ACK: [PATCH V2 1/3] acpi: madt: Use 20 as the GIC ITS node length

[Colin Ian King]

On 21/07/16 20:58, Jeffrey Hugo wrote:
> The ACPI 6.0 spec technically defines the GIC ITS node length as 16,
> however the conbination of all defined fields is 20 bytes, so the length is
> really 20.  The spec defined length is corrected in following revisions of
> the spec.
> 
> 20 is what ACPI implementations are expected to be using in practice, so
> updating the test to use 20 despite the spec definition has the test match
> the real world, updates the test implementation to match the test
> documentation, and allows the test to correctly find the start of any nodes
> following a GIC ITS node in the MADT table.
> 
> Signed-off-by: Jeffrey Hugo <jhugo at codeaurora.org>
> ---
>  src/acpi/madt/madt.c | 10 +++++++++-
>  1 file changed, 9 insertions(+), 1 deletion(-)
> 
> diff --git a/src/acpi/madt/madt.c b/src/acpi/madt/madt.c
> index e260222..141405d 100644
> --- a/src/acpi/madt/madt.c
> +++ b/src/acpi/madt/madt.c
> @@ -183,7 +183,15 @@ static struct acpi_madt_subtable_lengths spec_info[] = {
>  		.madt_version = 3,
>  		.num_types = 16,
>  		.lengths = { 8, 12, 10, 8, 6, 12, 16, SUBTABLE_VARIABLE,
> -			     16, 16, 12, 80, 24, 24, 16, 16 }
> +			     16, 16, 12, 80, 24, 24, 16, 20 }
> +		/*
> +		 * The spec technically defines the GIC ITS node length to be
> +		 * 16, but this is a clear mistake as the fields add up to be
> +		 * length 20, and the length was corrected in the next revision.
> +		 * Test for 20, as that is what is expected to be used in real,
> +		 * implementations and using length 20 ensures that the test can
> +		 * correctly find the start of any following nodes.
> +		 */
>  	},
>  	{ /* terminator */
>  		.major_version = 0,
> 
Thanks Jeffrey

Acked-by: Colin Ian King <colin.king at canonical.com>