wifi: rsi: fix endian conversions

Commit Message

Johannes Berg Feb. 23, 2024, 10:40 a.m. UTC

From: Johannes Berg <[e-mail]>

This really seems like a bug, endian conversions now happen
twice in this code. Remove one conversion to make sparse no
longer warn, not sure anyone can even test this on big endian
platforms.

Fixes: 0a60014b76f5 ("rsi: miscallaneous changes for 9116 and common")
Signed-off-by: Johannes Berg <[e-mail]>
---
 drivers/net/wireless/rsi/rsi_91x_usb.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

Patch

diff --git a/drivers/net/wireless/rsi/rsi_91x_usb.c b/drivers/net/wireless/rsi/rsi_91x_usb.c
index 10a465686439..0ce8c9aad1f1 100644
--- a/drivers/net/wireless/rsi/rsi_91x_usb.c
+++ b/drivers/net/wireless/rsi/rsi_91x_usb.c
@@ -222,7 +222,7 @@  static int rsi_usb_reg_write(struct usb_device *usbdev,
 			     u32 value,
 			     u16 len)
 {
-	u8 *usb_reg_buf;
+	__le32 *usb_reg_buf;
 	int status = -ENOMEM;
 
 	if (len > RSI_USB_CTRL_BUF_SIZE)
@@ -232,17 +232,14 @@  static int rsi_usb_reg_write(struct usb_device *usbdev,
 	if (!usb_reg_buf)
 		return status;
 
-	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;
+	usb_reg_buf[0] = cpu_to_le32(value);
 
 	status = usb_control_msg(usbdev,
 				 usb_sndctrlpipe(usbdev, 0),
 				 USB_VENDOR_REGISTER_WRITE,
 				 RSI_USB_REQ_OUT,
-				 ((cpu_to_le32(reg) & 0xffff0000) >> 16),
-				 (cpu_to_le32(reg) & 0xffff),
+				 upper_16_bits(reg),
+				 lower_16_bits(reg),
 				 (void *)usb_reg_buf,
 				 len,
 				 USB_CTRL_SET_TIMEOUT);