staging: r8188eu: Fix smatch warnings in os_dep/*.c
Hello Larry and Martin,
The patch 0d5e4bfe47ea: "staging: r8188eu: Fix smatch warnings in os_dep/*.c" from Aug 11, 2021, leads to the following Smatch static checker warning:
drivers/staging/r8188eu/os_dep/usb_intf.c:286 rtw_hw_resume() warn: variable dereferenced before check 'padapter' (see line 284)
drivers/staging/r8188eu/os_dep/usb_intf.c
281 int rtw_hw_resume(struct adapter *padapter)
282 {
283 struct pwrctrl_priv *pwrpriv;
284 struct net_device *pnetdev = padapter->pnetdev;
^^^^^^^^^^^^^^^^^Dereferenced
285
--> 286 if (!padapter)
^^^^^^^^^Checked too late.
This warning was silenced when Smatch could tell that "padapter" was non-NULL but Martin removed the last caller in commit c0a099b7341c ("staging: r8188eu: bHWPwrPindetect is always false").
So the correct fix is to just delete this function?
287 goto error_exit;
288 pwrpriv = &padapter->pwrctrlpriv;
289 DBG_88E("==> rtw_hw_resume\n");
290 _enter_pwrlock(&pwrpriv->lock);
291 pwrpriv->bips_processing = true;
292 rtw_reset_drv_sw(padapter);
293 regards,
dan carpenter
Hello Dan,
Thus wrote Dan Carpenter ([email protected]):
Quoted 21 lines
Hello Larry and Martin,
The patch
0d5e4bfe47ea: "staging: r8188eu: Fix smatch warnings in os_dep/*.c" from Aug 11, 2021, leads to the following Smatch static checker warning:drivers/staging/r8188eu/os_dep/usb_intf.c:286 rtw_hw_resume() warn: variable dereferenced before check 'padapter' (see line 284)
drivers/staging/r8188eu/os_dep/usb_intf.c281 int rtw_hw_resume(struct adapter *padapter) 282 { 283 struct pwrctrl_priv *pwrpriv; 284 struct net_device *pnetdev = padapter->pnetdev; ^^^^^^^^^^^^^^^^^Dereferenced
285 --> 286 if (!padapter) ^^^^^^^^^
Quoted 7 lines
Checked too late.
This warning was silenced when Smatch could tell that "padapter" was non-NULL but Martin removed the last caller in commit
c0a099b7341c("staging: r8188eu: bHWPwrPindetect is always false").So the correct fix is to just delete this function?
yes, rtw_hw_resume is no longer used. I submitted a patch to remove the function. I'll have to rebase and resend it.
Best regards,
Martin