Re: [PATCH] ARM: tegra114: add speedo-based process identification

[Stephen Warren <swarren-3lzwWm7+Weoh9ZMKESR00Q@public.gmane.org>]

On 03/14/2013 01:40 AM, Danny Huang wrote:
> Add speedo-based process identifictaion for Tegra114.
> 
> Based on the work by:
> Alex Frid <afrid-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>

This code is surprisingly quite a bit simpler than the existing
tegra30_speedo.c. Are you sure it's complete?

> diff --git a/arch/arm/mach-tegra/fuse.c b/arch/arm/mach-tegra/fuse.c

> @@ -137,6 +138,9 @@ void tegra_init_fuse(void)
>  		tegra_fuse_spare_bit = TEGRA30_FUSE_SPARE_BIT;
>  		tegra_init_speedo_data = &tegra30_init_speedo_data;
>  		break;
> +	case TEGRA114:
> +		tegra_init_speedo_data = &tegra114_init_speedo_data;
> +		break;

Don't you need to set tegra_fuse_spare_bit there just like all the other
paths do?

> diff --git a/arch/arm/mach-tegra/tegra114_speedo.c b/arch/arm/mach-tegra/tegra114_speedo.c

> +#define CORE_PROCESS_CORNERS_NUM	2
> +#define CPU_PROCESS_CORNERS_NUM		2
> +
> +enum {
> +	THRESHOLD_INDEX_0,
> +	THRESHOLD_INDEX_1,
> +	THRESHOLD_INDEX_COUNT,
> +};
> +
> +static const u32 core_process_speedos[THRESHOLD_INDEX_COUNT]
> +	[CORE_PROCESS_CORNERS_NUM] = {
> +	{1123,     UINT_MAX},
> +	{0,        UINT_MAX},
> +};
> +
> +static const u32 cpu_process_speedos[THRESHOLD_INDEX_COUNT]
> +	[CPU_PROCESS_CORNERS_NUM] = {
> +	{1695,     UINT_MAX},
> +	{0,        UINT_MAX},
> +};

Those enums/tables are a lot smaller than Tegra30. I'm surprised if we
end up making new chips simpler rather than more complex in this area.
Are those tables complete?

> +static void rev_sku_to_speedo_ids(int rev, int sku, int *threshold)
> +{
> +	u32 tmp;
> +
> +	switch (sku) {
> +	case 0x00:
> +	case 0x10:
> +	case 0x05:
> +	case 0x06:
> +		tegra_cpu_speedo_id = 1;
> +		tegra_soc_speedo_id = 0;
> +		*threshold = THRESHOLD_INDEX_0;
> +		break;
> +
> +	case 0x03:
> +	case 0x04:
> +		tegra_cpu_speedo_id = 2;
> +		tegra_soc_speedo_id = 1;
> +		*threshold = THRESHOLD_INDEX_1;
> +		break;
> +
> +	default:
> +		pr_err("Tegra114 Unknown SKU %d\n", sku);
> +		tegra_cpu_speedo_id = 0;
> +		tegra_soc_speedo_id = 0;
> +		*threshold = THRESHOLD_INDEX_0;
> +		break;
> +	}
> +
> +	if (rev == TEGRA_REVISION_A01) {
> +		tmp = tegra_fuse_readl(0x270) << 1;
> +		tmp |= tegra_fuse_readl(0x26c);
> +		if (!tmp)
> +			tegra_cpu_speedo_id = 0;
> +	}
> +}

That's also a lot simpler than Tegra30. Are all those SKUs really valid
for all chip revisions including A01 where the 0x270/0x26c fuses are clear?

If life really is this simple, then I should be happy:-) But I just want
to check that this code really is accurate.

> +void tegra114_init_speedo_data(void)

The Tegra30 code has a couple BUILD_BUG_ON() here to ensure that the
size of the {cpu,core}_process_speedos arrays match
THRESHOLD_INDEX_COUNT. It'd be good to be consistent here.


In general, the implementation of tegra114_init_speedo_data() is quite
different from that of the existing tegra30_init_speedo_data(). It'd be
nice if they were as similar as possible in structure so they could be
easily compared. Can you take a look at the two and see if any changes
are warranted in this patch?