stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Matthew Wilcox <willy@infradead.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Alexey Dobriyan <adobriyan@gmail.com>,
	Kees Cook <keescook@chromium.org>,
	linux-fsdevel@vger.kernel.org,
	Davidlohr Bueso <dave@stgolabs.net>,
	Miklos Szeredi <miklos@szeredi.hu>,
	Daniel Colascione <dancol@google.com>,
	Dave Chinner <david@fromorbit.com>,
	Randy Dunlap <rdunlap@infradead.org>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH AUTOSEL 4.9 55/87] genirq: Avoid summation loops for /proc/stat
Date: Wed, 27 Mar 2019 14:20:08 -0400	[thread overview]
Message-ID: <20190327182040.17444-55-sashal@kernel.org> (raw)
In-Reply-To: <20190327182040.17444-1-sashal@kernel.org>

From: Thomas Gleixner <tglx@linutronix.de>

[ Upstream commit 1136b0728969901a091f0471968b2b76ed14d9ad ]

Waiman reported that on large systems with a large amount of interrupts the
readout of /proc/stat takes a long time to sum up the interrupt
statistics. In principle this is not a problem. but for unknown reasons
some enterprise quality software reads /proc/stat with a high frequency.

The reason for this is that interrupt statistics are accounted per cpu. So
the /proc/stat logic has to sum up the interrupt stats for each interrupt.

This can be largely avoided for interrupts which are not marked as
'PER_CPU' interrupts by simply adding a per interrupt summation counter
which is incremented along with the per interrupt per cpu counter.

The PER_CPU interrupts need to avoid that and use only per cpu accounting
because they share the interrupt number and the interrupt descriptor and
concurrent updates would conflict or require unwanted synchronization.

Reported-by: Waiman Long <longman@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Waiman Long <longman@redhat.com>
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Davidlohr Bueso <dbueso@suse.de>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: linux-fsdevel@vger.kernel.org
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Miklos Szeredi <miklos@szeredi.hu>
Cc: Daniel Colascione <dancol@google.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Link: https://lkml.kernel.org/r/20190208135020.925487496@linutronix.de

8<-------------

v2: Undo the unintentional layout change of struct irq_desc.

 include/linux/irqdesc.h |    1 +
 kernel/irq/chip.c       |   12 ++++++++++--
 kernel/irq/internals.h  |    8 +++++++-
 kernel/irq/irqdesc.c    |    7 ++++++-
 4 files changed, 24 insertions(+), 4 deletions(-)

Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/linux/irqdesc.h |  1 +
 kernel/irq/chip.c       | 12 ++++++++++--
 kernel/irq/internals.h  |  8 +++++++-
 kernel/irq/irqdesc.c    |  7 ++++++-
 4 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h
index c9be57931b58..bb5547a83daf 100644
--- a/include/linux/irqdesc.h
+++ b/include/linux/irqdesc.h
@@ -61,6 +61,7 @@ struct irq_desc {
 	unsigned int		core_internal_state__do_not_mess_with_it;
 	unsigned int		depth;		/* nested irq disables */
 	unsigned int		wake_depth;	/* nested wake enables */
+	unsigned int		tot_count;
 	unsigned int		irq_count;	/* For detecting broken IRQs */
 	unsigned long		last_unhandled;	/* Aging timer for unhandled count */
 	unsigned int		irqs_unhandled;
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index f30110e1b8c9..9e745cc0726d 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -729,7 +729,11 @@ void handle_percpu_irq(struct irq_desc *desc)
 {
 	struct irq_chip *chip = irq_desc_get_chip(desc);
 
-	kstat_incr_irqs_this_cpu(desc);
+	/*
+	 * PER CPU interrupts are not serialized. Do not touch
+	 * desc->tot_count.
+	 */
+	__kstat_incr_irqs_this_cpu(desc);
 
 	if (chip->irq_ack)
 		chip->irq_ack(&desc->irq_data);
@@ -758,7 +762,11 @@ void handle_percpu_devid_irq(struct irq_desc *desc)
 	unsigned int irq = irq_desc_get_irq(desc);
 	irqreturn_t res;
 
-	kstat_incr_irqs_this_cpu(desc);
+	/*
+	 * PER CPU interrupts are not serialized. Do not touch
+	 * desc->tot_count.
+	 */
+	__kstat_incr_irqs_this_cpu(desc);
 
 	if (chip->irq_ack)
 		chip->irq_ack(&desc->irq_data);
diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h
index bc226e783bd2..22e3f29a30d8 100644
--- a/kernel/irq/internals.h
+++ b/kernel/irq/internals.h
@@ -199,12 +199,18 @@ static inline bool irqd_has_set(struct irq_data *d, unsigned int mask)
 
 #undef __irqd_to_state
 
-static inline void kstat_incr_irqs_this_cpu(struct irq_desc *desc)
+static inline void __kstat_incr_irqs_this_cpu(struct irq_desc *desc)
 {
 	__this_cpu_inc(*desc->kstat_irqs);
 	__this_cpu_inc(kstat.irqs_sum);
 }
 
+static inline void kstat_incr_irqs_this_cpu(struct irq_desc *desc)
+{
+	__kstat_incr_irqs_this_cpu(desc);
+	desc->tot_count++;
+}
+
 static inline int irq_desc_get_node(struct irq_desc *desc)
 {
 	return irq_common_data_get_node(&desc->irq_common_data);
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 77977f55dff7..5e0ea17d01a6 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -109,6 +109,7 @@ static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node,
 	desc->depth = 1;
 	desc->irq_count = 0;
 	desc->irqs_unhandled = 0;
+	desc->tot_count = 0;
 	desc->name = NULL;
 	desc->owner = owner;
 	for_each_possible_cpu(cpu)
@@ -880,11 +881,15 @@ unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
 unsigned int kstat_irqs(unsigned int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-	int cpu;
 	unsigned int sum = 0;
+	int cpu;
 
 	if (!desc || !desc->kstat_irqs)
 		return 0;
+	if (!irq_settings_is_per_cpu_devid(desc) &&
+	    !irq_settings_is_per_cpu(desc))
+	    return desc->tot_count;
+
 	for_each_possible_cpu(cpu)
 		sum += *per_cpu_ptr(desc->kstat_irqs, cpu);
 	return sum;
-- 
2.19.1


  parent reply	other threads:[~2019-03-27 18:41 UTC|newest]

Thread overview: 90+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-27 18:19 [PATCH AUTOSEL 4.9 01/87] CIFS: fix POSIX lock leak and invalid ptr deref Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 02/87] h8300: use cc-cross-prefix instead of hardcoding h8300-unknown-linux- Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 03/87] i2c: sis630: correct format strings Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 04/87] tracing: kdb: Fix ftdump to not sleep Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 05/87] gpio: gpio-omap: fix level interrupt idling Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 06/87] include/linux/relay.h: fix percpu annotation in struct rchan Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 07/87] sysctl: handle overflow for file-max Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 08/87] enic: fix build warning without CONFIG_CPUMASK_OFFSTACK Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 09/87] scsi: hisi_sas: Set PHY linkrate when disconnected Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 10/87] mm/cma.c: cma_declare_contiguous: correct err handling Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 11/87] mm/page_ext.c: fix an imbalance with kmemleak Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 12/87] mm/vmalloc.c: fix kernel BUG at mm/vmalloc.c:512! Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 13/87] mm/slab.c: kmemleak no scan alien caches Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 14/87] ocfs2: fix a panic problem caused by o2cb_ctl Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 15/87] f2fs: do not use mutex lock in atomic context Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 16/87] fs/file.c: initialize init_files.resize_wait Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 17/87] cifs: use correct format characters Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 18/87] dm thin: add sanity checks to thin-pool and external snapshot creation Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 19/87] cifs: Fix NULL pointer dereference of devname Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 20/87] fs: Make splice() and tee() take into account O_NONBLOCK flag on pipes Sasha Levin
2019-03-28 15:37   ` Slavomir Kaslev
2019-03-28 16:04     ` Steven Rostedt
2019-04-03 16:19       ` Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 21/87] jbd2: fix invalid descriptor block checksum Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 22/87] fs: fix guard_bio_eod to check for real EOD errors Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 23/87] tools lib traceevent: Fix buffer overflow in arg_eval Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 24/87] wil6210: check null pointer in _wil_cfg80211_merge_extra_ies Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 25/87] crypto: crypto4xx - add missing of_node_put after of_device_is_available Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 26/87] usb: chipidea: Grab the (legacy) USB PHY by phandle first Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 27/87] scsi: core: replace GFP_ATOMIC with GFP_KERNEL in scsi_scan.c Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 28/87] coresight: etm4x: Add support to enable ETMv4.2 Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 29/87] ARM: 8840/1: use a raw_spinlock_t in unwind Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 30/87] iommu/io-pgtable-arm-v7s: Only kmemleak_ignore L2 tables Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 31/87] mmc: omap: fix the maximum timeout setting Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 32/87] e1000e: Fix -Wformat-truncation warnings Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 33/87] mlxsw: spectrum: Avoid " Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 34/87] IB/mlx4: Increase the timeout for CM cache Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 35/87] ASoC: qcom: Fix of-node refcount unbalance in apq8016_sbc_parse_of() Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 36/87] scsi: megaraid_sas: return error when create DMA pool failed Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 37/87] perf test: Fix failure of 'evsel-tp-sched' test on s390 Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 38/87] SoC: imx-sgtl5000: add missing put_device() Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 39/87] media: sh_veu: Correct return type for mem2mem buffer helpers Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 40/87] media: s5p-jpeg: " Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 41/87] media: s5p-g2d: " Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 42/87] media: mx2_emmaprp: " Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 43/87] vfs: fix preadv64v2 and pwritev64v2 compat syscalls with offset == -1 Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 44/87] HID: intel-ish-hid: avoid binding wrong ishtp_cl_device Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 45/87] leds: lp55xx: fix null deref on firmware load failure Sasha Levin
2019-03-27 18:19 ` [PATCH AUTOSEL 4.9 46/87] iwlwifi: pcie: fix emergency path Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 47/87] ACPI / video: Refactor and fix dmi_is_desktop() Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 48/87] kprobes: Prohibit probing on bsearch() Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 49/87] ARM: 8833/1: Ensure that NEON code always compiles with Clang Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 50/87] ALSA: PCM: check if ops are defined before suspending PCM Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 51/87] usb: f_fs: Avoid crash due to out-of-scope stack ptr access Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 52/87] bcache: fix input overflow to cache set sysfs file io_error_halflife Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 53/87] bcache: fix input overflow to sequential_cutoff Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 54/87] bcache: improve sysfs_strtoul_clamp() Sasha Levin
2019-03-27 18:20 ` Sasha Levin [this message]
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 56/87] iw_cxgb4: fix srqidx leak during connection abort Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 57/87] fbdev: fbmem: fix memory access if logo is bigger than the screen Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 58/87] cdrom: Fix race condition in cdrom_sysctl_register Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 59/87] e1000e: fix cyclic resets at link up with active tx Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 60/87] ASoC: fsl-asoc-card: fix object reference leaks in fsl_asoc_card_probe Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 61/87] locking/lockdep: Add debug_locks check in __lock_downgrade() Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 62/87] efi/memattr: Don't bail on zero VA if it equals the region's PA Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 63/87] ARM: dts: lpc32xx: Remove leading 0x and 0s from bindings notation Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 64/87] soc: qcom: gsbi: Fix error handling in gsbi_probe() Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 65/87] mt7601u: bump supported EEPROM version Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 66/87] ARM: avoid Cortex-A9 livelock on tight dmb loops Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 67/87] tty: increase the default flip buffer limit to 2*640K Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 68/87] powerpc/pseries: Perform full re-add of CPU for topology update post-migration Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 69/87] media: mt9m111: set initial frame size other than 0x0 Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 70/87] hwrng: virtio - Avoid repeated init of completion Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 71/87] soc/tegra: fuse: Fix illegal free of IO base address Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 72/87] HID: intel-ish: ipc: handle PIMR before ish_wakeup also clear PISR busy_clear bit Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 73/87] Bluetooth: Verify that l2cap_get_conf_opt provides large enough buffer Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 74/87] hpet: Fix missing '=' character in the __setup() code of hpet_mmap_enable Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 75/87] dmaengine: imx-dma: fix warning comparison of distinct pointer types Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 76/87] dmaengine: qcom_hidma: assign channel cookie correctly Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 77/87] netfilter: physdev: relax br_netfilter dependency Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 78/87] media: s5p-jpeg: Check for fmt_ver_flag when doing fmt enumeration Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 79/87] regulator: act8865: Fix act8600_sudcdc_voltage_ranges setting Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 80/87] drm/nouveau: Stop using drm_crtc_force_disable Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 81/87] x86/build: Specify elf_i386 linker emulation explicitly for i386 objects Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 82/87] selinux: do not override context on context mounts Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 83/87] wlcore: Fix memory leak in case wl12xx_fetch_firmware failure Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 84/87] x86/build: Mark per-CPU symbols as absolute explicitly for LLD Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 85/87] dmaengine: tegra: avoid overflow of byte tracking Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 86/87] drm/dp/mst: Configure no_stop_bit correctly for remote i2c xfers Sasha Levin
2019-03-27 18:20 ` [PATCH AUTOSEL 4.9 87/87] ACPI / video: Extend chassis-type detection with a "Lunch Box" check Sasha Levin

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=20190327182040.17444-55-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=adobriyan@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=dancol@google.com \
    --cc=dave@stgolabs.net \
    --cc=david@fromorbit.com \
    --cc=keescook@chromium.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=rdunlap@infradead.org \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=willy@infradead.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).