public inbox for linux-kernel@vger.kernel.org
 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, Yunfeng Ye <yeyunfeng@huawei.com>,
	Lai Jiangshan <jiangshanlai@gmail.com>, Tejun Heo <tj@kernel.org>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 4.4 02/38] workqueue: Kick a worker based on the actual activation of delayed works
Date: Mon, 11 Jan 2021 14:00:34 +0100	[thread overview]
Message-ID: <20210111130032.588245581@linuxfoundation.org> (raw)
In-Reply-To: <20210111130032.469630231@linuxfoundation.org>

From: Yunfeng Ye <yeyunfeng@huawei.com>

[ Upstream commit 01341fbd0d8d4e717fc1231cdffe00343088ce0b ]

In realtime scenario, We do not want to have interference on the
isolated cpu cores. but when invoking alloc_workqueue() for percpu wq
on the housekeeping cpu, it kick a kworker on the isolated cpu.

  alloc_workqueue
    pwq_adjust_max_active
      wake_up_worker

The comment in pwq_adjust_max_active() said:
  "Need to kick a worker after thawed or an unbound wq's
   max_active is bumped"

So it is unnecessary to kick a kworker for percpu's wq when invoking
alloc_workqueue(). this patch only kick a worker based on the actual
activation of delayed works.

Signed-off-by: Yunfeng Ye <yeyunfeng@huawei.com>
Reviewed-by: Lai Jiangshan <jiangshanlai@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 kernel/workqueue.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 3fb2d45c0b42f..6b293804cd734 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -3361,17 +3361,24 @@ static void pwq_adjust_max_active(struct pool_workqueue *pwq)
 	 * is updated and visible.
 	 */
 	if (!freezable || !workqueue_freezing) {
+		bool kick = false;
+
 		pwq->max_active = wq->saved_max_active;
 
 		while (!list_empty(&pwq->delayed_works) &&
-		       pwq->nr_active < pwq->max_active)
+		       pwq->nr_active < pwq->max_active) {
 			pwq_activate_first_delayed(pwq);
+			kick = true;
+		}
 
 		/*
 		 * Need to kick a worker after thawed or an unbound wq's
-		 * max_active is bumped.  It's a slow path.  Do it always.
+		 * max_active is bumped. In realtime scenarios, always kicking a
+		 * worker will cause interference on the isolated cpu cores, so
+		 * let's kick iff work items were activated.
 		 */
-		wake_up_worker(pwq->pool);
+		if (kick)
+			wake_up_worker(pwq->pool);
 	} else {
 		pwq->max_active = 0;
 	}
-- 
2.27.0




  parent reply	other threads:[~2021-01-11 13:02 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-11 13:00 [PATCH 4.4 00/38] 4.4.251-rc1 review Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 01/38] kbuild: dont hardcode depmod path Greg Kroah-Hartman
2021-01-11 13:00 ` Greg Kroah-Hartman [this message]
2021-01-11 13:00 ` [PATCH 4.4 03/38] lib/genalloc: fix the overflow when size is too big Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 04/38] depmod: handle the case of /sbin/depmod without /sbin in PATH Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 05/38] atm: idt77252: call pci_disable_device() on error path Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 06/38] net: dcb: Validate netlink message in DCB handler Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 07/38] ipv4: Ignore ECN bits for fib lookups in fib_compute_spec_dst() Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 08/38] net: hns: fix return value check in __lb_other_process() Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 09/38] net: hdlc_ppp: Fix issues when mod_timer is called while timer is running Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 10/38] CDC-NCM: remove "connected" log message Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 11/38] vhost_net: fix ubuf refcount incorrectly when sendmsg fails Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 12/38] net: sched: prevent invalid Scell_log shift count Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 13/38] virtio_net: Fix recursive call to cpus_read_lock() Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 14/38] ethernet: ucc_geth: fix use-after-free in ucc_geth_remove() Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 15/38] video: hyperv_fb: Fix the mmap() regression for v5.4.y and older Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 16/38] usb: gadget: enable super speed plus Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 17/38] USB: cdc-acm: blacklist another IR Droid device Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 18/38] usb: chipidea: ci_hdrc_imx: add missing put_device() call in usbmisc_get_init_data() Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 19/38] USB: xhci: fix U1/U2 handling for hardware with XHCI_INTEL_HOST quirk set Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 20/38] usb: uas: Add PNY USB Portable SSD to unusual_uas Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 21/38] USB: serial: iuu_phoenix: fix DMA from stack Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 22/38] USB: serial: option: add LongSung M5710 module support Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 23/38] USB: yurex: fix control-URB timeout handling Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 24/38] USB: usblp: fix DMA to stack Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 25/38] ALSA: usb-audio: Fix UBSAN warnings for MIDI jacks Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 26/38] usb: gadget: select CONFIG_CRC32 Greg Kroah-Hartman
2021-01-11 13:00 ` [PATCH 4.4 27/38] usb: gadget: f_uac2: reset wMaxPacketSize Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 28/38] usb: gadget: function: printer: Fix a memory leak for interface descriptor Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 29/38] USB: gadget: legacy: fix return error code in acm_ms_bind() Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 30/38] usb: gadget: Fix spinlock lockup on usb_function_deactivate Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 31/38] usb: gadget: configfs: Preserve function ordering after bind failure Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 32/38] USB: serial: keyspan_pda: remove unused variable Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 33/38] x86/mm: Fix leak of pmd ptlock Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 34/38] ALSA: hda/conexant: add a new hda codec CX11970 Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 35/38] Revert "device property: Keep secondary firmware node secondary by type" Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 36/38] netfilter: ipset: fix shift-out-of-bounds in htable_bits() Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 37/38] netfilter: xt_RATEEST: reject non-null terminated string from userspace Greg Kroah-Hartman
2021-01-11 13:01 ` [PATCH 4.4 38/38] x86/mtrr: Correct the range check before performing MTRR type lookups Greg Kroah-Hartman
2021-01-11 16:23 ` [PATCH 4.4 00/38] 4.4.251-rc1 review Pavel Machek
2021-01-11 21:52 ` Guenter Roeck
2021-01-11 23:41 ` Shuah Khan
2021-01-12  9:04 ` Naresh Kamboju

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=20210111130032.588245581@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=jiangshanlai@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tj@kernel.org \
    --cc=yeyunfeng@huawei.com \
    /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