From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Sumit Garg <sumit.garg@linaro.org>,
Johannes Berg <johannes.berg@intel.com>
Subject: [PATCH 5.6 53/71] mac80211: fix race in ieee80211_register_hw()
Date: Mon, 20 Apr 2020 14:39:07 +0200 [thread overview]
Message-ID: <20200420121519.805021247@linuxfoundation.org> (raw)
In-Reply-To: <20200420121508.491252919@linuxfoundation.org>
From: Sumit Garg <sumit.garg@linaro.org>
commit 52e04b4ce5d03775b6a78f3ed1097480faacc9fd upstream.
A race condition leading to a kernel crash is observed during invocation
of ieee80211_register_hw() on a dragonboard410c device having wcn36xx
driver built as a loadable module along with a wifi manager in user-space
waiting for a wifi device (wlanX) to be active.
Sequence diagram for a particular kernel crash scenario:
user-space ieee80211_register_hw() ieee80211_tasklet_handler()
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
| | |
|<---phy0----wiphy_register() |
|-----iwd if_add---->| |
| |<---IRQ----(RX packet)
| Kernel crash |
| due to unallocated |
| workqueue. |
| | |
| alloc_ordered_workqueue() |
| | |
| Misc wiphy init. |
| | |
| ieee80211_if_add() |
| | |
As evident from above sequence diagram, this race condition isn't specific
to a particular wifi driver but rather the initialization sequence in
ieee80211_register_hw() needs to be fixed. So re-order the initialization
sequence and the updated sequence diagram would look like:
user-space ieee80211_register_hw() ieee80211_tasklet_handler()
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
| | |
| alloc_ordered_workqueue() |
| | |
| Misc wiphy init. |
| | |
|<---phy0----wiphy_register() |
|-----iwd if_add---->| |
| |<---IRQ----(RX packet)
| | |
| ieee80211_if_add() |
| | |
Cc: stable@vger.kernel.org
Signed-off-by: Sumit Garg <sumit.garg@linaro.org>
Link: https://lore.kernel.org/r/1586254255-28713-1-git-send-email-sumit.garg@linaro.org
[Johannes: fix rtnl imbalances]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/mac80211/main.c | 24 +++++++++++++-----------
1 file changed, 13 insertions(+), 11 deletions(-)
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1051,7 +1051,7 @@ int ieee80211_register_hw(struct ieee802
local->hw.wiphy->signal_type = CFG80211_SIGNAL_TYPE_UNSPEC;
if (hw->max_signal <= 0) {
result = -EINVAL;
- goto fail_wiphy_register;
+ goto fail_workqueue;
}
}
@@ -1113,7 +1113,7 @@ int ieee80211_register_hw(struct ieee802
result = ieee80211_init_cipher_suites(local);
if (result < 0)
- goto fail_wiphy_register;
+ goto fail_workqueue;
if (!local->ops->remain_on_channel)
local->hw.wiphy->max_remain_on_channel_duration = 5000;
@@ -1139,10 +1139,6 @@ int ieee80211_register_hw(struct ieee802
local->hw.wiphy->max_num_csa_counters = IEEE80211_MAX_CSA_COUNTERS_NUM;
- result = wiphy_register(local->hw.wiphy);
- if (result < 0)
- goto fail_wiphy_register;
-
/*
* We use the number of queues for feature tests (QoS, HT) internally
* so restrict them appropriately.
@@ -1198,9 +1194,9 @@ int ieee80211_register_hw(struct ieee802
goto fail_flows;
rtnl_lock();
-
result = ieee80211_init_rate_ctrl_alg(local,
hw->rate_control_algorithm);
+ rtnl_unlock();
if (result < 0) {
wiphy_debug(local->hw.wiphy,
"Failed to initialize rate control algorithm\n");
@@ -1254,6 +1250,12 @@ int ieee80211_register_hw(struct ieee802
local->sband_allocated |= BIT(band);
}
+ result = wiphy_register(local->hw.wiphy);
+ if (result < 0)
+ goto fail_wiphy_register;
+
+ rtnl_lock();
+
/* add one default STA interface if supported */
if (local->hw.wiphy->interface_modes & BIT(NL80211_IFTYPE_STATION) &&
!ieee80211_hw_check(hw, NO_AUTO_VIF)) {
@@ -1293,17 +1295,17 @@ int ieee80211_register_hw(struct ieee802
#if defined(CONFIG_INET) || defined(CONFIG_IPV6)
fail_ifa:
#endif
+ wiphy_unregister(local->hw.wiphy);
+ fail_wiphy_register:
rtnl_lock();
rate_control_deinitialize(local);
ieee80211_remove_interfaces(local);
- fail_rate:
rtnl_unlock();
+ fail_rate:
fail_flows:
ieee80211_led_exit(local);
destroy_workqueue(local->workqueue);
fail_workqueue:
- wiphy_unregister(local->hw.wiphy);
- fail_wiphy_register:
if (local->wiphy_ciphers_allocated)
kfree(local->hw.wiphy->cipher_suites);
kfree(local->int_scan_req);
@@ -1353,8 +1355,8 @@ void ieee80211_unregister_hw(struct ieee
skb_queue_purge(&local->skb_queue_unreliable);
skb_queue_purge(&local->skb_queue_tdls_chsw);
- destroy_workqueue(local->workqueue);
wiphy_unregister(local->hw.wiphy);
+ destroy_workqueue(local->workqueue);
ieee80211_led_exit(local);
kfree(local->int_scan_req);
}
next prev parent reply other threads:[~2020-04-20 12:57 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-20 12:38 [PATCH 5.6 00/71] 5.6.6-rc1 review Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 01/71] amd-xgbe: Use __napi_schedule() in BH context Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 02/71] hsr: check protocol version in hsr_newlink() Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 03/71] l2tp: Allow management of tunnels and session in user namespace Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 04/71] net: dsa: mt7530: fix tagged frames pass-through in VLAN-unaware mode Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 05/71] net: ipv4: devinet: Fix crash when add/del multicast IP with autojoin Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 06/71] net: ipv6: do not consider routes via gateways for anycast address check Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 07/71] net: macsec: fix using wrong structure in macsec_changelink() Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 08/71] net: phy: micrel: use genphy_read_status for KSZ9131 Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 09/71] net: qrtr: send msgs from local of same id as broadcast Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 10/71] net: revert default NAPI poll timeout to 2 jiffies Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 11/71] net: tun: record RX queue in skb before do_xdp_generic() Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 12/71] net: dsa: mt7530: move mt7623 settings out off the mt7530 Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 13/71] net: ethernet: mediatek: " Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 14/71] net/mlx5: Fix frequent ioread PCI access during recovery Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 15/71] net/mlx5e: Add missing release firmware call Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 16/71] net/mlx5e: Fix nest_level for vlan pop action Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 17/71] net/mlx5e: Fix pfnum in devlink port attribute Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 18/71] net: icmp6: do not select saddr from iif when route has prefsrc set Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 19/71] net: mscc: ocelot: fix untagged packet drops when enslaving to vlan aware bridge Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 20/71] net: stmmac: dwmac-sunxi: Provide TX and RX fifo sizes Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 21/71] net: phy: marvell: Fix pause frame negotiation Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 22/71] Revert "ACPI: EC: Do not clear boot_ec_is_ecdt in acpi_ec_add()" Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 23/71] ovl: fix value of i_ino for lower hardlink corner case Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 24/71] scsi: ufs: Fix ufshcd_hold() caused scheduling while atomic Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 25/71] platform/chrome: cros_ec_rpmsg: Fix race with host event Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 26/71] jbd2: improve comments about freeing data buffers whose page mapping is NULL Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 27/71] acpi/nfit: improve bounds checking for func Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 28/71] perf report: Fix no branch type statistics report issue Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 29/71] pwm: pca9685: Fix PWM/GPIO inter-operation Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 30/71] net: stmmac: xgmac: Fix VLAN register handling Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 31/71] net/bpfilter: remove superfluous testing message Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 32/71] ext4: fix incorrect group count in ext4_fill_super error message Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 33/71] ext4: fix incorrect inodes per group in " Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 34/71] HID: lg-g15: Do not fail the probe when we fail to disable F# emulation Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 35/71] clk: at91: sam9x60: fix usb clock parents Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 36/71] clk: at91: usb: use proper usbs_mask Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 37/71] ARM: dts: imx7-colibri: fix muxing of usbc_det pin Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 38/71] arm64: dts: librem5-devkit: add a vbus supply to usb0 Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 39/71] usb: dwc3: gadget: Dont clear flags before transfer ended Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 40/71] ASoC: Intel: mrfld: fix incorrect check on p->sink Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 41/71] ASoC: Intel: mrfld: return error codes when an error occurs Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 42/71] ALSA: hda: Allow setting preallocation again for x86 Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 43/71] ALSA: hda/realtek - Enable the headset mic on Asus FX505DT Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 44/71] ALSA: usb-audio: Filter error from connector kctl ops, too Greg Kroah-Hartman
2020-04-20 12:38 ` [PATCH 5.6 45/71] ALSA: usb-audio: Dont override ignore_ctl_error value from the map Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 46/71] ALSA: usb-audio: Dont create jack controls for PCM terminals Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 47/71] ALSA: usb-audio: Check mapping at creating connector controls, too Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 48/71] arm64: vdso: dont free unallocated pages Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 49/71] keys: Fix proc_keys_next to increase position index Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 50/71] tracing: Fix the race between registering snapshot event trigger and triggering snapshot operation Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 51/71] btrfs: check commit root generation in should_ignore_root Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 52/71] nl80211: fix NL80211_ATTR_FTM_RESPONDER policy Greg Kroah-Hartman
2020-04-20 12:39 ` Greg Kroah-Hartman [this message]
2020-04-20 12:39 ` [PATCH 5.6 54/71] mac80211_hwsim: Use kstrndup() in place of kasprintf() Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 55/71] net/mlx5e: Encapsulate updating netdev queues into a function Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 56/71] net/mlx5e: Rename hw_modify to preactivate Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 57/71] net/mlx5e: Use preactivate hook to set the indirection table Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 58/71] drm/amd/powerplay: force the trim of the mclk dpm_levels if OD is enabled Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 59/71] drm/amdgpu/gfx9: add gfxoff quirk Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 60/71] drm/amdgpu: fix the hw hang during perform system reboot and reset Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 61/71] i2c: designware: platdrv: Remove DPM_FLAG_SMART_SUSPEND flag on BYT and CHT Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 62/71] drm/nouveau/sec2/gv100-: add missing MODULE_FIRMWARE() Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 63/71] drm/i915/perf: Do not clear pollin for small user read buffers Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 64/71] rcu: Dont acquire lock in NMI handler in rcu_nmi_enter_common() Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 65/71] ext4: do not zeroout extents beyond i_disksize Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 66/71] irqchip/ti-sci-inta: Fix processing of masked irqs Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 67/71] proc, time/namespace: Show clock symbolic names in /proc/pid/timens_offsets Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 68/71] x86/resctrl: Preserve CDP enable over CPU hotplug Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 69/71] x86/resctrl: Fix invalid attempt at removing the default resource group Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 70/71] x86/microcode/AMD: Increase microcode PATCH_MAX_SIZE Greg Kroah-Hartman
2020-04-20 12:39 ` [PATCH 5.6 71/71] scsi: target: iscsi: calling iscsit_stop_session() inside iscsit_close_session() has no effect Greg Kroah-Hartman
2020-04-20 17:31 ` [PATCH 5.6 00/71] 5.6.6-rc1 review Naresh Kamboju
2020-04-20 19:53 ` Guenter Roeck
2020-04-21 9:55 ` Greg Kroah-Hartman
2020-04-21 9:55 ` Jon Hunter
2020-04-21 13:22 ` Greg Kroah-Hartman
2020-04-21 16:49 ` shuah
2020-04-21 18:42 ` Greg Kroah-Hartman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200420121519.805021247@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=johannes.berg@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
--cc=sumit.garg@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).