[PATCH v2] staging: rtl8723au: Fix sparse errors in rtl8723a_cmd.c

Al Viro viro at ZenIV.linux.org.uk
Tue Oct 6 23:11:34 UTC 2015


On Tue, Oct 06, 2015 at 12:32:28AM -0400, Jacob Kiefer wrote:

>  int rtl8723a_set_rssi_cmd(struct rtw_adapter *padapter, u8 *param)
>  {
> -	*((u32 *)param) = cpu_to_le32(*((u32 *)param));
> +	__le32 leparam;
> 
> -	FillH2CCmd(padapter, RSSI_SETTING_EID, 3, param);
> +	leparam = cpu_to_le32(*((u32 *)param));
> +
> +	FillH2CCmd(padapter, RSSI_SETTING_EID, 3, (u8 *)&leparam);

Why not fill the thing we are passing already with little-endian?  There's
only one caller, after all...

>  int rtl8723a_set_raid_cmd(struct rtw_adapter *padapter, u32 mask, u8 arg)
>  {
>  	u8 buf[5];
> +	__le32 lemask;
> 
>  	memset(buf, 0, 5);
> -	mask = cpu_to_le32(mask);
> -	memcpy(buf, &mask, 4);
> +	lemask = cpu_to_le32(mask);
> +	memcpy(buf, &lemask, 4);
>  	buf[4]  = arg;
> 
>  	FillH2CCmd(padapter, MACID_CONFIG_EID, 5, buf);

Ugh...
	struct macid_config_eid {__le32 mask; u8 arg;} buf = {
		.mask = cpu_to_le32(mask), .arg = arg
	};

	FillH2CCmd(padapter, MACID_CONFIG_EID, 5, &buf);

Why bother with memcpy/memset/whatnot when all you are trying to do is to
initialize a temporary structure?  And no, it's not going to have any gaps...


More information about the devel mailing list