[PATCH] dmi: dmi_decode: check acpi battery if type 22 is present.

[Colin Ian King]

On 31/01/12 13:43, Alex Hung wrote:
> Signed-off-by: Alex Hung<alex.hung at canonical.com>
> ---
>   src/dmi/dmi_decode/dmi_decode.c |   14 ++++++++++++--
>   1 files changed, 12 insertions(+), 2 deletions(-)
>
> diff --git a/src/dmi/dmi_decode/dmi_decode.c b/src/dmi/dmi_decode/dmi_decode.c
> index 00565d6..c922ee4 100644
> --- a/src/dmi/dmi_decode/dmi_decode.c
> +++ b/src/dmi/dmi_decode/dmi_decode.c
> @@ -49,6 +49,7 @@
>   #define DMI_ILLEGAL_MAPPED_ADDR_RANGE	"DMIIllegalMappedAddrRange"
>   #define DMI_MGMT_CTRL_HOST_TYPE		"DMIMgmtCtrlHostType"
>   #define DMI_INVALID_ENTRY_LENGTH	"DMIInvalidEntryLength"
> +#define DMI_INVALID_HARDWARE_ENTRY	"DMIInvalidHardwareEntry"
>
Once this patch is accepted please make sure to add 
DMIInvalidHardwareEntry to the table in 
https://wiki.ubuntu.com/Kernel/Reference/fwts/dmi_decode


>   #define GET_UINT16(x) (uint16_t)(*(const uint16_t *)(x))
>   #define GET_UINT32(x) (uint32_t)(*(const uint32_t *)(x))
> @@ -241,6 +242,8 @@ static void dmi_decode_entry(fwts_framework *fw,
>   	int	i;
>   	int 	len;
>   	int	failed_count = fw->minor_tests.failed;
> +	int	battery_count;
> +	int	ret;
>
>   	switch (hdr->type) {
>   		case 0: /* 7.1 */
> @@ -644,7 +647,7 @@ static void dmi_decode_entry(fwts_framework *fw,
>   					"(range allowed 0x01..0x08, 0xa0..0xa2) "
>   					"while accessing '%s', field '%s', offset 0x%2.2x",
>   					data[0x5], table, "Interface", 0x5);
> -}
> +			}
>   			break;
>
>   		case 22: /* 7.23 */
> @@ -666,8 +669,15 @@ static void dmi_decode_entry(fwts_framework *fw,
>   				break;
>   			if (data[0x09] == 0x02)
>   				dmi_str_check(fw, table, addr, "SBDS Device Chemistry", hdr, 0x14);
> -			break;
>
> +			ret = fwts_battery_get_count(fw,&battery_count);
> +			if (ret != FWTS_OK || battery_count<  1) {
> +				fwts_failed(fw, LOG_LEVEL_MEDIUM, DMI_INVALID_HARDWARE_ENTRY,
> +					"Invalid Hardware Configuration "
> +					"(no battery found at '%s') ",
> +					FWTS_PROC_ACPI_BATTERY);

Looks, OK, however, sorry to be pedantic, but informing the user at the 
battery has not been found at FWTS_PROC_ACPI_BATTERY is not entirely 
correct. This interface will be deprecated in the future (replaced by 
FWTS_SYS_CLASS_POWER_SUPPLY).  Perhaps it is best to just state: "(no 
battery found)".


> +			}
> +			break;
>   		case 23: /* 7.24 */
>   			table = "System Reset (Type 23)";
>   			if (hdr->length<  0x0D)