[06/13] staging: r8188eu: make rtl8188eu_inirp_init a void function
Commit Message
rtl8188eu_inirp_init's return value is not checked by its caller. Make
rtl8188eu_inirp_init a void function.
Signed-off-by: Martin Kaiser <martin@kaiser.cx>
---
drivers/staging/r8188eu/hal/usb_halinit.c | 14 +++-----------
drivers/staging/r8188eu/include/hal_intf.h | 2 +-
2 files changed, 4 insertions(+), 12 deletions(-)
Comments
Hi Martin,
Martin Kaiser <martin@kaiser.cx> says:
> rtl8188eu_inirp_init's return value is not checked by its caller. Make
> rtl8188eu_inirp_init a void function.
>
> Signed-off-by: Martin Kaiser <martin@kaiser.cx>
> ---
> drivers/staging/r8188eu/hal/usb_halinit.c | 14 +++-----------
> drivers/staging/r8188eu/include/hal_intf.h | 2 +-
> 2 files changed, 4 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/staging/r8188eu/hal/usb_halinit.c b/drivers/staging/r8188eu/hal/usb_halinit.c
> index d28b4dc2a767..afa24a59fbb3 100644
> --- a/drivers/staging/r8188eu/hal/usb_halinit.c
> +++ b/drivers/staging/r8188eu/hal/usb_halinit.c
> @@ -851,29 +851,21 @@ u32 rtl8188eu_hal_deinit(struct adapter *Adapter)
> return _SUCCESS;
> }
>
> -unsigned int rtl8188eu_inirp_init(struct adapter *Adapter)
> +void rtl8188eu_inirp_init(struct adapter *Adapter)
Hm, shouldn't we actually check return value on caller side?
This thing is called from netdev_open and issues urbs to read data from
the device. So let's imagine that we fail on 1st iteration (for some
reason): netdev_open() says all is OK, but driver does not communicate
with the device.
Maybe these urbs are not that important, tho..
With regards,
Pavel Skripkin
Hi Pavel,
Thus wrote Pavel Skripkin (paskripkin@gmail.com):
> Martin Kaiser <martin@kaiser.cx> says:
> > rtl8188eu_inirp_init's return value is not checked by its caller. Make
> > rtl8188eu_inirp_init a void function.
> Hm, shouldn't we actually check return value on caller side?
> This thing is called from netdev_open and issues urbs to read data from the
> device. So let's imagine that we fail on 1st iteration (for some reason):
> netdev_open() says all is OK, but driver does not communicate with the
> device.
your're right. It makes sense to relay the return value to _netdev_open.
We'd have to update/remove the intf_start pointer and usb_intf_start.
I'll resend the series without this patch and submit new patches for
relaying the error code.
Thanks & best regards,
Martin
@@ -851,29 +851,21 @@ u32 rtl8188eu_hal_deinit(struct adapter *Adapter)
return _SUCCESS;
}
-unsigned int rtl8188eu_inirp_init(struct adapter *Adapter)
+void rtl8188eu_inirp_init(struct adapter *Adapter)
{
u8 i;
struct recv_buf *precvbuf;
- uint status;
struct recv_priv *precvpriv = &Adapter->recvpriv;
- status = _SUCCESS;
-
/* issue Rx irp to receive data */
precvbuf = (struct recv_buf *)precvpriv->precv_buf;
for (i = 0; i < NR_RECVBUFF; i++) {
- if (!rtw_read_port(Adapter, (unsigned char *)precvbuf)) {
- status = _FAIL;
- goto exit;
- }
+ if (!rtw_read_port(Adapter, (unsigned char *)precvbuf))
+ return;
precvbuf++;
precvpriv->free_recv_buf_queue_cnt--;
}
-
-exit:
- return status;
}
/* */
@@ -26,7 +26,7 @@ void UpdateHalRAMask8188EUsb(struct adapter *adapt, u32 mac_id, u8 rssi_level);
int rtl8188e_IOL_exec_cmds_sync(struct adapter *adapter,
struct xmit_frame *xmit_frame, u32 max_wating_ms, u32 bndy_cnt);
-unsigned int rtl8188eu_inirp_init(struct adapter *Adapter);
+void rtl8188eu_inirp_init(struct adapter *Adapter);
uint rtw_hal_init(struct adapter *padapter);
uint rtw_hal_deinit(struct adapter *padapter);