Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 3 Jul 2018 10:48:16 +0800
From:   Jisheng Zhang <Jisheng.Zhang@synaptics.com>
To:     Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Jiri Slaby <jslaby@suse.com>, <linux-serial@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>,
        <linux-arm-kernel@lists.infradead.org>
Subject: Re: [PATCH 0/2] serial: 8250_dw: add fractional divisor support
Message-ID: <20180703104816.031abbfc@xhacker.debian>
In-Reply-To: <20180703102257.678db6a8@xhacker.debian>
References: <20180702180428.331b36c5@xhacker.debian>
        <fef0bf484ead5e0998666168ced86438bf935ce5.camel@linux.intel.com>
        <59b14f97234271a7d859fa07f27bd66ec252ccc9.camel@linux.intel.com>
        <20180703102257.678db6a8@xhacker.debian>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Received-SPF: None (protection.outlook.com: synaptics.com does not designate
 permitted sender hosts)
SpamDiagnosticOutput: 1:99
SpamDiagnosticMetadata: NSPM
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 03 Jul 2018 02:50:07.2988 (UTC)
X-MS-Exchange-CrossTenant-Network-Message-Id: d2d5317a-d318-43ac-5ca1-08d5e08fb098
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 335d1fbc-2124-4173-9863-17e7051a2a0e
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM5PR03MB2634
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Tue, 3 Jul 2018 10:22:57 +0800 Jisheng Zhang wrote:

> Hi,
> 
> On Mon, 2 Jul 2018 14:51:03 +0300 Andy Shevchenko wrote:
> 
> > On Mon, 2018-07-02 at 13:18 +0300, Andy Shevchenko wrote:  
> > > On Mon, 2018-07-02 at 18:04 +0800, Jisheng Zhang wrote:    
> > > > For Synopsys DesignWare 8250 uart which version >= 4.00a, there's a
> > > > valid divisor latch fraction register. The fractional divisor width
> > > > is
> > > > 4bits ~ 6bits.
> > > >     
> > > 
> > > There are several serial IPs that have fractional divider built-in.
> > > None
> > > is using any specific hooks. Why do you need in your case, esp. taking
> > > into consideration that we have a custom ->set_termios() callback?    
> > 
> > Okay, I see that in 8250 we have hooks embedded into 8250_port.c which
> > is not the best solution.
> > 
> > For example it prevents better splitting Exar code.
> > So, we would need these hooks, but better to integrate them in the same
> > way like it's done for the rest of 8250 ones, i.e.
> > - rename existing to have a "do" word
> > - create new functions which would be a replacement that choose between
> > "do" variant and custom one
> > - not sure if we need to export "do" variants (at least for now)  
> 
> So you mean add the support as following:
> 
> 1.rename current serial8250_set_divisor as serial8250_do_set_divisor
> 
> 2.add a new serial8250_set_divisor which will be as simple as
> 
> static void serial8250_set_divisor(struct uart_port *port, ...)
> {
> 	struct uart_8250_port *up = up_to_u8250p(port);
> 
> 	if (up->set_divisor)
> 		up->set_divisor(...);
> 	else
> 		serial8250_do_set_divisor(...);
> }
> 
> could you please confirm?
> 
> Another issue is I'm not sure which struct to add the hook,
> struct uart_port or struct uart_8250_port? Currently, it seems that only
> uart_8250_port needs this hook, sure, adding the hook to struct uart_port
> is fine either. Could you please kindly give some suggestions?

patching struct uart_port seems a bit overhead. After reading the code
again, I propose another solution, similar as what dl_write() is used in
8250 core:

1.introduce the hook to struct uart_8250_port as my previous patches do, 

2.rename current serial8250_set_divisor() as default_serial_get_divisor()
then introduce a new serial8250_set_divisor() as:
static inline void serial8250_set_divisor(struct uart_8250_port *up,....)
{
	up->set_divisor();
}

and point up->set-divisor to default_serial_get_divisor

what do you think about this solution?

Thanks

> 
> Thanks,
> Jisheng