linux-kernel - Re: [PATCH v3 12/14] tty: serial: atmel: Only divide Clock Divisor if the IP is USART

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: <701a3931-6718-48cf-5e85-03e94c8f2671@microchip.com>
Date:   Wed, 14 Sep 2022 12:17:26 +0000
From:   <Sergiu.Moga@...rochip.com>
To:     <ilpo.jarvinen@...ux.intel.com>
CC:     <lee@...nel.org>, <robh+dt@...nel.org>,
        <krzysztof.kozlowski+dt@...aro.org>, <Nicolas.Ferre@...rochip.com>,
        <alexandre.belloni@...tlin.com>, <Claudiu.Beznea@...rochip.com>,
        <radu_nicolae.pirea@....ro>, <richard.genoud@...il.com>,
        <gregkh@...uxfoundation.org>, <broonie@...nel.org>,
        <mturquette@...libre.com>, <sboyd@...nel.org>,
        <jirislaby@...nel.org>, <admin@...iphile.com>,
        <Kavyasree.Kotagiri@...rochip.com>, <Tudor.Ambarus@...rochip.com>,
        <devicetree@...r.kernel.org>,
        <linux-arm-kernel@...ts.infradead.org>,
        <linux-kernel@...r.kernel.org>, <linux-spi@...r.kernel.org>,
        <linux-serial@...r.kernel.org>, <linux-clk@...r.kernel.org>
Subject: Re: [PATCH v3 12/14] tty: serial: atmel: Only divide Clock Divisor if
 the IP is USART

On 14.09.2022 14:01, Ilpo Järvinen wrote:
> On Tue, 13 Sep 2022, Sergiu Moga wrote:
> 
>> Make sure that the driver only divides the clock divisor if the
>> IP handled at that point is USART, since UART IP's do not support
>> implicit peripheral clock division. Instead, in the case of UART,
>> go with the highest possible clock divisor.
>>
>> Signed-off-by: Sergiu Moga <sergiu.moga@...rochip.com>
>> ---
>>
>>
>> v1 -> v2:
>> - Nothing, this patch was not here before and is mainly meant as both cleanup
>> and as a way to introduce a new field into struct atmel_uart_port that will be
>> used by the last patch to diferentiate between USART and UART regarding the
>> location of the Baudrate Clock Source bitmask.
>>
>>
>>
>> v2 -> v3:
>> - Use ATMEL_US_CD instead of 65535
>> - Previously [PATCH 10]
> 
> 
>> @@ -2283,10 +2287,21 @@ static void atmel_set_termios(struct uart_port *port,
>>                cd = uart_get_divisor(port, baud);
>>        }
>>
>> -     if (cd > 65535) {       /* BRGR is 16-bit, so switch to slower clock */
>> +     /*
>> +      * If the current value of the Clock Divisor surpasses the 16 bit
>> +      * ATMEL_US_CD mask and the IP is USART, switch to the Peripheral
>> +      * Clock implicitly divided by 8.
>> +      * If the IP is UART however, keep the highest possible value for
>> +      * the CD and avoid needless division of CD, since UART IP's do not
>> +      * support implicit division of the Peripheral Clock.
>> +      */
>> +     if (atmel_port->is_usart && cd > ATMEL_US_CD) {
>>                cd /= 8;
>>                mode |= ATMEL_US_USCLKS_MCK_DIV8;
>> +     } else {
>> +             cd &= ATMEL_US_CD;
> 
> Now that I read the above comment with more thought, would:
>                  cd = min(cd, ATMEL_US_CD);
> be more appropriate here if "the highest possible value" is sought for?
> 
> 
> --
>   i.
> 



Yes, you are right, that would actually be the correct operation. I 
somehow only thought of frequencies lower than ATMEL_US_CD when writing 
that operation, although the commentary does say my intention. Thanks :).


Regards,
	Sergiu