] ptp: clockmatrix: support 32-bit address
 space

On Wed, Nov 15, 2023 at 10:10:53AM -0500, Min Li wrote:
> From: Min Li <min.li.xe@...esas.com>
> 
> We used to assume 0x2010xxxx address. Now that
> we need to access 0x2011xxxx address, we need
> to support read/write the whole 32-bit address space.

Hi Min Li,

I think it would be appropriate to include a patch in this series
that makes use of 0x2011xxxx addresses.

> Signed-off-by: Min Li <min.li.xe@...esas.com>
> ---
> - Drop MAX_ABS_WRITE_PHASE_PICOSECONDS advised by Rahul
> - Apply SCSR_ADDR to scrach register in idtcm_load_firmware advised by Simon
> - Apply u32 to base in idtcm_output_enable advised by Simon
> - Correct sync_ctrl0/1 parameter position for idtcm_write advised by Simon

Thanks for the updates.

>  drivers/ptp/ptp_clockmatrix.c    |  71 ++--
>  drivers/ptp/ptp_clockmatrix.h    |  32 +-
>  include/linux/mfd/idt8a340_reg.h | 542 ++++++++++++++++---------------
>  3 files changed, 331 insertions(+), 314 deletions(-)
> 
> diff --git a/drivers/ptp/ptp_clockmatrix.c b/drivers/ptp/ptp_clockmatrix.c
> index f6f9d4adce04..1d5da77502e6 100644
> --- a/drivers/ptp/ptp_clockmatrix.c
> +++ b/drivers/ptp/ptp_clockmatrix.c
> @@ -41,7 +41,7 @@ module_param(firmware, charp, 0);
>  static int _idtcm_adjfine(struct idtcm_channel *channel, long scaled_ppm);
>  
>  static inline int idtcm_read(struct idtcm *idtcm,
> -			     u16 module,
> +			     u32 module,
>  			     u16 regaddr,
>  			     u8 *buf,
>  			     u16 count)

...

> @@ -570,27 +571,27 @@ static int _sync_pll_output(struct idtcm *idtcm,
>  	if (qn_plus_1)
>  		val |= SYNCTRL1_Q1_DIV_SYNC_TRIG;
>  
> -	err = idtcm_write(idtcm, 0, sync_ctrl1, &val, sizeof(val));
> +	err = idtcm_write(idtcm, sync_ctrl1, 0, &val, sizeof(val));
>  	if (err)
>  		return err;
>  
>  	/* PLL5 can have OUT8 as second additional output. */
>  	if (pll == 5 && qn_plus_1 != 0) {
> -		err = idtcm_read(idtcm, 0, HW_Q8_CTRL_SPARE,
> +		err = idtcm_read(idtcm, HW_Q8_CTRL_SPARE, 0,
>  				 &temp, sizeof(temp));

I feel that I am missing something obvious, but I have a question which I
would like to ask by way of an example.

Both before and after this patch idtcm_read() looks like this:

static inline int idtcm_read(struct idtcm *idtcm,
			     u16 module,
			     u16 regaddr,
			     u8 *buf,
			     u16 count) {
        return regmap_bulk_read(idtcm->regmap, module + regaddr, buf, count);
}

And so before this patch the above call to idtcm_read() ends up as
a call to regmap_bulk_read:

	regmap_bulk_read(idtcm->regmap, 0 + 0xa7d4, buf, count);

In particular, the 2nd argument is 0 + 0xa7d4 = 0xa7d4.

But after this patch the call to idtcm_read() becomes:

	regmap_bulk_read(idtcm->regmap,  0x2010a7d4 + 0, buf, count);

In particular, the 2nd argument is now 0x2010a7d4 + 0 = 0x2010a7d4.

My question is, how does this patch take into account the
change in this value from 0xa7d4 to 0x2010a7d4?

Or to the point, does the call to regmap_bulk_read() still
work with the new value?