linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org,
	Manikanta Pubbisetty <mpubbise@codeaurora.org>,
	Johannes Berg <johannes.berg@intel.com>,
	Sasha Levin <alexander.levin@microsoft.com>
Subject: [PATCH 4.4 21/70] mac80211: restrict delayed tailroom needed decrement
Date: Mon, 24 Sep 2018 13:52:20 +0200	[thread overview]
Message-ID: <20180924113100.886554248@linuxfoundation.org> (raw)
In-Reply-To: <20180924113058.420454070@linuxfoundation.org>

4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Manikanta Pubbisetty <mpubbise@codeaurora.org>

[ Upstream commit 133bf90dbb8b873286f8ec2e81ba26e863114b8c ]

As explained in ieee80211_delayed_tailroom_dec(), during roam,
keys of the old AP will be destroyed and new keys will be
installed. Deletion of the old key causes
crypto_tx_tailroom_needed_cnt to go from 1 to 0 and the new key
installation causes a transition from 0 to 1.

Whenever crypto_tx_tailroom_needed_cnt transitions from 0 to 1,
we invoke synchronize_net(); the reason for doing this is to avoid
a race in the TX path as explained in increment_tailroom_need_count().
This synchronize_net() operation can be slow and can affect the station
roam time. To avoid this, decrementing the crypto_tx_tailroom_needed_cnt
is delayed for a while so that upon installation of new key the
transition would be from 1 to 2 instead of 0 to 1 and thereby
improving the roam time.

This is all correct for a STA iftype, but deferring the tailroom_needed
decrement for other iftypes may be unnecessary.

For example, let's consider the case of a 4-addr client connecting to
an AP for which AP_VLAN interface is also created, let the initial
value for tailroom_needed on the AP be 1.

* 4-addr client connects to the AP (AP: tailroom_needed = 1)
* AP will clear old keys, delay decrement of tailroom_needed count
* AP_VLAN is created, it takes the tailroom count from master
  (AP_VLAN: tailroom_needed = 1, AP: tailroom_needed = 1)
* Install new key for the station, assume key is plumbed in the HW,
  there won't be any change in tailroom_needed count on AP iface
* Delayed decrement of tailroom_needed count on AP
  (AP: tailroom_needed = 0, AP_VLAN: tailroom_needed = 1)

Because of the delayed decrement on AP iface, tailroom_needed count goes
out of sync between AP(master iface) and AP_VLAN(slave iface) and
there would be unnecessary tailroom created for the packets going
through AP_VLAN iface.

Also, WARN_ONs were observed while trying to bring down the AP_VLAN
interface:
(warn_slowpath_common) (warn_slowpath_null+0x18/0x20)
(warn_slowpath_null) (ieee80211_free_keys+0x114/0x1e4)
(ieee80211_free_keys) (ieee80211_del_virtual_monitor+0x51c/0x850)
(ieee80211_del_virtual_monitor) (ieee80211_stop+0x30/0x3c)
(ieee80211_stop) (__dev_close_many+0x94/0xb8)
(__dev_close_many) (dev_close_many+0x5c/0xc8)

Restricting delayed decrement to station interface alone fixes the problem
and it makes sense to do so because delayed decrement is done to improve
roam time which is applicable only for client devices.

Signed-off-by: Manikanta Pubbisetty <mpubbise@codeaurora.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/mac80211/cfg.c |    2 +-
 net/mac80211/key.c |   24 +++++++++++++++---------
 2 files changed, 16 insertions(+), 10 deletions(-)

--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -286,7 +286,7 @@ static int ieee80211_del_key(struct wiph
 		goto out_unlock;
 	}
 
-	ieee80211_key_free(key, true);
+	ieee80211_key_free(key, sdata->vif.type == NL80211_IFTYPE_STATION);
 
 	ret = 0;
  out_unlock:
--- a/net/mac80211/key.c
+++ b/net/mac80211/key.c
@@ -647,11 +647,15 @@ int ieee80211_key_link(struct ieee80211_
 {
 	struct ieee80211_local *local = sdata->local;
 	struct ieee80211_key *old_key;
-	int idx, ret;
-	bool pairwise;
-
-	pairwise = key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE;
-	idx = key->conf.keyidx;
+	int idx = key->conf.keyidx;
+	bool pairwise = key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE;
+	/*
+	 * We want to delay tailroom updates only for station - in that
+	 * case it helps roaming speed, but in other cases it hurts and
+	 * can cause warnings to appear.
+	 */
+	bool delay_tailroom = sdata->vif.type == NL80211_IFTYPE_STATION;
+	int ret;
 
 	mutex_lock(&sdata->local->key_mtx);
 
@@ -679,14 +683,14 @@ int ieee80211_key_link(struct ieee80211_
 	increment_tailroom_need_count(sdata);
 
 	ieee80211_key_replace(sdata, sta, pairwise, old_key, key);
-	ieee80211_key_destroy(old_key, true);
+	ieee80211_key_destroy(old_key, delay_tailroom);
 
 	ieee80211_debugfs_key_add(key);
 
 	if (!local->wowlan) {
 		ret = ieee80211_key_enable_hw_accel(key);
 		if (ret)
-			ieee80211_key_free(key, true);
+			ieee80211_key_free(key, delay_tailroom);
 	} else {
 		ret = 0;
 	}
@@ -874,7 +878,8 @@ void ieee80211_free_sta_keys(struct ieee
 		ieee80211_key_replace(key->sdata, key->sta,
 				key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE,
 				key, NULL);
-		__ieee80211_key_destroy(key, true);
+		__ieee80211_key_destroy(key, key->sdata->vif.type ==
+					NL80211_IFTYPE_STATION);
 	}
 
 	for (i = 0; i < NUM_DEFAULT_KEYS; i++) {
@@ -884,7 +889,8 @@ void ieee80211_free_sta_keys(struct ieee
 		ieee80211_key_replace(key->sdata, key->sta,
 				key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE,
 				key, NULL);
-		__ieee80211_key_destroy(key, true);
+		__ieee80211_key_destroy(key, key->sdata->vif.type ==
+					NL80211_IFTYPE_STATION);
 	}
 
 	mutex_unlock(&local->key_mtx);



  parent reply	other threads:[~2018-09-24 11:54 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-09-24 11:51 [PATCH 4.4 00/70] 4.4.158-stable review Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 01/70] iommu/arm-smmu-v3: sync the OVACKFLG to PRIQ consumer register Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 02/70] ALSA: msnd: Fix the default sample sizes Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 03/70] ALSA: usb-audio: Fix multiple definitions in AU0828_DEVICE() macro Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 04/70] xfrm: fix passing zero to ERR_PTR() warning Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 05/70] gfs2: Special-case rindex for gfs2_grow Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 06/70] clk: imx6ul: fix missing of_node_put() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 07/70] kbuild: add .DELETE_ON_ERROR special target Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 08/70] dmaengine: pl330: fix irq race with terminate_all Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 09/70] MIPS: ath79: fix system restart Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 10/70] media: videobuf2-core: check for q->error in vb2_core_qbuf() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 11/70] mtd/maps: fix solutionengine.c printk format warnings Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 12/70] fbdev: omapfb: off by one in omapfb_register_client() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 13/70] video: goldfishfb: fix memory leak on driver remove Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 14/70] fbdev/via: fix defined but not used warning Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 15/70] perf powerpc: Fix callchain ip filtering when return address is in a register Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 16/70] fbdev: Distinguish between interlaced and progressive modes Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 17/70] ARM: exynos: Clear global variable on init error path Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 18/70] perf powerpc: Fix callchain ip filtering Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 19/70] powerpc/powernv: opal_put_chars partial write fix Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 20/70] MIPS: jz4740: Bump zload address Greg Kroah-Hartman
2018-09-24 11:52 ` Greg Kroah-Hartman [this message]
2018-09-24 11:52 ` [PATCH 4.4 22/70] xen-netfront: fix queue name setting Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 23/70] arm64: dts: qcom: db410c: Fix Bluetooth LED trigger Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 24/70] s390/qeth: fix race in used-buffer accounting Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 25/70] s390/qeth: reset layer2 attribute on layer switch Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 26/70] platform/x86: toshiba_acpi: Fix defined but not used build warnings Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 27/70] crypto: sharah - Unregister correct algorithms for SAHARA 3 Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 28/70] xen-netfront: fix warn message as irq device name has / Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 29/70] RDMA/cma: Protect cma dev list with lock Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 30/70] pstore: Fix incorrect persistent ram buffer mapping Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 31/70] xen/netfront: fix waiting for xenbus state change Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 32/70] IB/ipoib: Avoid a race condition between start_xmit and cm_rep_handler Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 33/70] Tools: hv: Fix a bug in the key delete code Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 34/70] misc: hmc6352: fix potential Spectre v1 Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 35/70] usb: Dont die twice if PCI xhci host is not responding in resume Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 36/70] USB: Add quirk to support DJI CineSSD Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 37/70] usb: Avoid use-after-free by flushing endpoints early in usb_set_interface() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 38/70] usb: host: u132-hcd: Fix a sleep-in-atomic-context bug in u132_get_frame() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 39/70] USB: add quirk for WORLDE Controller KS49 or Prodipe MIDI 49C USB controller Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 40/70] USB: net2280: Fix erroneous synchronization change Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 41/70] USB: serial: io_ti: fix array underflow in completion handler Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 42/70] usb: misc: uss720: Fix two sleep-in-atomic-context bugs Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 43/70] USB: yurex: Fix buffer over-read in yurex_write() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 44/70] usb: cdc-wdm: Fix a sleep-in-atomic-context bug in service_outstanding_interrupt() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 45/70] cifs: prevent integer overflow in nxt_dir_entry() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 46/70] CIFS: fix wrapping bugs in num_entries() Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 47/70] binfmt_elf: Respect error return from `regset->active Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 48/70] audit: fix use-after-free in audit_add_watch Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 49/70] mtdchar: fix overflows in adjustment of `count` Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 50/70] MIPS: loongson64: cs5536: Fix PCI_OHCI_INT_REG reads Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 51/70] ARM: hisi: handle of_iomap and fix missing of_node_put Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 52/70] ARM: hisi: fix error handling and " Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 53/70] ARM: hisi: check of_iomap and fix " Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 54/70] drm/nouveau: tegra: Detach from ARM DMA/IOMMU mapping Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 55/70] parport: sunbpp: fix error return code Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 56/70] coresight: Handle errors in finding input/output ports Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 57/70] coresight: tpiu: Fix disabling timeouts Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 58/70] gpiolib: Mark gpio_suffixes array with __maybe_unused Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 59/70] drm/amdkfd: Fix error codes in kfd_get_process Greg Kroah-Hartman
2018-09-24 11:52 ` [PATCH 4.4 60/70] rtc: bq4802: add error handling for devm_ioremap Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 61/70] ALSA: pcm: Fix snd_interval_refine first/last with open min/max Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 62/70] selftest: timers: Tweak raw_skew to SKIP when ADJ_OFFSET/other clock adjustments are in progress Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 63/70] drm/panel: type promotion bug in s6e8aa0_read_mtp_id() Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 64/70] IB/nes: Fix a compiler warning Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 65/70] pinctrl: qcom: spmi-gpio: Fix pmic_gpio_config_get() to be compliant Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 66/70] USB: serial: ti_usb_3410_5052: fix array underflow in completion handler Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 67/70] mei: bus: type promotion bug in mei_nfc_if_version() Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 68/70] drivers: net: cpsw: fix segfault in case of bad phy-handle Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 69/70] MIPS: VDSO: Match data page cache colouring when D$ aliases Greg Kroah-Hartman
2018-09-24 11:53 ` [PATCH 4.4 70/70] MIPS: VDSO: Drop gic_get_usm_range() usage Greg Kroah-Hartman
2018-09-24 18:06 ` [PATCH 4.4 00/70] 4.4.158-stable review Nathan Chancellor
2018-09-24 21:06 ` Dan Rue
2018-09-24 22:15 ` Shuah Khan
2018-09-25  9:14   ` Greg Kroah-Hartman
2018-09-25  9:08 ` Greg Kroah-Hartman
2018-09-25 20:41 ` Guenter Roeck

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=20180924113100.886554248@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=alexander.levin@microsoft.com \
    --cc=johannes.berg@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mpubbise@codeaurora.org \
    --cc=stable@vger.kernel.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).