stable.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, kernel test robot <oliver.sang@intel.com>,
	Shile Zhang <shile.zhang@linux.alibaba.com>,
	Joerg Roedel <jroedel@suse.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	Borislav Petkov <bp@suse.de>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	Andy Lutomirski <luto@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	"Rafael J . Wysocki" <rafael.j.wysocki@intel.com>
Subject: [PATCH 4.19 51/65] x86/mm: split vmalloc_sync_all()
Date: Tue, 24 Mar 2020 14:11:12 +0100	[thread overview]
Message-ID: <20200324130803.309144605@linuxfoundation.org> (raw)
In-Reply-To: <20200324130756.679112147@linuxfoundation.org>

From: Joerg Roedel <jroedel@suse.de>

commit 763802b53a427ed3cbd419dbba255c414fdd9e7c upstream.

Commit 3f8fd02b1bf1 ("mm/vmalloc: Sync unmappings in
__purge_vmap_area_lazy()") introduced a call to vmalloc_sync_all() in
the vunmap() code-path.  While this change was necessary to maintain
correctness on x86-32-pae kernels, it also adds additional cycles for
architectures that don't need it.

Specifically on x86-64 with CONFIG_VMAP_STACK=y some people reported
severe performance regressions in micro-benchmarks because it now also
calls the x86-64 implementation of vmalloc_sync_all() on vunmap().  But
the vmalloc_sync_all() implementation on x86-64 is only needed for newly
created mappings.

To avoid the unnecessary work on x86-64 and to gain the performance
back, split up vmalloc_sync_all() into two functions:

	* vmalloc_sync_mappings(), and
	* vmalloc_sync_unmappings()

Most call-sites to vmalloc_sync_all() only care about new mappings being
synchronized.  The only exception is the new call-site added in the
above mentioned commit.

Shile Zhang directed us to a report of an 80% regression in reaim
throughput.

Fixes: 3f8fd02b1bf1 ("mm/vmalloc: Sync unmappings in __purge_vmap_area_lazy()")
Reported-by: kernel test robot <oliver.sang@intel.com>
Reported-by: Shile Zhang <shile.zhang@linux.alibaba.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Tested-by: Borislav Petkov <bp@suse.de>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>	[GHES]
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20191009124418.8286-1-joro@8bytes.org
Link: https://lists.01.org/hyperkitty/list/lkp@lists.01.org/thread/4D3JPPHBNOSPFK2KEPC6KGKS6J25AIDB/
Link: http://lkml.kernel.org/r/20191113095530.228959-1-shile.zhang@linux.alibaba.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/x86/mm/fault.c      |   26 ++++++++++++++++++++++++--
 drivers/acpi/apei/ghes.c |    2 +-
 include/linux/vmalloc.h  |    5 +++--
 kernel/notifier.c        |    2 +-
 mm/nommu.c               |   10 +++++++---
 mm/vmalloc.c             |   11 +++++++----
 6 files changed, 43 insertions(+), 13 deletions(-)

--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -273,7 +273,7 @@ static inline pmd_t *vmalloc_sync_one(pg
 	return pmd_k;
 }
 
-void vmalloc_sync_all(void)
+static void vmalloc_sync(void)
 {
 	unsigned long address;
 
@@ -300,6 +300,16 @@ void vmalloc_sync_all(void)
 	}
 }
 
+void vmalloc_sync_mappings(void)
+{
+	vmalloc_sync();
+}
+
+void vmalloc_sync_unmappings(void)
+{
+	vmalloc_sync();
+}
+
 /*
  * 32-bit:
  *
@@ -402,11 +412,23 @@ out:
 
 #else /* CONFIG_X86_64: */
 
-void vmalloc_sync_all(void)
+void vmalloc_sync_mappings(void)
 {
+	/*
+	 * 64-bit mappings might allocate new p4d/pud pages
+	 * that need to be propagated to all tasks' PGDs.
+	 */
 	sync_global_pgds(VMALLOC_START & PGDIR_MASK, VMALLOC_END);
 }
 
+void vmalloc_sync_unmappings(void)
+{
+	/*
+	 * Unmappings never allocate or free p4d/pud pages.
+	 * No work is required here.
+	 */
+}
+
 /*
  * 64-bit:
  *
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -201,7 +201,7 @@ static int ghes_estatus_pool_expand(unsi
 	 * New allocation must be visible in all pgd before it can be found by
 	 * an NMI allocating from the pool.
 	 */
-	vmalloc_sync_all();
+	vmalloc_sync_mappings();
 
 	return gen_pool_add(ghes_estatus_pool, addr, PAGE_ALIGN(len), -1);
 }
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -107,8 +107,9 @@ extern int remap_vmalloc_range_partial(s
 
 extern int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
 							unsigned long pgoff);
-void vmalloc_sync_all(void);
- 
+void vmalloc_sync_mappings(void);
+void vmalloc_sync_unmappings(void);
+
 /*
  *	Lowlevel-APIs (not for driver use!)
  */
--- a/kernel/notifier.c
+++ b/kernel/notifier.c
@@ -552,7 +552,7 @@ NOKPROBE_SYMBOL(notify_die);
 
 int register_die_notifier(struct notifier_block *nb)
 {
-	vmalloc_sync_all();
+	vmalloc_sync_mappings();
 	return atomic_notifier_chain_register(&die_chain, nb);
 }
 EXPORT_SYMBOL_GPL(register_die_notifier);
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -446,10 +446,14 @@ void vm_unmap_aliases(void)
 EXPORT_SYMBOL_GPL(vm_unmap_aliases);
 
 /*
- * Implement a stub for vmalloc_sync_all() if the architecture chose not to
- * have one.
+ * Implement a stub for vmalloc_sync_[un]mapping() if the architecture
+ * chose not to have one.
  */
-void __weak vmalloc_sync_all(void)
+void __weak vmalloc_sync_mappings(void)
+{
+}
+
+void __weak vmalloc_sync_unmappings(void)
 {
 }
 
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -1755,7 +1755,7 @@ void *__vmalloc_node_range(unsigned long
 	 * First make sure the mappings are removed from all page-tables
 	 * before they are freed.
 	 */
-	vmalloc_sync_all();
+	vmalloc_sync_unmappings();
 
 	/*
 	 * In this function, newly allocated vm_struct has VM_UNINITIALIZED
@@ -2300,16 +2300,19 @@ int remap_vmalloc_range(struct vm_area_s
 EXPORT_SYMBOL(remap_vmalloc_range);
 
 /*
- * Implement a stub for vmalloc_sync_all() if the architecture chose not to
- * have one.
+ * Implement stubs for vmalloc_sync_[un]mappings () if the architecture chose
+ * not to have one.
  *
  * The purpose of this function is to make sure the vmalloc area
  * mappings are identical in all page-tables in the system.
  */
-void __weak vmalloc_sync_all(void)
+void __weak vmalloc_sync_mappings(void)
 {
 }
 
+void __weak vmalloc_sync_unmappings(void)
+{
+}
 
 static int f(pte_t *pte, pgtable_t table, unsigned long addr, void *data)
 {



  parent reply	other threads:[~2020-03-24 13:15 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-24 13:10 [PATCH 4.19 00/65] 4.19.113-rc1 review Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 01/65] drm/mediatek: Find the cursor plane instead of hard coding it Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 02/65] spi: qup: call spi_qup_pm_resume_runtime before suspending Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 03/65] powerpc: Include .BTF section Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 04/65] ARM: dts: dra7: Add "dma-ranges" property to PCIe RC DT nodes Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 05/65] spi: pxa2xx: Add CS control clock quirk Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 06/65] spi/zynqmp: remove entry that causes a cs glitch Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 07/65] drm/exynos: dsi: propagate error value and silence meaningless warning Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 08/65] drm/exynos: dsi: fix workaround for the legacy clock name Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 09/65] drivers/perf: arm_pmu_acpi: Fix incorrect checking of gicc pointer Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 10/65] altera-stapl: altera_get_note: prevent write beyond end of key Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 11/65] dm bio record: save/restore bi_end_io and bi_integrity Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 12/65] dm integrity: use dm_bio_record and dm_bio_restore Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 13/65] riscv: avoid the PIC offset of static percpu data in module beyond 2G limits Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 14/65] drm/amd/display: Clear link settings on MST disable connector Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 15/65] drm/amd/display: fix dcc swath size calculations on dcn1 Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 16/65] xenbus: req->body should be updated before req->state Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 17/65] xenbus: req->err " Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 18/65] block, bfq: fix overwrite of bfq_group pointer in bfq_find_set_group() Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 19/65] parse-maintainers: Mark as executable Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 20/65] USB: Disable LPM on WD19s Realtek Hub Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 21/65] usb: quirks: add NO_LPM quirk for RTL8153 based ethernet adapters Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 22/65] USB: serial: option: add ME910G1 ECM composition 0x110b Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 23/65] usb: host: xhci-plat: add a shutdown Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 24/65] USB: serial: pl2303: add device-id for HP LD381 Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 25/65] usb: xhci: apply XHCI_SUSPEND_DELAY to AMD XHCI controller 1022:145c Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 26/65] ALSA: line6: Fix endless MIDI read loop Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 27/65] ALSA: seq: virmidi: Fix running status after receiving sysex Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 28/65] ALSA: seq: oss: " Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 29/65] ALSA: pcm: oss: Avoid plugin buffer overflow Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 30/65] ALSA: pcm: oss: Remove WARNING from snd_pcm_plug_alloc() checks Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 31/65] iio: st_sensors: remap SMO8840 to LIS2DH12 Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 32/65] iio: trigger: stm32-timer: disable master mode when stopping Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 33/65] iio: magnetometer: ak8974: Fix negative raw values in sysfs Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 34/65] iio: adc: at91-sama5d2_adc: fix differential channels in triggered mode Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 35/65] mmc: rtsx_pci: Fix support for speed-modes that relies on tuning Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 36/65] mmc: sdhci-of-at91: fix cd-gpios for SAMA5D2 Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 37/65] staging: rtl8188eu: Add device id for MERCUSYS MW150US v2 Greg Kroah-Hartman
2020-03-24 13:10 ` [PATCH 4.19 38/65] staging: greybus: loopback_test: fix poll-mask build breakage Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 39/65] staging/speakup: fix get_word non-space look-ahead Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 40/65] intel_th: Fix user-visible error codes Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 41/65] intel_th: pci: Add Elkhart Lake CPU support Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 42/65] rtc: max8907: add missing select REGMAP_IRQ Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 43/65] xhci: Do not open code __print_symbolic() in xhci trace events Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 44/65] btrfs: fix log context list corruption after rename whiteout error Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 45/65] drm/amd/amdgpu: Fix GPR read from debugfs (v2) Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 46/65] drm/lease: fix WARNING in idr_destroy Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 47/65] memcg: fix NULL pointer dereference in __mem_cgroup_usage_unregister_event Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 48/65] mm: slub: be more careful about the double cmpxchg of freelist Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 49/65] mm, slub: prevent kmalloc_node crashes and memory leaks Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 50/65] page-flags: fix a crash at SetPageError(THP_SWAP) Greg Kroah-Hartman
2020-03-24 13:11 ` Greg Kroah-Hartman [this message]
2020-03-24 13:11 ` [PATCH 4.19 52/65] USB: cdc-acm: fix close_delay and closing_wait units in TIOCSSERIAL Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 53/65] USB: cdc-acm: fix rounding error " Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 54/65] iio: light: vcnl4000: update sampling periods for vcnl4200 Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 55/65] kbuild: Disable -Wpointer-to-enum-cast Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 56/65] futex: Fix inode life-time issue Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 57/65] futex: Unbreak futex hashing Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 58/65] Revert "vrf: mark skb for multicast or link-local as enslaved to VRF" Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 59/65] Revert "ipv6: Fix handling of LLA with VRF and sockets bound " Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 60/65] ALSA: hda/realtek: Fix pop noise on ALC225 Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 61/65] arm64: smp: fix smp_send_stop() behaviour Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 62/65] arm64: smp: fix crash_smp_send_stop() behaviour Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 63/65] drm/bridge: dw-hdmi: fix AVI frame colorimetry Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 64/65] staging: greybus: loopback_test: fix potential path truncation Greg Kroah-Hartman
2020-03-24 13:11 ` [PATCH 4.19 65/65] staging: greybus: loopback_test: fix potential path truncations Greg Kroah-Hartman
2020-03-24 15:53 ` [PATCH 4.19 00/65] 4.19.113-rc1 review Chris Paterson
2020-03-24 19:51 ` shuah
2020-03-24 20:55 ` Guenter Roeck
2020-03-25  4:50 ` 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=20200324130803.309144605@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=akpm@linux-foundation.org \
    --cc=bp@suse.de \
    --cc=dave.hansen@linux.intel.com \
    --cc=jroedel@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=mingo@redhat.com \
    --cc=oliver.sang@intel.com \
    --cc=peterz@infradead.org \
    --cc=rafael.j.wysocki@intel.com \
    --cc=shile.zhang@linux.alibaba.com \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.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).