mwifiex: fix a crash in extended scan event processing
[113.967694] Unable to handle kernel NULL pointer dereference at virtual address 00000020 ............ [113.967859] PC is at mwifiex_update_rxreor_flags+0xfc/0x430 ............ [113.968110] mwifiex_update_rxreor_flags+0xfc/0x430 [113.968129] mwifiex_handle_event_ext_scan_report+0x1e4/0x21c [113.968148] mwifiex_process_sta_event+0x410/0x508 [113.968165] mwifiex_process_event+0x184/0x1e0 [113.968181] mwifiex_main_process+0x220/0x48c [113.968197] mwifiex_sdio_interrupt+0xc8/0x1cc [113.968210] sdio_irq_thread+0x11c/0x290 In case of legacy scan, adapter->curr_cmd is guranteed to be non-NULL in check_next_scan_cmd. This may not be case in extended scan where scan command response would come earlier and set curr_cmd to NULL. Extended scan event comes later and while trying to complete IOCTL for scan, driver would crash in dereferencing adapter->curr_cmd->wait_q_enabled. Avoid this by completing IOCTL in case of legacy scans only. Internal scan would be completed while handling extended scan command response. Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Showing
Please register or sign in to comment