All of lore.kernel.org
 help / color / mirror / Atom feed
From: Leonardo Bras <leonardo@linux.ibm.com>
To: linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Paul Mackerras <paulus@samba.org>,
	Michael Ellerman <mpe@ellerman.id.au>,
	Arnd Bergmann <arnd@arndb.de>,
	"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>,
	Christophe Leroy <christophe.leroy@c-s.fr>,
	Andrew Morton <akpm@linux-foundation.org>,
	Dan Williams <dan.j.williams@intel.com>,
	Nicholas Piggin <npiggin@gmail.com>,
	Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Richard Fontana <rfontana@redhat.com>,
	Ganesh Goudar <ganeshgr@linux.ibm.com>,
	Allison Randal <allison@lohutok.net>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Mike Rapoport <rppt@linux.ibm.com>,
	YueHaibing <yuehaibing@huawei.com>,
	Ira Weiny <ira.weiny@intel.com>, Jason Gunthorpe <jgg@ziepe.ca>,
	John Hubbard <jhubbard@nvidia.com>,
	Keith Busch <keith.busch@intel.com>
Subject: Re: [PATCH v2 00/11] Introduces new count-based method for monitoring lockless pagetable wakls
Date: Fri, 20 Sep 2019 17:12:00 -0300	[thread overview]
Message-ID: <1f5d9380418ad8bb90c6bbdac34716c650b917a0.camel@linux.ibm.com> (raw)
In-Reply-To: <20190920195047.7703-1-leonardo@linux.ibm.com>

[-- Attachment #1: Type: text/plain, Size: 3862 bytes --]

If a process (qemu) with a lot of CPUs (128) try to munmap() a large
chunk of memory (496GB) mapped with THP, it takes an average of 275
seconds, which can cause a lot of problems to the load (in qemu case,
the guest will lock for this time).

Trying to find the source of this bug, I found out most of this time is
spent on serialize_against_pte_lookup(). This function will take a lot
of time in smp_call_function_many() if there is more than a couple CPUs
running the user process. Since it has to happen to all THP mapped, it
will take a very long time for large amounts of memory.

By the docs, serialize_against_pte_lookup() is needed in order to avoid
pmd_t to pte_t casting inside find_current_mm_pte(), or any lockless
pagetable walk, to happen concurrently with THP splitting/collapsing.

It does so by calling a do_nothing() on each CPU in mm->cpu_bitmap[],
after interrupts are re-enabled. 
Since, interrupts are (usually) disabled during lockless pagetable
walk, and serialize_against_pte_lookup will only return after
interrupts are enabled, it is protected.

So, by what I could understand, if there is no lockless pagetable walk
running, there is no need to call serialize_against_pte_lookup().

So, to avoid the cost of running serialize_against_pte_lookup(), I
propose a counter that keeps track of how many find_current_mm_pte()
are currently running, and if there is none, just skip 
smp_call_function_many().

The related functions are:
start_lockless_pgtbl_walk(mm)
	Insert before starting any lockless pgtable walk
end_lockless_pgtbl_walk(mm)
	Insert after the end of any lockless pgtable walk
	(Mostly after the ptep is last used)
running_lockless_pgtbl_walk(mm)
	Returns the number of lockless pgtable walks running


On my workload (qemu), I could see munmap's time reduction from 275
seconds to 418ms.

> Leonardo Bras (11):
>   powerpc/mm: Adds counting method to monitor lockless pgtable walks
>   asm-generic/pgtable: Adds dummy functions to monitor lockless pgtable
>     walks
>   mm/gup: Applies counting method to monitor gup_pgd_range
>   powerpc/mce_power: Applies counting method to monitor lockless pgtbl
>     walks
>   powerpc/perf: Applies counting method to monitor lockless pgtbl walks
>   powerpc/mm/book3s64/hash: Applies counting method to monitor lockless
>     pgtbl walks
>   powerpc/kvm/e500: Applies counting method to monitor lockless pgtbl
>     walks
>   powerpc/kvm/book3s_hv: Applies counting method to monitor lockless
>     pgtbl walks
>   powerpc/kvm/book3s_64: Applies counting method to monitor lockless
>     pgtbl walks
>   powerpc/book3s_64: Enables counting method to monitor lockless pgtbl
>     walk
>   powerpc/mm/book3s64/pgtable: Uses counting method to skip serializing
> 
>  arch/powerpc/include/asm/book3s/64/mmu.h     |  3 +++
>  arch/powerpc/include/asm/book3s/64/pgtable.h |  5 +++++
>  arch/powerpc/kernel/mce_power.c              | 13 ++++++++++---
>  arch/powerpc/kvm/book3s_64_mmu_hv.c          |  2 ++
>  arch/powerpc/kvm/book3s_64_mmu_radix.c       | 20 ++++++++++++++++++--
>  arch/powerpc/kvm/book3s_64_vio_hv.c          |  4 ++++
>  arch/powerpc/kvm/book3s_hv_nested.c          |  8 ++++++++
>  arch/powerpc/kvm/book3s_hv_rm_mmu.c          |  9 ++++++++-
>  arch/powerpc/kvm/e500_mmu_host.c             |  4 ++++
>  arch/powerpc/mm/book3s64/hash_tlb.c          |  2 ++
>  arch/powerpc/mm/book3s64/hash_utils.c        |  7 +++++++
>  arch/powerpc/mm/book3s64/mmu_context.c       |  1 +
>  arch/powerpc/mm/book3s64/pgtable.c           | 20 +++++++++++++++++++-
>  arch/powerpc/perf/callchain.c                |  5 ++++-
>  include/asm-generic/pgtable.h                |  9 +++++++++
>  mm/gup.c                                     |  4 ++++
>  16 files changed, 108 insertions(+), 8 deletions(-)
> 

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

WARNING: multiple messages have this Message-ID (diff)
From: Leonardo Bras <leonardo@linux.ibm.com>
To: linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org
Cc: Jason Gunthorpe <jgg@ziepe.ca>,
	Thomas Gleixner <tglx@linutronix.de>,
	Arnd Bergmann <arnd@arndb.de>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	YueHaibing <yuehaibing@huawei.com>,
	Keith Busch <keith.busch@intel.com>,
	Nicholas Piggin <npiggin@gmail.com>,
	Mike Rapoport <rppt@linux.ibm.com>,
	Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>,
	Richard Fontana <rfontana@redhat.com>,
	Paul Mackerras <paulus@samba.org>,
	John Hubbard <jhubbard@nvidia.com>,
	"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>,
	Ganesh Goudar <ganeshgr@linux.ibm.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Ira Weiny <ira.weiny@intel.com>,
	Dan Williams <dan.j.williams@intel.com>,
	Allison Randal <allison@lohutok.net>
Subject: Re: [PATCH v2 00/11] Introduces new count-based method for monitoring lockless pagetable wakls
Date: Fri, 20 Sep 2019 17:12:00 -0300	[thread overview]
Message-ID: <1f5d9380418ad8bb90c6bbdac34716c650b917a0.camel@linux.ibm.com> (raw)
In-Reply-To: <20190920195047.7703-1-leonardo@linux.ibm.com>

[-- Attachment #1: Type: text/plain, Size: 3862 bytes --]

If a process (qemu) with a lot of CPUs (128) try to munmap() a large
chunk of memory (496GB) mapped with THP, it takes an average of 275
seconds, which can cause a lot of problems to the load (in qemu case,
the guest will lock for this time).

Trying to find the source of this bug, I found out most of this time is
spent on serialize_against_pte_lookup(). This function will take a lot
of time in smp_call_function_many() if there is more than a couple CPUs
running the user process. Since it has to happen to all THP mapped, it
will take a very long time for large amounts of memory.

By the docs, serialize_against_pte_lookup() is needed in order to avoid
pmd_t to pte_t casting inside find_current_mm_pte(), or any lockless
pagetable walk, to happen concurrently with THP splitting/collapsing.

It does so by calling a do_nothing() on each CPU in mm->cpu_bitmap[],
after interrupts are re-enabled. 
Since, interrupts are (usually) disabled during lockless pagetable
walk, and serialize_against_pte_lookup will only return after
interrupts are enabled, it is protected.

So, by what I could understand, if there is no lockless pagetable walk
running, there is no need to call serialize_against_pte_lookup().

So, to avoid the cost of running serialize_against_pte_lookup(), I
propose a counter that keeps track of how many find_current_mm_pte()
are currently running, and if there is none, just skip 
smp_call_function_many().

The related functions are:
start_lockless_pgtbl_walk(mm)
	Insert before starting any lockless pgtable walk
end_lockless_pgtbl_walk(mm)
	Insert after the end of any lockless pgtable walk
	(Mostly after the ptep is last used)
running_lockless_pgtbl_walk(mm)
	Returns the number of lockless pgtable walks running


On my workload (qemu), I could see munmap's time reduction from 275
seconds to 418ms.

> Leonardo Bras (11):
>   powerpc/mm: Adds counting method to monitor lockless pgtable walks
>   asm-generic/pgtable: Adds dummy functions to monitor lockless pgtable
>     walks
>   mm/gup: Applies counting method to monitor gup_pgd_range
>   powerpc/mce_power: Applies counting method to monitor lockless pgtbl
>     walks
>   powerpc/perf: Applies counting method to monitor lockless pgtbl walks
>   powerpc/mm/book3s64/hash: Applies counting method to monitor lockless
>     pgtbl walks
>   powerpc/kvm/e500: Applies counting method to monitor lockless pgtbl
>     walks
>   powerpc/kvm/book3s_hv: Applies counting method to monitor lockless
>     pgtbl walks
>   powerpc/kvm/book3s_64: Applies counting method to monitor lockless
>     pgtbl walks
>   powerpc/book3s_64: Enables counting method to monitor lockless pgtbl
>     walk
>   powerpc/mm/book3s64/pgtable: Uses counting method to skip serializing
> 
>  arch/powerpc/include/asm/book3s/64/mmu.h     |  3 +++
>  arch/powerpc/include/asm/book3s/64/pgtable.h |  5 +++++
>  arch/powerpc/kernel/mce_power.c              | 13 ++++++++++---
>  arch/powerpc/kvm/book3s_64_mmu_hv.c          |  2 ++
>  arch/powerpc/kvm/book3s_64_mmu_radix.c       | 20 ++++++++++++++++++--
>  arch/powerpc/kvm/book3s_64_vio_hv.c          |  4 ++++
>  arch/powerpc/kvm/book3s_hv_nested.c          |  8 ++++++++
>  arch/powerpc/kvm/book3s_hv_rm_mmu.c          |  9 ++++++++-
>  arch/powerpc/kvm/e500_mmu_host.c             |  4 ++++
>  arch/powerpc/mm/book3s64/hash_tlb.c          |  2 ++
>  arch/powerpc/mm/book3s64/hash_utils.c        |  7 +++++++
>  arch/powerpc/mm/book3s64/mmu_context.c       |  1 +
>  arch/powerpc/mm/book3s64/pgtable.c           | 20 +++++++++++++++++++-
>  arch/powerpc/perf/callchain.c                |  5 ++++-
>  include/asm-generic/pgtable.h                |  9 +++++++++
>  mm/gup.c                                     |  4 ++++
>  16 files changed, 108 insertions(+), 8 deletions(-)
> 

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

  parent reply	other threads:[~2019-09-20 20:12 UTC|newest]

Thread overview: 72+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-20 19:50 [PATCH v2 00/11] Introduces new count-based method for monitoring lockless pagetable wakls Leonardo Bras
2019-09-20 19:50 ` Leonardo Bras
2019-09-20 19:50 ` [PATCH v2 01/11] powerpc/mm: Adds counting method to monitor lockless pgtable walks Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-23 20:42   ` John Hubbard
2019-09-23 20:42     ` John Hubbard
2019-09-23 20:50     ` Leonardo Bras
2019-09-23 20:50       ` Leonardo Bras
2019-09-20 19:50 ` [PATCH v2 02/11] asm-generic/pgtable: Adds dummy functions " Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-23 20:39   ` John Hubbard
2019-09-23 20:39     ` John Hubbard
2019-09-23 20:48     ` Leonardo Bras
2019-09-23 20:48       ` Leonardo Bras
2019-09-23 20:53       ` John Hubbard
2019-09-23 20:53         ` John Hubbard
2019-09-20 19:50 ` [PATCH v2 03/11] mm/gup: Applies counting method to monitor gup_pgd_range Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-23 20:27   ` John Hubbard
2019-09-23 20:27     ` John Hubbard
2019-09-23 21:01     ` Leonardo Bras
2019-09-23 21:01       ` Leonardo Bras
2019-09-23 21:09       ` John Hubbard
2019-09-23 21:09         ` John Hubbard
2019-09-20 19:50 ` [PATCH v2 04/11] powerpc/mce_power: Applies counting method to monitor lockless pgtbl walks Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-20 19:50 ` [PATCH v2 05/11] powerpc/perf: " Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-20 19:50 ` [PATCH v2 06/11] powerpc/mm/book3s64/hash: " Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-20 19:50 ` [PATCH v2 07/11] powerpc/kvm/e500: " Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-20 19:50 ` [PATCH v2 08/11] powerpc/kvm/book3s_hv: " Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-23 20:47   ` John Hubbard
2019-09-23 20:47     ` John Hubbard
2019-09-20 19:50 ` [PATCH v2 09/11] powerpc/kvm/book3s_64: " Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-20 19:50 ` [PATCH v2 10/11] powerpc/book3s_64: Enables counting method to monitor lockless pgtbl walk Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-20 19:50 ` [PATCH v2 11/11] powerpc/mm/book3s64/pgtable: Uses counting method to skip serializing Leonardo Bras
2019-09-20 19:50   ` Leonardo Bras
2019-09-20 20:11   ` John Hubbard
2019-09-20 20:28     ` Leonardo Bras
2019-09-20 20:28       ` Leonardo Bras
2019-09-20 21:15       ` John Hubbard
2019-09-21  0:48       ` John Hubbard
2019-09-21  0:48         ` John Hubbard
2019-09-23 17:25         ` Leonardo Bras
2019-09-23 17:25           ` Leonardo Bras
2019-09-23 18:14           ` John Hubbard
2019-09-23 18:14             ` John Hubbard
2019-09-23 19:40             ` Leonardo Bras
2019-09-23 19:40               ` Leonardo Bras
2019-09-23 19:58               ` John Hubbard
2019-09-23 19:58                 ` John Hubbard
2019-09-23 20:23                 ` Leonardo Bras
2019-09-23 20:23                   ` Leonardo Bras
2019-09-23 20:26                   ` John Hubbard
2019-09-23 20:26                     ` John Hubbard
2019-09-20 19:56 ` [PATCH v2 00/11] Introduces new count-based method for monitoring lockless pagetable wakls Leonardo Bras
2019-09-20 19:56   ` Leonardo Bras
2019-09-20 20:12 ` Leonardo Bras [this message]
2019-09-20 20:12   ` Leonardo Bras
2019-09-20 21:24   ` John Hubbard
2019-09-20 21:24     ` John Hubbard
2019-09-23 20:51   ` John Hubbard
2019-09-23 20:51     ` John Hubbard
2019-09-23 20:58     ` Leonardo Bras
2019-09-23 20:58       ` Leonardo Bras
2019-09-24 20:31 Leonardo Bras
2019-09-24 20:31 ` Leonardo Bras

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=1f5d9380418ad8bb90c6bbdac34716c650b917a0.camel@linux.ibm.com \
    --to=leonardo@linux.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=allison@lohutok.net \
    --cc=aneesh.kumar@linux.ibm.com \
    --cc=arnd@arndb.de \
    --cc=benh@kernel.crashing.org \
    --cc=christophe.leroy@c-s.fr \
    --cc=dan.j.williams@intel.com \
    --cc=ganeshgr@linux.ibm.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=ira.weiny@intel.com \
    --cc=jgg@ziepe.ca \
    --cc=jhubbard@nvidia.com \
    --cc=keith.busch@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mahesh@linux.vnet.ibm.com \
    --cc=mpe@ellerman.id.au \
    --cc=npiggin@gmail.com \
    --cc=paulus@samba.org \
    --cc=rfontana@redhat.com \
    --cc=rppt@linux.ibm.com \
    --cc=tglx@linutronix.de \
    --cc=yuehaibing@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.