From mboxrd@z Thu Jan  1 00:00:00 1970
From: Thomas Chou <thomas@wytron.com.tw>
Date: Tue, 29 Sep 2015 15:18:05 +0800
Subject: [U-Boot] [PATCH v2] nios2: convert nios2 cpu to driver model
In-Reply-To: <201509290445.28107.marex@denx.de>
References: <1443400318-19151-1-git-send-email-thomas@wytron.com.tw>
 <1443490081-19526-1-git-send-email-thomas@wytron.com.tw>
 <201509290445.28107.marex@denx.de>
Message-ID: <560A3B2D.9000404@wytron.com.tw>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Hi Marek,

On 09/29/2015 10:45 AM, Marek Vasut wrote:
>> --- a/arch/Kconfig
>> +++ b/arch/Kconfig
>> @@ -64,6 +64,9 @@ config NIOS2
>>   	select HAVE_GENERIC_BOARD
>>   	select SYS_GENERIC_BOARD
>>   	select SUPPORT_OF_CONTROL
>> +	select OF_CONTROL
>> +	select DM
>> +	select CPU
>
> What's this CONFIG_CPU for please ?

drivers/cpu/Kconfig

It is "Enable CPU drivers using Driver Model". I think it should have 
been named as CONFIG_DM_CPU to avoid confusion.

>> -int arch_cpu_init(void)
>> +int arch_cpu_init_dm(void)
>>   {
>> -	gd->cpu_clk = CONFIG_SYS_CLK_FREQ;
>> +	struct udevice *dev;
>> +
>> +	uclass_first_device(UCLASS_CPU, &dev);
>> +	if (!dev)
>> +		return -ENODEV;
>> +
>> +	gd->cpu_clk = fdtdec_get_int(gd->fdt_blob, dev->of_offset,
>> +		"clock-frequency", 0);
>> +	gd->arch.dcache_line_size = fdtdec_get_int(gd->fdt_blob, dev->of_offset,
>> +		"dcache-line-size", 0);
>> +	gd->arch.icache_line_size = fdtdec_get_int(gd->fdt_blob, dev->of_offset,
>> +		"icache-line-size", 0);
>> +	gd->arch.dcache_size = fdtdec_get_int(gd->fdt_blob, dev->of_offset,
>> +		"dcache-size", 0);
>> +	gd->arch.icache_size = fdtdec_get_int(gd->fdt_blob, dev->of_offset,
>> +		"icache-size", 0);
>> +	gd->arch.reset_addr = fdtdec_get_int(gd->fdt_blob, dev->of_offset,
>> +		"altr,reset-addr", 0);
>> +	gd->arch.exception_addr = fdtdec_get_int(gd->fdt_blob, dev->of_offset,
>> +		"altr,exception-addr", 0);
>> +	gd->arch.has_mmu = fdtdec_get_int(gd->fdt_blob, dev->of_offset,
>> +		"altr,has-mmu", 0);
>
> Shouldn't there be some sort of return value checking here ?

There might be some. Though I would like to depend upon the sopc2dts 
which generates the dts. I don't think these values should be tweaked by 
a human. :)

>
>> +	info->features = 1 << CPU_FEAT_L1_CACHE
>> +		| (gd->arch.has_mmu ? 1 << CPU_FEAT_MMU : 0);
>
> I'd add parenthesis around the bitshifts, for the sake of clarity.
> Also, please put the ORR operator at the end of the line.

Will do as suggested.

Thanks a lot for your review.

Best regards,
Thomas Chou