[PATCH 4.20] ath10k: don't assume 'vif' is non-NULL in flush()

[PATCH 4.20] ath10k: don't assume 'vif' is non-NULL in flush()

[Brian Norris]

mac80211 may call us with vif == NULL, if the station is not currently
active (e.g., not associated). It is trivially easy to reproduce a crash
by suspending the system when not connected to an AP:

[   65.533934] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
...
[   65.574521] pc : ath10k_flush+0x30/0xd0 [ath10k_core]
[   65.574538] lr : __ieee80211_flush_queues+0x180/0x244 [mac80211]
[   65.599680] Process kworker/u12:1 (pid: 57, stack limit = 0x(____ptrval____))
[   65.599682] Call trace:
[   65.599695]  ath10k_flush+0x30/0xd0 [ath10k_core]
[   65.642064]  __ieee80211_flush_queues+0x180/0x244 [mac80211]
[   65.642079]  ieee80211_flush_queues+0x34/0x40 [mac80211]
[   65.642095]  __ieee80211_suspend+0xfc/0x47c [mac80211]
[   65.658611]  ieee80211_suspend+0x30/0x3c [mac80211]
[   65.658627]  wiphy_suspend+0x15c/0x3a8 [cfg80211]
[   65.672810]  dpm_run_callback+0xf0/0x1f0
[   65.672814]  __device_suspend+0x3ac/0x4f8
[   65.672819]  async_suspend+0x34/0xbc
[   65.684096]  async_run_entry_fn+0x54/0x104
[   65.684099]  worker_thread+0x4cc/0x72c
[   65.684102]  kthread+0x134/0x13c
[   65.684105]  ret_from_fork+0x10/0x18

Fixes: 9de4162f0999 ("ath10k: add peer flush in ath10k_flush for STATION")
Signed-off-by: Brian Norris <briannorris@chromium.org>
---
 drivers/net/wireless/ath/ath10k/mac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index a1c2801ded10..7e49342bae38 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -6867,7 +6867,7 @@ static void ath10k_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	u32 bitmap;
 
 	if (drop) {
-		if (vif->type == NL80211_IFTYPE_STATION) {
+		if (vif && vif->type == NL80211_IFTYPE_STATION) {
 			bitmap = ~(1 << WMI_MGMT_TID);
 			list_for_each_entry(arvif, &ar->arvifs, list) {
 				if (arvif->vdev_type == WMI_VDEV_TYPE_STA)
-- 
2.19.1.930.g4563a0d9d0-goog

Re: [PATCH 4.20] ath10k: don't assume 'vif' is non-NULL in flush()

[Kalle Valo]

Brian Norris <briannorris@chromium.org> wrote:

> mac80211 may call us with vif == NULL, if the station is not currently
> active (e.g., not associated). It is trivially easy to reproduce a crash
> by suspending the system when not connected to an AP:
> 
> [   65.533934] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
> ...
> [   65.574521] pc : ath10k_flush+0x30/0xd0 [ath10k_core]
> [   65.574538] lr : __ieee80211_flush_queues+0x180/0x244 [mac80211]
> [   65.599680] Process kworker/u12:1 (pid: 57, stack limit = 0x(____ptrval____))
> [   65.599682] Call trace:
> [   65.599695]  ath10k_flush+0x30/0xd0 [ath10k_core]
> [   65.642064]  __ieee80211_flush_queues+0x180/0x244 [mac80211]
> [   65.642079]  ieee80211_flush_queues+0x34/0x40 [mac80211]
> [   65.642095]  __ieee80211_suspend+0xfc/0x47c [mac80211]
> [   65.658611]  ieee80211_suspend+0x30/0x3c [mac80211]
> [   65.658627]  wiphy_suspend+0x15c/0x3a8 [cfg80211]
> [   65.672810]  dpm_run_callback+0xf0/0x1f0
> [   65.672814]  __device_suspend+0x3ac/0x4f8
> [   65.672819]  async_suspend+0x34/0xbc
> [   65.684096]  async_run_entry_fn+0x54/0x104
> [   65.684099]  worker_thread+0x4cc/0x72c
> [   65.684102]  kthread+0x134/0x13c
> [   65.684105]  ret_from_fork+0x10/0x18
> 
> Fixes: 9de4162f0999 ("ath10k: add peer flush in ath10k_flush for STATION")
> Signed-off-by: Brian Norris <briannorris@chromium.org>

Patch applied to wireless-drivers.git, thanks.

d987f783a480 ath10k: don't assume 'vif' is non-NULL in flush()

-- 
https://patchwork.kernel.org/patch/10673301/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches