staging: rtl8188eu: Fix a sleep-in-atomic-context bug in issue_deauth_ex()

The driver may sleep with holding a spinlock.
The function call path (from bottom to top) in Linux-4.17 is:

[FUNC] msleep
drivers/staging/rtl8188eu/core/rtw_mlme_ext.c, 1536:
	msleep in issue_deauth_ex
drivers/staging/rtl8188eu/core/rtw_mlme_ext.c, 5110:
	issue_deauth_ex in disconnect_hdl
drivers/staging/rtl8188eu/core/rtw_cmd.c, 521:
	disconnect_hdl in rtw_disassoc_cmd
drivers/staging/rtl8188eu/core/rtw_ioctl_set.c, 352:
	rtw_disassoc_cmd in rtw_set_802_11_infrastructure_mode
drivers/staging/rtl8188eu/os_dep/ioctl_linux.c, 1002:
	rtw_set_802_11_infrastructure_mode in rtw_wx_set_wap
drivers/staging/rtl8188eu/os_dep/ioctl_linux.c, 988:
	spin_lock_bh in rtw_wx_set_wap

To fix this bug, msleep() is replaced with mdelay().

This bug is found by my static analysis tool DSAC.

Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Jia-Ju Bai 2018-09-15 10:57:11 +08:00 committed by Greg Kroah-Hartman
parent 7af5bd0ca4
commit 2051c6ae92

View File

@ -1513,7 +1513,7 @@ static int issue_deauth_ex(struct adapter *padapter, u8 *da,
break; break;
if (i < try_cnt && wait_ms > 0 && ret == _FAIL) if (i < try_cnt && wait_ms > 0 && ret == _FAIL)
msleep(wait_ms); mdelay(wait_ms);
} while ((i < try_cnt) && ((ret == _FAIL) || (wait_ms == 0))); } while ((i < try_cnt) && ((ret == _FAIL) || (wait_ms == 0)));
if (ret != _FAIL) { if (ret != _FAIL) {