On Wed, 2019-04-03 at 09:43 +0530, Siva Rebbagondla wrote:
>
> static int rsi_usb_reg_write(struct usb_device *usbdev,
> u32 reg,
> - u16 value,
> + u32 value,
> u16 len)
> {
> u8 *usb_reg_buf;
> @@ -226,17 +226,17 @@ static int rsi_usb_reg_write(struct usb_device *usbdev,
> if (!usb_reg_buf)
> return status;
>
> - usb_reg_buf[0] = (value & 0x00ff);
> - usb_reg_buf[1] = (value & 0xff00) >> 8;
> - usb_reg_buf[2] = 0x0;
> - usb_reg_buf[3] = 0x0;
> + usb_reg_buf[0] = (cpu_to_le32(value) & 0x00ff);
> + usb_reg_buf[1] = (cpu_to_le32(value) & 0xff00) >> 8;
> + usb_reg_buf[2] = (cpu_to_le32(value) & 0x00ff0000) >> 16;
> + usb_reg_buf[3] = (cpu_to_le32(value) & 0xff000000) >> 24;
The cpu_to_le32() was wrong here, it's already contained in the shifting
that happened here ...
Easier to just write that as
__le32 *usb_reg_buf;
...
usb_reg_buf[0] = cpu_to_le32(value);
however.
> status = usb_control_msg(usbdev,
> usb_sndctrlpipe(usbdev, 0),
> USB_VENDOR_REGISTER_WRITE,
> RSI_USB_REQ_OUT,
> - ((reg & 0xffff0000) >> 16),
> - (reg & 0xffff),
> + ((cpu_to_le32(reg) & 0xffff0000) >> 16),
> + (cpu_to_le32(reg) & 0xffff),
This change also seems completely wrong, and we should probably revert
it?
johannes
