linux-kernel - Re: [PATCH] staging: pi433: remove rf69_get

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: <20220228063238.GA2794@kadam>
Date:   Mon, 28 Feb 2022 09:32:38 +0300
From:   Dan Carpenter <dan.carpenter@...cle.com>
To:     Paulo Miguel Almeida <paulo.miguel.almeida.rodenas@...il.com>
Cc:     gregkh@...uxfoundation.org, realwakka@...il.com,
        linux-staging@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] staging: pi433: remove rf69_get_flag function resolving
 enum conflict

Looks good.

Reviewed-by: Dan Carpenter <dan.carpenter@...cle.com>

On Sat, Feb 26, 2022 at 11:40:33AM +1300, Paulo Miguel Almeida wrote:
> The reason why rf69_get_flag() existed was to provide a high-level way
> to obtain values out of 1 (of 2) flags registers using bit masking. The
> idea was to map the possible flag values found in the data sheet like
> shown in page 70 of the RFM69HCW datasheet.
> 
> However, due to the fact that enums values in C must be unique, there
> was a naming conflict on 'fifo_not_empty' which is used by the
> tx_start_condition enum. So the author decided to create a 'fifo_empty'
> one which would negate the value that comes from the flag register as
> the solution to that conflict (which is very confusing).
> 
> this patch removes rf69_get_flag function which subsequently solves the
> enum redeclaration problem so kernel developers can follow the data
> sheet more easily.
> 
> Signed-off-by: Paulo Miguel Almeida <paulo.miguel.almeida.rodenas@...il.com>
> ---
>  drivers/staging/pi433/pi433_if.c  |  8 +++---
>  drivers/staging/pi433/rf69.c      | 44 -------------------------------
>  drivers/staging/pi433/rf69.h      |  1 -
>  drivers/staging/pi433/rf69_enum.h | 20 --------------
>  4 files changed, 4 insertions(+), 69 deletions(-)

You don't really need to write a long commit message for a commit which
deletes 69 - 4 = 65 lines.  Just say "Remove pointless rf69_get_flag()
function and call rf69_read_reg() directly.  This cleanup removes 65
lines of code and it more obvious to read."

> 
> diff --git a/drivers/staging/pi433/pi433_if.c b/drivers/staging/pi433/pi433_if.c
> index 069255f023c8..3f3e863e6cc8 100644
> --- a/drivers/staging/pi433/pi433_if.c
> +++ b/drivers/staging/pi433/pi433_if.c
> @@ -434,7 +434,7 @@ static int pi433_receive(void *data)
>  		return retval;
>  
>  	/* now check RSSI, if low wait for getting high (RSSI interrupt) */
> -	while (!rf69_get_flag(dev->spi, rssi_exceeded_threshold)) {
> +	while (!(rf69_read_reg(spi, REG_IRQFLAGS1) & MASK_IRQFLAGS1_RSSI)) {
>  		/* allow tx to interrupt us while waiting for high RSSI */
>  		dev->interrupt_rx_allowed = true;
>  		wake_up_interruptible(&dev->tx_wait_queue);
> @@ -442,8 +442,8 @@ static int pi433_receive(void *data)
>  		/* wait for RSSI level to become high */
>  		dev_dbg(dev->dev, "rx: going to wait for high RSSI level\n");
>  		retval = wait_event_interruptible(dev->rx_wait_queue,
> -						  rf69_get_flag(dev->spi,
> -								rssi_exceeded_threshold));
> +						  rf69_read_reg(spi, REG_IRQFLAGS1)
> +						  & MASK_IRQFLAGS1_RSSI);

The & character should go on the first line.

						  rf69_read_reg(spi, REG_IRQFLAGS1) &
						  MASK_IRQFLAGS1_RSSI);

But that can be done in a follow on patch if you want.  Or you can
leave it as-is.

regards,
dan carpenter