cw1200: Revert unnecessary patches that fix unreal use-after-free bugs
Subsystems (3)
PATCH 1/1 cw1200: Revert unnecessary patches that fix unreal use-after-free bugs
A previous commit 4f68ef64cd7f ("cw1200: Fix concurrency use-after-free bugs in cw1200_hw_scan()") tried to fix a seemingly use-after-free bug between cw1200_bss_info_changed() and cw1200_hw_scan(), where the former frees a sk_buff pointed to by frame.skb, and the latter accesses the sk_buff pointed to by frame.skb. However, this issue should be a false alarm because:
(1) "frame.skb" is not a shared variable between the above two functions, because "frame" is a local function variable, each of the two functions has its own local "frame" - they just happen to have the same variable name.
(2) the sk_buff(s) pointed to by these two "frame.skb" are also two different object instances, they are individually allocated by different dev_alloc_skb() within the two above functions. To free one object instance will not invalidate the access of another different one.
Based on these facts, the previous commit should be unnecessary. Moreover, it also introduced a missing unlock which was addressed in a subsequent commit 51c8d24101c7 ("cw1200: fix missing unlock on error in cw1200_hw_scan()"). Now that the original use-after-free is unreal, these two commits should be reverted. This patch performs the reversion.
--- drivers/net/wireless/st/cw1200/scan.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-)
drivers/net/wireless/st/cw1200/scan.c +7 -10
@@ -75,30 +75,27 @@ int cw1200_hw_scan(struct ieee80211_hw *hw,if (req->n_ssids >WSM_SCAN_MAX_NUM_OF_SSIDS )return -EINVAL ; -/* will be unlocked in cw1200_scan_work() */ -down (&priv->scan.lock); -mutex_lock (&priv->conf_mutex); - frame.skb =ieee80211_probereq_get (hw, priv->vif->addr,NULL ,0 , req->ie_len); -if (!frame.skb) { -mutex_unlock (&priv->conf_mutex); -up (&priv->scan.lock); +if (!frame.skb)return -ENOMEM ; - }if (req->ie_len)skb_put_data (frame.skb, req->ie, req->ie_len); +/* will be unlocked in cw1200_scan_work() */ +down (&priv->scan.lock); +mutex_lock (&priv->conf_mutex); + ret =wsm_set_template_frame (priv, &frame);if (!ret) {/* Host want to be the probe responder. */ ret =wsm_set_probe_responder (priv, true); }if (ret) { -dev_kfree_skb (frame.skb);mutex_unlock (&priv->conf_mutex);up (&priv->scan.lock); +dev_kfree_skb (frame.skb);return ret; }
@@ -120,8 +117,8 @@ int cw1200_hw_scan(struct ieee80211_hw *hw, ++priv->scan.n_ssids; } -dev_kfree_skb (frame.skb);mutex_unlock (&priv->conf_mutex); +dev_kfree_skb (frame.skb);queue_work (priv->workqueue, &priv->scan.work);return 0 ; }
Signature · 1 line
2.31.1
Thanks for fixing my previous mistake.
The patch looks good.
Quoted 75 lines · Hang Zhang
A previous commit
4f68ef64cd7f("cw1200: Fix concurrency use-after-free bugs incw1200_hw_scan()") tried to fix a seemingly use-after-free bug betweencw1200_bss_info_changed()andcw1200_hw_scan(), where the former frees ask_buffpointed to by frame.skb, and the latter accesses thesk_buffpointed to by frame.skb. However, this issue should be a false alarm because:(1) "frame.skb" is not a shared variable between the above two functions, because "frame" is a local function variable, each of the two functions has its own local "frame" - they just happen to have the same variable name.
(2) the
sk_buff(s)pointed to by these two "frame.skb" are also two different object instances, they are individually allocated by differentdev_alloc_skb()within the two above functions. To free one object instance will not invalidate the access of another different one.Based on these facts, the previous commit should be unnecessary. Moreover, it also introduced a missing unlock which was addressed in a subsequent commit
51c8d24101c7("cw1200: fix missing unlock on error incw1200_hw_scan()"). Now that the original use-after-free is unreal, these two commits should be reverted. This patch performs the reversion.Fixes 4f68ef64cd7f ("cw1200: Fix concurrency use-after-free bugs in cw1200_hw_scan()")Fixes 51c8d24101c7 ("cw1200: fix missing unlock on error in cw1200_hw_scan()")Signed-off-by Hang Zhang <[email protected]>--- drivers/net/wireless/st/cw1200/scan.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-)drivers/net/wireless/st/cw1200/scan.c +6
-9@@ -75,30 +75,27 @@ int cw1200_hw_scan(struct ieee80211_hw *hw,if (req->n_ssids >WSM_SCAN_MAX_NUM_OF_SSIDS )return -EINVAL ; -/* will be unlocked in cw1200_scan_work() */ -down (&priv->scan.lock); -mutex_lock (&priv->conf_mutex); - frame.skb =ieee80211_probereq_get (hw, priv->vif->addr,NULL ,0 , req->ie_len); -if (!frame.skb) { -mutex_unlock (&priv->conf_mutex); -up (&priv->scan.lock); +if (!frame.skb)return -ENOMEM ; - }if (req->ie_len)skb_put_data (frame.skb, req->ie, req->ie_len); +/* will be unlocked in cw1200_scan_work() */ +down (&priv->scan.lock); +mutex_lock (&priv->conf_mutex); + ret =wsm_set_template_frame (priv, &frame);if (!ret) {/* Host want to be the probe responder. */ ret =wsm_set_probe_responder (priv, true); }if (ret) { -dev_kfree_skb (frame.skb);mutex_unlock (&priv->conf_mutex);up (&priv->scan.lock); +dev_kfree_skb (frame.skb);return ret; }
drivers/net/wireless/st/cw1200/scan.c int cw1200_hw_scan(struct ieee80211_hw *hw, · line 124
Quoted 9 lines
drivers/net/wireless/st/cw1200/scan.c +1
-1@@ -120,8 +117,8 @@ int cw1200_hw_scan(struct ieee80211_hw *hw, ++priv->scan.n_ssids; } -dev_kfree_skb (frame.skb);mutex_unlock (&priv->conf_mutex); +dev_kfree_skb (frame.skb);queue_work (priv->workqueue, &priv->scan.work);return 0 ; }
Best wishes,
Jia-Ju Bai
Quoted 26 lines · Hang Zhang
A previous commit
4f68ef64cd7f("cw1200: Fix concurrency use-after-free bugs incw1200_hw_scan()") tried to fix a seemingly use-after-free bug betweencw1200_bss_info_changed()andcw1200_hw_scan(), where the former frees ask_buffpointed to by frame.skb, and the latter accesses thesk_buffpointed to by frame.skb. However, this issue should be a false alarm because:(1) "frame.skb" is not a shared variable between the above two functions, because "frame" is a local function variable, each of the two functions has its own local "frame" - they just happen to have the same variable name.
(2) the
sk_buff(s)pointed to by these two "frame.skb" are also two different object instances, they are individually allocated by differentdev_alloc_skb()within the two above functions. To free one object instance will not invalidate the access of another different one.Based on these facts, the previous commit should be unnecessary. Moreover, it also introduced a missing unlock which was addressed in a subsequent commit
51c8d24101c7("cw1200: fix missing unlock on error incw1200_hw_scan()"). Now that the original use-after-free is unreal, these two commits should be reverted. This patch performs the reversion.
Quoted 4 lines
Fixes 4f68ef64cd7f ("cw1200: Fix concurrency use-after-free bugs in cw1200_hw_scan()")Fixes 51c8d24101c7 ("cw1200: fix missing unlock on error in cw1200_hw_scan()")Signed-off-by Hang Zhang <[email protected]>Acked-by Jia-Ju Bai <[email protected]>
Patch applied to wireless-drivers-next.git, thanks.
3f60f4685699 cw1200: Revert unnecessary patches that fix unreal use-after-free bugs
Signature · 3 lines
https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/ https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches